BERGER MACROED CORRECTED 060305.DOC
6/6/2005 1:49 PM
SCIENCE FOR JUDGES IV INTRODUCTION
Margaret A. Berger∗
This issue of the Journal of Law and Policy contains articles
that had their inception as presentations made at a Science for
Judges program for federal and state judges. The conference, held
in November 2004, was the fourth in a series hosted by Brooklyn
Law School and funded by the Common Benefit Trust established
in the Silicone Breast Implant Products Liability Litigation.1 These
events are held under the auspices of Brooklyn Law School’s
Center for Health, Science and Public Policy, in collaboration with
the Federal Judicial Center, the National Center for State Courts,
and the Science, Technology and Law Panel of the National
Academies of Science.
Science for Judges IV examined the interaction of science and
law from a somewhat different perspective than previous
programs. The first session dealt exclusively with Agent Orange
and reviewed the scientific research that bears on whether Agent
Orange causes adverse health effects. The second session
considered research on human behavior that plays a role in judicial
proceedings. Both presentations pointed out the difficulties that
∗
Suzanne J. and Norman Miles Professor of Law, Brooklyn Law School.
Professor Berger is the Director of the Science for Judges Program.
1
Papers from previous Science for Judges programs can be found in 12 J.L.
& POL’Y 1, 1-53 (2003) (papers discussing the practice of epidemiology and the
science produced by administrative agencies); 12 J.L. & POL’Y 485, 485-639
(2004) (papers discussing toxicology and epidemiology); 13 J.L. & POL’Y 1, 1179 (2005) (papers discussing the integrity of scientific research and forensic
evidence in criminal proceedings); and 13 J.L. & POL’Y 499, 499-647 (2005)
(papers discussing Agent Orange and human behavior research). All papers are
available in electronic form at http://brooklaw.edu/centers/scienceforjudges/
papers.php.
499
BERGER MACROED CORRECTED 060305.DOC
500
6/6/2005 1:49 PM
JOURNAL OF LAW AND POLICY
arise in resolving controversies that encompass sophisticated
scientific questions, raise complex legal issues, and invoke
sensitive policy concerns.
The Agent Orange session illuminated how perplexing these
interrelated questions of science, law, and policy can be. Agent
Orange is the name given to herbicides sprayed as defoliants
during the Vietnam War. These herbicides, which were made by a
number of different manufacturers, were contaminated with
varying amounts of dioxin, a by-product of the manufacturing
process.2 Thirty years after the end of the Vietnam War, and
twenty years after the class action on behalf of Vietnam veterans
exposed to Agent Orange was settled, it is remarkable to see how
many unanswered questions remain.
Of course, considerably more scientific information is available
now than before the Vietnam War. The very limited knowledge
that was available before 1970 is illustrated by Dr. David Butler’s
paper, which paints a fascinating picture of the gradually growing
awareness of an association between health problems and
occupational exposures to dioxin.3 Drs. Jeanne Mager Stellman
and Steven Stellman describe the sophisticated model produced by
their research that permits an individual assessment for all Vietnam
veterans of their exposure to Agent Orange.4 These data on
exposure obviously provide a strong foundation, often missing in
toxic tort cases when exposure data are not available, for
epidemiologic studies on adverse health effects observed in
Vietnam veterans. Perhaps somewhat surprisingly, however, the
Stellmans tell us that meaningful epidemiologic research on these
veterans has never been conducted—a conclusion that is verified
by Dr. Irva Hertz-Picciotto, the chair of the Institute of Medicine
(IOM) Committee charged by Congress to review health
2
Irva Hertz-Picciotto, How Scientists View Causality and Assess Evidence:
A Study of the Institute of Medicine’s Evaluation of Health Effects in Vietnam
Veterans and Agent Orange, 13 J.L. & POL’Y 553 (2005).
3
David A. Butler, Connections: The Early History of Scientific and Medical
Research on “Agent Orange,” 13 J.L. & POL’Y 527 (2005).
4
Jeanne Mager Stellman & Steven D. Stellman, Characterization of
Exposure to Agent Orange in Vietnam Veterans As a Basis for Epidemiological
Studies, 13 J. L. & POL’Y 505 (2005).
BERGER MACROED CORRECTED 060305.DOC
SCIENCE FOR JUDGES IV
6/6/2005 1:49 PM
501
consequences in the Vietnam veterans who had been exposed to
herbicides. In her paper Dr. Hertz-Picciotto acknowledges the
paucity of studies actually conducted on Vietnam veterans and
explains that the bulk of the IOM Committee’s work focused on
occupational and environmental exposures to dioxin.5
The determinations reached by the IOM Committee are of
great interest to the legal community in part because the
Committee used standards for evaluating evidence that differ in
some respects from the legal standard for proving causation in a
judicial proceeding. The Committee’s standards, and the process
by which the Committee determined whether they were satisfied,
help to clarify some of the difficult issues courts face when
deciding whether to admit proffered expert testimony on causation
in a toxic tort case. In addition to explaining the work of the IOM
Committee and its conclusions regarding associations between
dioxin exposures and a variety of different health effects, Dr.
Hertz-Picciotto’s paper also furnishes a succinct guide to how
epidemiologists proceed in making inferences about causation.
This discussion should prove extremely valuable to members of
the legal community who desire a basic understanding of
epidemiology.
The last article on Agent Orange, by Dr. Mark Brown of the
U.S. Department of Veterans Affairs (VA), explains how the VA
has translated the available science into a compensation scheme for
Vietnam veterans.6 It spells out the statutory scheme by which
Congress created presumptions that authorize the VA to treat
certain illnesses as the result of a direct service connection.
Vietnam veterans who suffer from one of the presumptively
service-connected illnesses are then automatically entitled to
benefits. The VA’s list of diseases is based on the studies described
in Dr. Hertz-Picciotto’s article. Dr. Brown’s article also discusses
issues that have arisen in extending this approach to Gulf War
veterans who are claiming that military service affected their
health. The intersection of science with pressing national policy
5
Hertz-Picciotto, supra note 2, at 558-60.
Mark Brown, The Role of Science in Department of Veterans Affairs
Disability Compensation Policies for Environmental and Occupational Illnesses
and Injuries, 13 J.L. POL’Y 593 (2005).
6
BERGER MACROED CORRECTED 060305.DOC
502
6/6/2005 1:49 PM
JOURNAL OF LAW AND POLICY
considerations leads to problems that defy easy solutions. We will
undoubtedly see a new set of issues arising out of the war in Iraq.
On the program’s second day, a panel of scholars addressed a
very different topic: research on human behavior that may be
relevant in judicial proceedings. Edited and expanded versions of
two of the presentations, on gender stereotyping and predictions of
dangerousness, are included in this issue of the Journal. Any
possible argument that the test promulgated by the Supreme Court
in Daubert v. Merrill Dow Pharmaceuticals, Inc.7 for admitting
expert testimony applies only to the “hard” sciences was
obliterated by the Court’s subsequent opinion in Kumho Tire Co. v.
Carmichael.8 Kumho clearly established that Daubert’s relevancy
and reliability requirements apply to all expert testimony.
Consequently, in the federal courts and state courts that have
adopted Daubert and Kumho, testimony by psychologists may now
be subject to greater scrutiny. The articles by Drs. Eugene Borgida
and Edward Mulvey are therefore of interest not only because they
discuss cutting-edge research on interesting topics, but also
because they provide information that a court may need in deciding
whether an expert will be allowed to testify about this research.
The article by Dr. Borgida and his associates on gender
stereotyping research provides a helpful overview of the areas in
which research is being conducted and then discusses research
studies regarding women who self-promote in order to succeed in
their careers.9 The article reviews research on the effect of such
behavior on perceptions of the women’s likeability and
competency and examines the consequences that flow from these
perceptions. The article concludes with a discussion of how
testimony based on this research can be utilized by courts.
Dr. Mulvey’s article surveys research on predicting future
dangerousness—a finding that courts are asked to make in a wide
variety of legal contexts.10 His evaluation of the existing research
7
509 U.S. 579 (1993).
526 U.S. 139 (1999).
9
Eugene Borgida et al., On the Use of Gender Stereotyping Research in Sex
Discrimination Litigation, 13 J.L. & POL’Y 613 (2005).
10
Edward P. Mulvey, Assessing the Likelihood of Future Violence in
Individuals with Mental Illness: Current Knowledge and Future Issues, 13 J.L.
8
BERGER MACROED CORRECTED 060305.DOC
SCIENCE FOR JUDGES IV
6/6/2005 1:49 PM
503
on a variety of methodologies currently used in assessing the
likelihood of future violence provides an excellent starting point
for judges and lawyers who must deal with these issues. Although
Dr. Mulvey reports progress in understanding some of the
associations between mental disorders and violence, he urges
caution in assigning too much weight to the actuarial instruments
now being developed by researchers to be used in making
predictions.
I hope that these highly abbreviated descriptions of the
contents of the articles that follow in this issue of the Journal of
Law and Policy convey some of the challenges and complexities
that judges encounter in handling cases that require an
understanding of cutting-edge issues of science. Both of the topics
discussed at Science for Judges IV also illustrate that scientific and
behavioral research takes time, and obtaining needed knowledge
often is an extremely slow process.
& POL’Y 629 (2005).
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
CHARACTERIZATION OF EXPOSURE TO
AGENT ORANGE IN VIETNAM VETERANS
AS A BASIS FOR EPIDEMIOLOGICAL
STUDIES
Jeanne Mager Stellman, Ph.D. & Steven D. Stellman, Ph.D.,
M.P.H.∗
INTRODUCTION
Between 1961 and 1970, the U.S. military engaged in massive
chemical defoliation and crop destruction operations in Southeast
Asia.1 In 1985, nearly two decades after the spraying had ceased, a
landmark tort settlement was reached between a class of Vietnam
veterans and the chemical manufacturers that had supplied the
Agent Orange and other military herbicides to the U.S. Department
of Defense.2 It is notable that, at the time of the settlement, there
∗
Jeanne Mager Stellman, Ph.D., and Steven D. Stellman, Ph.D., MPH, are
professors of clinical health policy and management and clinical epidemiology,
respectively, at the Mailman School of Public Health at Columbia University.
The research described was supported by the National Academy of Sciences
Subcontract NAS-VA-5124-98-001 and USPHS Grants CA-68384, CA-17613
from the National Cancer Institute. The authors thank Col. Richard Christian
(ret.) for his collaboration on all aspects of the research and Donald Hakenson,
Director, U.S. Armed Services Center for Research of Unit Records, and his
staff for extensive technical support and access to military records. Carrie
Tomasallo, Tracy Weber, and Francine Benjamin contributed significantly to the
development of the geographic information system.
1
WILLIAM A. BUCKINGHAM, JR., OPERATION RANCH HAND: THE AIR
FORCE AND HERBICIDES IN SOUTHEAST ASIA 1961-1971 (Office of U.S. Air
Force History 1982).
2
In re “Agent Orange” Prod. Liab. Litig., 611 F. Supp. 1396 (E.D.N.Y.
1985).
505
STELLMAN MACROED CORRECTED 070505.DOC
506
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
was little convincing epidemiological evidence available, either
positive or negative, on the health consequences to veterans of the
herbicide operations. Today, more than three decades after this
massive environmental exposure, there is still a dearth of
epidemiological data on the extent to which adverse health
consequences resulted from the use, storage, and disposal of the
herbicides in Vietnam.
This paucity of epidemiological data stands in stark contrast to
the extensive amount of experimental data available on dioxin, an
important contaminant in about 60% of the herbicide sprayed.3
Much laboratory data convincingly demonstrate dioxin’s extreme
toxicity.4 The scientific literature also is growing with respect to
the carcinogenicity of the organic arsenical that was a primary
component of the Agent Blue used to destroy enemy food crops.5
Many epidemiological studies have been carried out on other,
much smaller populations exposed to the same chemicals. Indeed,
when the Institute of Medicine (IOM) conducts its biennial review
of the scientific literature and provides the Department of Veterans
3
The chemical composition of three major herbicides used in Vietnam and
the quantity dispersed are as follows:
Agent
Composition
Gallons
Agent Orange
2,4-D, 2,4,5-T*
12,066,840
Agent White
Picloram, 2,4-D
5,430,462
Agent Blue
Dimethylarsinic acid
1,252,541
(Synonym: Cacodylic acid)
*Contaminated with 2,3,7,8-TCDD (dioxin)
More detailed data are given in Jeanne Mager Stellman et al., The Extent and
Patterns of Usage of Agent Orange and Other Herbicides in Vietnam, 422
NATURE 681 (2003) [hereinafter Stellman et al., The Extent and Patterns of
Usage of Agent Orange and Other Herbicides in Vietnam].
4
See, e.g., OFFICE OF HEALTH AND ENVIRONMENTAL ASSESSMENT, U.S.
ENVTL. PROTECTION AGENCY, NO. EPA/600/8-84/014F, HEALTH ASSESSMENT
DOCUMENT FOR POLYCHLORINATED DIBENZO-P-DIOXINS, FINAL REPORT (1985).
5
Hideki Wanibuchi et al., Carcinogenicity of an Organic Arsenical,
Dimethylarsinic Acid and Related Arsenicals in Rat Urinary Bladder, 40 PROC.
OF THE AM. ASS’N FOR CANCER RES. 349 (1999); Min Wei et al., Urinary
Bladder Carcinogenicity of Dimethylarsinic Acid in Male F344 Rats, 20
CARCINOGENESIS 1873 (1999).
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 507
Affairs (VA) with a summary, including its appraisal of the
relationship between herbicide exposures and a list of health
outcomes, it relies to a large extent on studies carried out on nonveteran populations to support its conclusions.6 The degree to
which these other studies correctly estimate health effects in
Vietnam veterans is not known. Thus there continue to be practical
ramifications to the paucity of definitive epidemiological studies
on a sufficiently large exposed population of either veterans or
Vietnamese citizens.7
6
In accordance with the Agent Orange Act of 1991, Pub. L. No. 102-4, 105
Stat. 11 (1991) (codified as amended at 38 U.S.C. § 1116), the Committee to
Review the Health Effects in Vietnam Veterans of Exposure to Herbicides was
asked “to determine (to the extent that available data permit meaningful
determinations)” the following regarding associations between specific health
outcomes and exposure to TCDD and other chemical compounds in herbicides:
A) whether a statistical association with herbicide exposure exists,
taking into account the strength of the scientific evidence and the
appropriateness of the statistical and epidemiological methods used to
detect the association; B) the increased risk of the disease among those
exposed to herbicides during service in the Republic of Vietnam during
the Vietnam era; and C) whether there exists a plausible biological
mechanism or other evidence of a causal relationship between herbicide
exposure and the disease.
38 U.S.C. § 1116. See COMMITTEE TO REVIEW THE HEALTH EFFECTS IN
VIETNAM VETERANS OF EXPOSURE TO HERBICIDES, INSTITUTE OF MEDICINE,
VETERANS AND AGENT ORANGE: HEALTH EFFECTS OF HERBICIDES USED IN
VIETNAM (1994) [hereinafter 1994 IOM REPORT], available at http://books.nap.
edu/books/0309048877/html/index.html. The Institute of Medicine publishes
biennial updates based upon the deliberations of its Committee to Review the
Health Effects in Vietnam Veterans of Exposure to Herbicides. The series is
VETERANS AND AGENT ORANGE of which there are 1996, 1998, 2000, 2002 and
2004 publications.
7
Epidemiological studies can be used to support arguments that a disease
or dysfunction is more likely than not to have arisen from a particular causative
agent. Epidemiological studies examine the statistical distribution of a disease
(or other outcome) in two populations: one that was “exposed” to the agent or
condition under study and another “control” population not exposed and as alike
as possible in every other way to the exposed group. If the rate of disease
observed in the exposed population is greater than in the control population, and
if the rate differences satisfy certain statistical requirements, the rate difference
will be called “significant.”
STELLMAN MACROED CORRECTED 070505.DOC
508
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
This article discusses some of the factors that have contributed
to the lack of epidemiological evidence on military herbicide
operations. Part I of this article will provide a brief overview of the
purposes and methodology of environmental epidemiological
studies. Part II will discuss the application of this methodology to
exposed Vietnam veterans. In particular, this section will examine
the use of exposure opportunity measures in epidemiological
studies as well as recent successful work on the development and
use of military records for estimating exposure opportunity to
military herbicides in Vietnam. This article concludes that while
there are sufficiently large populations available for study and
appropriate methodologies to carry out such studies, these muchneeded epidemiological investigations remain unfunded and
undone, so that both legal and public policy decisions must
continue to be made with inadequate scientific data.
I. ENVIRONMENTAL EPIDEMIOLOGY
A valid environmental epidemiology study relating an
exposure to subsequent risk of disease requires a biologically
reasonable hypothesis, an exposed population, and either an
unexposed population or a set of disease rates in a reference
population to which the rate of disease in the study group can be
compared. Usually there are experimental laboratory studies or
clinical reports of adverse health effects in individuals that can be
used to generate a “null” hypothesis of the form: “Exposure to
agent XYZ is not related to development of disease ABC.” The
purpose of the epidemiological study is to test the null hypothesis.
If the null hypothesis is rejected (i.e., a statistically significant
difference in rates is observed between the exposed and
unexposed), then the study is considered positive and a relationship
between the exposure and the outcome is supported.8
8
Note the use of the word “supported.” Epidemiological studies do not
establish cause and effect. Rather, they indicate that there is a statistical
likelihood that a relationship between the exposure and the outcome exists.
Generally, epidemiologists require a 95% certainty that the relationship is not
compatible with chance in order to consider an outcome significant. Failure to
meet this criterion is a type 1 error.
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 509
A successful study of this nature requires a suitably large
population with sufficiently great exposure to agent XYZ to elicit
the measured health effect.9 Additionally, the population’s
exposure to the agent and the subsequent study must be adequately
spaced over time to have permitted XYZ an opportunity to induce
disease ABC. In fulfilling these requirements, epidemiological
studies must endeavor to identify all members of the at-risk
population, to successfully find and enroll these individuals in the
study, and to take into account other exposures (e.g., cigarette
smoking or occupational exposures) that also could lead to the
disease under study. It is also critically important that the exposed
population truly consist of exposed individuals. Many
environmental epidemiology studies have been compromised by
the inclusion of non-exposed individuals, for example, nonexposed clerical or management staff in a study of chemical plant
workers. Such dilution of a truly exposed population with
unexposed individuals is known as “differential misclassification”
of exposure and can result in an underestimate of the true
association between exposure and disease.10 There are many
instances of such misclassification errors in existing studies of
Vietnam veterans.11
In all cases, a necessity for the successful design and execution
of an environmental epidemiology study is the ability to define
“exposure.” Poorly defined population exposures can lead to two
9
The population size, the anticipated effect size (i.e. the environmental
agent’s potency as measured by the difference between the disease risk in the
exposed group and that in an unexposed reference group), and the desired level
of statistical certainty all contribute to the statistical “power” to actually observe
an effect if it is present. If there are too few exposed people or the effect size is
very small, an epidemiologic study may be useless and failure to reject the null
hypothesis (a negative result) non-informative. This is known as a type 2 error.
10
KENNETH J. ROTHMAN & SANDER GREENLAND, MODERN
EPIDEMIOLOGY 126-27 (1998).
11
JEANNE MAGER STELLMAN & STEVEN D. STELLMAN, INSTITUTE OF
MEDICINE, SUBCONTRACT VA-5124-98-0019, CHARACTERIZING EXPOSURE OF
VETERANS TO AGENT ORANGE AND OTHER HERBICIDES IN VIETNAM: FINAL
REPORT 109 tbl.32 (2003) [hereinafter STELLMAN & STELLMAN,
CHARACTERIZING EXPOSURE OF VETERANS TO AGENT ORANGE AND OTHER
HERBICIDES IN VIETNAM].
STELLMAN MACROED CORRECTED 070505.DOC
510
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
sources of misclassification error: unexposed people are
considered exposed or exposed people are considered not exposed.
Both types of error may occur in a given study and may lead to
unpredictable errors in estimates of exposure-disease associations.
For example, in the Air Force Health Study of Vietnam veterans
who were assigned to the herbicide spray operations, many in the
reference comparison group in fact had elevated blood levels of
dioxin, while many in the study population had non-detectable
levels.
Assigning exposure levels in epidemiology studies of chronic
exposures almost always poses great methodological challenges.12
Unlike “acute” exposures to an agent suspected of causing a health
effect (e.g., reactions to an implanted medical device or exposure
to environmental agents arising from industrial accidents or nonindustrial events, such as carbon monoxide poisoning from faulty
heaters), most chronic environmental exposures are characterized
by poor, incomplete, or even nonexistent measurements of actual
exposure levels. The “exposed” population may also have been
exposed to a host of other agents that could potentially cause the
same outcome and will have spent discontinuous—and usually
undocumented—periods of time being “exposed.” Studies are
often carried out years after the exposure has ended, making it
difficult, if not impossible, to find extant biological evidence of the
12
A valid metric for assigning exposures is a necessary element of an
epidemiology study—without it one cannot differentiate the exposed from the
controls. In addition, epidemiologists place higher confidence in studies that
demonstrate that the higher the dose of the exposure, the more likely the
outcome. For example, a cigarette smoker with a lifetime history of smoking one
pack per day, on average, has a relative risk of lung cancer eight to ten times that
of never-smokers, while a two pack per day smoker has a risk twenty times that
of a nonsmoker. Steven D. Stellman et al., Lung Cancer Risk in White and Black
Americans, 13 ANNALS OF EPIDEMIOLOGY 294, 298 (2003). Thus, it is desirable
to have a metric that permits the exposure to be more than simply ever/never,
but rather quantified so that a dose-response relationship can be tested and the
risk at high doses compared with that at low doses. For a discussion of the
importance of dose-response relationships in epidemiological studies, see Leslie
Stayner et al., Sources of Uncertainty in Dose-Response Modeling of
Epidemiological Data for Cancer Risk Assessment, 895 ANNALS OF THE NEW
YORK ACADEMY OF SCIENCES 212 (1999).
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 511
exposure among individuals thought to have been exposed and, for
many environmental exposures, no adequate biomarkers exist.13
Ubiquitous environmental agents complicate the job of finding a
truly unexposed control population.
II. EPIDEMIOLOGY AND VIETNAM VETERANS
Whether the military use of herbicides in Vietnam lends itself
to the basic requirements for an environmental epidemiology study
merits examination. Such a study could test the following
generalized null hypothesis, “Exposure to military herbicides used
during the Vietnam War did not lead to adverse health outcomes
among the exposed populations or their offspring,” or a related,
more general hypothesis that “military service in areas sprayed by
military herbicides did not lead to adverse health outcomes.”
It is clear from the IOM’s summaries of available experimental
evidence that there exist sufficient toxicological and clinical data
to justify undertaking major epidemiology studies for a variety of
disease outcomes. In particular, the large quantity of herbicide that
the United States sprayed in Vietnam as well as the vast amounts
13
For many environmental agents, even if one had a scientifically valid
exposure metric, the extent of exposure might not yield a population that is large
enough for a successful epidemiological study to be carried out because the
number of exposed persons is small, the intensity of the exposures is low, or,
with the passage of time, the amount of chemical in exposed individuals’ bodies
declines due to metabolic processes. In any of these cases, the ability to detect
an association between exposure and disease is very limited, as expressed
numerically by the concept of statistical power. Statistical power is defined as
the probability that a statistical test will yield a significant result. JACOB COHEN,
STATISTICAL POWER ANALYSIS FOR THE BEHAVIORAL SCIENCES 1 (Academic
Press 1977). Thus, a weak environmental agent that does indeed cause a disease
(small effect size) would require a very large population N for the observed
difference in rates to reach significance. If a legal criterion demands an effect
size of a twofold difference between the exposed and non-exposed, the size of
the populations studied would also have to be expanded in relation to a criterion
which demanded a 50% increase (i.e. twofold relative risk). The necessary
population size for the study will also depend on the rate at which the disease is
observed in the unexposed population. Agents that cause rare diseases may be
less likely to be identified than those that cause common diseases because too
few cases ever occur to satisfy the requirements of statistical power.
STELLMAN MACROED CORRECTED 070505.DOC
512
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
that it manufactured would provide epidemiological studies with a
sufficiently large sample of affected individuals and chemical
resources. Nearly 20 million gallons of military herbicides were
sprayed in Vietnam and Laos between 1961 and 1971. Agent
Orange accounted for more than 12 million gallons of these
herbicides. The chemical consists of a 50:50 mixture of two
phenoxyherbicides, 2,4-D and 2,4,5-T, with much of the 2,4,5-T
component contaminated with dioxins and dibenzofurans. The
most notorious and deadly of these contaminants is 2,3,7,8tetrachloro-p-dibenodioxin, usually simply called dioxin, although
the precise levels of contamination are not known.14 These
chemicals are unwanted byproducts of the manufacturing process
and are inevitably present unless manufacturing conditions are
carefully controlled.15 Even prior to the 1985 Agent Orange tort
settlement, there was no question regarding dioxin’s deadly
effects.16
In order to proceed with an epidemiological study of Agent
Orange, there must be a suitable population available for study.
The authors have calculated that between 3 and 5 million
Vietnamese citizens were directly in the spray path of the
herbicide.17 Many areas, so-called “hotspots,” are still highly
contaminated with TCDD residues and a variety of bioassays have
found that Vietnamese individuals have higher-than-normal levels
of dioxin in their tissue,18 although there is little data on the
14
Stellman et al., The Extent and Patterns of Usage of Agent Orange and
Other Herbicides in Vietnam, supra note 3, at 682.
15
ALASTAIR HAY, THE CHEMICAL SCYTHE: LESSONS OF 2,4,5-T AND
DIOXIN (Plenum Press 1982).
16
Indeed, in its decision with respect to the Agent Orange Class action, the
court stated:
As to the poisonous nature of dioxin and its ability to cause harm to
mammals, including homo sapiens, there is no doubt. The form of
dioxin implicated in Agent Orange is a dangerous, stable, long lasting
chemical. . . . Dioxin is one of the most powerful poisons known . . . .
In re “Agent Orange” Prod. Liab. Litig., 597 F. Supp. 740, 777 (E.D.N.Y.
1984).
17
Stellman et al., The Extent and Patterns of Usage of Agent Orange and
Other Herbicides in Vietnam, supra note 3, at 684.
18
Arnold Schecter, Food As a Source of Dioxin Exposure in the Residents
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 513
relationship between body burden and putative exposure. While
the precise number is not known, it is estimated that about 3
million American soldiers served in the Vietnam theatre.19 Soldiers
directly charged with carrying out the Air Force Operation Ranch
Hand, the name for the military operation that carried out the great
majority of aerial defoliation and crop destruction missions, were
potentially exposed to herbicides.20 Similarly, some, but certainly
not all, of those belonging to the Army Chemical Corps, another
group that has been studied, were also potentially were exposed. It
would be erroneous to classify all of these individuals as
potentially exposed.21 Some Army troops were herbicide handlers
or backpack sprayers, or were engaged in missions to keep base
camp perimeters free from vision-blocking foliage. There is
evidence that those whose missions brought them into recently
defoliated areas absorbed the herbicides.22 Finally, military unit
of Bien Hoa City, Vietnam, 45 J. OF OCCUPATIONAL & ENVTL. MED., 781, 78182 (2003).
19
Sharon R. Cohany, The Vietnam-Era Cohort: Employment and Earnings,
115 MONTHLY LABOR REV. 3, 5 (1992).
20
The Air Force Health Study often called the Ranch Hand study, a 20year longitudinal study examining health, mortality, and reproductive outcomes,
has found associations between Ranch Hand service and subsequent risk of
prostate cancer and Type II diabetes. Matthew P. Longnecker & Joel E.
Michalek, Serum Dioxin Level in Relation to Diabetes Mellitus among Air Force
Veterans with Background Levels of Exposure, 11 EPIDEMIOLOGY 44 (2000).
This study, however, because of its unavoidably small size, is not informative on
rarer cancers.
21
Serum dioxin levels in the Air Force Health Study for the comparison
group reach a level nearly twice that of the Ranch Hand low category group in
the study population. See Akhtar et al., Cancer in US Air Force Veterans of the
Vietnam War, 46 J. OF OCCUPATIONAL & ENVTL. MED. 123, 127 (2004). Some
Air Force personnel not directly assigned to the herbicide operational units did,
in fact, have work assignments that brought them into contact with herbicides,
while some flight personnel, notably pilots, who never handled herbicides, had
access to shower and laundry facilities and flew in air pressurized cabins. See
BUCKINGHAM, supra note 1 (providing an extensive history of the military
herbicide program in Vietnam). The exposure misclassification of the
comparison group as unexposed would, however, tend to strengthen our
confidence in the positive cancer findings. See supra note 20.
22
Peter C. Kahn et al., Dioxins and Dibenzofurans in Blood and Adipose
Tissue of Agent Orange-Exposed Vietnam Veterans and Matched Controls, 259
STELLMAN MACROED CORRECTED 070505.DOC
514
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
history records show that a sufficiently large number of units were
directly sprayed during Operation Ranch Hand to justify largescale studies.23
The extent to which soldiers entering into previously sprayed
areas or living in base camps in which the perimeters were
regularly cleared with defoliants received a biologically significant
dose of herbicides or their contaminants is not clear. Because so
many years have passed since the exposure, measurement of the
current body burden of dioxin is subject to serious
misclassification errors, and biomarkers are not available for
herbicide formulations that were not contaminated with TCDD.24
JAMA 1661 (1988) (showing that the leaders of jungle patrols in heavily
sprayed areas, so-called “pointmen,” had elevated levels of dioxin compared to a
matched unexposed control population).
23
See STELLMAN & STELLMAN, CHARACTERIZING EXPOSURE OF VETERANS
TO AGENT ORANGE AND OTHER HERBICIDES IN VIETNAM, supra note 11, at 48
tbl.11. Table 11 demonstrates numerous documented instances in which combat
units were subject to “direct hits” from herbicide spray. The direct spraying of
combat units has been a contentious issue for several decades. The history of the
controversy is well described in the IOM’s 1994 report. Both the White House
Agent Orange Working Group and the Centers for Disease Control, Centers for
Disease Control Veterans Health Studies, Serum 2,3,7,8-Tetrachloro-PDibenzo-P-Dioxin Levels in U.S. Army Vietnam-Era Veterans, 260 JAMA 1249
(1988) [hereinafter CDC Veterans Health Studies], have declared that ground
troops were not exposed to herbicides and that only those troops with duties that
involved the handling and spraying operations were exposed. Examination of
military archives by the U.S. General Accounting Office, COMPTROLLER
GENERAL OF THE U.S., U.S. GENERAL ACCOUNTING OFFICE, U.S. GROUND
TROOPS IN SOUTH VIETNAM WERE IN AREAS SPRAYED WITH HERBICIDE
ORANGE (1979), available at http://161.203.16.4/f0302/110930.pdf, and by the
CDC itself found a significant number of troops to have been located directly
under the spray path. Centers for Disease Control, AGENT ORANGE STUDY:
EXPOSURE ASSESSMENT: PROCEDURES AND STATISTICAL ISSUES (CDC Agent
Orange Project, Agent Orange Projects Interim Report, Draft, Feb. 1985).
24
The Institute of Medicine has specifically addressed the difficulties in
using current body burden measurements of dioxin to reflect past exposures in
Vietnam. First, during its oversight of the original CDC Agent Orange Study,
the IOM rejected the CDC proposal to “validate” military records of troop
location by using serum dioxin samples obtained at least two decades postexposure. See ADVISORY COMMITTEE ON THE CDC STUDY OF THE HEALTH OF
VIETNAM VETERANS, INSTITUTE OF MEDICINE, FIFTH LETTER REPORT, REVIEW
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 515
For these soldiers, service in an area with a history of herbicide
spraying would be the relevant measure of exposure (called an
“exposure opportunity index,” or EOI) rather than a measure of
biological dose. Such studies would seek to test the hypothesis that
military service in defoliated areas increased the risk that soldiers
would develop the diseases under study and that the risk was
proportional to the soldiers’ proximity in time and space to the
spraying.25
A. Agent Orange and Measures of Exposure
Methodological difficulties in assigning relative Agent Orange
COMPARISON OF SERUM LEVELS OF 2,3,7,8-TCDD WITH INDIRECT
ESTIMATES OF AGENT ORANGE EXPOSURE IN VIETNAM VETERANS (1987). The
CDC continued its “validation study” notwithstanding, CDC VETERANS HEALTH
STUDIES, supra, and, as a result of a purported lack of correlation between serum
dioxin and a records-based exposure index, abandoned the large Agent Orange
Study of III Corps Army combat battalions already underway. Agreeing with the
aforementioned IOM critique, a second IOM committee reviewed the conceptual
underpinnings of the CDC validation study, and, in contradistinction to the CDC
and Agent Orange Working Group conclusion, recommended that a study be
conducted on the utility of the historical reconstruction of military records for
characterizing exposure to military herbicides in Vietnam. See 1994 IOM
REPORT, supra note 6.
25
It should be noted that broad studies on the health of Vietnam veterans
are not informative with respect to the health effects of Agent Orange because of
serious misclassification errors that arise from considering the fact of service in
Vietnam to be equivalent to having served in a sprayed area. The large-scale
study undertaken by the Centers for Disease Control on the health of troops
assigned to Vietnam, CENTERS FOR DISEASE CONTROL, U.S. DEPT. OF HEALTH
AND HUMAN SERVICES, HEALTH STATUS OF VIETNAM VETERANS: VIETNAM
EXPERIENCE STUDY (1989) (Vols. I-V, Supplements A-C), for example, reveals
nothing about Agent Orange and other herbicides, nor was it the intent of the
study to address this issue. The same is true for the Selected Cancer Study. See
SELECTED CANCERS COOPERATIVE STUDY GROUP, ASS’N OF SELECTED
CANCERS WITH SERVICE IN THE U.S. MILITARY IN VIETNAM, II. SOFT-TISSUE
AND OTHER SARCOMAS, 150 ARCHIVES OF INTERNAL MED. 2485 (1990);
SELECTED CANCERS COOPERATIVE STUDY GROUP, ASS’N OF SELECTED
CANCERS WITH SERVICE IN THE U.S. MILITARY IN VIETNAM, III. HODGKIN’S
DISEASE, NASAL CANCER, NASOPHARYNGEAL CANCER, AND PRIMARY LIVER
CANCER, 150 ARCHIVES OF INTERNAL MED. 2495 (1990).
OF
STELLMAN MACROED CORRECTED 070505.DOC
516
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
exposure levels to Vietnam veterans have been a major roadblock
to carrying out large-scale epidemiology studies of the relationship
between exposure to military herbicides and adverse health
outcomes. Indeed, the Centers for Disease Control and Prevention
(CDC) and the Agent Orange Working Group, a subcommittee of
the White House Domestic Policy Council, declared that military
records could not be used to reconstruct past exposures, and the
CDC Agent Orange Study was abruptly halted, with unused funds
being returned to the Treasury.26
In 1994, however, the IOM again did not concur with the
federal scientists’ conclusions that any epidemiological study was
ipso facto impossible because of the inability to classify exposure
based on military records. The IOM recommended that a
methodological study be undertaken to determine whether methods
involving the historical reconstruction of military records could be
used for characterizing exposure to herbicides in Vietnam and as
the basis for epidemiology studies of Vietnam veteran health. The
National Academy of Sciences (NAS) subsequently received a
contract from the VA to seek independent researchers to develop
an appropriate methodology to conduct the investigation. The
exposure opportunity methodology described in this article is the
result of a subcontract from the NAS undertaken by the authors for
that purpose.27 In 1998, a project was begun to refine and validate
an EOI methodology that had previously been used in the exposure
assessment of claimants to the Agent Orange Veterans Payment
Program28 and in studies of Vietnamese citizens29 and American
26
The abandonment of the Agent Orange Study was the subject of
unsuccessful litigation by the American Legion and the Vietnam Veterans of
America, who sought to have the congressionally mandated study reinstated. See
American Legion v. Derwinski, 54 F.3d 789 (D.C. Cir. 1995); American Legion
v. Derwinski, 827 F. Supp. 805 (D.D.C. 1993).
27
See 1994 IOM REPORT, supra note 6; COMMITTEE ON THE ASSESSMENT
OF WARTIME EXPOSURE TO HERBICIDES IN VIETNAM, INSTITUTE OF MEDICINE,
CHARACTERIZING EXPOSURE OF VETERANS TO AGENT ORANGE AND OTHER
HERBICIDES USED IN VIETNAM: SCIENTIFIC CONSIDERATIONS REGARDING A
REQUEST FOR PROPOSALS FOR RESEARCH (1997).
28
The original methodology was the basis for assessing exposure eligibility
for the Agent Orange Veterans Payment Program, established in the Agent
Orange class action settlement. In re “Agent Orange” Prod. Liab. Litig., 611 F.
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 517
Vietnam veterans.30
EOI models, which are becoming increasingly common in
epidemiological studies, typically represent exposure as a function
of proximity in time and space to a toxic agent.31 An EOI is often
used, for example, in occupational or environmental studies as a
surrogate estimator of historical exposure where current
environmental or biomarker measurements are inadequate
estimators of past exposures and where measurements were never
made in the past or are unavailable.32 The EOI concept is
complementary to traditional exposure methodologies based upon
toxicological models and measures. Exposure opportunity is not in
itself a toxicological measure, but EOI scores can be incorporated
into toxicological models as “presentation” dosages. Such dosages
are intended for use in large-scale studies in which a location
history is the principal source of information about an individual or
group, such as a military unit. These models are especially
applicable to studies in which body burden measurements are
impractical or unlikely to reflect exposures in the distant past.
To perfect an EOI model for military herbicides in Vietnam, a
Geographical Information System (GIS) for the former Republic of
Vietnam was created. The GIS is a relational database whose
Supp. 1396 (E.D.N.Y. 1985).
29
Marie-Catherine Ha et al., Agent Orange and the Risk of Gestational
Trophoblastic Disease in Vietnam, 51 ARCHIVES ENVTL. HEALTH 368 (1996).
30
A cross-sectional study of American Legionnaires utilized an earlier
version of the EOI methods described here. See Steven D. Stellman & Jeanne
Mager Stellman, Estimation of Exposure to Agent Orange and Other Defoliants
Among American Troops in Vietnam: A Methodological Approach, 9 AM. J.
INDUS. MED. 305 (1986), for the methodology and Steven D. Stellman et al.,
Combat and Herbicide Exposure in Vietnam Among American Legionnaires, 47
ENVTL. RESEARCH 112, 120-21 (1988), for the distribution of EOIs in the
cohort.
31
See, e.g., Kirk R. Smith, Place Makes the Poison: Wesolowski Award
Lecture – 1999, 12 J. EXPOSURE ANALYSIS. & ENVTL. EPIDEMIOLOGY 167
(2002); Mary H. Ward et al., Identifying Populations Potentially Exposed to
Agricultural Pesticides Using Remote Sensing and a Geographic Information
System, 108 ENVTL. HEALTH PERSPECTIVES 5 (2000).
32
See, e.g., E. S. Schaeffner et al., Use of an Asbestos Exposure Score and
the Presence of Pleural and Parenchymal Abnormalities in a Lung Cancer Case
Series, 7 INT’L J. OF OCCUPATIONAL & ENVTL. HEALTH 14 (2001).
STELLMAN MACROED CORRECTED 070505.DOC
518
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
component tables (the “layers”) contain data on herbicide
application, military troop location, and other geographically
encoded data resources that are designed to be utilized in the
assessment of exposure to herbicides and exposure-related health
risks for specific populations. Table 1 contains an abbreviated list
of data layers that are currently included in the GIS.33 The GIS is
built around two interrelated concepts: the partitioning of Vietnam
into 0.01° x 0.01° “square” grids and the association of the
geographic center of each grid with a continuous EOI and a vector
of four proximity “hit” scores.34 Data in each layer have been
geocoded in a manner compatible with our Vietnam grid system.
Unique grid identifiers serve to link data between cartographic
layers.
33
Adapted from Jeanne Mager Stellman et al., A Geographic Information
System for Characterizing Exposure to Agent Orange and Other Herbicides in
Vietnam, 111 ENVTL. HEALTH PERSPECTIVES 321, 322 (2003).
34
The EOI takes into account entry into areas sprayed in the past as well as
being present during an actual spray mission. A conservative first-order decay
model is used to simulate the decay of herbicide in the environment. The term
“hit” is applied when an individual actually was located in or near the spray path
during a mission. The mathematical representations of these two models are
given in Steven D. Stellman & Jeanne Mager Stellman, Exposure Opportunity
Models for Agent Orange, Dioxin, and Other Military Herbicides Used in
Vietnam, 1961-1971, 14 J. EXPOSURE, ANALYSIS & ENVTL. EPIDEMIOLOGY 354
(2004).
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 519
Table 1. GIS data tables (“layers”) of location and herbicide
spray data that can be linked to create exposure opportunity
scores.35
Type of activity or
Examples
data
HERBS file
Flight paths and other details of herbicide
spray applications
Civilian habitations
Cities, towns, villages, hamlets, plantations
Vietnam land and
Soil typology, land topography, rivers,
water features
streams
Civil structures
Roadways, utility lines, rail lines, canals,
air fields
Military structures
Military bases, base camps, landing zones,
air fields
Troop locations
Headquarters, base camps, depots, and
other locations assigned to support and
combat support units; tracked locations for
combat troops
Operation Ranch
Specific areas designated for defoliation
Hand Targets
and crop destruction by an elaborate
approval mechanism
Herbicide storage,
Locations of known “incidents” such as
transport, and
spills, dumps, and crashes
unplanned dispersal
At the heart of exposure assessment is a comprehensive
database, known as the HERBS file, that describes all documented
herbicide applications that were carried out by the U.S. military
during the Vietnam War. This database was compiled from a wide
variety of archival sources under a contract from the NAS. The
HERBS file contains information consisting of one or more
records that collectively describe the spray coordinates of single or
multiple aircraft (known as sorties) during 9,141 missions. The
majority of spray (about 18 million gallons) was applied by
specially equipped C-123 transport aircraft in Operation Ranch
Hand. The chief herbicide uses were defoliation and crop
35
See Stellman et al., supra note 33, for more details.
STELLMAN MACROED CORRECTED 070505.DOC
520
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
destruction. During the work undertaken for the NAS, the HERBS
file was extensively corrected and validated.36 U.S. Army (not Air
Force) personnel sprayed tens of thousands of gallons along base
camp perimeters, waterways, and communication lines by
helicopter, backpack, truck, and boat. Each of these modalities was
calibrated to spray the phenoxyherbicides at a rate of 3 gallons per
acre. A large percentage of these missions were entered into a
second HERBS file, sometimes called the Services-HERBS. A
major data cleaning and reconciliation effort was undertaken to
eliminate redundancies in these two files, and the current version
of the HERBS file contained in the GIS reflects those quality
control changes.
More than 98% of all herbicide spraying was by fixed-wing
aircraft. Key to the usefulness of the HERBS file is the fact that it
describes the actual flight paths taken by the Ranch Hand aircraft.
For example, the HERBS file contains “leg designators” that
permit the reconstruction of the contiguous flight paths of 5,215
fixed-wing Ranch Hand missions, most with multiple sorties.
Although the GIS is a useful tool for visualizing locations of
individuals or military units in relation to herbicide applications,
the sheer quantity of data in both the herbicide and unit location
databases and the almost limitless possibilities for temporal and
spatial variation make calculation of exposure opportunity scores a
formidable challenge. To reduce this task to manageable
proportions, a user-friendly software system called Herbicide
Exposure Assessment – Vietnam (HEA-V) was created.37 The
software accepts as input a “location history” database in
Microsoft Excel or Access format and produces as output a set of
“hits” scores and the EOI score for each input record. The input
may pertain equally to a specific military unit that traveled from
one place to another or to an individual who belonged to that unit.
It may also pertain to a fixed location, such as a village, hamlet, or
36
Details of the process whereby the HERBS file was corrected are given
in STELLMAN & STELLMAN, CHARACTERIZING EXPOSURE OF VETERANS TO
AGENT ORANGE AND OTHER HERBICIDES IN VIETNAM, supra note 11, at 5-7.
37
Stellman & Greene Consulting, Herbicide Exposure AssessmentVietnam (HEA-V), software manual and appendices (Brooklyn, N.Y., Found.
for Worker, Veteran and Envtl. Health, Inc. 2003) (on file with author).
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 521
other geographical entity. It is thus equally suited for use in both
epidemiological and ecological studies.
B. Military Unit Location Database Core Data Layer
In the course of providing exposure analyses for the Agent
Orange Veterans Payment Program (AOVPP), a database was
compiled (at the battalion level) reflecting those locations at which
AOVPP claimants were stationed. Further troop location data for
all Army combat support units assigned to Vietnam have now been
compiled from a variety of primary and secondary military
sources.38 Through a Freedom of Information Act request, the
Special Master to the AOVPP obtained data files that contained the
locations of many Army combat military units. The files had been
created in the course of the CDC’s aforementioned abandoned
Agent Orange Study. The data were compiled by the Department
of Defense Environmental Support Group, which had identified
and tracked more than sixty combat battalions stationed in the III
Corps Tactical Zone, a very heavily sprayed region extending from
the southern coast of Vietnam to the Cambodian border and
including Saigon. The Support Group tracked the daily activities
and locations of individual companies in these battalions between
1967 and 1969 utilizing a wide variety of data sources, such as
daily journals and ORLLs (Operations Reports and Lessons
Learned). These data have now been updated and extensively
“cleaned” to remove obvious typographical errors; the GIS now
38
In general, approximately five out of six troops serve in such support
units, which we call “stable” units because troops are stationed at specific base
camps and are not required to move frequently. The ratio in Vietnam appears to
have been lower, with proportionately more troops assigned to combat. There
were more than 1,650 “stable” Army units, which together had an average
authorized troop strength of just under 200,000. Nearly 1,000 additional units,
whose authorized total troop strength was about 162,000, were also largely
stationary but had “mobile elements” who routinely left base camps to carry out
their missions. These units included Aviation, Engineering, Ordnance, Signal,
Transportation, and Medical Corps and Military Police. The stable units
provided support for more than 400 highly mobile units, such as Infantry,
Armor, Cavalry, and Artillery battalions, whose strength averaged more than
120,606.
STELLMAN MACROED CORRECTED 070505.DOC
522
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
contains detailed location data for sixty-three combat battalions for
the time period between 1967 and 1969.39 In addition, nonexhaustive databases for the U.S. Air Force, Navy, and Marines
also have been compiled.40
With knowledge of the unit to which an individual was
assigned and the individual’s dates of assignment, it is thus
possible to link the individual to various locations over time. These
locations and dates then become input data for the exposure
opportunity calculations. Thus, for any given location, military
unit, or individual, researchers can calculate an EOI as a
quantitative spatio-temporal representation of that individual’s
proximity to a toxic agent. The EOI model takes into account three
independent factors that determine an individual’s exposure:
concentration of the toxicologically active substance, distance from
the spray application, and the time during which the exposure may
have taken place. Details of this exposure methodology have been
published.41
Extensive calculations have been carried out to validate EOI
measurements. Those locations at which military units were found
to have high EOI scores coincide closely with the “hot spots”
indicated in EOI surface plots of Vietnam. The log-normal
distributions of exposure scores, especially those that show
39
The CDC had asserted that these extracted files contained location gaps
that invalidate them as a data source for epidemiological studies. The data
cleaning carried out during the course of our research found many gaps to be the
result of clerical error rather than missing data and that sufficient data are
available to construct study populations of a size suitable for valid
epidemiological studies.
40
Exposure estimation for these branches of the military is usually simpler
because, for example, most Naval units (with known exceptions such as
Riverine units) were located offshore and thus had no opportunity for exposure.
In addition, there were a limited number of Air Force installations and the
Marines belonged to a comparatively small number of units, mostly assigned to
I Corps (the northern region) in comparatively restricted areas.
41
See Jeanne Mager Stellman et al., A Geographic Information System for
Characterizing Exposure to Agent Orange and Other Herbicides in Vietnam,
supra note 33; Steven D. Stellman & Jeanne Mager Stellman, Exposure
Opportunity Models for Agent Orange, Dioxin, and Other Military Herbicides
Used in Vietnam, supra note 34.
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 523
systematically higher exposure for the combat units whose
locations were in the heavily sprayed areas of III Corps described
above, indicate both face and content validity.
Two studies have compared the EOIs calculated in the GIS
with serum dioxin. These comparisons provide an objective
measure of body burden when samples are taken sufficiently close
in time to the exposure event and when the kinetics of metabolism
are taken into account. In 1989, Dr. Sylvaine Cordier of France’s
National Institute of Health and Medical Research (INSERM)
requested that the authors help evaluate exposures for a series of
twenty-seven patients admitted for abdominal surgery to the Cho
Ray Hospital in Ho Chi Minh City, for whom adipose tissue was
being collected for subsequent dioxin assay. EOIs were estimated
through our then-current algorithms using the subjects’ residential
locations. Five patients’ levels were at background. For the
remaining twenty-two patients, the Pearson correlation
coefficient42 was 0.50 for association between the log of serum
dioxin and the log of the EOI.43 The second biomarker study was a
pilot project that was part of a collaboration between the authors
and the International Agency for Research on Cancer (IARC),
which had carried out a case-control study of non-Hodgkin’s
lymphoma and soft tissue sarcomas in Vietnamese civilians in Ho
Chi Minh City between 1993 and 1996.44 This study also produced
a significant correlation.
Apart from this quantitative validation, a qualitative
concordance has been observed between extremely high dioxin
concentrations in samples of soil taken at an abandoned U.S. air
42
The Pearson correlation coefficient is a measure of association between
two variables. Its value ranges from -1 to +1. A zero coefficient indicates no
association, while a ±1 is either a perfectly direct or perfectly inverse
relationship. Squaring the correlation coefficient approximates the degree of
variation “explained” by the association. A statistically significant correlation of
0.50 is generally considered strong in environmental studies.
43
Pierre Verger et al., Correlation between Dioxin Levels in Adipose Tissue
and Estimated Exposure to Agent Orange in South Vietnamese Residents, 65
ENVTL. RES. 226 (1994).
44
Eva Kramarova et al., Exposure to Agent Orange and Occurrence of
Soft-Tissue Sarcomas or Non-Hodgkin Lymphomas: An Ongoing Study in
Vietnam, 106 ENVTL. HEALTH PERSPECTIVES 671, 671 (1998).
STELLMAN MACROED CORRECTED 070505.DOC
524
7/5/2005 2:37 PM
JOURNAL OF LAW AND POLICY
base in the Ashau Valley45 and the HERBS file locations of
sprayings of Agent Purple at the same location. Agent Purple was
an early herbicide with a dioxin contamination level estimated at
10 to 100 times that of Agent Orange.46
CONCLUSION
In its review of the methodological work on the GIS described
in this article, the IOM confirmed that the exposure opportunity
methodology and the resulting GIS system made epidemiological
studies possible and, moreover, urged that epidemiological studies
be undertaken immediately.47 The editors of Nature similarly
agreed that the work on the GIS and on the revised inventory of
spraying (the HERBS file) enabled the performance of urgently
needed studies on the effects of Agent Orange.48
45
L. Wayne Dwernychuk et al., Dioxin Reservoirs in Southern Viet Nam: A
Legacy of Agent Orange, 47 CHEMOSPHERE 117, 121 (2002).
46
Recently the possibility that the elevated dioxin could be attributed to
storage of herbicide at Special Forces base was raised. See L. Wayne
Dwernychuk, Dioxin Hotspots in Vietnam, CHEMOSPHERE (forthcoming). This is
highly unlikely since operational records specifically state that all defoliation
was to be carried out by C-123 spray mission because tree height made hand
spraying impractical and the loading of spray planes is documented to have
taken place at Tan Son Nhut Air Force base and not at the camp itself. The camp
was only in operation for a relatively brief period of time because it proved to be
ineffective against the Viet Cong insurgency, thus making it likely that
documentation of spraying is complete. U.S. DEPARTMENT OF DEFENSE.
RECORDS OF THE U.S. FORCES IN SOUTHEAST ASIA, HEADQUARTERS, MILITARY
ASSISTANCE COMMAND VIETNAM (MACV), ASSISTANT CHIEF OF STAFF FOR
OPERATIONS (J3), CHEMICAL OPERATIONS DIVISION (MACJ-3–09). Herbicide
Operations Plans (1966–1967) series, Record Group 472 (National Archives
and Records Administration, College Park, MD; 1950–75).
47
COMMITTEE ON THE ASSESSMENT OF WARTIME EXPOSURE TO
HERBICIDES IN VIETNAM, INSTITUTE OF MEDICINE, CHARACTERIZING EXPOSURE
OF VETERANS TO AGENT ORANGE AND OTHER HERBICIDES USED IN VIETNAM:
REPORT AND RECOMMENDATIONS (National Academy of Sciences Press,
Washington, D.C., 2003).
48
Some of the work reported here appeared as a cover article in Nature,
which was accompanied by the following legend: “[T]his work has provided a
geographic information system that will allow epidemiologists to piece together
health effects that may exist in the region as they now have a much clearer idea
STELLMAN MACROED CORRECTED 070505.DOC
7/5/2005 2:37 PM
EPIDEMIOLOGICAL STUDIES AND AGENT ORANGE 525
It is of interest that the work undertaken on behalf of the NAS
built upon the exposure methodologies adopted by the Special
Master for the Agent Orange Veterans Payment Program as a
means for determining whether a deceased or disabled claimant
met the court-established criteria for exposure. The court had
reasoned that the NAS previously had considered the HERBS file
to be a unique and valid source of specific information on the
military spraying49 and that “geographic and temporal limits must
be set to determine whether a veteran who was in a location near a
sprayed area at or subsequent to the time of spraying will be
considered exposed.”50 This reasoning has now been affirmed by
the IOM.
The IOM recommendations were strongly endorsed with
bipartisan support by both the House and Senate Veterans Affairs
Committees, which requested that the VA initiate such studies
immediately.51 The VA responded that the studies were still
premature, but that “in-house” validation studies would be carried
out.52 Such an internal study by the VA is, in fact, explicitly
disallowed by the Agent Orange Act of 1991, which sought to
avoid potential conflicts of interest by mandating that a major
epidemiological study be carried out by non-governmental
researchers. Further correspondence from the VA to the American
Legion53 at the time of this writing indicates that the VA has taken
no further steps to launch an external investigation and plans to
continue with its internal studies until at least 2007.
about the distribution of the agents (and dioxin) and about the ‘hot spots.’”
49
COMMITTEE ON THE EFFECTS OF HERBICIDES IN VIETNAM, NATIONAL
RESEARCH COUNCIL, THE EFFECTS OF HERBICIDES IN SOUTH VIETNAM; PART A.
SUMMARY AND CONCLUSIONS (National Academy of Sciences Press,
Washington, D.C., 1974).
50
In re “Agent Orange” Prod. Liab. Litig., 611 F. Supp. 1396, 1417
(E.D.N.Y. 1985).
51
Letter from U.S. Congress House and Senate Veterans’ Affairs
Committees to Honorable Anthony Principi (Nov. 24, 2003) (on file with
author).
52
Letter from Honorable Anthony Principi to U.S. Congress House and
Veterans Affairs Committees (Dec. 18, 2003) (on file with author).
53
Letter from Jonathan B. Perlin, Acting Undersecretary for Health, Dep’t
of Veterans Affairs, to John Sommer (Nov. 12, 2004) (on file with author).
20313_blp_13-2 Sheet No. 18 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
CONNECTIONS: THE EARLY HISTORY OF
SCIENTIFIC AND MEDICAL RESEARCH ON
“AGENT ORANGE”
David A. Butler, Ph.D.
INTRODUCTION
The author is Senior Program Officer at the National Academy of
Sciences, Institute of Medicine in Washington, D.C., and has directed a number
of the National Academies’ studies of Vietnam veterans’ health.
1
This paper uses the term “Agent Orange” as a catch-all to refer to all of
the herbicides employed in Vietnam and the controversy that resulted from that
action. The quotation marks are removed when referencing the particular
chemical formulation that was given that name.
C M
Y K
06/23/2005 10:19:08
527
20313_blp_13-2 Sheet No. 18 Side A
Scientists, accustomed to working with and advancing the state
of the art, are apt to forget that those in earlier times did not
possess the information they take for granted. Individuals in the
legal profession tend to have a better appreciation for this reality
since their work is grounded in the evolutionary process that
defines precedent; indeed, questions of “who knew what and
when” are central to some cases. However, lawyers and judges too
can benefit from a better understanding of the facts and mindsets
that have informed decisions made in the past.
Historical perspective is especially important for understanding
environmental health questions because the field has advanced
quickly in the past several decades, and the attitudes of exposed
populations, industry, government, and medical, scientific, and
legal professionals have changed accordingly. “Agent Orange”1
litigation issues, for example, are the subject of many present-day
challenges and fascinations. The events that preceded these
challenges are no less interesting. Unfortunately, they remain
20313_blp_13-2 Sheet No. 18 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
528
largely inaccessible to the legal community, in part because
historical accounts of salient science and health issues tend to be
focused on discrete subissues rather than the larger phenomena at
hand2 and because they tend to appear in journals or venues3 not
frequented by legal scholars. This paper’s goal is to address this
gap by providing a lay summary of the primary events and
advances that defined the “Agent Orange” issue up to the time that
the U.S. government decided to suspend herbicide spray operations
in Vietnam. In so doing, this paper provides a context for
understanding some of the past actions of the parties to this longrunning controversy.
I. THE 1870S-1900S: SETTING THE STAGE FOR THE MODERN
INDUSTRIAL CHEMICAL AND HERBICIDE INDUSTRIES
2
C M
Y K
06/23/2005 10:19:08
See generally Orvin C. Burnside, The History of 2,4-D and Its Impact on
Development of the Discipline of Weed Science in the United States, in U.S.
DEP’T AGRIC., No. 1-PA-96, NAT’L AGRIC. PESTICIDE IMPACT ASSESSMENT
PROGRAM (NAPIAP) 5 (1996) (explaining the history of 2,4-D development);
WILLIAM A. BUCKINGHAM JR., U.S. AIR FORCE, OPERATION RANCH HAND: THE
AIR FORCE AND HERBICIDES IN SOUTHEAST ASIA, 1961-1971 (1982), available
at https://www.airforcehistory.hq.af.mil/Publications/fulltext/operation_ranch_
hand.pdf (detailing the Vietnam aerial spray program).
3
See generally NAT’L AGRIC. LIBR., The Alvin L. Young Collection on
Agent Orange, available at www.nal.usda.gov/speccoll/findaids/agentorange/
(providing numerous references used in this paper).
4
V. Merz & W. Weith, Zur Kenntnis des Perchlorphenols [On the
Characteristics of Pentachlorophenol], 5 BERICHTE DER DEUTSCHEN
CHEMISCHEN GESELLSCHAFT 458 (1872).
20313_blp_13-2 Sheet No. 18 Side B
The modern chemical industry evolved in the late 1800s as
researchers invented and refined mass production processes. These
advances led not only to the more-efficient manufacture of existing
chemicals, but also to a significant expansion of the variety and
properties of new compounds. Among the myriad papers relating
experimental results, many of them produced by the formidable
German chemical industry complex, is an 1872 account of
experiments on wood preservatives.4 The researchers included a
description of a byproduct of potassium pentachlorophenol
production: octachlorodibenzo-p-dioxin (OCDD). This is the
20313_blp_13-2 Sheet No. 19 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
6/6/2005 12:56 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 529
5
C M
Y K
06/23/2005 10:19:08
See infra App. A (presenting a primer on dioxins).
NAT’L INST. ENVTL. HEALTH SCI., NAT’L INST. HEALTH, DIOXIN
RESEARCH AT THE NATIONAL INSTITUTE OF ENVIRONMENTAL HEALTH SCIENCES
(NIEHS), available at http://www.niehs.nih.gov/oc/factsheets/dioxin.htm (last
modified Nov. 18, 2002).
7
The halogens are any of five related nonmetallic elements—fluorine,
chlorine, bromine, iodine, or astatine—that readily form negative ions and are
thus highly reactive.
8
Karl Herxheimer, Über Chlorakne [Regarding Chloracne], 46
MÜNCHENER MEDIZINISCHE WOCHENSCHRIFT 278 (1899).
9
See infra App. B (presenting a primer on chloracne).
10
James R. Troyer, In the Beginning: The Multiple Discovery of the First
Hormone Herbicides, 49 WEED SCI. 290, 291 (2001) (citing Theophil Ciesielski,
6
20313_blp_13-2 Sheet No. 19 Side A
earliest known reference to “dioxin” in the scientific literature. In
brief, dioxins5 are a class of chemical compounds that are a
modern-day public health concern because they accumulate in fat,
exhibit high toxicity in some experimental animals, and have been
associated with health problems in humans.6
An important innovation of this time was the implementation
of electrolytic production methods to make chemicals. These
methods, which use electric current to catalyze reactions, were far
more efficient than earlier techniques and, as a result, led to a
significant expansion of the commercial production of chlorinebased and other halogenated compounds.7 It appears in hindsight
that this expansion had an unintended consequence: for the first
time in the medical literature, descriptions and case reports of an
acne-like occupational illness associated with exposure to halogens
appear. Karl Herxheimer8 is responsible for the original account of
this disease or was at least the first to refer to the condition by the
name that it is now known—chlorakne or, in English, chloracne.
The condition he described resulted from exposure to tars that were
generated during an electrolytic process producing chlorine.9
A separate, and at this point distinct, line of scientific inquiry
also produced knowledge that would later contribute to the “Agent
Orange” story. James Troyer notes that three late nineteenth
century botanists (Theophil Ciesielski, Charles Darwin, and
Hermann Vöchting) independently established that a stimulus to
one part of a plant might result in a response in a different part of
the plant.10 Research conducted in the early twentieth century
20313_blp_13-2 Sheet No. 19 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
530
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
established that this phenomenon was governed by chemical
signaling and that plant growth was among the effects that were
controlled in this manner. Frits Warmolt Went11 was the first to
isolate a growth-controlling substance—3-indoleacetic acid
(IAA)—from plant tissues. This class of chemicals later came to be
called plant “hormones” because they function in an analogous
manner to animal hormones, that is, they produce an effect on the
activity of cells remote from their point of origin. Meanwhile, a
French viticulturist observed in 1896 that bouillie bordelaise
(“Bordeaux mixture”)—a fungicide used in vineyards to control
powdery mildew—also eradicated certain weeds.12 He and other
researchers in Germany and the United States pursued this finding
and established that copper salts13 and other inorganic compounds
would kill broadleaf weeds, but not cereals growing in the same
area.14 This was the first indication that some chemicals had
selective herbicidal properties, and the work stimulated
investigations of herbicides that could be used in agricultural
applications.
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 19 Side B
Untersuchungen über die Abwärtskrümmung der Wurzel [Investigations of the
Downward Curvature of Roots], 1 BEITRAEGE ZUR BIOLOGIE DER PFLANZEN 1
(1872); CHARLES DARWIN, THE POWER OF MOVEMENT IN PLANTS 592 (1880);
Hermann Vöchting, Über die Lichtstellung der Laubblätter [Regarding the
Position of Leaves in Response to Light], 46 BOTANISCHE ZEITUNG 501 (1888)).
11
Frits Warmolt Went, On Growth-Accelerating Substances in the
Coleoptile of Avena Sativa, 30 PROCEEDINGS OF THE KONINKLIJKE
NEDERLANDSE AKADEMIE VAN WETENSCHAPPEN 10 (1926); Frits Warmolt
Went, Wuchsstoff und Wachstum [Growth-substance and Growth], 24 RECUEIL
DES TRAVAUX BOTANIQUES NÉERLANDAIS 1 (1928).
12
COMM. ON THE EFFECTS OF HERBICIDES IN VIETNAM, NAT’L ACAD. SCI.,
The Effects of Herbicides in South Vietnam: Part A - Summary and Conclusions,
II-1 (1974), available at http://www.nal.usda.gov/speccoll/findaids/agentorange/
text/00181.pdf.
13
Copper sulfate is an active ingredient in Bordeaux mixture. Id. at II-1.
14
Burnside, supra note 2, at 5; see also A. Robert Mazur, Weeds and Their
Control, USGA GREEN SECTION RECORD, Sept. 1969, at 1, available at
http://turf.lib.msu.edu/1960s/1969/690901.pdf.
20313_blp_13-2 Sheet No. 20 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
6/6/2005 12:57 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 531
II. THE 1900-1940S: THE FIRST REPORTS OF U.S. HEALTH
PROBLEMS FROM EXPOSURE TO DIOXIN-LIKE CHEMICALS
Many of the early reports of chloracne and other diseases now
associated with acute dioxin intoxication in occupational settings
were associated with the use of chlorinated naphthalenes.15
Although contemporaneous documentation is lacking, David E.
Wells and Jacob de Boer noted that occupational exposure to
chlorinated naphthalenes used in the production of gas masks
during World War I was associated with numerous incidents of the
disease.16
In 1927, Ludwig Teleky published (in German) an
investigation of occupational chloracne, which reported that the
fewer the chlorines present in the chlorinated naphthalene, the less
the acne in the exposed subject.17 The first comprehensive study
published in English18 was stimulated by health problems in
15
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 20 Side A
Polychlorinated naphthalenes (PCNs) consist of two benzene rings and
comprise 75 congeners with between 1 and 8 chlorines; however, in
naphthalenes, the rings are directly joined at two adjacent carbons. Their
physical and chemical properties are largely similar to those of polychlorinated
biphenyls (PCBs), and some exhibit dioxin-like toxicity.
16
David E. Wells & Jacob de Boer, Polychlorinated Biphenyls, Dioxins
and Other Polyhalogenated Hydrocarbons as Environmental Contaminants in
Food, in ENVIRONMENTAL CONTAMINANTS IN FOOD 305, 305-06 (Colin F.
Moffat & Kevin J. Whittle eds., 1999).
17
Ludwig Teleky, Die Pernakrankheit [Chloracne], 6 KLINISCHE
WOCHENSCHRIFT 845 (1927); compare Cecil K. Drinker, Report to the
Monsanto Chemical Company (Sept. 15, 1938) (evaluating, among other topics,
the
toxicity
of
“chlorinated
diphenyl
#1268”),
available
at
http://www.chemicalindustryarchives.org/search/pdfs/anniston/19380915_545.p
df. This compound—now called Aroclor 1268 or PCB-1268—is highly (68%)
chlorinated. Drinker found that it was less toxic than diphenyls that were not as
highly chlorinated, exposing a flaw in Teleky’s earlier presumption. Id. at 9-12.
18
Barry Commoner, The Political History of Dioxin, Keynote Address at
the Second Citizens Conference on Dioxin (July 30, 1994), available at
http://www.greens.org/s-r/078/07-03.html (noting, but not providing a citation
for, a 1936 case history in the Archives of Dermatology and Syphilology that
describes a Monsanto worker who “began work in the distillation of chlorinated
diphenyl in April 1930”). The employee experienced severe chloracne and other
symptoms that were present more than three years after the initial exposure. Id.
20313_blp_13-2 Sheet No. 20 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
532
employees of the Halowax Corporation, which used chlorinated
naphthalenes and PCBs (then called chlorinated diphenyls) to
produce coatings for insulating, waterproofing, and fireproofing
wires. Cecil Drinker, a physician researcher at Harvard University,
was asked by Halowax in the spring of 1936 to investigate these
problems, which included three fatal cases of jaundice.19 In their
1937 paper, Drinker and his colleagues discussed the findings of
their laboratory investigations (using a rat model) into the effect of
these chlorinated hydrocarbons on the liver. They concluded that
systemic effects were possible, but that high-level exposure was
required to trigger this result. Accordingly, they concluded that
“[c]ompared to benzene . . . and many other compounds, these
substances are very little toxic and operations employing them can
easily be safeguarded.”20 The Drinker et al. paper includes what is
presented as a transcript of a colloquium convened to discuss the
findings. Colloquium participants included the chief of the U.S.
Public Health Service’s Division of Industrial Hygiene; state health
officials from Connecticut, Massachusetts, and Ohio;
representatives of the General Electric and Monsanto Chemical
companies; and the president of Halowax Corporation.21
A number of English-language papers on chloracne and the
health impacts of chlorinated naphthalenes followed Drinker’s
seminal effort.22 Of particular interest is Leonard Greenburg and
C M
Y K
06/23/2005 10:19:08
Cecil K. Drinker et al., The Problem of Possible Systemic Effects from
Certain Chlorinated Hydrocarbons, 19 J. INDUS. HYGIENE & TOXICOLOGY 283
(1937). These cases were previously described by Flinn and Jarvik, who were
using laboratory animals to investigate the toxicity of chlorinated naphthalenes.
F.B. Flinn & N.E. Jarvik, Action of Certain Chlorinated Napthalenes on the
Liver, 35 PROCEEDINGS OF THE SOC’Y FOR EXPERIMENTAL BIOLOGY & MED.
118, 119 (1936).
20
Drinker, supra note 19, at 299.
21
Id. at 300-11.
22
See May R. Mayers & Mabel G. Silverberg, Skin Conditions Resulting
From Exposure to Certain Chlorinated Hydrocarbons, 20 J. INDUS. HYGIENE &
TOXICOLOGY 244 (1938); Cecil K. Drinker, Further Observations on the
Possible Systemic Toxicity of Certain of the Chlorinated Hydrocarbons with
Suggestions for Permissible Concentrations in the Air of Workrooms, 21 J.
INDUS. HYGIENE & TOXICOLOGY 155 (1939); Halowax Acne (Queries and
Minor Notes), 12 JAMA 471 (1943); Lewis Schwartz & Samuel M. Peck,
20313_blp_13-2 Sheet No. 20 Side B
19
20313_blp_13-2 Sheet No. 21 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
6/6/2005 12:58 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 533
colleagues’ account of three young adults (two males and one
female) “known to have been working with chlorinated
naphthalenes and diphenyls” who died after presenting with
jaundice and other symptoms.23 These investigators recommended
“conscientious reporting by physicians of all illnesses occurring
among workers exposed to chlorinated naphthalenes and
diphenyls, particularly cases that have been worked up, so that the
clinical disease entities resulting from such exposures can be
further clarified and thus more readily recognized in the future.”24
It is important to note that while these early research efforts
associated various health problems with exposure to specific
chemicals, they might reflect an incomplete evaluation of the full
range of the subjects’ exposures. Furthermore, it is unlikely that
any tests for contaminants were conducted. While some
compounds that workers were exposed to—notably Halowaxes—
are strong chloracnegens themselves, impurities in others may
have caused or contributed to the reported health problems.
III. THE 1930S-1940S: THE ADVANCEMENT OF PLANT HORMONE
RESEARCH
C M
Y K
06/23/2005 10:19:08
Occupational Acne, 43 N.Y. STATE J. OF MED. 1711 (1943); Thomas P.
Connelly & William C. Marsh, Chloracne (Cable Rash), 42 U.S. NAVY MED.
BULLETIN 403 (1944).
23
Leonard Greenburg et al., The Systemic Effects Resulting from Exposure
to Certain Chlorinated Hydrocarbons, 21 J. INDUS. HYGIENE & TOXICOLOGY
29, 36 (1939).
24
Id. at 38 (emphasis in original).
25
Fritz Kögl et al., Über den Einfluss der auxine auf das Wurzelwachstum
und die chemische Natur des Auxins der Graskoleoptilen [Regarding the
Influence of Auxins on Root Growth and the Chemical Nature of the Auxins
Found in Maize Coleoptiles], 228 ZEITSCHRIFT FÜR PHYSIOLOGISCHE CHEMIE
104, 121 (1934).
26
Folke Skoog & Kenneth V. Thimann, Further Experiments on the
Inhibition of the Development of Lateral Buds by Growth Hormone, 20
20313_blp_13-2 Sheet No. 21 Side A
The burgeoning research on plant hormones in the 1930s is
now recognized as having laid the foundation for the modern
herbicides industry. Fritz Kögl and colleagues25 and Folke Skoog
and Kenneth Thimann26 are credited with first observing that IAA
20313_blp_13-2 Sheet No. 21 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
534
6/6/2005 12:59 PM
JOURNAL OF LAW AND POLICY
C M
Y K
06/23/2005 10:19:08
PROCEEDINGS OF THE NAT’L ACAD. OF SCIENCES 480, 482-83 (1934).
27
Burnside, supra note 2, at 5-6; Troyer, supra note 10, at 291.
28
From the Greek auxeinҗ, meaning “to grow.” See http://dictionary.
reference.com/search?q=auxin.
29
Percy W. Zimmerman & Frank Wilcoxon, Several Chemical Growth
Substances Which Cause Initiation of Roots and Other Responses in Plants, 7
CONTRIBUTIONS FROM BOYCE THOMPSON INST. 209, 225-26 (1935).
30
Burnside, supra note 2, at 5-6.
31
William G. Templeman, The Effect of Some Plant Growth-Substances on
Dry-Matter Production in Plants, 7 EMPIRE J. OF EXPERIMENTAL AGRIC. 76
(1939).
20313_blp_13-2 Sheet No. 21 Side B
could either promote or inhibit growth, depending on the
concentration used. Several researchers soon identified natural
compounds in urine and synthesized others that were chemically
similar to IAA and had the same properties.27 The term “auxin”28
was coined to signify both natural and synthetic substances with
these properties. The discovery by Percy Zimmerman and Frank
Wilcoxon29 that certain phenylacetic acids and naphthylacetic
acids (NAA) acted as auxins was particularly important, as this
generated interest in aryloxyacetic acids as potential plant growth
regulators.30
The aryloxyacetic acids that were to become the primary
constituents of Agent Orange were first synthesized in the late
1930s. William Gladstone Templeman’s work in 1936 and 1937,
which he documented in a 1939 paper, established that both IAA
and NAA could have toxic effects in plants.31 However, these
compounds were expensive to isolate and not stable in the
environment, making them unsuitable for widespread use.
Templeman, who was affiliated with Imperial Chemical Industries
(ICI), tapped that company’s expertise to synthesize a number of
chemically similar compounds. Among these was 2,4dichlorophenoxyacetic acid (2,4-D), for which a patent application
was filed in April 1941. Robert Pokorny, an American working for
the specialty chemical company C.B. Dolge, is responsible for the
first scientific publication that mentions 2,4-D and the other Agent
Orange component: 2,4,5-trichlorophenoxyacetic acid (2,4,5-T).
He disseminated one-paragraph descriptions of their synthesis, one
after the other, in a March 27, 1941 submission to a section of the
20313_blp_13-2 Sheet No. 22 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 535
Journal of the American Chemical Society called “New
Compounds.”32
Other researchers were apparently monitoring these
developments or working along similar lines because several
advances regarding 2,4-D and 2,4,5-T took place over the next few
years. Notable among the publications were Percy Zimmerman and
Alfred Hitchcock’s 1942 findings that 2,4-D was an auxin
(establishing its mode of action) and that it was 300 times more
powerful than indolebutyric acid, the most widely used growth
promoter at the time.33
IV. THE 1940S: HERBICIDES AND THE MILITARY
In addition to this publicly available work, other research was
being conducted that would only become known after wartime
secrecy concerns had passed. Gale Peterson34 relates that the chair
of the University of Chicago’s botany department, Ezra Kraus, was
central to this work. Kraus is credited with being the first to note
that auxins might be useful as herbicides if applied at sufficiently
high doses. An early 1941 letter sent by Kraus to two of his former
graduate students—John Mitchell and Charles Hamner—
documented the suggestion and led to their initiating research on
32
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 22 Side A
Robert Pokorny, New Compounds: Some Chlorophenoxyacetic Acids, 63
J. OF THE AM. CHEM. SOC’Y 1768 (1941), available at http://pubs.acs.org/cgibin/abstract.cgi/jacsat/1941/63/i06/f-pdf/f_ ja01851a601.pdf.
33
Percy W. Zimmerman & Alfred E. Hitchcock, Substituted Phenoxy and
Benzoic Acid Growth Substances and the Relation of Structure to Physiological
Activity, 12 CONTRIBUTIONS FROM BOYCE THOMPSON INST. 321 (1942).
34
Gale E. Peterson, The Discovery and Development of 2,4-D, 41 AGRIC.
HISTORY 243 (1967). Peterson’s history is summarized and in some cases
supplemented by the Midwest Research Institute’s study of the ecological
effects of repeated use of herbicides. See MIDWEST RESEARCH INST.,
ASSESSMENT OF ECOLOGICAL EFFECTS OF EXTENSIVE OR REPEATED USE OF
HERBICIDES (1967) (project monitored by the Department of the Army under
contract No. DAHC15-68-C-0119; ARPA Order No. 1086) [hereinafter MRI],
available at http://www.nal.usda.gov/speccoll/findaids/agentorange/text/03632.
pdf; see also Burnside, supra note 2; Troyer supra note 10. Troyer details the
near simultaneous discovery of many of the important properties of phenoxy
herbicides by a number of researchers.
20313_blp_13-2 Sheet No. 22 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
536
the topic later that year.35 Meanwhile, in October 1941, Secretary
of War Henry Stimson asked the National Academy of Sciences
(NAS) to form a committee to assess the state of knowledge in the
field of biological warfare. In a late 1942 or early 1943
presentation to the “War Bureau of Consultants” committee,36
Kraus suggested that “the toxic properties of growth-regulating
substances for the destruction of crops or the limitation of crop
production” might have military application and should be
investigated.37 In response to these comments and
recommendations from the NAS committee, research was greatly
accelerated. Camp Detrick, Maryland (later named Fort Detrick),
which had recently been established as a center for weapons
research, began work on herbicides and the U.S. Army poured
funding into Kraus’s and other investigators’ studies.38 Of the more
than 1,000 potential agents screened, 2,4-D39 and 2,4,5-T were
found to be the most promising.40 Field trials, including aerial
spraying, were conducted in the Florida Everglades, and an
ammonium thiocyanate formulation was developed for possible
use in the Pacific Theater. However, that war ended without
herbicides being used in a military application. It was not until the
so-called “Malayan Emergency”—an insurgency that attempted to
overthrow the British colonial administration of Malaysia—that
herbicides were employed as a weapon. British troops used aerial
dissemination of herbicides for defoliation and crop destruction
C M
Y K
06/23/2005 10:19:08
Peterson, supra note 34, at 245 & n.12.
Id. at 246. The committee may have instead been the “ABC committee,”
a successor working group given a nonsense name to hide the nature of its work.
37
Ezra J. Kraus & John W. Mitchell, Growth-Regulating Substances as
Herbicides, 108 BOTANICAL GAZETTE 301, 302 (1947).
38
Peterson, supra note 34, at 248.
39
Id.; Paul C. Marth & John W. Mitchell, 2,4-Dichlorophenoxyacetic Acid
as a Differential Herbicide, 106 BOTANICAL GAZETTE 224 (1944) (noting the
one property that would make 2,4-D among the most popular of the commercial
herbicides—it selectively killed dicotyledon plants (dichots, which include
broadleaf plants like most weeds) while sparing monocotyledons (monocots,
including grasses and many food plants like cereals and grains)). This meant it
could be applied on fields and lawns without laying waste to desirable growth.
2,4,5-T also has this property.
40
MRI, supra note 34.
36
20313_blp_13-2 Sheet No. 22 Side B
35
20313_blp_13-2 Sheet No. 23 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 537
between 1951 and 1953.41
V. THE LATE 1940S AND 1950S: THE BEGINNING OF LARGE-SCALE
PHENOXY HERBICIDE PRODUCTION AND THE CONTINUATION OF
CIVILIAN SECTOR RESEARCH
41
VALERIE ADAMS, CHEMICAL WARFARE, CHEMICAL DISARMAMENT 74
(1990).
42
C M
Y K
06/23/2005 10:19:08
2,4,5-trichlorophenol (abbreviated as trichlorophenol or TCP) is a
chemical made in an intermediate step of the production process for 2,4,5-T and
some other biocides.
43
William F. Ashe & Raymond R. Suskind, Reports on Chloracne Cases,
Monsanto Chemical Company, Nitro, WV (Oct. 1949 & Apr. 1950)
(unpublished reports for Dep’t of Envtl. Health, College of Medicine, Univ. of
20313_blp_13-2 Sheet No. 23 Side A
The commercialization and widespread use of 2,4-D and 2,4,5T-based herbicides began in the years following the end of World
War II, spurred in part by the results of wartime research efforts.
As production processes were established and the industry geared
up, large numbers of workers were exposed for the first time to the
herbicides, their precursors, and unintentional contaminants. This
led in 1949 to what is now regarded as the first mass health
incident specifically attributable to the dioxin 2,3,7,8tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD or TCDD). The
incident began on March 8, 1949 at a Nitro, West Virginia, plant
operated by Monsanto Chemical Corporation. When the
temperature and pressure in a reactor vessel (called an autoclave)
that was used to make 2,4,5-trichlorophenol42 exceeded limits, a
safety valve released and vented the vessel’s contents into the air
outside the building. Workers tasked with the cleanup (there was
no decontamination) and repair of the unit and those who were in
the vicinity in the days after the incident reported a number of
symptoms, including eye and respiratory tract irritation, headache,
dizziness, nausea, and severe skin irritation. Within weeks,
chloracne, hyperpigmentation, liver function impairment, muscle
pain, and a variety of central nervous system disturbances were
observed. These health problems were documented in
contemporaneous unpublished reports by physicians working for
Monsanto,43 but were not addressed in a peer-reviewed journal
20313_blp_13-2 Sheet No. 23 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
538
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
until 1980.44 Interestingly, Marion Moses et al. indicated that while
117 workers developed chloracne as a result of the incident, an
almost equal number (111) showed signs of chloracne prior to this
exposure.45
There were at least four other exposure incidents associated
with TCP in facilities in Germany during the next five years: at a
chemical plant in Nordrhein-Westfalen in 1949;46 at two C.H.
Boehringer Sohn facilities located in the middle Rhine in 1952 and
1953;47 at a BASF (Badische Anilin- & Soda-Fabrik) factory in
Ludwigshafen in 1953;48 and at a Hamburg-Moorfleet plant
working under contract to Boehringer in 1954.49 The BASF
incident also was the result of an autoclave explosion, and
involved not only plant workers and persons performing the
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 23 Side B
Cincinnati); Raymond R. Suskind & William F. Ashe, A Clinical and
Environmental Survey, Monsanto Chemical Company, Nitro, W.V. (July 1953)
(unpublished report of the Kettering Laboratory, Univ. of Cincinnati) (the
authors of this report are also cited as Atkins, Davis, and Suskind in some
sources).
44
Judith A. Zack & Raymond R. Suskind, The Mortality Experience of
Workers Exposed to Tetrachlorodibenzodioxin in a Trichlorophenol Process
Accident, 22 J. OCCUPATIONAL MED. 11 (1980).
45
Marion Moses et al., Health Status of Workers with Past Exposure to
2,3,7,8-Tetrachlorodibenzo-p-dioxin in the Manufacture of 2,4,5Trichlorophenoxyacetic Acid: Comparison of Findings With and Without
Chloracne, 5 AM. J. INDUS. MED. 161, 164 tbl.1 (1984).
46
E. W. Baader & H. J. Bauer, Industrial Intoxication Due to
Pentachlorophenol, 20 INDUS. MED. & SURGERY 286 (1951); ALVIN L. YOUNG
ET AL., THE TOXICOLOGY, ENVIRONMENTAL FATE AND HUMAN RISK OF
HERBICIDE ORANGE AND ITS ASSOCIATED DIOXIN V-7 (1978).
47
H. Bauer et al., Berufliche Vergiftungen bei der Herstellung von
Chlorphend-Verbindungen [Occupational Poisonings in the Production of
Chlorophenol Compounds], 18 ARCHIV FÜR GEWERBEPATHOLOGIE UND
GEWERBEHYGIENE 538 (1961).
48
P. J. Goldman, Schwerste akute Chloracne, eine Massenintoxikation
durch 2,3,6,7-Tetrachlorodibenzodioxin [Severe Acute Chloracne, A Mass
Intoxication Due to 2,3,6,7-Tetrachlorodibenzodioxin] 24 DER HAUTARZT 149
(1973); A. M. Thiess et al., Mortality Study of Persons Exposed to Dioxin in a
Trichlorophenol-Process Accident That Occurred in the BASF/AG on November
17, 1953, 3 AM. J. INDUS. MED. 179, 179-80 (1953).
49
A. Manz et al., Cancer Mortality Among Workers in Chemical Plant
Contaminated with Dioxin, 338 LANCET 959 (1991).
20313_blp_13-2 Sheet No. 24 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 539
cleanup, but also a child of one of the employees. The child
apparently developed chloracne as a result of coming into contact
with a contaminated towel and scarf.50
Preliminary toxicologic research on the herbicides was also
being conducted. LW Kephart reports that Kraus experimented on
himself and found no ill effect from ingesting 0.5 grams of 2,4-D
per day for three weeks.51 In 1946, John Mitchell and colleagues
published the first animal studies of 2,4-D, stating that cows and
sheep grazing on pastures treated with the compound did not
manifest toxic effects.52 Other early studies include that of Nancy
Bucher,53 who found that repeated injections of 2,4-D did not
influence the growth rate of sarcomas in mice, and that of Edwin
Hill and Harold Carlisle,54 who examined acute and subacute
effects of ingestion in a variety of mammals and determined toxic
doses.
The first toxicity studies of 2,4,5-T were conducted in the
1950s by Dow Chemical Company (Dow)55 and the first
publication of results was made in 1953.56 The authors, who tested
50
Goldman, supra note 48.
L. W. Kephart, Panel Discussion on New Chemicals for Weed Control, 2
PROCEEDINGS OF THE N. CENT. WEED CONTROL CONF. 68 (November 26-28,
1945).
52
J.W. Mitchell et al., Tolerance of Farm Animals to Food Containing 2,4Dichlorophenoxyacetic Acid 5 J. ANIMAL SCI. 226, 232 (1946).
53
Nancy L. R. Bucher, Effects of 2,4-Dichlorophenoxyacetic Acid on
Experimental Animals, 63 PROCEEDINGS OF THE SOC’Y FOR EXPERIMENTAL
BIOLOGY & MED. 204, 205 (1946).
54
Edwin V. Hill & Harold Carlisle, Toxicity of 2,4-Dichlorophenoxyacetic
Acid for Experimental Animals, 29 J. INDUS. HYGIENE & TOXICOLOGY 85, 94-95
(1947).
55
COLIN M. MACLEOD, EXECUTIVE OFFICE OF THE PRESIDENT, REPORT ON
2,4,5-T: A REPORT OF THE PANEL ON HERBICIDES OF THE PRESIDENT’S SCIENCE
ADVISORY COMMITTEE 39 (1971).
56
Victor A. Drill & Tomiharu Hiratzka, Toxicity of 2,4Dichlorophenoxyacetic Acid and 2,4,5-Trichlorophenoxyacetic Acid: A Report
on Their Acute and Chronic Toxicity in Dogs, 7 A.M.A. ARCHIVES INDUS.
HYGIENE & OCCUPATIONAL MED. 61 (1953). ENVTL. PROT. AGENCY, REP. OF
THE ADVISORY COMM. ON 2,4,5-T TO THE ADMIN. OF THE ENVTL PROT.
AGENCY, (1971) (incorrectly asserting that the first 2,4,5-T toxicology paper
was not published until 1963) [hereinafter REPORT OF THE ADVISORY
51
20313_blp_13-2 Sheet No. 24 Side A
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 24 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
540
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
the effects of ingestion on dogs, found that their highest dose level
(20 mg/kg/day) resulted in death, but that animals that survived
lower dose levels did not develop significant lesions in the liver or
other organs. The extent to which the 2,4,5-T used in these
experiments may have been contaminated with dioxins is not
known.57
VI. 1957: THE CONNECTION IS MADE
C M
Y K
06/23/2005 10:19:08
COMMITTEE], available at http://www.nal.usda.gov/speccoll/findaids/agent
orange/text/01152.pdf.
57
Later research would show that there are unusually large variations
between and among species in susceptibility to TCDD’s effects. For example,
several thousand times more TCDD is needed to kill a hamster than a guinea
pig, when doses are equalized by body weight. AGENCY FOR TOXIC SUBSTANCES
& DISEASE REGISTRY, U.S. DEP’T HEALTH & HUMAN SERVICES,
TOXICOLOGICAL PROFILE FOR CHLORINATED DIBENZO-P-DIOXINS 11 (1998)
[hereinafter ATSDR], available at http://www.atsdr.cdc.gov/toxprofiles/ tp104.
pdf. This characteristic makes it difficult to translate animal tests of dioxin
toxicity to humans.
58
Bo Holmstedt, Prolegomena to Seveso Ecclesiastes I 18, 44 ARCHIVES
OF TOXICOLOGY 211, 216 (1980).
20313_blp_13-2 Sheet No. 24 Side B
Thus, by the mid-1950s, there had been a series of incidents in
which workers exposed to 2,4,5-trichlorophenol had experienced a
set of health problems that had previously been seen in industries
using polychlorinated naphthalenes (PCDDs) and polychlorinated
biphenyls (PCBs). The physicians treating these workers had
advanced various theories to explain the source of the health
problems but had yet to identify a mechanism. The set of events
that would clarify this connection began in 1956. Karl H. Schulz, a
physician in the Department of Dermatology at the Eppendorfer
Krankenhaus in Hamburg, was referred a patient from the
Boehringer Hamburg-Moorfleet factory who was suffering from
chloracne.58 The case stimulated Schulz to begin studying
chloracne problems at the facility. In cooperation with a chemist
there named Georg Sorge, Schulz obtained samples of 2,4,5-T that
had been manufactured at various levels of purity. Tests performed
on rabbit ears showed that the highest purity samples (formulated
for use in experimentation) did not elicit a skin reaction, while the
20313_blp_13-2 Sheet No. 25 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
6/6/2005 1:01 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 541
so-called “technical grade” samples (formulated for use in
chemical processes such as 2,4,5-T production) did.59 This
suggested that some impurity in lower-grade TCP was
chloracnegenic. At about the same time, a team of researchers led
by Wilhelm Sandermann of the University of Hamburg
synthesized TCDD while investigating new chlorophenols for use
as wood preservatives. An assistant on the team developed
chloracne when some of the dried compound accidentally blew
into his face.60 He was referred to Schulz, who recognized the
possible connection with his other chloracne cases and contacted
Sorge.61 The chemist isolated TCDD from technical grade TCP
and independently synthesized the dioxin; both yielded positive
results in rabbit ear tests conducted by Schulz. The connection was
confirmed when Schulz applied a 0.01% solution of TCDD to his
forearm and developed a chloracneform lesion within the next
several days.62
In 1957, Schulz and a colleague published three papers (in
German) implicating TCDD as the causative agent in the chloracne
observed in persons exposed to chlorinated hydrocarbons.63
Sandermann et al. also released their description of TCDD
synthesis, again in German.64 Michael Gough reports that although
59
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 25 Side A
Id. at 217. There was no economic incentive to eliminate trace
contaminants that did not affect the performance of the final product.
60
MICHAEL GOUGH, DIOXIN, AGENT ORANGE: THE FACTS 31 (1986).
61
Id.
62
Id. at 33; Moses, supra note 45, at 161-62.
63
J. Kimmig & K. H. Schulz, Berufliche Akne (sog. Chlorakne) Durch
Chlorierte Aromatische Zyklische Äther [Occupational Acne (So-Called
Chloracne) Due to Chlorinated Aromatic Cyclic Ether], 115 DERMATOLOGIA
540 (1957); J. Kimmig & K. H. Schulz, Chlorierte Aromatische Cyclische Äther
als Ursachen der Sogenannten Chlorakne [Chlorinated Aromatic Cyclic Ethers
as the Cause of Chloracne], 44 NATURWISSENSCHAFTEN 337 (1957); K. H.
Schultz, Klinische und Experimentelle Untersuchungen zur Ätiologie der
Chlorakne [Clinical and Experimental Investigations into the Etiology of
Chloracne] 206 ARCHIV FÜR KLINISCHE UND EXPERIMENTELLE DERMATOLOGIE
589 (1957).
64
W. Sandermann et al., Über die Pyrolyse des Pentachlorphenols
[Regarding the Pyrolysis of Pentacholorphenols] 90 CHEMISCHE BERICHTE
(BERICHTE DER DEUTSCHEN CHEMISCHEN GESELLSCHAFT) 690 (1957).
20313_blp_13-2 Sheet No. 25 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
542
Sorge did not turn out any papers, his research was applied by
Boehringer to reduce TCDD contamination in the production
process.65
VII. THE LATE 1950S-1960S: MILITARY USES OF HERBICIDES
AGAIN COME TO THE FORE
U.S. military interest in herbicides did not end with the WWII
research effort, although the pace of work diminished after the end
of the conflict. Wartime herbicide development and candidate
screening programs were terminated in 1950, and a much lower
level effort was carried on between 1950 and 1957 under the
general rubric of anticrop warfare research.66 Alvin Young and
colleagues indicate that one of the findings made by Fort Detrick
investigators (in 1951) was that mixtures of the butyl esters of 2,4D and 2,4,5-T were the chemicals of choice for vegetation
control.67 Research also included some refinement of aerial
dissemination hardware and the testing of formulations in Puerto
Rico.68
All herbicide research funding at Fort Detrick ended in fiscal
year 1958 and the facility’s remaining stock was turned over to the
U.S. Department of Agriculture (USDA). However, only months
later, the facility was asked to help clear vegetation from an
C M
Y K
06/23/2005 10:19:08
Gough, supra note 60, at 33.
WILLIAM F. WARREN, A REVIEW OF THE HERBICIDE PROGRAM IN SOUTH
VIETNAM (1968), available at http://www.nal.usda.gov/speccoll/findaids/agent
orange/text/03121.pdf.
67
ALVIN L. YOUNG ET AL., THE TOXICOLOGY, ENVIRONMENTAL FATE AND
HUMAN RISK OF HERBICIDE ORANGE AND ITS ASSOCIATED DIOXIN (1978),
available at http://www.nal.usda.gov/speccoll/findaids/agentorange/text/01165.
pdf.
68
J. van Overbeek & Ismael Vélez, Use of 2,4-Dichlorophenoxyacetic Acid
as a Selective Herbicide in the Tropics, 103 SCIENCE 472 (1946); F. P. HUDDLE,
A TECHNOLOGY ASSESSMENT OF THE VIETNAM DEFOLIANT MATTER, REPORT TO
THE SUBCOMMITTEE ON SCIENCE, RESEARCH, AND DEVELOPMENT OF THE
COMMITTEE ON SCIENCE AND ASTRONAUTICS, 91ST CONG. (1st Sess. 1969),
available at http://www.nal.usda.gov/speccoll/findaids/agentorange/text/05167.
pdf.
66
20313_blp_13-2 Sheet No. 25 Side B
65
20313_blp_13-2 Sheet No. 26 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 543
artillery test range at Fort Drum in New York.69 Some of the stock
was retrieved and the operation was carried out in June 1959. An
undiluted 1:1 mixture of butyl esters of 2,4-D and 2,4,5-T (this
formulation was later called “Purple,” and then “Agent Purple”)
was sprayed from booms attached to a helicopter70 over a four
square mile area at a concentration of approximately 6 lbs/acre.71
Operation director James W. Brown, a researcher in the Crops
Division at Fort Detrick, reported that the leaves of the hardwood
trees (primarily sugar maples) in the testing area turned color
shortly after the application and appeared brown and dead after
about a month.72
The operation was deemed a success, and in May 1961, the
Department of Defense’s (DOD’s) Advanced Research Projects
Agency tasked Brown and other Fort Detrick researchers to travel
to Vietnam and evaluate the effect of various herbicides,
formulations, and application methods on indigenous trees and
foodstuffs.73 Of the eighteen formulations tested, the Agent Purple
mix of 2,4-D and 2,4,5-T and one other formulation were deemed
the best.74 This determination was based not only on efficacy, but
also on cost, availability of large quantities, and “known or proven
safety in regard to their toxicity to humans and animals.”75
In the first aerial herbicide spray mission, which was conducted
on January 10, 1962, Agent Purple was applied to an area adjacent
C M
Y K
06/23/2005 10:19:08
KENT R. IRISH ET AL., DEP’T ARMY, INFORMATION MANUAL FOR
VEGETATION CONTROL IN SOUTHEAST ASIA (Dec. 1969), available at
http://www.nal.usda.gov/speccoll/findaids/agentorange/text/00073.pdf.
70
JAMES W. BROWN, U.S. ARMY CHEM. CORPS RESEARCH & DEV.
COMMAND, VEGETATIONAL SPRAY TESTS IN SOUTH VIETNAM, SUPPLEMENT (U)
26 (1962), available at http://www.nal.usda.gov/speccoll/findaids/agentorange/
text/00337.pdf.
71
Id. at 9.
72
Id. at 10.
73
Id. at 4, 9.
74
Id. at 18-20. The other was Agent Blue, which was composed of the acid
and sodium salt of cacodylic acid (also called dimethylarsinic acid). Id. at 19
tbl.1. Agent Blue was used against rice crops, where 2,4-D and 2,4,5-T were
ineffective.
75
Id. at 9.
20313_blp_13-2 Sheet No. 26 Side A
69
20313_blp_13-2 Sheet No. 26 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
544
to a target that had been sprayed with Agent Pink76 during a test
mission a few weeks earlier. Several available accounts of the
program (known as “Operation Ranch Hand”) detail the conduct of
the missions and the range of herbicide agents used.77
VIII. THE MID- TO LATE-1960S: CONCERNS OVER HEALTH IMPACTS
FROM WARTIME HERBICIDES EXPOSURE
Domestic manufacturers struggled to meet the great demand
for herbicides that stemmed from the Vietnam conflict.78 Among
their efforts was an initiative by Dow in a Midland, Michigan,
plant to expand its 2,4,5-T capacity through a new production
process. Chloracne cases started appearing in the workforce soon
after production began, and Dow responded by shutting down the
facility and alerting other manufacturers to the problem. TCDD
was implicated as the causative agent. When the plant reopened in
1966, it used a revamped production process that limited TCDD
contamination to less than 1 part per million (ppm).79
The mid-1960s was also a time of burgeoning interest in
studies of the mutagenic, carcinogenic, and reproductive effects of
chemicals. The National Cancer Institute launched an investigation
76
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 26 Side B
Id. at 87. Agent Pink was a 60:40 mix of the n-butyl and isobutyl esters
of 2,4,5-T. The formulation was used only sparingly during the war. Id. at 19
tbl.1.
77
See, e.g., BUCKINGHAM, supra note 2; INSTITUTE OF MEDICINE,
VETERANS AND AGENT ORANGE: HEALTH EFFECTS OF HERBICIDES USED IN
VIETNAM (1994) [hereinafter IOM]. There are numerous subsequent
publications in the VETERANS AND AGENT ORANGE report series, including
1996, 1998, 2000, 2002, and 2004 Updates. See also Jeanne Mager Stellman et
al., The Extent and Patterns of Usage of Agent Orange and Other Herbicides in
Vietnam, 422 NATURE 681 (2003).
78
The need for Agent Orange and its chemical precursors exceeded the
total domestic production capacity in late 1966 and the Defense Production Act
of 1950 was invoked in March 1967 to accelerate delivery of supplies. See
Jonathan Glasser, The Government Contract Defense: Is Sovereign Immunity A
Necessary Prerequisite?, 52 BROOK. L. REV. 495, 512-14 (1986) (addressing
some of the legal implications of invoking the Defense Production Act of 1950
for manufacturers of Agent Orange—a major issue in the In re Agent Orange
litigation).
79
MACLEOD, supra note 55, at 17.
20313_blp_13-2 Sheet No. 27 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 545
80
C M
Y K
06/23/2005 10:19:08
“Viet Cong”—short for Viet Nam Cong San (Vietnamese communists)—
is term used in the United States to refer to the rebels that fought for the
National Front for the Liberation of Southern Vietnam, which was also known
as the National Liberation Front. See http://www.encyclopedia.com/html/v/
vietc1ong.asp (last visited Apr. 10, 2005).
81
CHARLES V. COLLINS, HERBICIDE OPERATIONS IN SOUTHEAST ASIA JULY
1961-JUNE 1967 54 (1967) available at http://www.nal.usda.gov/speccoll/
findaids/agentorange/text/03116.pdf.
82
Id. at 47.
20313_blp_13-2 Sheet No. 27 Side A
of the tumorogenic, mutagenic, and teratogenic potential of a
number of insecticides and herbicides in 1965 and gave the
contract to Bionetics Research Laboratories. Aerial spray
operations in Vietnam significantly increased in 1966 and 1967,
and there is evidence that the Viet Cong (VC)80 became concerned
about the health effects associated with exposure to herbicides. An
Air Force tactical evaluation written in 1967 and declassified in
1974 stated:
Even some [Viet Cong] leaders have misconceptions of the
effects of the herbicides. VC medical officers instruct
members of units not to eat the contaminated food as it
would “damage their health and cause stomach and liver
disorders”. . . . Another official VC document . . . directs
the units to not allow livestock to graze in sprayed areas or
to be given food that has been sprayed. VC officials also
instructed the men to wear homemade or issued gas masks
as “bodily contact would cause physical harm or in some
cases even death.”81
The report also stated that “[i]t must be remembered that the
chemical spray is non-toxic and has had no effects on aircrews or
[south Vietnamese] regular troops, nor have there been any illeffects reported during use of these chemicals in the United
States,” and attributed claims to the contrary among the
Vietnamese to mistaken beliefs or propaganda.82
Opposition to the wartime use of herbicides within the
academic community in the United States led to a series of position
statements and petitions from individuals and organizations. These
culminated in a September 13, 1967 letter from the president of the
American Association for the Advancement of Science (AAAS) to
20313_blp_13-2 Sheet No. 27 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
546
Secretary of Defense Robert McNamara suggesting that an
independent study be initiated to evaluate the short- and long-term
consequences of the military’s use of herbicides in Vietnam.83 The
DOD had already commissioned the Midwest Research Institute to
do such a study, which was released that December.84 The report
summarized the then-available literature on ecologic studies of the
“extensive or repeated use of herbicides” and on toxicologic
effects.85 The authors recommended that additional research be
undertaken into the possible teratogenic and carcinogenic activity
of the active ingredient in Agent Blue, but did not offer any
suggestions regarding 2,4,5-T or the other herbicides used in agent
formulations.86 Dioxin is not mentioned in the report’s more than
370 pages of text.
Two events in the summer of 1969 changed the outlooks of
many with regard to the safety of wartime herbicide spraying.
Between June 26 and July 5, 1969, newspapers in Vietnam
published accounts of increased rates of birth defects in areas in
which spraying had taken place.87 These reports raised public
concern and prompted two surveys of hospital records.88 Results of
2,4,5-T testing, conducted as part of the Bionetics study, also
started circulating in public that summer. The study, which was the
first to address an outcome related to low-level exposure, found
that ingested and injected 2,4,5-T could cause birth defects in some
C M
Y K
06/23/2005 10:19:08
See, e.g., BUCKINGHAM, supra note 2.
See MRI, supra note 34, at 139.
85
Id.
86
Id. at 164.
87
REPORT OF THE ADVISORY COMMITTEE, supra note 56, at 51.
88
The results of one of the surveys were published in December 1970.
ROBERT T. CUTTING ET AL., DEP’T DEF., CONGENITAL MALFORMATIONS,
HYDATIDIFORM MOLES AND STILLBIRTHS IN THE REPUBLIC OF VIETNAM, 19601969 (1970), available at http://www.nal.usda.gov/speccoll/findaids/agent
orange/text/00579.pdf. Findings from the second were presented a few months
after the spray missions had been discontinued. MATTHEW S. MESELSON ET AL.,
HERBICIDE ASSESSMENT COMM’N AM. ASS’N FOR THE ADVANCEMENT OF SCI.,
BACKGROUND MATERIAL RELEVANT TO PRESENTATIONS AT THE 1970 ANNUAL
MEETING OF THE AAAS 23-25 (1971), available at http://www.nal.usda.gov/
speccoll/findaids/agentorange/text/00900.pdf.
84
20313_blp_13-2 Sheet No. 27 Side B
83
20313_blp_13-2 Sheet No. 28 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 547
laboratory animals.89 These events compelled the White House to
order a partial curtailment of the use of Agent Orange in Vietnam
on October 29, 1969.90 Concurrently, restrictions were placed on
the use of 2,4,5-T on food crops.91 The Secretary of Health,
Education, and Welfare,92 and the Secretaries of Agriculture and
the Interior jointly announced the suspension of many uses of the
herbicide on April 15, 1970; the DOD temporarily suspended the
military use of 2,4,5-T the same day.93 There were limited stocks
of the other agent formulations and supplies quickly ran out. The
last fixed-wing defoliation mission was flown on May 9, 1970,
although limited crop destruction missions were carried out
throughout the rest of the year. On January 7, 1971, aerial spray
missions came to an end. Research on the longer term and lower
exposure level effects of Agent Orange herbicides intensified soon
thereafter, and investigations into dioxin’s biological effects and
health impacts began in earnest. This work and its influence on
policy and other issues are addressed in accompanying papers in
this volume.
CONCLUSION
The early history of “Agent Orange” may therefore be viewed
as a series of connections between what began in the late 1800s as
distinct lines of scientific, technological, and medical inquiry.94
C M
Y K
06/23/2005 10:19:08
A paper reporting salient results appeared in the journal Science on May
15, 1970. K. Diane Courtney et al., Teratogenic Evaluation of 2,4,5-T, 168
SCIENCE 864 (1970). It included a note (apparently added on 9 April of that
year) that stated that the 2,4,5-T used in testing was contaminated with ~30ppm
TCDD. Id. at 866.
90
REPORT OF THE ADVISORY COMMITTEE, supra note 56, at 3.
91
MACLEOD, supra note 55, at 8.
92
The U.S. Department of Health, Education, and Welfare was renamed
the U.S. Department of Health and Human Services in 1980 upon the passage of
the Department of Education Organization Act of 1979, 20 U.S.C. § 3441
(2005), which created a separate Department of Education. See http://www.hhs.
gov/about/hhshist.html.
93
BUCKINGHAM, supra note 2, at 166.
94
This inquiry takes its inspiration from Connections (BBC and Time Life
1978), a series of documentaries on the history of major scientific and technical
20313_blp_13-2 Sheet No. 28 Side A
89
20313_blp_13-2 Sheet No. 28 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
548
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
While much remains to be learned, we now have more than thirtyfive years of relatively intense research to consult when making
decisions about the effects of herbicide and dioxin exposure on
humans. The antecedents addressed here yield perspective on the
decisions made by industry executives, government officials, and
military planners by shedding light on when they may have been
aware of particular elements of the “Agent Orange” story.
20313_blp_13-2 Sheet No. 28 Side B
06/23/2005 10:19:08
discoveries by the author and broadcaster James Burke.
C M
Y K
20313_blp_13-2 Sheet No. 29 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 549
Appendix: A Dioxin Primer
“Dioxin” is the general name given to organic compounds
containing carbon, oxygen, and hydrogen with one to eight
chlorine atoms. Chlorinated dibenzo-p-dioxins (CDDs) are a
family of seventy-five different compounds (called congeners) that
share the parent chemical structure illustrated by the stylized
diagram below:
ATSDR, supra note 57, at 11 (stating that 2,3,7,8-TCDD is the “most
extensively studied CDD and it has been shown to cause a large number of
adverse health effects in animals”).
C M
Y K
06/23/2005 10:19:08
95
20313_blp_13-2 Sheet No. 29 Side A
A chemist would describe this structure as comprising two
benzene rings (hence, dibenzo) joined via two oxygen bridges
(dioxin) at adjacent carbons on the benzene rings; the “-p-” in the
chemical name is an abbreviation for para, the nomenclature for
the position of the oxygens with respect to one another. The seven
polychlorinated dioxins (PCDDs) that have chlorine atoms at the 2,
3, 7, and 8 positions are more toxic than other congeners. Among
these, the congener with four chlorine atoms at these positions
only—2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD)—is the
most potent and by far the most studied.95 2,3,7,8-TCDD is often
(and confusingly) shorted to “dioxin” in the popular literature.
20313_blp_13-2 Sheet No. 29 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
550
Researchers often include two structurally-related compounds
in their consideration of health effects: chlorinated dibenzofurans
(CDFs) and so-called “dioxin-like” polychlorinated biphenyls
(PCBs):
dibenzofuran parent structure
biphenyl parent structure
96
C M
Y K
06/23/2005 10:19:08
This is different than the number of PCDDs with this configuration
because the structure of PCDFs allows combinations of chlorine atoms that are
not physically possible with PCDDs.
97
Martin Van den Berg et al., Toxic Equivalency Factors (TEFs) for PCBs,
PCDDs, PCDFs for Humans and Wildlife, 106 ENVTL. HEALTH PERSPECTIVES
775 (1998).
20313_blp_13-2 Sheet No. 29 Side B
CDFs differ from CDDs by having one direct carbon-carbon
bond substitute for one of the oxygen bridges. There are ten
polychlorinated dibenzofurans (PCDFs) with chlorine atoms at the
2, 3, 7, and 8 positions out of the 135 congeners.96 A single
carbon-carbon bond unites the two benzene rings that make up the
basic structure of PCBs. This bond allows the rings to rotate
relative to one another, which means that many do not have the
planar structure that contributes to the toxic potential of PCDDs
and PCDFs. However, the World Health Organization identifies
two coplanar (also called “non-ortho substituted”) PCB
congeners—3,3’,4,4’,5-pentachlorobiphenyl
(PCB-126)
and
3,3’,4,4’,5,5’
hexachlorobiphenyl
(PCB-169)—as
having
considerable dioxin-like toxicity and ten similar congeners as
having some dioxin-like toxic potential.97
When evaluating the dioxin-related toxicity of a chemical
20313_blp_13-2 Sheet No. 30 Side A
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
THE HISTORY OF RESEARCH ON “AGENT ORANGE” 551
exposure, health researchers often use toxic equivalents (TEQs).
TEQ is a means of calculating the toxicity-weighted masses of
exposures consisting of mixtures of dioxins. It employs toxic
equivalency factors (TEFs) that indicate the potency of a congener
with relation to 2,3,7,8-TCDD, which has the reference value of 1.
TEFs are consensus values derived from the results of in vivo
(animal) and in vitro (cellular) studies and are different for humans
and other animals because species vary greatly in their sensitivity
to dioxin. The TEQ of a mixture is the sum of the concentrations of
the congeners in it multiplied by their TEFs.
APPENDIX B: A CHLORACNE PRIMER
98
C M
Y K
06/23/2005 10:19:08
CANADIAN CENTRE FOR OCCUPATIONAL HEALTH AND SAFETY
(CCOHS), What is Occupational Acne?, available at http://www.ccohs.ca/
oshanswers/diseases/acne.html (last updated Dec. 1, 1997).
99
IOM, supra note 77, at 262.
100
P. D. HOWE ET AL., WORLD HEALTH ORGANIZATION, CHLORINATED
NAPHTHALENES, CONCISE INT’L CHEM. ASSESSMENT DOC. 34, 27 (2001),
available at http://www.who.int/ipcs/publications/cicad/en/cicad34.pdf.
20313_blp_13-2 Sheet No. 30 Side A
Chloracne is an acne-like rash characterized by the formation
of large numbers of comedones (plugs of sebaceous and dead skin
material stuck in the opening of a hair follicle) on the face. In more
severe cases, cysts also form on the shoulders, chest, back,
abdomen, and extremities. Dioxin-associated chloracne can result
from skin contact, ingestion, or inhalation exposure.98 However,
chloracne is not considered to be a good biomarker for dioxin
exposure because individuals vary widely in the dose required to
induce a skin reaction.99 Over the years, several names have been
used to refer to what is here termed “chloracne,” including
Pernakrankheit or Perna disease (because it was caused by
exposure to perchloronaphthaline), chloric or chlorine acne,
halogen acne, Halowax acne, and cable rash.100 The mechanism by
which dioxin causes skin lesions is not understood. Early
researchers thought that skin exposure to dioxin-containing
chemicals plugged the pores, but because the effect also occurs in
inhalation and ingestion exposures, this explanation is clearly
inadequate. Acute dioxin intoxication has other health
20313_blp_13-2 Sheet No. 30 Side B
06/23/2005 10:19:08
BUTLER MACROED 051705.DOC
5/23/2005 10:24 PM
JOURNAL OF LAW AND POLICY
552
consequences, including gastrointestinal symptoms (nausea,
vomiting, epigastric pain, and loss of appetite) and liver damage.101
As early as 1919 it was known that these symptoms appeared in
animals fed chlorinated naphthalenes.102 Epidemiologic studies
also indicate long-term effects from low-level exposure. A full
discussion of this sometimes-controversial topic is beyond the
scope of this paper. However, several sources address it, including
the Institute of Medicine’s Veterans and Agent Orange series of
reports and the U.S. Environmental Protection Agency’s ongoing
dioxin reassessment.103
C M
Y K
06/23/2005 10:19:08
Alexandra Geusau et al., Severe 2,3,7,8-Tetrachlorodibenzo-p-dioxin
(TCDD) Intoxication: Clinical and Laboratory Effects, 109 ENVTL. HEALTH
PERSPECTIVES 865, 865-66 (2001).
102
K.B. LEHMANN, KURZES LEHRBUCH DER ARBEITS - UND
GEWERBEHYGIENE [SHORT TEXTBOOK OF WORK AND TRADE HYGIENE] (1919).
103
IOM, supra note 77; see also U.S. ENVTL. PROT. AGENCY, EXPOSURE
AND HUMAN HEALTH REASSESSMENT OF 2,3,7,8-TETRACHLORODIBENZO-PDIOXIN (TCDD) AND RELATED COMPOUNDS, REVIEW DRAFT, PART I (Dec.
2003), available at http://www.epa.gov/ncea/pdfs/dioxin/nas-review/.
20313_blp_13-2 Sheet No. 30 Side B
101
20313_blp_13-2 Sheet No. 31 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
HOW SCIENTISTS VIEW CAUSALITY AND
ASSESS EVIDENCE: A STUDY OF THE
INSTITUTE OF MEDICINE’S EVALUATION
OF HEALTH EFFECTS IN VIETNAM
VETERANS AND AGENT ORANGE
Irva Hertz-Picciotto, Ph.D., M.P.H.
INTRODUCTION
Professor of Epidemiology, Department of Public Health Sciences,
University of California, Davis; Chair, 2000 and 2002 Institute of
Medicine/National Academy of Sciences Committee on the Health Effects in
Vietnam Veterans of Exposure to Agent Orange and Other Herbicides.
C M
Y K
06/23/2005 10:19:08
553
20313_blp_13-2 Sheet No. 31 Side A
The courts are often called upon to settle disputes in which
health damages are alleged to have been caused by environmental
exposures to chemical, physical, or biologic agents. Similarly,
health scientists are often called upon to assess evidence regarding
diseases or developmental injuries that might be regarded as
resulting from specific exposures. The overarching purpose of this
paper is to familiarize readers with the way in which scientists
assess data and view evidence about causality, using the example
of herbicide and related exposures incurred by U.S. military
personnel during service in Vietnam.
One mechanism by which governmental or regulatory agencies
at the international, national, or regional levels seek advice from
scientists is by convening expert panels. These panels or advisory
boards may be assembled as part of an ongoing program that
reviews the state of the scientific literature on a topic or in
response to specific needs. For example, panels may be assembled
to help formulate a regulatory standard for a chemical in drinking
water, ambient air, or the workplace environment. Thus, the
20313_blp_13-2 Sheet No. 31 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
554
Agent Orange Act of 1991, Pub. L. No. 102-4, 105 Stat. 11 (codified as
amended at 38 U.S.C. § 1116) [hereinafter Agent Orange Act].
C M
Y K
06/23/2005 10:19:08
1
20313_blp_13-2 Sheet No. 31 Side B
documents produced by expert committees may become the
foundation for the development of health-related policies.
The Institute of Medicine Committee to Review the Health
Effects in Vietnam Veterans of Exposure to Herbicides (“IOM
Committee” or “Committee”) is one such panel. This Committee
was formed under the mandate of Public Law 102-4 (better known
as the Agent Orange Act)1 to provide reports on a biannual basis to
the Department of Veterans Affairs (VA), beginning in 1994.
These reports were concerned with the potential adverse effects
that might have been experienced by those who served in Vietnam
because of exposures to herbicides, particularly the mixture
dubbed Agent Orange, or contaminants found in this mixture,
including the well-known chemical compound commonly referred
to as “dioxin.”
Part I of this article introduces the charge to the Committee, the
process the Committee followed in order to reach conclusions
about the evidence, the types of studies it considered, and the
evidentiary categories it established for classifying specific health
outcomes. Part II provides context for the decisions of the IOM
Committee through a discussion of the principles that guided the
Committee’s evaluative process and a presentation of the scientific
concepts that constitute the foundation for inferences about
causation in biomedical research. Part III explains the approach
used by scientists, specifically, epidemiologists, for conducting
studies in populations, estimating causal effects, and examining
hypotheses. It also focuses more concretely on the obstacles to
inferences about causation, specifically, imprecision, which is the
uncertainty that arises from studying small samples, and bias,
which is the uncertainty that derives from imperfections in study
methodology. Part IV narrows this discussion to a description of
the major types of bias—confounding, information, selection, and
statistical bias.
In contrast to the preceding sections, which focus on individual
epidemiologic studies, Part V delineates the process by which
scientists reach consensus and presents the framework commonly
20313_blp_13-2 Sheet No. 32 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
555
used in weighing a body of evidence involving sometimes dozens
of studies. Part VI returns to the work of the IOM Committee and
provides a detailed discussion of the evidence the Committee
reviewed regarding the four outcomes mentioned above, taking
into consideration the concepts presented in Parts III through V.
I. VETERANS AND AGENT ORANGE: THE INSTITUTE OF MEDICINE
COMMITTEE
A. Charge to the Committee
2
C M
Y K
06/23/2005 10:19:08
Id.
Id. §§ 2-3.
4
Id. § 3(g)(1) (requiring that a report be submitted to the Secretary of
Veteran Affairs “at least once every two years”).
5
Id. § 3(d)(1). See also COMMITTEE TO REVIEW THE HEALTH EFFECTS IN
VIETNAM VETERANS OF EXPOSURE TO HERBICIDES, INSTITUTE OF MEDICINE OF
THE NATIONAL ACADEMIES, VETERANS AND AGENT ORANGE: HEALTH EFFECTS
OF HERBICIDES USED IN VIETNAM 221 (1994) [hereinafter VAO 1994].
3
20313_blp_13-2 Sheet No. 32 Side A
In light of growing concern about the health of Vietnam
veterans, Congress enacted Public Law 102-4, the Agent Orange
Act of 1991.2 Through this Act, Congress directed the Secretary of
Veterans Affairs to request from the National Academy of
Sciences (NAS) a comprehensive evaluation of the potential health
effects from exposure to Agent Orange, a chemical compound used
as a defoliant by the U.S. military during the Vietnam War.3 This
legislation also called for reviews of newly available information
on a biannual basis for a period of ten years.4 The Institute of
Medicine (IOM) of the NAS convened a Committee to carry out
this work. The charge to the Committee was to determine “to the
extent that available scientific data permit meaningful
determinations” the answers to three questions regarding specific
health outcomes and their relationships to Agent Orange
exposure.5 The first was “whether a statistical association with
herbicide exposure exists, taking into account the strength of the
scientific evidence and the appropriateness of the statistical and
20313_blp_13-2 Sheet No. 32 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
556
epidemiologic methods used to detect the association.”6 The
Committee also was charged with determining “the increased risk
of the disease among those exposed to herbicides during service in
the Republic of Vietnam and during the Vietnam era.”7 Further, the
Committee was asked to assess “whether there exists a plausible
biologic mechanism or other evidence of a causal relationship
between herbicide exposure and the disease” in question.8 Finally,
Congress charged the Committee with making recommendations
6
Agent Orange Act § 3(d)(1)(A); see also VAO 1994, supra note 5, at 221.
Agent Orange Act § 3(d)(1)(B); see also VAO 1994, supra note 5, at 221
(stating “the increased risk of each disease in question among those exposed to
herbicides during Vietnam service”).
8
Agent Orange Act § 3(d)(1)(C); see also VAO 1994, supra note 5, at 221.
Some authors have argued that the first charge does not mandate the
Committee’s examination of “cause” or “causal association,” but instead
requires only that the Committee look into a possible “statistical association.”
However, the third question indeed requests that the Committee evaluate the
existence of “evidence of a causal relationship.” Notably, any determination
about the existence of “statistical association” that takes into account “strength”
of the evidence and “appropriateness” of the methods examines the same
concerns that enter into a consideration of evidence for causation. These
concerns (for example, the strength of the association and the methods used)
give rise to issues such as bias and confounding, which are defined in detail in
Parts II through IV. Thus, although the Committee was not charged with
drawing a conclusion about causation, the combination of responses to questions
one and three effectively results in a lengthy consideration of virtually all of the
issues that would be discussed if such a conclusion were required. As stated in a
recent update issued by the 2002 Committee:
The evaluation of evidence to reach conclusions about statistical
associations goes beyond quantitative procedures at several stages:
assessing the relevance and validity of individual reports; deciding on
the possible influence of error, bias, confounding, or chance on the
reported results; integrating the overall evidence within and between
diverse fields of research; and formulating the conclusions themselves.
Those aspects of the committee’s review required thoughtful
consideration of alternative approaches at several points and could not
be accomplished by adherence to a narrowly prescribed formula.
COMM. TO REVIEW THE HEALTH EFFECTS IN VIETNAM VETERANS OF EXPOSURE
TO HERBICIDES, INSTITUTE OF MEDICINE OF THE NATIONAL ACADEMIES,
VETERANS AND AGENT ORANGE: UPDATE 2002 28 (2003) [hereinafter VAO
UPDATE 2002].
7
20313_blp_13-2 Sheet No. 32 Side B
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 33 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
557
9
C M
Y K
06/23/2005 10:19:08
Agent Orange Act § 3(e) (directing the National Academy of Sciences to
“make any recommendations it has for additional scientific studies to resolve
areas of continuing scientific uncertainty relating to herbicide exposure”); see
also VAO 1994, supra note 5, at 15. One of these recommendations was to
commission an historical exposure reconstruction. Id. at 17-18. This
recommendation led to the studies of Agent Orange exposure in Vietnam
described by Dr. Jean Mager Stellman in this issue. See Jeanne Mager Stellman
& Steven D. Stellman, Characterization of Exposure to Agent Orange in
Vietnam Veterans As a Basis for Epidemiological Studies, 13 J.L. & POL’Y 505
(2005).
20313_blp_13-2 Sheet No. 33 Side A
for areas in which further study might help answer the questions of
concern.9
Faced with the above mandates, the Committee first clarified
the exposures to be evaluated. The Committee focused specifically
on exposure to Agent Orange. Agent Orange and the other
defoliants used in Vietnam were comprised of combinations of one
or more of four herbicides: 2,4-D (dichlorophenoxyacetic acid),
2,4,5-T
(trichlorophenoxyacetic
acid),
4-amino-3,5,6trichloropiclorinic acid (picloram), and dimethylarsenic acid
(DMA or cacodylic acid). Mixtures containing 2,4-D or 2,4,5-T
were contaminated by chemicals formed during the production
process, including 2,3,7,8-trichlorodibenzodioxin (2,3,7,8-TCDD).
Although other dioxins and dibenzofurans were also formed,
2,3,7,8-TCDD is considered the most toxic and, as such, was the
compound reviewed most extensively by the Committee.
In the course of examining the effects of Agent Orange
exposure, the Committee never evaluated the claim of any
individual veteran, as this was not its charge. Indeed, the Agent
Orange Act specified that such decisions would be made by the
VA. Moreover, the Committee was instructed not to consider the
issue of potential compensation in its deliberations.
Before beginning its work, the members of the Committee
were required to disclose potential conflicts of interest or biases, or
anything that might create the appearance of a conflict of interest.
These included financial holdings, consulting activities,
government service, areas of research, and professional affiliations
as well as any public statements or intellectual positions relevant to
the topic of the Committee. Committee members served without
20313_blp_13-2 Sheet No. 33 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
558
compensation, except for reimbursement of expenses.
Because the scope of the review is broad, the Committee
includes health scientists representing expertise in a wide range of
fields covering epidemiology, oncology, neurology, reproductive
health, and toxicology. The IOM staff assists with the review,
conducting library searches that begin with hundreds, if not a few
thousand, of articles, and works with the Committee to
progressively narrow them down to those articles that are pertinent
to the questions at hand.
B. Types of Evidence Reviewed
10
C M
Y K
06/23/2005 10:19:08
2,3,7,8-TCDD is produced as a by-product of the bleaching process
Such an event occurred in Seveso, Italy, when an explosion at a chemical
plant caused 2,3,7,8-TCDD to contaminate a wide residential area. See Pier
Alberto Bertazzi et al., Health Effects of Dioxin Exposure: A 20-year Mortality
Study, 153 AM. J. EPIDEMIOLOGY 1031 (2001).
11
20313_blp_13-2 Sheet No. 33 Side B
For the scientists on the Committee, it was obvious that any
findings of health effects from these same exposures could serve as
evidence for potential effects in the Vietnam veterans, even if the
results were obtained in other populations. Scientists consider
biological systems in human beings to be sufficiently similar
throughout the world that a high proportion of research findings,
especially those that have been replicated in several studies, can be
“generalized” to much larger populations beyond those that were
studied. Because few studies actually were conducted on Vietnam
veterans, other data sources were frequently used as the basis for
the Committee’s decisions regarding the first and third questions
posed to it by the Act.
The three main sources of epidemiologic data used to address
the first question were studies conducted in: (a) occupational
groups with exposures in the workplace, such as chemical
manufacturing, farming, application of herbicides, or paper and
pulp manufacturing (where 2,3,7,8-TCDD is produced as a byproduct of the bleaching process);10 (b) populations with
environmental exposures, which typically result from accidents
that contaminate residential or recreational areas,11 or alternatively,
from residence in agricultural regions in which herbicides are
20313_blp_13-2 Sheet No. 34 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
559
12
C M
Y K
06/23/2005 10:19:08
See, e.g., Marilyn A. Fingerhut et al., Cancer Mortality in Workers
Exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin, 324 NEW ENG. J. MED. 212,
212 (1991); Kyle Steenland et al., Cancer, Heart Disease, and Diabetes in
Workers Exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 91 J. NAT’L CANCER
INST. 779 (1999).
13
Manolis Kogevinas et al., Cancer Mortality in Workers Exposed to
Phenoxy Herbicides, Chlorophenols, and Dioxins: An Expanded and Updated
International Cohort Study, 145 AM. J. EPIDEMIOLOGY 1061, 1061(1997).
14
See Bertazzi, supra note 11.
15
SCI. APPLICATIONS INT’L CORP., AIR FORCE HEALTH STUDY, AN
EPIDEMIOLOGIC INVESTIGATION OF HEALTH EFFECTS IN AIR FORCE PERSONNEL
FOLLOWING EXPOSURE TO HERBICIDES (1997), FOLLOW-UP EXAMINATION
RESULTS (2000).
20313_blp_13-2 Sheet No. 34 Side A
widely used; and (c) veterans who served in Vietnam, including
not only the U.S. armed forces, but also those from Australia.
Although hundreds of studies have been reviewed each time
the Committee has been convened (every two years), there are
some cohorts of exposed persons that have played a prominent role
in the deliberations. These cohorts had high exposures and were
evaluated numerous times, often for different health outcomes
(such as cancer, heart disease, diabetes, neurologic disorders,
immune function and allergies, reproductive events, etc.), each
time contributing more information to our knowledge base. Some
of the most important of these were the National Institute of
Occupational Safety and Health (NIOSH) cohort of workers
employed after 1942 at twelve plants that manufactured chemicals
containing 2,3,7,8-TCDD;12 a similar multinational cohort from
more than half a dozen European countries, assembled by the
International Agency for Research on Cancer (IARC);13 the cohort
exposed to the explosion of a chemical plant in Seveso, Italy, in
1976 that released 2,3,7,8-TCDD over an area populated by more
than 200,000 persons;14 and the Air Force Health Study (AFHS),
also known as the “Ranch Hand” study, of U.S. Air Force
personnel responsible for flying spraying missions to defoliate
North Vietnam using Agent Orange (these missions were termed
“Operation Ranch Hand”).15
Each of these cohorts was characterized by higher than usual
exposures. For instance, in the two occupational cohorts, a subset
of the workers had experienced chloracne, an acute reaction of skin
20313_blp_13-2 Sheet No. 34 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
560
eruptions that follows high exposures to 2,3,7,8-TCDD. In addition
to the examination of multiple health endpoints, numerous subsets
of the cohorts also were examined more extensively. Furthermore,
in each of these four cohorts exposure was, at some point,
measured in blood drawn from a subset of participants.
The Ranch Hand study was the most extensive study of
veterans. The Air Force initiated this cohort study, which
attempted to recruit about 1,200 servicemen who were identified as
Ranch Hand personnel and 1,700 Air Force personnel who were
assigned to duty in Southeast Asia, but were not exposed
occupationally to herbicides. In 1982, a baseline examination was
conducted of both groups of men, and follow-up took place every
five years thereafter.
The Committee also held public hearings and invited written
submissions. These provided the Committee members with an
opportunity to hear from those most familiar with the conditions
and sequelae of service in Vietnam, as well as the authors of
relevant papers, including some that were in press, but not yet
published.
C. The Process
16
18
See VAO 1994, supra note 5, at 246.
Id.
Id. at 247.
06/23/2005 10:19:08
17
20313_blp_13-2 Sheet No. 34 Side B
To provide a framework for its decisions, the initial committee,
which began meeting in 1992, defined four categories of
evidence.16 The first category consists of those health outcomes for
which the available data provide sufficient evidence of an
association.17 This category applies when multiple studies are
consistent in showing an association, and bias, confounding, or
random variation are not likely to explain the findings. The second
category consists of those health outcomes for which the available
research provides limited or suggestive evidence of an
association.18 This category may apply when multiple studies
observe an association, but the magnitude is sufficiently small that
bias, confounding, or random variation cannot be ruled out.
C M
Y K
20313_blp_13-2 Sheet No. 35 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
561
20
C M
Y K
Id.
Id.
06/23/2005 10:19:08
19
20313_blp_13-2 Sheet No. 35 Side A
Alternatively, there may be one or more reasonably high quality
studies showing an association that other studies do not confirm.
The third category is used to identify those health outcomes for
which the literature provides inadequate or insufficient evidence
from which to determine whether an association exists.19 This
category is used when there are very few studies, none of which is
definitive, or where there are many studies, but the quality is
inadequate (the studies might have failed to control confounding)
or the findings are inconsistent. Finally, the last category is used to
designate those health outcomes for which the extant research
provides limited or suggestive evidence of no association.20 This
category is used when there are numerous studies of reasonably
high quality, and they consistently show no association between
the exposure and the outcome.
The above categories were applied to the first question with
which Congress charged the Committee. With regard to the second
question, the paucity of data on those who served in Vietnam
precluded, for the most part, making a determination about the
magnitude of increased risk. First, the inability to assign exposures
to individual veterans, including the vast majority of those who
participated in the research studies that were conducted, was seen
as an enormous obstacle. When an agent induces a response, it is
recognized that the magnitude of the response, or the likelihood of
developing a disease, tends to increase as the exposure gets larger.
This phenomenon is referred to as “dose-response.” The
Committee concluded that, without knowledge of the exposure
level, the size of the risk could not be quantified. Even if an
average exposure level were known, it would still be difficult to
estimate an average risk because the existing research, whether in
veterans, exposed workers, or accidentally-exposed populations,
usually could not establish, that is, did not quantify, the doseresponse relationship. Given the lack of information about how
steeply the risk for each of the health outcomes evaluated would be
expected to rise, the Committee concluded that it was unable to
answer the second question regarding the “increased risk of each
20313_blp_13-2 Sheet No. 35 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
562
21
C M
Y K
06/23/2005 10:19:08
The term “risk” will be further defined later in this article but, for now, it
shall suffice to say that, in the judgment of the Committee, the number of cases
of each disease among Vietnam veterans due to herbicide exposure could not be
estimated with any reasonable accuracy.
20313_blp_13-2 Sheet No. 35 Side B
disease among those exposed” with any specificity.21
The responses to the third question—whether a biologically
plausible mechanism or other evidence supporting a causal
association existed—expanded the work of the Committee beyond
epidemiology and engaged the Committee in the review of a broad
spectrum of studies. The premise of this question is that any
determination about a causal relationship is strengthened when the
mechanism of action is understood. For example, researchers
might find that blood pressure in persons who have taken a certain
drug is generally lower than in those who have not. If in addition
researchers determine that this drug relaxes the smooth muscles in
the arterial walls known to be inversely related to blood pressure
(muscle contraction increases pressure, but relaxation reduces it),
then the plausibility of the drug having a causal effect (in this case
a protective one) is enhanced. The data used in evaluating biologic
plausibility may derive from a wide range of scientific fields,
including toxicokinetics, which examines how and at what rate
compounds are absorbed into the body, distributed to different
organs, chemically metabolized, and excreted; whole animal
toxicology, which addresses the pathologic and homeostatic
responses of the organism, often in rodents but also in other
species, including humans; and molecular and cellular biology,
which seeks to understand the biochemical alterations that result
from an exposure and the subsequent consequences for cell
functioning. Thus, biologic evidence from experimental studies in
humans, other animals, and test systems such as cell cultures is
used to determine whether a plausible mechanism exists. Such
evidence is considered to provide support for inferring causation
when statistical associations have been observed in human studies.
In evaluating the evidence pertaining to its congressional
mandate, the Committee made decisions regarding the relationship
between Agent Orange exposure and dozens of health outcomes.
Four of these will be reviewed in detail: (a) non-Hodgkin’s
lymphoma, (b) Type II diabetes, (c) prostate cancer, and (d) the
20313_blp_13-2 Sheet No. 36 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
563
presumptive period for respiratory cancer. The first three are health
outcomes, while the final one is an issue of timing and causation.
The Committee concluded that the evidence for an association with
non-Hodgkin’s lymphoma was sufficient and that biologic
plausibility was established. For Type II diabetes and for prostate
cancer, the Committee concluded that there was limited but
suggestive evidence for an association. The VA had ruled that
respiratory cancer could be considered service-related only if it
manifested within a period of thirty years after the end of service in
Vietnam, termed the “presumptive period.” Charged with
determining whether this presumptive period had a scientific basis,
the Committee concluded that it did not. To provide the reader
with a background for understanding how the Committee reached
each of these conclusions, this article now turns to an exposition of
key concepts in scientific and epidemiologic research.
II. CAUSAL EFFECTS
20313_blp_13-2 Sheet No. 36 Side A
The concept of causation is fundamental to scientific inquiry,
which seeks to understand cause and effect relationships of
physical, chemical, or biological phenomena. Within biomedical
sciences, potential causes of disease and developmental disorders
are studied using a variety of tools, including epidemiology and
toxicology. In these fields, however, the concept of “cause” differs
from that which courts use in settling individual or even class
action cases.
To study the causal effects of an exposure, two identical
“units” must be compared, one exposed and one unexposed. A unit
might be, among other things, a person, a laboratory animal, a cell,
or a piece of tissue. In order for the study to produce results about
causal effects, it is essential that the two studied units be absolutely
identical, which is to say that they differ only with regard to the
exposure. Each unit is evaluated for some response, such as
growth, chemical or electrical activity, or structural or functional
change. The difference in response between the exposed unit and
the unexposed unit represents the “causal effect.”
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 36 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
564
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
Studies can be classified in many ways, but one significant
distinction is between experimental and observational studies.
Table 1 compares these two types of studies:
Table 1
Experiment
Observational Study
Identical units, such as a single
strain of laboratory mice
Units not identical
Scientist manipulates exposure
and determines which units are
or are not exposed
Exposure occurs beyond the
control of the scientist
Scientist determines (measures)
exposure levels
Scientist measures exposure
(measurements may be subject
to greater error than in
experiment)
Outcome is measured:
reliability will vary with the
nature of the outcome and the
quality of the protocol for its
measurement
Outcome is measured:
reliability will vary with the
nature of the outcome and the
quality of the protocol for its
measurement
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 36 Side B
Notably, it is easier to ensure the use of identical “units” in an
experimental study than in an observational study. However, the
more fundamental difference is that, in an experimental study, the
exposure is controlled by the investigator; that is, the investigator
decides which unit will receive the exposure and which will not.
Typically, this decision is made randomly and either of the units
could be the chance recipient of the exposure. The investigator also
determines the level of exposure each unit receives and may assign
the units to different amounts or intensities of exposure. In an
observational study, by contrast, exposure is not assigned, but
rather, occurs for reasons that have nothing to do with any
20313_blp_13-2 Sheet No. 37 Side A
06/23/2005 10:19:08
HERTZ MACROED CORRECTED.DOC
HEALTH EFFECTS OF AGENT ORANGE
6/6/2005 1:06 PM
565
22
C M
Y K
06/23/2005 10:19:08
It should be noted that even in double-blind, placebo-controlled trials,
inferring general causation can be problematic, due largely to the fact that
persons who participate in these studies and who are compliant with the
treatment regimes are often a select group. Additionally, despite randomization,
the exposed and unexposed may differ in unmeasured ways, by chance,
particularly in small trials. Randomization reduces the likelihood of confounding
but does not eliminate it. See infra Part III.
23
See Donald B. Rubin & Roderick J. Little, Causal Effects in Clinical and
Epidemiological Studies Via Potential Outcomes: Concepts and Analytical
Approaches, 21 ANN. REV. PUBLIC HEALTH 121 (2000).
24
A response might be a continuous measurement such as blood pressure,
or a binary outcome, such as the presence or absence of disease.
20313_blp_13-2 Sheet No. 37 Side A
researcher’s actions. Historical, social, political, and physical
forces, as well as individual choices that may also be shaped by
any of the above factors, will determine when and where exposure
occurs. As a result, the exposed unit is rarely identical to the
unexposed unit. For this reason, the concept of causation in
observational studies has been more elusive than in experimental
studies.22
Recently however, a conceptual paradigm has been developed
that aids in the understanding of the conditions under which causal
inferences can be made from observational investigations.23 The
underlying concept is the “counterfactual” that contrasts two
scenarios. Under the first scenario, the individual unit (usually a
person, but also possibly a non-laboratory animal) is exposed and
its response is measured.24 Under scenario two, we suppose that
the individual is not exposed and, therefore, we can measure the
response that would have occurred had the individual, counter to
fact, not been exposed. We call this the counterfactual response.
Thus, the individual causal effect in an observational setting is the
difference between the actual and the counterfactual response.
Unfortunately, the individual causal effect can never be known
since researchers can never observe both the factual and
counterfactual experience. Epidemiologists, however, strive to
measure the group-level causal effect, which represents a type of
average of the individual-level effects, under the assumption that
the two groups (exposed and unexposed) each represent the
counterfactual experience, on average, of the other. In order to do
so, epidemiologists must first define the following terms: risk (R),
20313_blp_13-2 Sheet No. 37 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
566
risk factor (RF), risk difference (RD), and risk ratio (RR). Risk is
defined as the probability of disease, while risk factor refers to an
exposure or characteristic that increases risk or serves as a
surrogate for a factor that increases risk. Risk difference is
calculated by subtracting the “risk if exposed” value from the “risk
if unexposed” value. Finally, the risk ratio, also known as relative
risk, is defined as “risk if exposed” divided by “risk if unexposed.”
The importance of the counterfactual assumption cannot be
overemphasized. In any study, it is possible to make
measurements. In many studies, the risk difference or risk ratio can
be measured. However, defining these terms or measuring them
does not in itself make them meaningful in terms of causation.
Epidemiologists and other scientists often say that, “association
does not necessarily imply causation.” One can gain additional
insight into the source of various conditions and diseases through
the use of the sufficient causes model. Figure 1 provides an
example of the application of this paradigm:
Figure 1
Q
P
R
P
U
S
A: Single cause is
sufficient (rare
situation)
B: Two causes are
required and sufficient
T
C: Multiple causes are
required (common
situation)
20313_blp_13-2 Sheet No. 37 Side B
V
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 38 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
567
25
C M
Y K
06/23/2005 10:19:08
MODERN EPIDEMIOLOGY 8-12 (Kenneth J. Rothman & Sander Greenland
eds., 1998).
26
Sarah J. Spence, The Genetics of Autism, 11 SEMINAR PEDIATRIC
NEUROLOGY 196, 198 (2004).
27
Irva Hertz-Picciotto et al., Synergism Between Occupational Arsenic
Exposure and Smoking in Lung Cancer Induction, 3 EPIDEMIOLOGY 23, 28
(1992).
20313_blp_13-2 Sheet No. 38 Side A
Each circle, or pie, represents a set of sufficient causes.25 In
circle A, a single cause will result in the disease. This cause might
be, for example, the measles virus. The virus alone causes the
clinical entity we call measles. In circle B, a second factor is
needed; this example might apply if not all individuals exposed to
the virus actually exhibited the clinical symptoms of the disease,
that is, if some individuals lacked susceptibility to the virus. Thus,
in circle B, P is the virus, but R is also necessary because neither P
alone nor R alone results in the disease. In circle C, five factors are
necessary to cause disease. This type of scenario corresponds to
most chronic, or non-infectious, diseases, for which multiple
factors are likely to play a role in any individual case. By way of
example, it has been suggested that there may be more than ten
genes for autism.26 However, each child with autism probably does
not require all such genes to develop this disorder, and there may
be several environmental factors also involved. Note also that for
any given disease there may be several different sets of sufficient
causes; some individuals will require one set and others will
require a different, though possibly overlapping, set. For example,
among workers who smoke and are exposed to arsenic at their
workplace, some might develop respiratory cancer from the
cigarette smoke alone, while others might develop cancer from the
arsenic alone, and still others might develop respiratory cancer
only because they received both exposures.27
The sufficient causes model is also instructive in terms of
inferring individual causes. Knowledge about the presence or
absence of other known risk factors changes the probability that a
suspect risk factor was causal for an individual case. If an
individual who has never smoked, whose parents never smoked,
and who is not married to a smoker develops lung cancer, the
probability that this cancer was caused by some other known lung
20313_blp_13-2 Sheet No. 38 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
568
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
carcinogen is increased. If, for instance, that individual is known to
have high exposure to radon, the likelihood that the radon caused
the cancer is higher than it would be for another individual with the
same high exposure to radon who smoked or was exposed
passively to tobacco smoke.
III. STUDIES IN GROUPS: ESTIMATION AND PRECISION
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 38 Side B
Researchers prefer to enroll groups for their studies rather than
rely on individuals, primarily because individuals almost never
provide definitive evidence about causal effects. In recognition of
this limitation, measurements are made on an enrolled group with
the idea that the results can be extrapolated to the population from
which the group arose, and hence, to other individuals who were
not participants in the study. The group that is studied is termed a
“sample,” and any measure on the sample is considered an
“estimate” of the parameter (risk ratio, for instance) for the
complete population.
For example, researchers concerned that adolescents with
symptoms of depression may engage in binge drinking of alcohol
might sample a group of high school students. In the sample,
researchers may determine what proportion of tenth and eleventh
graders attending one high school selected at random from all high
schools in a metropolitan school district exhibit depressive
symptoms (perhaps by use of a questionnaire). This result provides
an estimate of the true proportion of high school students with
depressive symptoms in that school district and perhaps in that
metropolitan area, that state, the country, or all similar countries.
If researchers also found out how many of those high school
students engaged in binge drinking, they could estimate the risk
ratio for binge drinking by comparing those with depressive
symptoms to those without such symptoms. The resulting risk ratio
would be an estimate of the risk ratio in the population. If the risk
ratio were 1.5, it would mean that high school students in the study
who had depressive symptoms were one and one-half times more
likely to engage in binge drinking than those who did not. If the
risk ratio were 1.0, it would mean that each group of high school
students had the same risk of engaging in binge drinking.
20313_blp_13-2 Sheet No. 39 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
569
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 39 Side A
However, epidemiologists recognize that the sample studied
might be different from or unrepresentative of the complete
population and thus they also construct a range around this
estimate. This range is known as a confidence interval and
represents a range of values that, on average and under certain
conditions, is expected to include the true population value. The
width of this interval is roughly a function of the study size, a good
indication of a more technical quantity known as statistical power.
In a study in which less than half of the population is exposed and
disease is not common, this power is mostly determined by the
number of exposed persons with disease. As the number of
exposed persons who develop disease increases, the confidence in
the estimate increases and the interval becomes tighter around the
estimate. For example, a small study with an RR of 1.5 might have
a 95% confidence interval of 0.5 to 4.5, in which case we would
say that the precision is low. A much larger study that also had an
RR of 1.5 might have a 95% confidence interval of 1.3 to 1.7,
indicating very high precision.
The above exposition emphasizes the estimation of effects. The
precision of these estimates (reflected in confidence intervals) is
related to another concept used by scientists and invoked in recent
court decisions regarding admissibility of scientific evidence,
namely “statistical hypothesis testing.” A common practice in
many scientific fields is to construct a “null hypothesis,” which
states that there is no association between the exposure and the
outcome. Once the study has been conducted, the result is
compared with the null hypothesis. If the study result is extremely
different from what is predicted by the “null hypothesis,” then,
assuming the data are reliable, one may conclude that the null
hypothesis is not supported because if it were true, then large
deviations from the null would be improbable. To quantify the
improbability of the result, one calculates its probability of
occurring under the dual assumptions of no association and
complete absence of any other information. The resulting
probability is called a “p-value.” It is sometimes referred to as an
“error” rate.
The merits and misuses of p-values have been the subject of
20313_blp_13-2 Sheet No. 39 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
570
28
C M
Y K
06/23/2005 10:19:08
One website lists “326 Articles/Books Questioning the Indiscriminate
Use of Statistical Hypothesis Tests in Observational Studies.” See http://www.
cnr.colostate.edu/~anderson/thompson1.html.
29
Note that if the 95% confidence interval includes the null value (0 for a
risk difference, 1 for a risk ratio), then the p-value will be greater than 0.05.
30
See infra Part V.
31
See infra Part IV.
20313_blp_13-2 Sheet No. 39 Side B
considerable debate within the scientific community.28 One
criticism of the p-value relates to the convention of using a
cutpoint of 0.05 to determine whether a finding is “significant”
(the designation when the p-value is less than 0.05, that is, when
the probability of the result is less than one in twenty if the null
hypothesis were true), and declaring all results with p-values above
0.05 as “nonsignificant.”29 The result of a statistical hypothesis test
is a decision of whether to “reject” the null hypothesis. In practice,
it is difficult to argue that results with a p-value of 0.051 are
qualitatively different from those with a p-value of 0.049. Another
problem is that the p-value combines two different aspects of the
study result: the magnitude of the association and its precision. For
instance, one study may have a p-value of 0.04 when the RR is 8.0;
this will be a less precise estimate (the confidence interval will be
wider) than another study with a p-value of 0.04 and an RR of 2.5.
To address this concern, many epidemiologists have preferred to
express their results with “estimates” and “confidence limits,”
thereby keeping these two aspects of the study findings clear and
separate.
It has also been noted that although a p-value provides
information about the consistency between the “null hypothesis”
and the data collected, it provides no information at all about any
other hypothesis. If one wanted to hypothesize that a risk is
doubled for individuals who are exposed, one would not calculate
a p-value. Similarly, if previous studies have already suggested
that the null hypothesis may not be true, then it may be of greater
interest to evaluate whether the new data are consistent with the
previously published findings rather than whether they are
consistent with a null effect. In fact, p-values do not provide the
means for placing findings in context,30 or for considering the
possibility of biases.31 Instead, they are calculated by either
20313_blp_13-2 Sheet No. 40 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
571
32
C M
Y K
06/23/2005 10:19:08
See Janet M. Lang et al., That Confounded P-Value, 9 EPIDEMIOLOGY 7,
8 (1998); The Editors, The Value of P, 12 EPIDEMIOLOGY 286 (2001).
33
I will not presume to guess what probability would correspond to the
criterion of “beyond a reasonable doubt.”
20313_blp_13-2 Sheet No. 40 Side A
assuming no other information or by explicitly ignoring it.
Overall, it is unwise to make decisions on the basis of a single
set of data, a practice that is encouraged by the use of p-values.
Science does not actually proceed in the manner implied by
statistical hypothesis testing and, rather than relying on decisions at
the end of each study, scientists gather and review the body of
evidence as a whole. It has been suggested that the practice of
hypothesis testing detracts from scientific thinking; indeed, one
journal in the field of epidemiology strongly discourages the use of
p-values to summarize results32 and frequently asks authors to
remove them as a condition of accepting a paper.
Further critiques point out that the common use of p<0.05 as a
criterion for deciding to reject the null hypothesis is based on the
implicit assumption that there is a high cost to mistakenly rejecting
the null hypothesis and thereby “finding an association.” In other
words, this convention presumes that such a conclusion should be
made only very cautiously (society cannot afford to make this
mistake more than 5% of the time). In some circumstances, this
implicit assumption may prove problematic. If the harm from an
exposure is severe, a regulatory body, for example, may wish to err
on the side of protecting public health. This, however, would
require the use of different criteria. In the courtroom, a “more
likely than not” standard is used in some circumstances as the bar
against which to evaluate evidence. In a single study (absent any
other research), a p<0.50 means the probability is less than 50%
that the data (or more extreme data) arose from a population in
which exposure and disease are not associated. Hence, a p<0.50
would be much closer to the criterion of “more likely than not” for
evaluating whether the data arose from a population in which
exposure does not cause disease.33
While epidemiologists strive to conduct studies that produce
precise estimates, there is always the possibility that the estimate
could be wrong, not because of chance “sampling error” that
occurs with small or even moderate-sized samples, but because of
20313_blp_13-2 Sheet No. 40 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
572
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
a more systematic problem known as bias. When bias is present,
not only is the estimated association incorrect, but the p-value does
not represent the purported “error rate.” As we shall see, bias is
usually a greater concern than errors due to random fluctuations
that produce these error rates in observational studies.
IV. STUDIES IN GROUPS: BIAS
Bias is present when, on average, the estimates tend to be
either too high or too low relative to the true population parameter.
Figure 2 displays how bias can distort a relative risk by creating
either artificial effects or masking true effects:
Figure 2: Upward and Downward Bias
20313_blp_13-2 Sheet No. 40 Side B
The base of the wedge at the start of each arrow is the true
value of the risk ratio, and the arrowhead is at the biased
(observed) value. In case 1, an increased risk due to a harmful
exposure will be presumed to be smaller than it truly is. In case 2,
an exposure with no effect will be presumed to reduce risk. In case
3, an exposure that is beneficial (reduces risk) will appear to be
more beneficial than it really is, whereas in case 4, a beneficial
exposure will appear as though it has a small harmful effect. In
case 5, an exposure that has no effect will be presumed to be
harmful and, in case 6, a slightly harmful exposure will be
presumed to be more harmful than it is.
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 41 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
HEALTH EFFECTS OF AGENT ORANGE
2.0
1
2
3
4
573
5
6
1.5
No effect
1.0
0.5
Cases 1-3 show “downward” bias, and
cases 4-6 show “upward” bias.
20313_blp_13-2 Sheet No. 41 Side A
The main types of epidemiologic bias are selection bias,
information bias, confounding bias, and statistical bias. Selection
bias occurs when the subjects in the study sample do not represent
the targeted population with regard to the exposure and the disease.
Consider Figure 3, the epidemiologic two-by-two table in which
each individual falls into one of four cells: exposed with disease,
exposed without disease, unexposed with disease, or unexposed
without disease. A completely representative sample will take
approximately the same proportion from the population out of each
of the four cells. (This means that if 90% of the population is in the
cell for unexposed without disease, then 90% of the sample also
would be from that cell.) It is also possible to intentionally sample
at a different rate from one column or one row and still obtain an
unbiased estimate, but only if the investigator ensures that both
cells in that row or column are sampled in the same proportion.
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 41 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
574
Figure 3: The Two-by-Two Table
Exposed:
Unexposed:
Diseased:
Exposed with
disease
Unexposed with
disease
Total diseased
Not diseased:
Exposed without
disease
Unexposed without
disease
Total without
disease
Total exposed
Total unexposed
Grand Total
34
C M
Y K
06/23/2005 10:19:08
This point was made in a commentary by Irva Hertz-Picciotto &
Penelope P. Howards, Invited Commentary: Hot Tubs and Miscarriage:
Methodological & Substantive Reasons Why the Case Is Weak, 158 AM. J.
EPIDEMIOLOGY 938 (2003) (critiquing Li De-Kun et al., Hot Tub Use during
Pregnancy and the Risk of Miscarriage, 158 AM. J. EPIDEMIOLOGY 931 (2003)).
35
See case 6 in Figure 2.
20313_blp_13-2 Sheet No. 41 Side B
More concretely, suppose that a study is conducted to examine
the hypothesis that the use of hot tubs by pregnant women
increases the risk of spontaneous abortion. Suppose further that
women who use hot tubs are more likely to participate in the study
because they have more leisure time and that women who have
spontaneous abortions are also more likely to participate because
they are concerned about why they lost their pregnancies. In
essence, a larger percentage of the population in the upper left cell
of the epidemiologic two-by-two table participated in the study
than the population percentage in the other cells. In other words,
proportionately fewer women who did not use a hot tub or who did
not spontaneously abort would participate in the study.34 This
would lead to an upward bias in the estimated RR. Thus, if, for the
sake of argument, the true risk ratio for spontaneous abortion from
hot tub use were 1.2, in this study we might see an estimated risk
ratio of 1.5.35 If the true risk ratio were 1.0, we might, for example,
20313_blp_13-2 Sheet No. 42 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
575
36
38
C M
Y K
See case 5 in Figure 2.
See case 1 in Figure 2.
See case 2 in Figure 2.
06/23/2005 10:19:08
37
20313_blp_13-2 Sheet No. 42 Side A
observe a risk ratio of 1.2 or greater.36
In other examples, the bias might occur in the opposite
direction. For instance, researchers who studied high fat diets and
diabetes might find that persons eating high fat diets and diabetics
would be less likely to participate. In this case, the upper left cell
in Figure 3 would be underrepresented as compared to the
population at large. Therefore, if the true RR were 1.8, then one
might observe an RR of, say, 1.2;37 alternately, if the true RR were
1.0, one might observe a lower RR of, for instance, 0.7.38 In the
former case, researchers might incorrectly conclude that there is
only a small detrimental effect of the high fat diet when it is quite
harmful, and, in the latter case, the study incorrectly suggests a
protective effect, that is, a lower risk of diabetes among those who
eat high fat diets. In short, selection bias can lead one to draw the
wrong conclusion.
Information bias, by comparison, occurs when information
about the disease diagnosis differs between those who are exposed
and those who are unexposed. For example, bias might result
where individuals of low socioeconomic status who do participate
are more likely to be exposed than those at higher socioeconomic
levels, but less likely to be diagnosed because they lack health
insurance and rarely see a physician. Thus, persons with the
disease may be misclassified as healthy because they are not yet
diagnosed. Information bias also might occur when data on
exposure differs with regard to those who have the disease and
those who do not. For example, in a study of the possible
connection between pesticide use around the home and incidence
of childhood leukemia, parents may be asked to recall what
pesticide products they used and when. The parents of affected
children might be more likely to recall every insecticide or
fungicide used in or around their house than the parents of healthy
children. This would result in a specific type of information bias
termed reporting bias or recall bias, which usually results in
upward bias.
20313_blp_13-2 Sheet No. 42 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
576
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 42 Side B
Another type of bias, confounding bias, occurs when an
alternative risk factor for the disease (one that is not the exposure
of interest for the study) happens to occur more or less frequently
in the exposed as compared with the unexposed. Consider, for
example, a study to examine the hypothesis that exposure to
polychlorinated biphenyls (PCBs) during infancy adversely affects
the cognitive development of children. Suppose that a major source
of PCBs to infants is breast milk. Then suppose that mothers who
breastfeed their infants are more educated and are more likely to
read to their children or offer other intellectual stimulation. Notice
that an experiment to test the hypothesis that PCBs adversely affect
cognitive development would randomly assign some mothers to
breastfeed and others to give formula. However, in the real world,
women who choose to breastfeed are not the same as those who do
not elect to breastfeed and hence cannot serve as the
“counterfactual” experience for those who do not breastfeed. The
result is confounding bias: children with a higher exposure to
PCBs were given greater intellectual stimulation. In this example,
the RR would be biased downward, but it is possible that in other
examples the RR could be biased upward.
A fourth type of bias is statistical bias. Statistical bias occurs as
a result of errors in statistical analysis or limitations in data.
Sometimes the methods used for analysis do not match the
conditions in which the data were collected or the variables as
defined by the investigator; hence, bias results. In other instances,
the adjustment for confounders is done incorrectly, and bias is
introduced inadvertently. Thus, to avoid statistical bias,
epidemiologists, in addition to having an intimate understanding of
the subject they are studying, must be knowledgeable about both
statistical methods and proper confounder selection strategies.
It is important to keep in mind that all of these types of bias
could be in either the positive (upward) or negative (downward)
direction. However, one cannot dismiss the results of a study
simply because there is a possibility of bias or confounding.
Frequently, one can glean information that bears on the direction
of bias. For example, if the factors tending toward downward bias
are stronger than those that would magnify the association between
exposure and disease, one would expect the true relationship to be
20313_blp_13-2 Sheet No. 43 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
577
V. REACHING CONSENSUS
In practice, epidemiologic studies are never perfect, and even
the best studies only approximately meet the necessary conditions
for risk ratios or risk differences to be interpreted as causal effects.
For this reason, it is nearly always true that causation cannot be
40
C M
Y K
This ensures a low probability of selection bias.
This acts to reduce information bias.
06/23/2005 10:19:08
39
20313_blp_13-2 Sheet No. 43 Side A
stronger than the one observed in the study.
This discussion about bias and precision can now be used to
answer the problem of how to measure causal effects in groups.
The key is that when certain conditions or requirements are met,
the association between exposure and disease may be interpreted
as a causal one or at least one can conclude that such an inference
probably does not stray far from the truth. These conditions and
requirements may be satisfied when (a) study subjects have been
properly sampled and recruited;39 (b) exposures and disease have
been measured or diagnosed accurately;40 (c) confounder data are
complete and adequately measured; and (d) the appropriate
multivariable statistical techniques have been used to analyze the
data. Under these conditions, once all confounders have been
accounted for, the unexposed group provides a good representation
of the counterfactual experience of the exposed group and the
analysis properly compares the group responses.
In other words, as long as the quality of data is reliable and the
analysis is statistically correct and appropriately takes account of
confounders, then the two groups (exposed and unexposed) can be
validly compared. In this scenario, one can infer that the study RD
or RR will be a measure of the causal effect of exposure. Of
course, this measured causal effect may or may not be a precise
estimate, as that will depend on whether the study has an adequatesized sample. The quality of an individual study, therefore,
depends on there being (1) minimum bias, which is achieved
through careful design, sound methods of data collection and
measurement of exposures and disease, and appropriate statistical
treatment of the data; and (2) an adequate-sized study sample.
20313_blp_13-2 Sheet No. 43 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
578
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 43 Side B
inferred from a single study, but rather, must be examined in a
multitude of studies. The problem is less acute when experimental
(randomized) studies are possible, such as for the evaluation of
drugs that are believed to impart a benefit to those taking them. In
this situation, the evaluation of evidence is more straightforward
than it would be for exposures for which it would be unethical to
conduct such research (such as cigarette smoking or asbestos
exposure). It is these latter, allegedly harmful exposures that have
generated discussion about how to infer causation. This discussion
has focused on how epidemiologists should evaluate a body of
evidence from multiple studies, including human epidemiologic
investigations, experimental data from whole animals, and
mechanistic research in which cells or tissues are manipulated to
understand physiologic or biochemical processes believed to be
related to pathogenesis in the human body.
As the body of evidence grows and new hypotheses are
proposed, the research community begins the process of reaching
consensus regarding which studies and ideas it finds convincing.
Arriving at a consensus can take months, years, or decades. For
example, consensus regarding the role of the human
immunodeficiency virus (HIV) in AIDS took a relatively short
time, whereas the environmental contribution to breast cancer still
remains contentious.
Consensus does not require and is not synonymous with
unanimity. Even today it is possible to find some who are
unconvinced about the relationship between HIV and AIDS or
between smoking and lung cancer. That being said, the reaching of
consensus often follows a typical pattern, in which evidence
accrues and scientific opinion shifts. For example, consider a study
that finds a previously unstudied association in which exposure E
is related to an increased risk of disease D. To receive attention,
the study often would have observed a strong association.
Frequently, these first findings are based on a small sample size.
Some scientists may reject these findings because they object to
the study’s methodology. Other researchers will then attempt to
replicate the finding using improved methodology and maybe
larger study samples, but it is possible that only some of the studies
will confirm the original result. Over time, the weight of the
20313_blp_13-2 Sheet No. 44 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
579
Austin Bradford Hill, The Environment and Disease: Association or
Causation?, 58 PROC. ROYAL SOC’Y MED. 295 (1965).
C M
Y K
06/23/2005 10:19:08
41
20313_blp_13-2 Sheet No. 44 Side A
evidence will tend to fall on one side or the other. At some point, a
meta-analysis or “quantitative review,” which is a combined
analysis of multiple studies, will be conducted. For this, it is
preferable to use high quality studies as this type of analysis is
more effective in addressing the precision of results than the
biases. Meanwhile, toxicologic or other basic science studies may
or may not demonstrate a plausible mechanism. Thus, the
consensus will build either in support or in contradiction of a
causal effect.
Although ideally scientists will evaluate evidence in a valuefree context, it is increasingly recognized that it is impossible for
scientists to be totally “objective” because individuals are
unavoidably influenced by their particular cultures and personal
experiences. Studies have documented how these experiences
influence the way in which individuals assess scientific studies and
place greater weight on certain studies or lines of evidence as
compared with others. It should be noted that the IOM, in
assembling its committees, consciously seeks to achieve not only
diversity of fields of expertise, but also “balance” among possible
biases on its committees.
Although criteria have existed for inferences about microbial
causes of infections for more than a century, the discussion about
causal inference for chronic diseases is more recent. In the 1960s,
the debate as to whether cigarette smoking causes lung cancer
provided the impetus for the development of a specific set of
guidelines for inferring causality. These were summarized by Sir
Bradford Hill41 and include the following primary considerations:
1. Temporality: A cause must precede an effect.
2. Strength of Association: A high RR or RD provides
greater weight than a low one.
3. Coherence: Evidence from other fields should support,
not contradict, the causal hypothesis.
4. Biologic Plausibility: Known biologic facts should
support, not contradict, the proposed causal effect.
20313_blp_13-2 Sheet No. 44 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
580
5. Consistency: Multiple studies using different designs
and/or different populations should confirm the finding.
6. Dose-response: The greater the exposure, the greater
should be the likelihood of a response.
7. Specificity: The outcome should be less frequent in the
absence of exposure or after removal of the exposure.
These considerations are not formal criteria and Hill himself
cautioned against using them as such, although such misuse is
often found in the scientific literature.42 Moreover, it can be shown
that failure to observe several of these facets of an association does
not necessarily detract from the conclusion of causality. In fact, it
has been argued that only temporality is truly required.43
VI. THE IOM COMMITTEE AND ITS EVALUATION OF SPECIFIC
OUTCOMES FOR VIETNAM VETERANS
42
C M
Y K
06/23/2005 10:19:08
See, e.g., Carl V. Phillips & Karen J. Goodman, The Missed Lessons of
Sir Austin Bradford Hill, 1 EPIDEMIOLOGIC PERSPECTIVES & INNOVATIONS 3
(2004), available at http://www.epi-perspectives.com/content/pdf/1742-5573-13.pdf. These considerations have now spilled over into the courts as well. See
Joe G. Hollingsworth & Eric G. Lasker, The Case Against Differential
Diagnosis: Daubert, Medical Causation Testimony, and the Scientific Method,
37 J. HEALTH L. 85 (2004).
43
Mervyn Susser, Falsification, Verification, and Causal Inference in
Epidemiology: Reconsiderations in the Light of Sir Karl Popper’s Philosophy,
in CAUSAL INFERENCE (Kenneth J. Rothman ed., 1988).
44
See VAO 1994, supra note 5.
20313_blp_13-2 Sheet No. 44 Side B
Applying the epidemiological principles discussed earlier in
this article, this section reviews the evidence and reasoning behind
the decisions reached by the Committee with regard to nonHodgkin’s lymphoma, Type II diabetes, prostate cancer, and the
presumptive period for respiratory cancer.
In the case of non-Hodgkin’s lymphoma, the first Committee,
which completed its review in 1994, concluded that the evidence
was sufficient regarding an association with herbicides or their
contaminants.44 A sizable number of studies in occupational
cohorts had been conducted, and although many showed either no
association (RR=1.0) or very slight associations, quite a few
20313_blp_13-2 Sheet No. 45 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
6/12/2005 2:54 PM
581
studies showed an elevated risk of non-Hodgkin’s lymphoma.
These included studies of Swedish workers who were exposed to
phenoxy herbicides;45 forest conservationists who worked for the
U.S. Department of Agriculture (RR=2.5, 95% CI=1.0 to 6.3);46
farmers in Kansas who had used herbicides for more than twenty
days per year (RR=6.0, 95% CI=1.9 to 19.5);47 Canadian farmers
who applied pesticides to more than 250 acres (RR=2.2, 95%
CI=1.0 to 4.6);48 Washington State forestry herbicide appliers
(RR=4.8, 95% CI=1.2 to 19.4);49 and Italian farmers licensed to
use pesticides (RR=1.8, 95% CI=1.2 to 2.5).50 In addition to these
studies of occupational exposures, non-Hodgkin’s lymphoma was
increased among male residents of Italian provinces in
contaminated areas (RR=2.2, 95% CI=1.4 to 3.5),51 and in a
Finnish community in which the water supply was contaminated
with chlorophenols (RR=2.8, with a 95% CI=1.4 to 5.6).52 Also,
unlike many of the other health outcomes examined, nonHodgkin’s lymphoma was observed at a higher rate in Vietnam
45
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 45 Side A
See Bodil Persson et al., Malignant Lymphomas and Occupational
Exposures, 46 BR. J. IND. MED. 516 (1989); Lennart Hardell, Malignant
Lymphoma and Exposure to Chemical Substances, in Particular Organic
Solvents, Chlorphenol and Phenoxyacetates, 77 LAKARTIDNINGEN 208 (1980).
46
See Michael C. Alavanja et al., Mortality Among Forest and Soil
Conservationists, 44 ARCHIVES ENVTL. HEALTH 94 (1989).
47
See Shelia Hoar et al., Agricultural Herbicide Use and Risk of
Lymphoma and Soft-Tissue Sarcoma, 256 JAMA 1141 (1986), erratum, 256
JAMA 3351 (1986).
48
See Donald T. Wigle et al., Mortality Study of Canadian Male Farm
Operators: Non-Hodgkin’s Lymphoma Mortality and Agricultural Practices in
Saskatchewan, 82 J. NAT’L CANCER INST. 575, 579 Tbl.7 (1990).
49
See James S. Woods & L. Polissar, Non-Hodgkin’s Lymphoma among
Phenoxy Herbicide-Exposed Farmworkers in Western Washington State, 18
CHEMOSPHERE 401 (1987).
50
See G. Corrao et al., Cancer Risk in a Cohort of Licensed Pesticide
Users, 15 SCANDINAVIAN J. WORK, ENV’T & HEALTH 203 (1989).
51
Paolo Vineis et al., Incidence Rates of Lymphomas and Soft-Tissue
Sarcomas and Environmental Measurements of Phenoxy Herbicides, 83 J.
NAT’L CANCER INST. 362 (1991).
52
P. Lampi et al., Cancer Incidence Following Chlorophenol Exposure in a
Community in Southern Finland, 47 ARCHIVES ENVTL. HEALTH 167, 171 Tbl.5
(1992).
20313_blp_13-2 Sheet No. 45 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
582
veterans than in the general population. As of the review
conducted by the first IOM Committee, an excess of nonHodgkin’s lymphoma cases had been observed in several studies
of U.S. Navy personnel (RR =2.2, 95% CI=1.2 to 3.9),53 or Marine
personnel (RR=2.1, 95% CI=1.2 to 3.854 and RR=3.2 95% CI =1.4
to 7.455).
In total, more than two dozen studies showed some indication
of excess mortality or incidence from non-Hodgkin’s lymphoma.
Not all of these studies were of the highest quality and there were
some studies that showed no excess risk, that is, no significant
departures from the expected level of risk. Although many of the
studies cited above adjusted for potential confounders, such
variables could have created the appearance of an association
(increased the estimated RR) or could have obscured an
association (reduced the estimated RR). In some of the studies, the
definition of exposure was extremely broad and probably included
a high proportion of individuals who were not exposed to any of
the herbicides that were used in Vietnam, resulting in “information
bias.” In such circumstances, it would be easy to underestimate the
effect of an exposure. In light of what might be an expected
“downward” bias, the replication across quite a number of
investigations that had an adequate sample size was impressive.
Neither the Seveso cohort56 nor the chemical production
workers57 experienced increased risks for non-Hodgkin’s
C M
Y K
06/23/2005 10:19:08
The Selected Cancers Cooperative Study Group, The Association of
Selected Cancers with Service in the U.S. Military in Vietnam III, Centers for
Disease Control, Hodgkin’s Disease, Nasal Cancer, Nasopharyngeal Cancer,
and Primary Liver Cancer, 150 ARCHIVES INTERNAL MED. 2495 (1990).
54
Patricia Breslin et al., Proportionate Mortality Study of U.S. Army and
U.S. Marine Corps Veterans of the Vietnam War, 30 J. OCCUPATIONAL MED.
412, 416 Tbl.6 (1988).
55
PATRICIA BRESLIN ET AL., VETERAN’S ADMINISTRATION, NONHODGKIN’S LYMPHOMA AMONG VIETNAM VETERANS (1987).
56
See Pier Alberto Bertazzi et al., Ten-Year Mortality Study of the
Population Involved in the Seveso Incident in 1976, 129 AM. J. EPIDEMIOLOGY
1187 (1989); Angela C. Pesatori et al., Cancer Morbidity in the Seveso Area,
1976-1986, 25 CHEMOSPHERE 209 (1992).
57
Fingerhut et al., supra note 12, at 216; Andreas Zober et al., Thirty-FourYear Mortality Follow-Up of BASF Employees Exposed to 2,3,7,8-TCDD after
20313_blp_13-2 Sheet No. 45 Side B
53
20313_blp_13-2 Sheet No. 46 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
583
lymphoma. As these groups were most heavily exposed to 2,3,7,8TCDD, with little or no exposure to the herbicides in Agent
Orange, the epidemiologic data tended to suggest that the
associations were more likely due to 2,4-D and 2,4,5-T. However,
the Committee did not attempt to make the case that these
compounds were the causal agents.
Biologic plausibility that Agent Orange was capable of
producing this type of cancer was supported by a study that
produced lymphoma in female mice after the administration of
2,3,7,8-TCDD.58 However, the Committee noted that the
herbicides contained in Agent Orange, including 2,4-D, 2,4,5-T,
picloram, and cacodylic acid, had been inadequately tested in
animals.
The conclusion of sufficient evidence drew on a set of studies
that showed fair consistency. While not all studies could
definitively exclude bias or confounding, it was unlikely that all of
the studies were biased in the same direction. Moreover, in several
investigations, the groups with the best-documented or highest
probability of exposure showed the greatest increase in risk. Later
studies confirmed the findings of excess risk for non-Hodgkin’s
lymphoma in yet other populations.59
Type II diabetes and prostate cancer are both characterized as
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 46 Side A
the 1953 Accident, 62 INST. ARCH. OCCUPATIONAL ENVT’L HEALTH 139 (1990);
Alfred Manz et al., Cancer Mortality among Workers in Chemical Plant
Contaminated with Dioxin, 338 LANCET 959 (1991).
58
See James Huff et al., Long-Term Carcinogenesis Studies on 2,3,7,8tetrachlorodibenzo-p-dioxin and Hexachlorodibenzo-p-dioxins, 7 CELL
BIOLOGY AND TOXICOLOGY 67 (1991).
59
See, e.g., COMMITTEE TO REVIEW THE HEALTH EFFECTS IN VIETNAM
VETERANS OF EXPOSURE TO HERBICIDES, INSTITUTE OF MEDICINE OF THE
NATIONAL ACADEMIES, VETERANS AND AGENT ORANGE: UPDATE 1996 (1996)
[hereinafter VAO UPDATE 1996]; COMMITTEE TO REVIEW THE HEALTH EFFECTS
IN VIETNAM VETERANS OF EXPOSURE TO HERBICIDES, INSTITUTE OF MEDICINE
OF THE NATIONAL ACADEMIES, VETERANS AND AGENT ORANGE: UPDATE 1998
(1999) [hereinafter VAO UPDATE 1998]; COMMITTEE TO REVIEW THE HEALTH
EFFECTS IN VIETNAM VETERANS OF EXPOSURE TO HERBICIDES, INSTITUTE OF
MEDICINE OF THE NATIONAL ACADEMIES, VETERANS AND AGENT ORANGE:
UPDATE 2000 (2001) [hereinafter VAO UPDATE 2000]; VAO UPDATE 2002,
supra note 8.
20313_blp_13-2 Sheet No. 46 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
584
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
having “limited/suggestive” evidence of an association. For Type
II diabetes, data were considered inadequate at the time the first
three Committees evaluated the evidence. (The first two
Committees considered the broader grouping of metabolic
disorders as a whole, largely because little research had been
published relating diabetes to the herbicides used in Vietnam or
their contaminants.) Nevertheless, the third Committee, which
published its findings in Update 1998, noted that a number of
reports, including one on the Ranch Hand personnel, showed
altered glucose metabolism. The Update reported, “Further
analyses and full publication of existing studies may justify a
reevaluation of this conclusion.”60 A flurry of papers appeared
between 1996 and 2000 suggesting some association with 2,3,7,8TCDD (“dioxin”).61 As a result, the fourth Committee, which
published its result in 2000, determined that the evidence was
limited, but suggestive of an association with exposures incurred in
Vietnam. Among residents exposed to dioxin because of the
industrial accident in Seveso, deaths from diabetes occurred at a
higher rate than in the reference population that was not exposed,
particularly among females.62 Excess mortality from diabetes was
60
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 46 Side B
VAO UPDATE 1998, supra note 59, at 11.
61
See, e.g., Geoffrey M. Calvert et al., Evaluation of Diabetes Mellitus,
Serum Glucose, and Thyroid Function among United States Workers Exposed to
2,3,7,8-tetrachlorodibenzo-p-dioxin, 56 OCCUPATIONAL & ENVTL. MED. 270
(1999); Gary L. Henriksen et al., Serum Dioxin and Diabetes Mellitus in
Veterans of Operation Ranch Hand, 8 EPIDEMIOLOGY 252 (1997); Angela C.
Pesatori et al., Dioxin Exposure and Non-Malignant Health Effects: A Mortality
Study, 55 OCCUPATIONAL & ENVTL. MED. 126 (1998); John Vena et al.,
Exposure to Dioxin and Nonneoplastic Mortality in the Expanded IARC
International Cohort Study of Phenoxy Herbicide and Chlorophenol Production
Workers and Sprayers, 106 ENVTL. HEALTH PERSPECTIVE 645 (Supp. 2 1998),
available at http://ehp.niehs.nih.gov/members/1998/Suppl-2/645-653vena/vena.
html; COMMONWEALTH DEP’T OF VETERANS’ AFFAIRS, MORBIDITY OF VIETNAM
VETERANS: A STUDY OF THE HEALTH OF AUSTRALIA’S VIETNAM VETERAN
COMMUNITY, VOLUME 1: MALE VIETNAM VETERANS SURVEY AND COMMUNITY
COMPARISON OUTCOMES (1998) [hereinafter COMMONWEALTH STUDY]; Morris
F. Cranmer et al., Exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is
Associated with Hyperinsulinemia and Insulin Resistance, 56 TOXICOLOGICAL
SCIENCES 431, 433 (2000).
62
See Pier A. Bertazzi et al., The Seveso Studies on Early and Long-Term
20313_blp_13-2 Sheet No. 47 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
585
observed in a multinational European cohort of chemical
production workers,63 although the excess was not statistically
significant. No excess was observed by Steenland et al., who
studied the U.S. cohort of chemical workers assembled by
NIOSH.64
Typically, Type II diabetes is not fatal and is often not listed on
a death certificate, even if one of its complications is the cause of
death. For this reason, studies of mortality from diabetes would be
limited in their ability to detect associations with exposures. By
comparison, diagnoses among the living might provide a more
complete ascertainment of cases, and hence, studies on morbidity
would be considered more definitive. In one such study, selfreports of diabetes were substantially higher than expected in
Australian veterans who served in Vietnam.65 Among Air Force
personnel who participated in the “Ranch Hand” study, glucose
abnormalities and use of oral medications for diabetes were
elevated.66 Additionally, higher blood serum concentrations of
2,3,7,8-TCDD were associated with an elevated incidence of Type
II diabetes.67 Table 2 shows the risk ratios for men whose blood
serum TCDD was in the three upper quartiles as compared with
those whose blood serum TCDD was in the lowest quartile. The
data do not show a perfect trend of increasing risk, but the upper
two quartiles seem to be at higher risk than the lower two.
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 47 Side A
Effects of Dioxin Exposure: A Review, 106 ENVTL. HEALTH PERSPECTIVE 625
(Supp. 2 1998), available at http://ehp.niehs.nih.gov/members/1998/Suppl-2/
625-633bertazzi/bertazzi.html.
63
See Vena et al., supra note 60.
64
See Kyle Steenland et al., Cancer, Heart Disease, and Diabetes in
Workers Exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin, 91 J. NAT’L CANCER
INST. 779, 785 (1999).
65
See COMMONWEALTH STUDY, supra note 60.
66
See Henriksen et al., supra note 60.
67
See Matthew P. Longnecker et al., Serum Dioxin Level in Relation to
Diabetes Mellitus among Air Force Veterans with Background Levels of
Exposure, 11 EPIDEMIOLOGY 44 (2000).
20313_blp_13-2 Sheet No. 47 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
586
Table 2: Incidence of Type II diabetes among Air Force
Ranch Hand personnel according to blood serum
concentration of dioxin, in quartiles.
1st quartile
(lowest )
2nd quartile
3rd quartile
4th quartile
(highest)
Risk ratio*
1
0.9
1.9
1.7
95% confidence
interval
--
Serum dioxin
concentration:
(0.5, 1.6)
(1.1, 3.2)
(1.0, 2.9)
*adjusted for family history, age, race, and military occupation
06/23/2005 10:19:08
68
20313_blp_13-2 Sheet No. 47 Side B
The confidence intervals (CIs) indicate that the data are
consistent with anywhere between a rather small increased risk
(RR just slightly above 1.0) and a fairly substantial one (a nearly
three-fold higher risk). This study is notable in that the designation
of diabetes was based on a clinical examination, not self-reporting.
Additionally, a study conducted among residents near a hazardous
waste site with dioxin contamination showed elevated risks for
“high” fasting insulin if their serum TCDD concentration was
elevated.68 In general, the conclusion that the data showed
limited/suggestive evidence of association was based on both the
mortality and morbidity studies, with emphasis on the latter. The
fact that some of these associations occurred in Vietnam veterans
also weighed into the Committee’s deliberations. Nevertheless,
because many of the studies relied on self-reported illness,
therefore raising the possibility of bias, the evidence fell far short
of being sufficient.
A large number of studies have addressed the risk for prostate
See Cranmer et al., supra note 60, at 431-33.
C M
Y K
20313_blp_13-2 Sheet No. 48 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
587
69
C M
Y K
06/23/2005 10:19:08
See, e.g., VAO UPDATE 2002, supra note 8.
See Howard Morrison et al., Farming and Prostate Cancer Mortality,
137 AM. J. EPIDEMIOLOGY 270 (1993).
71
Bertazzi et al., supra note 56.
72
See VAO UPDATE 2000, supra note 59; INSTITUTE OF MEDICINE,
VETERANS AND AGENT ORANGE: HERBICIDE/DIOXIN EXPOSURE AND TYPE 2
DIABETES (2000).
70
20313_blp_13-2 Sheet No. 48 Side A
cancer.69 Evaluation of this health outcome is difficult for several
reasons. First, it is very common among elderly men, and second,
most of the risk ratios are small (approximately 1.2). This is likely
to occur when an outcome has multiple causes because no single
cause is responsible for a high proportion of cases. Another factor
to consider is the question of incidence versus mortality. Mortality
is influenced by the aggressiveness of a tumor, but also by several
other factors, including the quality of care, the treatment, and the
stage at which the disease was diagnosed. In turn, these factors are
affected by such variables as access to care and a patient’s
socioeconomic status. Thus, even if an exposure increases the
incidence of prostate cancer, it may not show an association with
mortality from prostate cancer because so much can intervene to
alter survival after the occurrence of disease. Some of the early
evidence used in the Committee’s decision came from a wellconducted investigation of farmers or herbicide applicators, where
greater exposures conferred higher risk,70 and a number of
occupational cohort studies in which risk was increased, but not
significantly so. Additionally, the exposed population in Seveso
showed an increased risk of prostate cancer.71 In subsequent
reviews of the evidence, the trend continued as many studies
produced slightly elevated risk ratios while a few studies suggested
a stronger association.
The Committee has, during updates of the reports, changed the
classification of some of the health outcomes. For example, as
mentioned above, diabetes was first classified as having inadequate
evidence and then categorized as having limited or suggestive
evidence of an association at the 2000 Update and by a separate
committee convened to address this question on its own.72
Although it has not happened yet, it is possible that the Committee
could find the evidence regarding some outcome to be inadequate
20313_blp_13-2 Sheet No. 48 Side B
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
5/23/2005 7:48 PM
JOURNAL OF LAW AND POLICY
588
73
C M
Y K
06/23/2005 10:19:08
See Disease Associated with Exposure to Certain Herbicide Agents
(Multiple Myeloma and Respiratory Cancers), 59 Fed. Reg. 29723-01 (June 9,
1994) (to be codified at 38 C.F.R. pt. 3).
74
The half-life is the time it takes for the concentration to decrease to half
of what it was.
75
See Dieter Flesch-Janys et al., Elimination of Polychlorinated Dibenzo-pdioxins and Dibenzofurans in Occupationally Exposed Persons, 47 J.
TOXICOLOGY ENVTL. HEALTH 363, 377 (1996); Joel E. Michalek & Ram C.
Tripathi, Pharmacokinetics of TCDD in Veterans of Operation Ranch Hand: 15-
20313_blp_13-2 Sheet No. 48 Side B
after that disease was in the limited or suggestive category if newer
studies were conducted that tended to show no association and
were of higher quality than the earlier ones.
The fourth and final example of how the Committee has
reviewed evidence concerns respiratory cancer and the
“presumptive period.” The VA had ruled that respiratory cancer
could be considered service related only if it manifested within
thirty years following one’s service in Vietnam.73 This thirty-year
period was referred to as the “presumptive period.” The Committee
was asked to determine whether there was a scientific basis for this
presumptive period. However, based on all of the empirical
evidence from Vietnam veterans and other exposed populations,
the question simply could not be answered. The analysis of time
since the beginning of employment in exposed jobs suggested that
the elevated risk for respiratory cancer might continue for at least
the third decade. But this analysis begs the question, how long
after an exposure ends will risk continue to be increased? Most
occupational studies had not analyzed the mortality among cohorts
of workers to determine whether excess risk of respiratory cancer
changed with time since exposure ended. For the Seveso cohort, an
insufficient period of time has elapsed to evaluate the thirty-year
presumptive period (the accident occurred in 1976, fewer than
thirty years prior to this writing). Thus, given the lack of pertinent
epidemiologic data, the Committee relied on toxicokinetic data
about how the chemicals of interest are stored in the body and on
current understanding of the biology of human cancer. Dioxin is
known to have a relatively long half-life in human tissues.74 This
TCDD half-life is estimated at between seven and nine years, but
this period depends on the amount of fat in the studied individual75
20313_blp_13-2 Sheet No. 49 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
589
C M
Y K
06/23/2005 10:19:08
Year Follow-Up, 57 J. TOXICOLOGY ENVTL. HEALTH 369, 376 (1999).
76
The period between the start of a disease process and the time it is
diagnosed is termed the “latent period.”
20313_blp_13-2 Sheet No. 49 Side A
and may differ between men and women. Hence, after external
exposure ends, the compound remains in fatty tissue, circulates in
the blood, and deposits itself in various organs. At any time during
this period, disease induction can occur even though external
exposure has ceased. In addition, disease detection may occur long
after induction.76
Cancer progresses through multiple stages, beginning with
initiation, the time at which a cell’s DNA is damaged. The
damaged cell then escapes the surveillance of the body’s repair
system and the immune system, which usually hunts out damaged
cells. Other changes, known as promotion, may occur until the cell
begins to divide unchecked, resulting in proliferation. Further
stages enable the tumor to develop its own blood supply.
The point at which diagnosis occurs is determined by biologic,
social, and individual psychologic factors. Biologic determinants
will include the aggressiveness of the tumor, age of the person, and
presence of other medical conditions that might influence
immunologic competence. The social factors will include access to
care, the quality of any screening program, and the skill and
vigilance of the health provider. Individual characteristics that
influence how early in the disease process a diagnosis is made
include the propensity to seek medical care, which is highly
variable in the population and is related to the degree of trust
placed in the medical profession, and the fear of a diagnosis of
cancer.
Given the above considerations, the Committee concluded that
there was no scientific justification for a presumptive period of
thirty years for respiratory cancer. The possibility that circulating
TCDD might result in the initiation of cancer decades after a
veteran’s service in Vietnam had ended could not be excluded. A
further consideration was the uncertain length of the latency period
between the initiation of the disease process and the diagnosis.
20313_blp_13-2 Sheet No. 49 Side B
06/23/2005 10:19:08
HERTZ MACROED CORRECTED.DOC
590
6/6/2005 1:05 PM
JOURNAL OF LAW AND POLICY
SUMMARY
06/23/2005 10:19:08
C M
Y K
20313_blp_13-2 Sheet No. 49 Side B
The IOM Committees were charged with determining whether
there were associations between health outcomes and herbicides
used in Vietnam or their contaminants. The IOM Committees
addressed three questions: whether there was a statistical
association between the exposures and any health outcomes, what
magnitude of increased risk Vietnam veterans would be expected
to experience for each of the health outcomes due to herbicide
exposures incurred while in Vietnam, and whether evidence
supported the biologic plausibility of a causal association. To
answer the first question, the Committees classified the outcomes
into four categories of evidence (sufficient, limited or suggestive,
inadequate, or limited evidence of no association) and adopted an
approach that weighed the body of evidence and took into
consideration the methodologic rigor of the studies. With regard to
the second question, that of quantifying the risk to Vietnam
veterans, the Committee concluded that the increased risk could
not be identified due to the lack of adequate data quantifying the
exposures of those who served in Vietnam. To address the third
question, that is, whether a plausible biologic mechanism exists
through which the herbicides and their contaminants could cause
specific health outcomes, the Committee evaluated a wide range of
data types, including toxicologic studies in humans and
experimental animals, and research on mechanisms that use tissues
and cell cultures.
In reviewing the Committees’ findings, it is important to
remember that most non-infectious diseases are caused by multiple
factors and that to determine the effects of exposure, causality is
defined in an individual, but can only be measured in groups.
Epidemiologists therefore study groups and, for ethical reasons,
frequently rely on observational rather than experimental methods.
The quality of observational studies depends on minimizing the
four types of bias and maximizing precision by using large sample
sizes (particularly with regard to the number of exposed cases of
disease). Statistical significance is a small part of evidence, and the
use of p-values for causal inference can result in faulty
conclusions. Single studies can add to or detract from evidence for
20313_blp_13-2 Sheet No. 50 Side A
06/23/2005 10:19:08
HERTZ MACROED 051905.DOC
HEALTH EFFECTS OF AGENT ORANGE
5/23/2005 7:48 PM
591
causality, but ultimately an inference of causality depends on
replication across studies that provide precise estimates of effects
and that are relatively free of bias. Accrual of epidemiologic
evidence over time, along with experimental studies in animals and
cell or tissue cultures that establish mechanisms, generally leads
towards a consensus as to whether an exposure causes a health
outcome, although this process often takes years or longer. The
evidence about health effects of herbicides used in Vietnam and
their contaminants was slow to accumulate, partially because a
concerted effort to study the veterans longitudinally, beginning
from the time of their return to the United States, was not
undertaken, and partially because it was technologically difficult to
study dioxin, as it is present in such small quantities.
20313_blp_13-2 Sheet No. 50 Side A
06/23/2005 10:19:08
C M
Y K
BROWN MACROED CORRECTED 060605.DOC
6/20/2005 7:07 PM
THE ROLE OF SCIENCE IN DEPARTMENT
OF VETERANS AFFAIRS DISABILITY
COMPENSATION POLICIES FOR
ENVIRONMENTAL AND OCCUPATIONAL
ILLNESSES AND INJURIES
Mark Brown, Ph.D.∗
INTRODUCTION
The U.S. Department of Veterans Affairs (VA) has responded
to the healthcare and disability compensation needs of veterans of
the Vietnam War for more than three decades. Ongoing concerns
by veterans, their families, Congress and others have made longterm health effects from exposure to Agent Orange and other
herbicides used in the Vietnam War a major focus of the VA’s
response.
Part I of this article will provide an overview of the VA’s
mandate with respect to the provision of federal benefits to
veterans and their families. This section also will discuss the
requirements that veterans applying for such benefits must meet
and the difficulties veterans face when their claims are based on
latent illnesses arising from hazardous exposures that occurred
years before. Part II will describe the VA’s approach to disability
compensation for Vietnam veterans exposed to Agent Orange as
well as the VA’s more recent attempts to duplicate that approach
∗
Mark Brown, Ph.D., is the Director of the Environmental Agents Service
of the U.S. Department of Veterans Affairs. This paper was presented to the
fourth program on Science for Judges, held at Brooklyn Law School on
November 4, 2004. The author would like to acknowledge Mr. David Barrans,
U.S. Department of Veterans Affairs, for his help in reviewing this manuscript
and for explaining legal citations.
593
BROWN MACROED CORRECTED 060605.DOC
594
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
for veterans from other combat missions, including the 1991 Gulf
War. This article concludes that, despite some unanticipated
consequences, the VA’s approach to Agent Orange has worked
well for establishing scientifically based and politically acceptable
compensation policies for Vietnam veterans. Efforts to apply this
approach to other groups of combat veterans with possible
environmental and occupational injuries and illnesses, however,
have not been successful.
I. PROVIDING DISABILITY COMPENSATION FOR VETERANS
The VA is responsible for providing a wide range of federal
benefits, including healthcare, disability compensation, education
and vocational training, home loans, pensions, rehabilitation, and
survivor and burial benefits, to nearly 26 million American
veterans and their families.1 The provision of this wide range of
services has made the VA the second largest of all Cabinet-level
departments in terms of budget and staff. The VA’s fiscal year
2004 spending was $62.1 billion—$29.1 billion for healthcare and
$32.4 billion for benefits, including disability compensation.2 The
VA’s national healthcare system includes 158 hospitals, with at
least one in each of the forty-eight contiguous states, Puerto Rico,
and the District of Columbia. In fiscal year 2003, the VA provided
healthcare to about 4.8 million veterans, who made more than 49.8
million outpatient clinic visits.3
The VA’s disability compensation program provides monthly
monetary benefits to veterans who are disabled by serviceconnected injuries or diseases, that is, for illnesses or injuries that
were incurred or aggravated during their active military service.4
The amount of a monthly disability check is established by
Congress and is based on the degree of the service-connected
disability in 10% increments (10%, 20%, etc.), as determined by a
1
Office of Public Affairs Media Relations, U.S. Dep’t of Veterans Affairs,
Facts about the Department of Veterans Affairs (Feb. 2005), available at
http://www.va.gov/OPA/fact/docs/vafacts.pdf.
2
Id.
3
Id.
4
Id.
BROWN MACROED CORRECTED 060605.DOC
6/20/2005 7:07 PM
VETERAN COMPENSATION POLICIES
595
VA disability rating specialist.5 Currently, a veteran with a 10%
service-connected disability receives $108 per month, while a
veteran with a 100% service-connected disability receives $2,299
monthly.6 In fiscal year 2003, the VA spent $26 billion providing
disability compensation, death compensation, and pensions to
about 3.4 million veterans, and to nearly 600,000 spouses,
children, and parents of deceased veterans.7
A. Direct Service Connection and Disability Compensation
Establishing service connection and degree of disability usually
requires a simple review for a veteran with an acute disease or
injury, for example, a bullet or shrapnel wound.8 Most veterans
receiving disability compensation from the VA have injuries that
may be assessed through such a direct service connection
approach. Establishing direct service connection can be more
contentious, however, when the illness or injury involves a chronic
disease possibly caused by an environmental or occupational
exposure that may have occurred decades in the past (for example,
exposure to benzene as a potential cause of leukemia).
1. The Legal Standard
Congress has provided the VA with statutory guidance for
evaluating a direct disability compensation claim. Pursuant to
statute, the VA is authorized to pay disability compensation only
for a “disability resulting from personal injury suffered or disease
contracted in line of duty, or for aggravation of a preexisting injury
suffered or disease contracted in line of duty, in the active military,
5
AND
U.S. DEP’T OF VETERANS AFFAIRS, FEDERAL BENEFITS FOR VETERANS
DEPENDENTS 73 (2005), available at http://www.vetwatchnw.com/Fedben.
pdf.
6
Id. at 73.
Office of Public Affairs Media Relations, supra note 1.
8
Compensation and Pension Service, U.S. Dep’t of Veterans Affairs,
Disability Compensation Benefits (Dec. 2004), available at http://www.vba.
va.gov/bln/21/Milsvc/Docs/Compeg.doc.
7
BROWN MACROED CORRECTED 060605.DOC
596
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
naval, or air service.”9
The required showing that a disease was incurred in service can
present significant difficulties in cases involving latent diseases
allegedly associated with exposures many years earlier. Courts
have held that claimants must establish a causal “nexus” between
their current diseases and some incident or exposure during
military service.10 The VA has, however, adopted a relatively
generous approach to resolving factual issues pertinent to this
inquiry. A VA regulation, 38 C.F.R. section 3.102, establishes
what is known as the “reasonable doubt” or “benefit of the doubt”
doctrine, which also has been referred to as the “tie goes to the
runner” rule based upon the baseball analogy. In relevant part, the
regulation provides:
It is the defined and consistently applied policy of the
Department of Veterans Affairs to administer the law under
a broad interpretation, consistent, however, with the facts
shown in every case. When, after careful consideration of
all procurable and assembled data, a reasonable doubt
arises regarding service origin, the degree of disability, or
any other point, such doubt will be resolved in favor of the
claimant. By reasonable doubt is meant one which exists
because of an approximate balance of positive and negative
evidence which does not satisfactorily prove or disprove
the claim. . . . The reasonable doubt doctrine is also
applicable even in the absence of official records,
particularly if the basic incident allegedly arose under
combat, or similarly strenuous conditions, and is consistent
9
38 U.S.C. § 1110 (2005).
Shedden v. Principi, 381 F.3d 1163, 1166-67 (Fed. Cir. 2004) (“[I]n
order to establish service connection . . . the veteran must show: (1) the
existence of a present disability; (2) in-service incurrence or aggravation of a
disease or injury; and (3) a causal relationship between the present disability and
the disease or injury incurred or aggravated during service.”); McCartt v. West,
12 Vet. App. 164, 168 (1999) (holding that, in the absence of an applicable
presumption of service connection, a claimant alleging disability due to Agent
Orange exposure had to submit “medical evidence of a nexus between Agent
Orange exposure and the appellant’s current [disease]”).
10
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
597
with the probable results of such known hardships.11
In other words, the VA adjudicates a veteran’s direct service
disability compensation claim based upon the merits of the
individual case and grants a claim when the VA determines that a
veteran’s illness or injury is at least as likely as not to have been
caused by the environmental or occupational exposure.
2. Four Key Categories of Evidence for Direct Service Connection
As a practical matter, establishing a “nexus” between a
current disease and a claimed environmental or occupational
exposure in service generally requires four key categories of
evidence:
1. Evidence of a Scientific Association. Veterans must show
credible scientific or medical evidence that the exposure
involved is accepted as being associated with their specific
illness or injury;
2. Evidence of Military Exposure. Veterans must show
evidence that the relevant environmental or occupational
exposure occurred during their active military duty;
3. Evidence of Temporal Plausibility. Veterans must show
that their illnesses or injuries were initiated or were
exacerbated during active military duty; and
4. Evidence of Exposure Magnitude. Veterans must show
evidence of an unusually large or prolonged exposure to
support the conclusion that the exposure was at least as
likely as not to have been the specific cause of their
illnesses or injuries, in comparison to all other potential
causes of those illnesses experienced before and after
military service.
The “Evidence of Exposure Magnitude” requirement means
that a minimal, short-term, or commonplace exposure to an
environmental hazard might support the possibility that an illness
or injury was caused by the exposure; however, it might fail to
cross the statutory threshold requiring that it be at least as likely as
11
38 C.F.R. § 3.102 (2005).
BROWN MACROED CORRECTED 060605.DOC
598
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
not to have been the cause when compared to all other possible
causes. For example, a veteran who served two years in the
military and was diagnosed with leukemia at age fifty may have
had forty-eight years of exposure to benzene (a component of
gasoline) as a civilian outside of his military service and only two
years of exposure during service. Evidence that might support an
unusually large benzene exposure could include documentation
that the veteran’s military occupation, for example, regular motor
vehicle maintenance duties, specifically involved contact with
benzene.
In practice, this point can represent a significant hurdle for
establishing direct service connection for an illness or injury
caused by an environmental or occupational exposure. For
example, benzene is only a single and probably relatively minor
cause of all leukemias, and leukemia from all causes is not an
uncommon disease. Further, virtually everyone experiences
constant, minor benzene exposure because benzene is a component
of gasoline and other common solvents. Despite these restrictions,
most disability claims for environmental or occupational injury or
illness are based upon this direct service connection approach.
B. An Alternate Route to Disability Compensation:
Presumptive Service Connection
The VA has the authority to bypass one or even all of the four
key categories of evidence required for a direct service claim by
establishing a “presumptive” or automatic service connection. For
example, 38 C.F.R. section 3.307 and 38 C.F.R. section 3.309 list
some relatively common chronic diseases, including arthritis,
leukemia, and Type II diabetes, that the VA can presume to be
service connected when they appear within a certain period after
separation from military service, even if available evidence is not
sufficient to support a direct service connection. These
presumptively service connected illnesses must lead to at least
10% disability and appear generally within one year from the date
of the claimant’s separation from military service.12 Similarly, in
12
38 C.F.R. § 3.307 (2005); 38 C.F.R. § 3.309 (2005).
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
599
1995, Congress authorized the VA’s compensation of veterans
with undiagnosed illnesses or with difficult to diagnose illnesses,
such as chronic fatigue syndrome, fibromyalgia, and irritable
bowel syndrome, that are defined by a cluster of signs or
symptoms.13 Under 38 U.S.C. section 1117, the VA is empowered
to provide compensation to Gulf War veterans who, for at least six
months, exhibit objective evidence of 10% or greater disability,
which may include disability due to fatigue, skin conditions,
headache, muscle and joint pain, sleep disturbances, abnormal
weight loss, menstrual disorders, and neurologic or
neuropsychological,
respiratory,
gastrointestinal,
and
14
cardiovascular illnesses. Approximately 3,200 veterans of the
1991 Gulf War have received compensation based upon this law.
These statutes give veterans the benefit of the doubt in cases in
which certain poorly understood illnesses manifest within a
defined period. Further, they effectively eliminate at least the
“Evidence of Temporal Plausibility” requirement of the direct
service connection test. Presumptive service connection for the
specified diseases is not, however, automatic, and the VA may
consider evidence in rebuttal of service connection, including “any
evidence of a nature usually accepted as competent to indicate the
time of existence or inception of disease . . . .”15 In fact, the VA’s
statutorily defined Agent Orange Vietnam Veteran compensation
policies, described in Part II of this article, have eliminated
essentially all four key categories. Not surprisingly, this has also
produced certain unexpected problems.
II. BACKGROUND ON THE VA’S PRESUMPTIVE SERVICE
CONNECTION FOR AGENT ORANGE
Vietnam veterans during the 1960s and 1970s voiced
increasing concerns about how exposure to herbicides and dioxin
had affected their health. Some veterans cited Agent Orange as the
source of various health problems that extended to birth defects
13
14
15
38 U.S.C. § 1117.
Id.
38 C.F.R. § 3.307 (2005); 38 C.F.R. § 3.309 (2005).
BROWN MACROED CORRECTED 060605.DOC
600
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
among their children. Initially, the VA had problems establishing
policies on Agent Orange disability compensation. Minimal
veteran exposure information and limited scientific understanding
of Agent Orange and dioxin health effects meant that in the face of
mounting concerns from veterans and others, essentially all four
necessary categories of evidence for direct service connection were
missing.
In response, in 1991, Congress passed Public Law 102-4, more
commonly known as the “Agent Orange Act.” The Act mandated a
new process for establishing presumptive service connections for
illnesses related to Vietnam veterans’ exposure to Agent Orange,
other herbicides, and the contaminant dioxin, including a
presumption of exposure to those agents.16 The new law
represented a significant breakthrough for establishing
compensation policies in this area for what remains a controversial
issue even today.
The Agent Orange Act directed the VA to contract with the
National Academy of Sciences (NAS) to conduct a comprehensive
review of all scientific and medical literature on the health effects
from exposure to Agent Orange and other herbicides used in
Vietnam, and to dioxin. The initial 1994 NAS report—an
exhaustive and thorough review of all published literature on
health effects from exposure to these agents—established the
pattern for all future reports. Most of the reviewed literature came
from studies of civilians exposed either through industrial
accidents or in the workplace rather than from veterans themselves.
A. NAS Science and VA Policy
The Agent Orange Act assigns the NAS the responsibility of
evaluating the relevant science.17 The VA, in turn, is given
responsibility for translating the NAS’s scientific conclusions into
16
Agent Orange Act of 1991, Pub. L. No. 102-4, 105 Stat. 11 (1991)
(codified as amended at 38 U.S.C. § 1116).
17
38 U.S.C. § 1116 note (stating that the Act authorizes the National
Academy of Sciences “to review and evaluate the available scientific evidence
regarding associations between diseases and exposure to dioxin and other
chemical compounds in herbicides”).
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
601
veteran compensation policy.18 In other words, the statute
expressly invites the Secretary of Veterans Affairs to consider
evidence in addition to that provided by the NAS. In practice, the
VA assembles an internal taskforce of scientists, medical doctors,
attorneys, and compensation experts to evaluate and recommend
possible policy options to the Secretary in response to an NAS
report.
The Act further provides for an automatic “Evidence of a
Scientific Association” for all Vietnam veterans, stating that “[a]n
association between the occurrence of a disease in humans and
exposure to an herbicide agent shall be considered to be positive
for the purposes of this section if the credible evidence for the
association is equal to or outweighs the credible evidence against
the association.”19 Based on the 1994 NAS report, the VA decided
to presumptively recognize a range of illnesses, including soft
tissue sarcoma, non-Hodgkin’s lymphoma, Hodgkin’s disease,
chloracne, porphyria cutanea tarda, multiple myeloma, and
respiratory cancers. The Agent Orange Act also requires the NAS
to update its reviews based on new science at least every two
years.20 Updated reviews, published in 1996, 1998, 2000, and
2002, have expanded the VA’s list of presumptively service18
38 U.S.C. § 1116(b)(1)-(2). In relevant part, the statute states:
Whenever the Secretary determines, on the basis of sound medical and
scientific evidence, that a positive association exists between (A) the
exposure of humans to an herbicide agent, and (B) the occurrence of a
disease in humans, the Secretary shall prescribe regulations providing
that a presumption of service connection is warranted for that disease
for the purposes of this section. In making determinations for the
purpose of this subsection, the Secretary shall take into account (A)
reports received by the Secretary from the National Academy of
Sciences under section 3 of the Agent Orange Act of 1991, and (B) all
other sound medical and scientific information and analyses available
to the Secretary. In evaluating any study for the purpose of making
such determinations, the Secretary shall take into consideration whether
the results are statistically significant, are capable of replication, and
withstand peer review.
Id.
19
20
38 U.S.C. § 1116(b)(3).
Id. § 1116 note.
BROWN MACROED CORRECTED 060605.DOC
602
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
connected illnesses to include acute or sub-acute peripheral
neuropathy, Type II diabetes, prostate cancer, and most recently,
chronic lymphocytic leukemia. Based upon the 1996 NAS update
and congressional action, Vietnam veterans’ children with spina
bifida are also eligible for certain compensation and other
services.21
The Act also prescribes a “presumption of exposure” that
effectively frees a Vietnam veteran from having to show
“Evidence of Military Exposure,” “Evidence of Temporal
Plausibility,” and “Evidence of Exposure Magnitude,” the second,
third, and fourth key categories of evidence required for
establishing a direct service connection.22 As a result of this
statutorily defined policy, an eligible veteran (i.e., a veteran with
any discharge other than dishonorable discharge) must only show
that he is a Vietnam veteran diagnosed with one of the diseases
presumptively recognized as service connected to herbicide
exposure. Once this showing is made, service connection becomes
automatic.
This process eliminates a significant burden for Vietnam
veterans trying to establish service connection and disability
compensation for illnesses related to herbicide exposure. However,
there have been unexpected consequences in terms of apparent
inequities and unanticipated costs.
21
38 C.F.R. § 3.814 (2005).
38 U.S.C. § 1116. Specifically, the statute provides:
For purposes of establishing service connection for a disability or death
resulting from exposure to a herbicide agent, including a presumption
of service-connection under this section, a veteran who, during active
military, naval, or air service, served in the Republic of Vietnam during
the period beginning on January 9, 1962, and ending on May 7, 1975,
shall be presumed to have been exposed during such service to an
herbicide agent containing dioxin or 2,4-dichlorophenoxyacetic acid,
and may be presumed to have been exposed during such service to any
other chemical compound in an herbicide agent, unless there is
affirmative evidence to establish that the veteran was not exposed to
any such agent during that service.
Id. § 1116(f) (2005).
22
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
603
1. Credibility and Independence
The strengths of the NAS scientific review process are its
breadth and thoroughness, and the NAS’s reputation for
independence and scientific prestige. Earlier efforts by the VA to
conduct its own scientific reviews on herbicide health effects were
viewed by many veterans as lacking credibility and
independence.23 Although veterans have not always been happy
with the NAS findings, the NAS’s credibility has remained intact.
Indeed, the NAS process has become an essential step in ensuring
that new service connection presumptions command scientific
credibility.
2. Compensation Inequities
Without statutorily defined presumptions, Vietnam veterans
would have difficulty establishing a direct service connection for
any illnesses related to herbicide or dioxin exposure. In particular,
it would be difficult for claimants to provide “Evidence of a
Scientific Association,” “Evidence of Military Exposure,” and
“Evidence of Temporal Plausibility.” Epidemiological studies of
Vietnam veterans suggest that herbicide and dioxin exposure play,
probably at most, only a minute role in overall mortality.24 Taken
together, these studies suggest that Vietnam veterans would have
pronounced difficulty establishing “Evidence of Exposure
Magnitude,” that is, evidence that their exposures were sufficiently
23
U.S. DEP’T OF VETERANS AFFAIRS, REVIEW OF LITERATURE ON
HERBICIDES, INCLUDING PHENOXY HERBICIDES AND ASSOCIATED DIOXINS
(1981-1992) (vols. 1 & 2 prepared by JRB Associates, McLean, Va.; vols. 3 to
18 prepared by Clement Int’l, Fairfax, Va.; vols. 19 & 20 prepared by Info.
Ventures, Phila., Pa.).
24
Numerous epidemiological studies of Vietnam veterans in general do not
show that this group has higher mortality or morbidity from most of the diseases
presumptively connected to herbicide exposure. An excellent summary of
mortality and morbidity research on Vietnam veterans compared to their nondeployed peers is available in “Veterans and Agent Orange: A Continuing
Medical Education Program,” an independent study course first published in
2002 by the Department of Veterans Affairs, Employee Education System,
available at www.va.gov/VHI/.
BROWN MACROED CORRECTED 060605.DOC
604
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
great or prolonged to make it at least as likely as not that the
exposures were the cause of the veterans’ illnesses or injuries when
compared to all other potential causes. A related problem is that
the Agent Orange Act has created a narrow focus on herbicides as
the key to Vietnam veterans’ illnesses when in fact they almost
certainly play only a very minor role.
Further, the Agent Orange Act applies only to Vietnam
veterans. Non-Vietnam veterans exposed to herbicides and dioxins
do not receive the benefit of presumptive service connections;
however, many non-Vietnam veterans have been exposed to these
agents, including U.S. troops serving during the Vietnam War but
only in nearby countries, including Cambodia, Laos, and Thailand,
or off-shore aboard ships. The U.S. military during 1968 and 1969
also used Agent Orange and similar herbicides to defoliate the
demilitarized zone between North and South Korea. The military
use of Agent Orange and related herbicides also was tested and
developed at U.S. bases located in the United States and abroad.
There is no obvious scientific or public health basis for
excluding these non-Vietnam War veterans from the presumptive
service connection offered to Vietnam veterans. Nevertheless, the
Agent Orange Act does not reference these veterans. To partially
address this apparent inequity, the VA has established the general
policy that when a non-Vietnam veteran is diagnosed with one of
the presumptively service-connected Agent Orange illnesses and
the veteran can provide evidence of exposure to Agent Orange,
then he can be granted service connection through a sort of
modified direct service connection route.
Even this approach may ultimately prove unmanageable
because, in fact, the majority of veterans could in principle claim
herbicide exposure during military service, and thus, service
connection for related illnesses. From the 1950s to the early 1970s,
Agent Orange and related herbicides, including those with dioxin
contaminants, were extensively used domestically for weed control
on lawns and golf courses (including those on military bases and
even VA hospitals), in forestry, and for weed control along fences,
borders, and roads. Thus, everyone living during that period likely
would have had some exposure.
Moreover, why should this policy be limited to military
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
605
personnel? Many domestic civilian workers used these same
chemical agents. The VA has received inquiries about the VA’s
Agent Orange compensation policies from employees of the U.S.
Department of Interior who were involved in spraying these
herbicides on U.S. forests during the 1960s and 1970s. Indeed, the
Government Accountability Office (GAO) recently asked the VA
why these same policies should not be applied to workers’
compensation claims filed with the U.S. Department of Labor.
Similarly, the Government of Vietnam has publicly insisted that
the U.S. government provide compensation to Vietnamese
civilians for Agent Orange-related injuries. (Thus far, the U.S.
government has denied this request.)
3. Cost Inequities
The economic implications of the Agent Orange Act may not
have been fully anticipated by Congress. In perhaps the most
dramatic and expensive example, the 2000 NAS Agent Orange
special report concluded that there was “limited/suggestive”
evidence associating herbicide exposure and Type II diabetes.25
(Previous NAS reports had not found such positive evidence, but
new scientific studies finally tipped the balance.) After reviewing
that finding, in 2001, the VA announced a new presumptive
service connection for Type II diabetes among Vietnam veterans.
This decision has significant economic implications. The VA
estimated that about 9%, or about 270,000 of the approximately 3
million Vietnam veterans, would have Type II diabetes based
solely upon on their age and other demographics. Diabetes often
involves prolonged disability, and treatment can be expensive. The
VA estimated that disability and treatment would cost several
billion dollars over the first five years of implementing this policy.
This is a significant portion of the VA’s overall disability
compensation budget for all veterans.
In fact, epidemiological studies do not show Vietnam veterans
25
INST. OF MEDICINE, NAT’L ACADEMY OF SCIENCES, VETERANS AND
AGENT ORANGE: HERBICIDE/DIOXIN EXPOSURE AND TYPE 2 DIABETES 2 (2000)
(Nat’l Academy Press, Wash., D.C.).
BROWN MACROED CORRECTED 060605.DOC
606
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
dying from Type II diabetes at greater rates than their nondeployed peers. This suggests that there are probably only a small
number of excess cases of Type II diabetes among Vietnam
veterans due to herbicide or dioxin exposure during military
service; indeed, the NAS noted that the biggest risk factors for
diabetes are lifestyle and obesity. Coupled with limited exposure
data, this means that few, if any, of these cases would have been
granted service connection via the direct service connection route.
In effect, the VA’s policy compensates a very large number of
veterans who would have been diagnosed with Type II diabetes
regardless of their military service in order to ensure coverage of
the few veterans who may have contracted the disease because of
it.
B. Applying the NAS Process to Gulf War Veterans
Despite these problems, the Agent Orange Act and the NAS
process it defined are generally acknowledged as successful
approaches to incorporating science into difficult and contentious
veteran compensation policy decisions. Nevertheless, more recent
attempts to apply the Agent Orange approach to emerging
environmental disabilities in veterans have suffered from the
unintended consequences associated with implementing the Agent
Orange Act for Vietnam veterans and thus far have produced little
or no benefit for veterans.
The 1991 Gulf War concluded fourteen years ago. In response
to the concerns of veterans and their families, and of Congress that
the health of Gulf War veterans might have been affected by
exposure to a wide variety of environmental hazards during the
war, Congress passed two statutes, Public Law 105-277 and Public
Law 105-368.26 These statutes were drawn directly from the Agent
Orange Act of 1991 and established the now-familiar formal NAS
process, which mandates regular and thorough reviews of the
scientific and medical literature relevant to health and Gulf War
26
Persian Gulf War Veterans Act of 1998, Pub. L. No. 105-277, 112 Stat.
2681 (1998); Veterans Programs Enhancement Act of 1998, Pub. L. No. 105368, 112 Stat. 3315 (1998).
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
607
exposures.
Major differences in the environmental exposures experienced
by individuals serving during the Vietnam and Gulf Wars led to
immediate problems with this approach. In contrast to the narrow
range of exposures composed of “herbicides used in Vietnam and
their dioxin contaminant,” the exposures related to service in the
Gulf War involved dozens of different and unrelated
environmental hazards. In fact, the two new statutes specified
thirty-three Gulf War-related environmental and occupational
hazards as well as broad categories of hazards to be evaluated
through this process. Consequently, although the NAS has
produced three major biannual reports, it has yet to complete even
an initial review of all of the statutorily defined Gulf War
hazards.27
In addition, in 1991, there was only a limited amount of
scientific literature on health effects of dioxins and Vietnamrelated herbicides for the NAS to review. In contrast, virtually all
of the Gulf War hazards were well known and characterized, with
an abundant health effects literature. The NAS reviews have thus
failed to produce any new insights into the health effects of
exposure to these hazards, as the NAS’s findings have mirrored
those found in any standard occupational health and toxicology
textbook.
The first NAS report, completed in 2000, reviewed health
effects from exposure to sarin, depleted uranium, pyridostigmine
bromide, and certain vaccines, including the anthrax vaccine.28 The
second report, completed in 2003, reviewed the health effects of
exposure to all of the insecticides and solvents used in the 1991
27
INST. OF MEDICINE, NAT’L ACADEMY OF SCIENCES, GULF WAR AND
HEALTH VOL. 1: DEPLETED URANIUM, PYRIDOSTIGMINE BROMIDE, SARIN,
VACCINES (2000) (Nat’l Academy Press, Washington, D.C.) [hereinafter GULF
WAR AND HEALTH VOL. 1] (on file with author); INST. OF MEDICINE, NAT’L
ACADEMY OF SCIENCES, GULF WAR AND HEALTH VOL. 2: INSECTICIDES AND
SOLVENTS (2003) (Nat’l Academy Press, Washington, D.C.) [hereinafter GULF
WAR AND HEALTH VOL. 2] (on file with author); INST. OF MEDICINE, NAT’L
ACADEMY OF SCIENCES, GULF WAR AND HEALTH VOL. 3: FUELS, COMBUSTION
PRODUCTS, AND PROPELLANTS (2004) (Nat’l Academy Press, Washington, D.C.)
[hereinafter GULF WAR AND HEALTH VOL. 3] (on file with author).
28
GULF WAR AND HEALTH VOL. 1, supra note 27, at 2.
BROWN MACROED CORRECTED 060605.DOC
608
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
Gulf War.29 The third report, completed in 2004, reviewed the
health effects of exposure to oil well fire air pollutants and certain
other chemicals associated with the 1991 Gulf War.30
Taken together, these recent NAS reports contain dozens of
somewhat predictable findings on health effects for dozens of
hazardous agents. As with the Agent Orange Act of 1991, the new
statutes direct the VA to find a positive association between a Gulf
War hazard and a specific illness “if the credible evidence for the
association is equal to or outweighs the credible evidence against
the association,”31 or to provide to Congress a report including “the
Secretary’s recommendations as to whether there is sufficient
evidence to warrant a presumption of service-connection for the
occurrence of a specified condition in Gulf War veterans.”32 When
the evidence supports an association with a particular disease more
than it does not, the statute requires the VA to develop regulations
defining a presumptive service connection for that disease among
Gulf War veterans.33
C. Policy Problems
The VA has had difficulty applying the NAS findings to Gulf
War veterans because nearly all of the reviewed Gulf War-related
29
GULF WAR AND HEALTH VOL. 2, supra note 27, at 2.
GULF WAR AND HEALTH VOL. 3, supra note 27 (forthcoming
publication).
31
Persian Gulf War Veterans Act of 1998 § 1602(b)(3), 112 Stat. 2681
(1998) (codified at 38 U.S.C. § 1118).
32
Veterans Programs Enhancement Act of 1998 § 101(i)(2), 112 Stat. 3315
(1998).
33
Persian Gulf War Veterans Act of 1998, Pub. L. No. 105-277, §
1602(c)(1), 112 Stat. 2681 (1998) (codified at 38 U.S.C. § 1118) (“Not later than
60 days after the date on which the Secretary receives a report from the National
Academy of Sciences . . . the Secretary shall determine whether or not a
presumption of service connection is warranted for each illness, if any, covered
by the report.”); id. § 1602(c)(2) (codified at 38 U.S.C. § 1118) (“If the
Secretary determines under this subsection that a presumption of service
connection is warranted, the Secretary shall, not later than 60 days after making
the determination, issue proposed regulations setting forth the Secretary’s
determination.”).
30
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
609
hazards represent common, well-characterized occupational
exposures that are experienced by virtually all Americans. It may
come as a surprise to learn that military environmental exposures
generally closely mirror the environmental exposures experienced
by all Americans.
1. Considering Exposure Magnitude
The VA has had particular difficulty applying the NAS’s
conclusions based on studies of civilian workers with occupational
exposures to the experience of deployed Gulf War veterans.
Although essentially all health effects reported by the NAS are
based on studies of workers with occupational exposures typically
occurring over years, and indeed decades, deployments during the
1991 Gulf War typically lasted about only three months. For
example, in its 2002 report, the NAS documented a slight increase
in risk for leukemia among chemical industry workers who have
large occupational exposure to benzene. These findings do little to
inform us about the potential increased risk of leukemia in the
typical Gulf War soldier experiencing an unremarkable benzene
exposure during the few months of deployment.
On the other hand, there are certainly examples of Gulf War
veterans with greater than everyday or commonplace benzene
exposure, for example, veterans who regularly worked on vehicle
maintenance. Those cases could involve benzene exposure at
levels more comparable to the typical civilian occupational
exposures that formed the bases of the studies reviewed by the
NAS.
Similarly, many NAS findings on long-term health effects
among civilian workers are reported only in cases that involved an
unusual exposure sufficiently large to cause immediate and serious
health effects. For example, certain long-term health effects from
common organophosphorus pesticides are well documented, but
only as the result of an exposure large enough to cause severe and
immediate initial poisoning, typically, an occupational exposure.
Exposures that do not cause immediate and serious effects have not
been associated with long-term effects. How should the VA apply
these NAS findings to the vast majority of veterans who
BROWN MACROED CORRECTED 060605.DOC
610
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
experienced only unexceptional exposures to these agents? In fact,
all Americans have experienced some small, but more or less
continuous exposure to common organophosphorus pesticides or to
benzene during their lifetimes. Still, these exposures are considered
to have negligible health risk significance, given the small
magnitude of the exposure.
The problem for the VA is that the underlying statutes
outlining this process eliminate the requirement for any “Evidence
of Exposure Magnitude” that would be necessary for a direct
service connection and offer little or no guidance for making
distinctions based on exposure magnitude or duration. In effect,
this creates the somewhat scientifically implausible result of
treating all exposures as equally likely to lead to an associated
long-term health effect. As a result, the VA could presumptively
service connect all of the illnesses positively associated by the
NAS with all of the associated Gulf War risk factors identified by
the NAS, no matter how universal or trivial the exposure
magnitude may have been for most or even all Gulf War veterans.
For these reasons, the VA has thus far been unable to develop
any new presumptive service connected disabilities for the wide
range of hazardous occupational and environmental exposures
associated with the 1991 Gulf War. Some of the NAS findings are
still under review.
2. Are New Presumptive Service Connection Polices Necessary?
Any presumptive service connection compensation policy will
bring certain negative consequences. Policymakers presumably
have determined that the advantages outweigh the drawbacks. In
this instance, it is not clear that new presumptive service
connections are necessary to provide equitable compensation to
Gulf War veterans for the environmental and occupational injuries
they sustained during their service.
Both short- and long-term health effects from most Gulf War
hazardous exposures were generally very well characterized, even
before that war began. That information, which is summarized in
the NAS reviews, provides Gulf War veterans with a strong basis
for pursuing disability compensation through the conventional
BROWN MACROED CORRECTED 060605.DOC
VETERAN COMPENSATION POLICIES
6/20/2005 7:07 PM
611
direct service connection route. The NAS reviews provide Gulf
War veterans with ready access to scientific information to support
the “Evidence of a Scientific Association” criterion. A Gulf War
veteran would still be required to show “Evidence of Military
Exposure” and “Evidence of Temporal Plausibility” to support a
direct service connection claim. However, today many
environmental hazards have been inexorably linked with service in
the 1991 Gulf War.
Perhaps the greatest obstacle for establishing a direct service
connection would be providing “Evidence of Exposure
Magnitude” that demonstrates that the veteran’s exposure was at
least as likely as not to have been the cause of her disability.
However, that is a commonplace and hardly insurmountable
obstacle for veteran compensation claims in general. For example,
in cases in which the NAS had documented long-term health
effects marked by immediate and serious health effects at the time
of exposure, a veteran’s military record or other lay evidence
would likely be sufficient to support the veteran’s claim. Indeed,
veterans might even prevail in cancer claims based on initial
exposures that did not involve immediate and obvious effects by
demonstrating, through the presentation of their specific military
occupation and service records, that they experienced long-term
moderate to high-level exposures that are generally associated with
cancer in the relevant occupational health literature and that these
exposures were “at least as likely as not” the cause of their cancer.
Thus, the direct service connection route could cover most
disability claims for illnesses that the NAS found to be associated
with Gulf War environmental and occupational exposures. More
simply, the conventional direct service connection process, based
on a review of the identity of the exposure and information about
its magnitude, would be sufficient to establish service connection
when warranted.
CONCLUSION
The majority of veteran disability compensation claims for
injuries related to environmental or occupational exposures are
evaluated on the merits of the individual case through the direct
BROWN MACROED CORRECTED 060605.DOC
612
6/20/2005 7:07 PM
JOURNAL OF LAW AND POLICY
service connection process. New presumptive service connection
policies inevitably bring unintended consequences, including a
perceived or actual disparity in access to disability benefits for
different groups of veterans, unanticipated costs, and scientifically
implausible or untenable policies.
Presumptive service connection policies may be useful for
specific situations in which it is impractical for veterans to develop
a direct service claim, for example, when a veteran is diagnosed
with an illness of unclear or unknown cause within a short period
following separation from military service. Nevertheless, before
considering new presumptive service connections that offer certain
veterans special presumptive service connection, the VA and
congressional policymakers should first determine that the
conventional direct service connection route is not adequate to the
task.
When a new presumptive service connection policy is
determined to be necessary, it should be implemented in a manner
that is considered fair and consistent with available science. In
virtually all cases this requires the use of an independent scientific
review body, such as the NAS, to ensure the credibility of the new
policy and the perception by all parties that the policy is
impartially based upon the best science.
Experience has shown that presumptive service connection
policies that eliminate consideration of the magnitude of exposure
should be avoided because they run the likely risk of inadvertently
extending eligibility to all veterans, regardless of how trivial or
commonplace their exposures may have been. Some of these issues
may require clarification by Congress in the form of legislative
fixes for those portions of statutes that have led to unanticipated
problems.
BORGIDA MACROED 051605.DOC
5/23/2005 10:26 PM
ON THE USE OF GENDER STEREOTYPING
RESEARCH IN SEX DISCRIMINATION
LITIGATION
Eugene Borgida, Ph.D., Corrie Hunt, and Anita Kim∗
INTRODUCTION
In cases in which expert scientific testimony may be helpful to
a jury, a witness who is qualified as an expert may “testify in the
form of an opinion or otherwise, if (1) the testimony is based upon
sufficient facts or data, (2) the testimony is the product of reliable
principles and methods, and (3) the witness has applied the
principles and methods reliably to the facts of the case.”1 In cases
involving claims of gender stereotyping, several courts have
permitted testimony by experts who are qualified to explain the
practice of stereotyping and the conditions under which it is more
or less likely to occur.2
∗
Eugene Borgida is Professor of Psychology and Law and the MorseAlumni Distinguished Teaching Professor of Psychology at the University of
Minnesota. He received his B.A. in 1971 from Wesleyan University and Ph.D.
from the University of Michigan in 1976. He has served as an expert witness in
a number of class action and single-plaintiff sex discrimination cases, and as a
science advisor to the U.S. Equal Employment Opportunity Commission from
1995-1999. Corrie Hunt received her B.S. from the University of Washington
(Seattle) in 2002 and is a graduate student in the social psychology Ph.D.
program. Anita Kim received her B.S. from the University of California at San
Diego in 1999 and is a graduate student in the social psychology Ph.D. program.
1
FED. R. EVID. 702; see also Daubert v. Merrell Dow Pharm. Inc., 509 U.S.
579, 593-94 (1993).
2
Butler v. Home Depot, Inc., 984 F. Supp. 1257, 1262-63 (N.D. Cal.
1997); Dukes v. Wal-Mart, Inc., 222 F.R.D. 189, 191-92 (N.D. Cal. 2004);
Jenson v. Eveleth Taconite Co., 824 F. Supp. 847, 882 (D. Minn. 1993); Beck v.
Boeing Company, 203 F.R.D. 459, 461 (D. Wash. 2001); Price Waterhouse v.
613
BORGIDA MACROED 051605.DOC
614
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
However, expert testimony on gender stereotyping has not
been without its critics. In Price Waterhouse v. Hopkins, for
example, the majority rejected the notion of expertise in gender
stereotyping in evaluating the testimony of the expert psychologist
presented by the plaintiff, Dr. Susan Fiske:
[W]e are tempted to say that Dr. Fiske’s expert testimony
was merely icing on Hopkins’ cake. It takes no special
training to discern sex stereotyping in a description of an
aggressive female as requiring “a course in charm school.”
Nor . . . does it require expertise in psychology to know
that, if an employee’s flawed “interpersonal skills” can be
corrected by a soft-hued suit or a new shade of lipstick,
perhaps it is the employee’s sex and not her interpersonal
skills that has drawn the criticism.3
Similarly, in overturning the admission of expert testimony on
gender stereotyping, the Minnesota Court of Appeals, in Ray v.
Miller Meester Advertising, Inc., offered the following rationale:
Information about and commentary on gender issues is so
abundant in our society that it has become a common
stereotype that women receive disparate and often unfairly
discriminatory treatment in the workplace. . . . Gender
stereotypes are the stuff of countless television situation
comedies and are the focus of numerous media treatments
on a nearly daily basis. It is unarguable that virtually all
adults in our society know about gender stereotypes.4
In addressing such criticism, Faigman, Kaye, Saks, and
Sanders5 have offered an astute perspective on the differences of
opinion regarding the validity of gender stereotyping testimony
and the value of social science evidence in the litigation context.
Hopkins, 490 U.S. 228, 255-56 (1989); Robinson v. Jacksonville Shipyards,
Inc., 760 F. Supp. 1486, 1505 (D. Fla. 1991); E.E.O.C. v. Morgan Stanley &
Co., 324 F.Supp.2d 451 (S.D.N.Y. 2004).
3
490 U.S. at 256.
4
Ray v. Miller Meester Advertising, Inc., 664 N.W.2d 355, 366 (Minn. Ct.
App. 2003).
5
David L. Faigman et al., Legal Issues, in 2 MODERN SCIENTIFIC
EVIDENCE: THE LAW AND SCIENCE OF EXPERT TESTIMONY 374 (David L.
Faigman et al. eds., 2002).
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
615
“One significant value of much social science research is that it
makes clearer what we only dimly perceive, if we perceive it at
all.”6 The researchers noted that individuals often respond to
reports of psychological findings by nonchalantly remarking, “‘of
course we knew this all along.’”7 However, individuals’ beliefs
about what they have always known often prove “not quite correct
or, more importantly, not quite correct in substantial detail.”8
While Faigman and colleagues suggest that research on eyewitness
identification is the paradigmatic example of this phenomenon,
they assert that research on gender stereotyping also fits this model
and “offers insights about gender relations . . . beyond what
experience alone can provide.”9
Overall, research in gender stereotyping has yielded a body of
knowledge that illustrates the complex nature of gender relations.
Although some of the findings may at first seem intuitive, the
effects of gender stereotyping are incredibly complicated and
nuanced. As will be discussed, the research demonstrates that
judgments of women are strongly influenced by stereotypes and
that this is especially true when women behave in
counterstereotypic ways.
I. OVERVIEW OF RESEARCH ON GENDER STEREOTYPING
Several reviews of the scientific literature on gender
stereotyping suggest that this body of knowledge reflects a
scientifically established and mature area of psychological science
with areas of scientific agreement and disagreement that provide
evidence-based insights into the nature of gender relations.10 Hunt
6
Id.
Id.
8
Id.
9
Id.
10
E.g., Diana Burgess & Eugene Borgida, Who Women Are, Who Women
Should Be: Descriptive and Prescriptive Gender Stereotyping in Sex
Discrimination, 5 PSYCHOL., PUB. POL’Y & L. 665 (1999); Susan T. Fiske et al.,
Social Science Research on Trial: Use of Sex Stereotyping Research in Price
Waterhouse v. Hopkins, 46 AM. PSYCHOL. 1049 (1991); Jennifer S. Hunt et al.,
Gender Stereotyping: Scientific Status, in 2 MODERN SCIENTIFIC EVIDENCE: THE
LAW AND SCIENCE OF EXPERT TESTIMONY 374 (D. L. Faigman et al. eds., 2002).
7
BORGIDA MACROED 051605.DOC
616
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
et al. suggest several areas of scientific agreement with regard to
gender stereotyping research.11 First, researchers have studied the
traits and attributes generally associated with women and men, and
the traits that are associated with subtypes of women and men
(e.g., businesswomen, feminists, and housewives, or construction
workers, businessmen, and jocks). With respect to the content of
stereotypes, Hunt et al. argue that the research has consistently
shown that men are seen as agentic and achievement oriented, and
women are seen as communal, nurturing, and passive. It is
noteworthy that stereotypically male traits are associated with
success in the business world and that stereotypically female traits
are not. Further, people not only associate individual traits with
men and women, but they also believe prescriptively that men and
women should behave in gender-consistent ways.
Second, Hunt et al. show that research has continually found
that gender stereotypes have a small but definite effect on social
judgments of women and men (e.g., evaluations, attributions, and
employment decisions), especially when women and men act in
ways that are inconsistent with existing stereotypes. Research has
shown that when forming initial impressions of women, people
frequently rely more on their stereotypes about women than on
information about the specific woman (individuating information).
Thus, preconceived stereotypes of businesswomen more strongly
influence a perceiver’s impressions of a specific businesswoman
than do that woman’s own attributes.
Research also has shown that the effects of using these
stereotypes are predictable and especially strong in contexts in
which individuals behave in gender-inconsistent ways. In one
meta-analysis, for example, Swim, Borgida, Maruyama, and Myers
found that, overall, there was only a slight tendency to evaluate
women more negatively than men.12 However, when only
stereotypically masculine and gender-neutral roles were examined,
this effect size increased substantially. The effect disappeared
when women were evaluated in stereotypically feminine roles,
11
Hunt et al., supra note 10, at 423-24.
Janet Swim et al., McKay vs. McKay: Is There a Case for Gender
Biased Evaluations?, 105 PSYCHOL. BULL. 409 (1989).
12
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
617
supporting the idea that women are penalized when they behave in
gender-inconsistent ways.13
A third area in which there is sufficient scientific agreement
involves research showing that stereotypes are more likely to exert
an influence in gender-salient situations, ambiguous situations, or
situations in which the perceiver is not motivated to make accurate
judgments.14 Finally, as noted by Hunt et al.,15 researchers also
have studied the psychological processes by which individuals use
stereotypes in their perceptions and evaluations of others. Gender
is a fundamental dimension of categorization. Once an individual
is categorized as belonging to a gender, the stereotypes of that
gender may quickly come to the perceiver’s mind, a process
known as stereotype activation. Once stereotypes are activated,
they are then available for the perceiver to apply in her thinking
about and evaluation of the target person. It is important to note
that categorization, stereotype activation, and stereotype
application can all occur outside of the perceiver’s awareness.
Moreover, some people are more likely to use stereotypes to
guide their judgment than others, and different motives seem to
affect the use of stereotypes differently. Right Wing Authoritarian
(RWA) personalities, for example, are characterized by
endorsement of traditional norms and values, submission to
authority, and aggression against those who defy traditional norms
and values. Research supports the idea that individuals with RWA
personalities endorse traditional gender roles and disapprove of
women who behave non-traditionally. One study reviewed by Hunt
et al., for example, found that men with RWA personalities rated
feminists significantly less favorably than housewives, did not
believe feminists promoted equality for women, and perceived
feminists as holding different values than their own.16
As discussed in the next section, the activation and application
of gender stereotypes can be nuanced and pose double-bind
impression management dilemmas for women in organizational
13
14
15
16
Id. at 423-24.
Hunt et al., supra note 10, at 408.
Id. at 412-19.
Id. at 420.
BORGIDA MACROED 051605.DOC
618
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
contexts.
II. A CASE STUDY IN GENDER STEREOTYPING: WOMEN WHO SELFPROMOTE
A relatively recent and important area of research in the field
of gender stereotyping illustrates how gendered prescriptions for
personality interact with what women may (and do) experience as
they attempt to climb the occupational hierarchy. Emphasizing the
conflicts women face in their dual gender and leadership roles, this
growing body of research demonstrates that in order to be
perceived as competent when compared to their male counterparts
in leadership roles, women must self-promote and explicitly
emphasize their prior successes. However, this self-promotion
involves an impression management dilemma for women: selfpromotion may increase perceptions of their qualifications, but
self-promotion also appears to increase the likelihood of social
rejection or what Rudman refers to as the “backlash effect.”17
Various experimental studies have been conducted18 to examine
the impact of this “backlash” effect on hiring decisions,
perceptions of likeability, social skills, and competence. This
approach relies on the presumption of communal conceptions of
women as friendly, unselfish, and other-focused, and as
expressively clashing with the masculine and agentic expectations
17
Laurie A. Rudman, Self-Promotion as a Risk Factor for Women: The
Costs and Benefits of Counterstereotypical Impression Management, 74 J. OF
PERSONALITY AND SOC. PSYCHOL. 629, 629 (1998) [hereinafter Rudman, SelfPromotion As a Risk Factor for Women].
18
See, e.g., id.; Laurie A. Rudman & Kimberly Fairchild, Reactions to
Counterstereotypic Behavior: The Role of Backlash in Cultural Stereotype
Maintenance, 87 J. OF PERSONALITY AND SOC. PSYCHOL. 157 (2004); Laurie A.
Rudman & P. Glick, Feminized Management and Backlash toward Agentic
Women: The Hidden Costs to Women of a Kinder, Gentler Image of Middle
Managers, 77 J. OF PERSONALITY AND SOC. PSYCHOL. 1004 (1999) [hereinafter
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women]; Laurie A. Rudman & P. Glick, Prescriptive Gender Stereotypes and
Backlash toward Agentic Women, 57 J. OF SOC. ISSUES 743 (2001) [hereinafter
Rudman & Glick, Prescriptive Gender Stereotypes and Backlash toward
Agentic Women].
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
619
associated with managerial leadership.19 The experimental
evidence generally suggests that because self-promotion is
incongruent with prescriptions for leadership traits, agentic women
are at a greater risk of experiencing a “backlash.”
A. Perceptions of Competence
The prediction that women are less likely to be perceived as
agentic has primarily been investigated in terms of competence
ratings for hypothetical female job applicants.20 In one experiment,
for example, undergraduate participants interviewed and evaluated
a female or male applicant (target) for a potential partner in a
Jeopardy-like game.21 Results indicated that when participants
interviewed a self-effacing applicant, they rated the man as more
competent than the woman. In contrast, participants rated both
self-promoting men and women equally in terms of task aptitude.
Insofar as modesty is considered to be a part of the feminine
stereotype and self-promotion is not,22 the previous experiment
provides some support for the idea that conformity to the
stereotype of a feminine personality may result in diminished
perceptions of a woman’s competence. Heilman, Wallen, Fuchs,
and Tamkins further examined how gender stereotypes affect
perceptions of competence of female job applicants by
experimentally manipulating the presence of specific and objective
information about the candidate’s competence for a managerial (or
19
Alice H. Eagly & Steven J. Karau, Role Congruity Theory of Prejudice
toward Female Leaders, 109 PSYCHOL. REV. 573, 574-75 (2002).
20
Rudman, Self-Promotion as a Risk Factor for Women, supra note 17, at
629-30; Rudman & Glick, Feminized Management and Backlash toward
Agentic Women, supra note 18, at 1004-06; Rudman & Glick, Prescriptive
Gender Stereotypes and Backlash toward Agentic Women, supra note 18, at
746-49; M. E. Heilman et. al., Penalties for Success: Reactions to Women Who
Succeed at Male Gender-Stereotyped Tasks, 89 J. OF APPLIED PSYCHOL. 416,
417-18 (2004) [hereinafter Heilman et al., Penalties for Success].
21
Rudman, Self-Promotion as a Risk Factor for Women, supra note 17, at
639-40.
22
Mary E. Wade, Women and Salary Negotiation: The Costs of SelfAdvocacy, 25 PSYCH. OF WOMEN Q. 65, 68-70 (2001).
BORGIDA MACROED 051605.DOC
620
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
leadership) position.23 In this study, participants read a job
description and background of three employees holding the same
position in a large organization. What is particularly interesting in
this study is that when participants did not read an evaluation of
the candidate’s prior job performance, they made the trait inference
that the male candidate was more competent than the female
candidate.24 Only when participants read that upper-level
management had evaluated the female candidate as extremely
competent did participants actually rate the two candidates equally
on the dimension of task aptitude.25
While the studies detailed here focused specifically on selfpromotion and on whether or not the applicant’s prior successes
were known, other studies have found that agentic applicants are
perceived as more competent for managerial positions than
communal26 and androgynous candidates.27 The finding that in
ambiguous situations people will infer that women are less capable
than men is consistent with the argument that descriptive
stereotypes of “how women are” decrease a woman’s chances of
being perceived as having the agentic qualities and attributes
required of successful managers and leaders. On this basis, it might
seem that in order to avoid these inferences, women should simply
make sure they self-promote and provide solid, credible evidence
of their job qualifications.
B. Perceptions of Likeability
Unfortunately, enhanced perceptions of competence for women
who self-promote seem to come at a social cost. Women who
speak strongly about their strengths and clearly delineate prior
successes are viewed as more qualified, but also as less socially
23
Heilman et al., Penalties for Success, supra note 20, at 418.
Id. at 419-20.
25
Id.
26
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women, supra note 18, at 1008.
27
Rudman & Glick, Prescriptive Gender Stereotypes and Backlash toward
Agentic Women, supra note 18, at 752.
24
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
621
attractive.28 For example, Rudman and Glick’s research explored
how women who present themselves as directive and assertive—or
as successful agentic leaders—are viewed as stepping outside of
their feminine gender roles and, thus, as no longer in possession of
stereotypically feminine interpersonal skills.29 In one experiment,
participants who viewed a videotaped interview or read an
application essay rated agentic males as more socially skilled than
agentic females.30 Supporting the hypothesis that this discrepancy
is due to perceived inconsistencies between agentic expectations
and feminine roles, ratings for the social skills of female
communal applicants were statistically indistinguishable from
those for male communal applicants.31 These applicants spoke or
wrote more modestly about their previous accomplishments and, as
a result, were viewed as more interpersonally skilled.32 However,
participants rated the women as less competent overall.33
Moreover, participants who automatically associated women with
communality and men with agency were more likely to view
agentic females as interpersonally deficient.34
In this light, the evidence rather clearly demonstrates that
gendered personality constructs can influence people’s perceptions
of job applicants and that this effect places women at a
disadvantage. Heilman et al., in the study described in the previous
section, found that when prior successes were clear, participants
28
Rudman, Self-Promotion as a Risk Factor for Women, supra note 17, at
635-36; Rudman & Glick, Feminized Management and Backlash toward
Agentic Women, supra note 18, at 1007-08; Rudman & Glick, Prescriptive
Gender Stereotypes and Backlash toward Agentic Women, supra note 18, at
752-53.
29
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women, supra note 18, at 1008-09; Rudman & Glick, Prescriptive Gender
Stereotypes and Backlash toward Agentic Women, supra note 18, at 757-59.
30
Rudman & Glick, Prescriptive Gender Stereotypes and Backlash toward
Agentic Women, supra note 18, at 749-53.
31
Id. at 753.
32
Id.
33
Id. at 752.
34
Rudman & Glick, Prescriptive Gender Stereotypes and Backlash toward
Agentic Women, supra note 18, at 758.
BORGIDA MACROED 051605.DOC
622
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
actually liked the female applicant less than the male applicant.35
More specifically, participants in the study assumed women who
demonstrated success in a masculine domain to be more hostile
than men and, in turn, less likeable.36 Interestingly, this effect was
attenuated in the female or neutral job condition.37 What is
particularly noteworthy, however, is that participants inferred that
the successful woman was interpersonally hostile, while they did
not infer that an equally qualified man was hostile toward others.38
This research demonstrates how descriptive and injunctive
stereotypes about women’s communality affect people’s
perceptions of female job applicants in the dimensions of
competence and likeability. Based on this evidence, the question
becomes whether these perceptions result in systematic
discrimination against women in terms of hirability and
promotions. In this regard, research has shown that unlikable
people are viewed as less worthy of salary increases and
promotions.39 Although this effect was found for both men and
women, it certainly has implications for agentic women who are
viewed as less likeable.
Gender differences for hiring recommendations indicate that
women should moderate their presentation according to the gender
of their evaluators. While female participants consistently
preferred a self-effacing female partner, male participants preferred
a self-promoting woman when told that they had a vested selfinterest in the performance of the applicant.40 In contrast, selfpromoting male applicants were consistently rated as very hirable,
regardless of the gender and motivation of the evaluators.
Extending these findings to gendered leadership positions, in
an experiment with a group of undergraduate students, Rudman
and Glick found patterns of discrimination against women
35
Heilman et al., Penalties for Success, supra note 20, at 419.
Id. at 423.
37
Id.
38
Id.
39
Id. at 425-26.
40
Rudman, Self-Promotion as a Risk Factor for Women, supra note 17, at
633-34.
36
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
623
competing for a masculine manager position.41 Specifically,
undergraduate participants recommended that communal female
applicants be hired for the position significantly less frequently
than they recommended the hiring of communal male applicants.
Moreover, participants with implicit gender stereotypes (of which
they were unaware) were more likely to prefer a male authority
figure.42
Although this research suggests that women will be
discriminated against as they try to enter male-dominated
professions, discrimination patterns have also been documented for
women applying to stereotypically feminine positions. Research
has demonstrated that people recommend agentic females
significantly less frequently than agentic males for a feminized
managerial job.43 This is consistent with the finding that success in
female-dominated occupations is associated with being gentle,
nurturing, and supportive.44 When women presented themselves as
possessing the agentic qualities deemed necessary for leadership
positions, they were viewed as lacking in the feminine “niceness”
presumably required for more feminine manager positions.
Notably, while communal females and males were rated
equivalently for hiring in a feminized job description, agentic
males were recommended for the job most often.45 Likewise, no
significant differences were found between agentic male and
female applicants in the masculine job condition. In combination
with the findings on the role of likeability in hiring decisions, this
41
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women, supra note 18, at 1008; Rudman & Glick, Prescriptive Gender
Stereotypes and Backlash toward Agentic Women, supra note 18, at 753.
42
Laurie A. Rudman & S. E. Kilianski, Implicit and Explicit Attitudes
toward Female Authority. 26 PERSONALITY AND SOC. PSYCHOL. BULL. 1315,
1325 (2000).
43
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women, supra note 18, at 1008; Rudman & Glick, Prescriptive Gender
Stereotypes and Backlash toward Agentic Women, supra note 18, at 753.
44
Mary Ann Cejka & Alice H. Eagly, Gender-Stereotypic Images of
Occupations Correspond to the Sex Segregation of Employment, 25
PERSONALITY AND SOC. PSYCHOL. BULL. 413, 418-19 (1999).
45
Rudman & Glick, Feminized Management and Backlash toward Agentic
Women, supra note 18, at 1008.
BORGIDA MACROED 051605.DOC
624
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
research suggests that women indeed face an impression
management dilemma when applying for jobs; they must choose
between being successful and being liked. Only when women selfpromote and behave in an overtly friendly and sociable manner do
the gender discrimination patterns seem to disappear.46 But
because men need only self-promote to be considered good job
candidates in all circumstances, the bar seems to be placed higher
for women.
III. EXAMINING THE EXTENSION OF THE SELF-PROMOTION
DILEMMA BEYOND THE LABORATORY
The question of whether the findings discussed in the previous
sections are generalizable to non-laboratory settings merits
consideration. This question of “overreaching” based on
experimental science is by no means limited to this social scientific
domain.47 But this is an especially important question in light of
organizational trends emphasizing the communal qualities of
managerial jobs (which, in turn, may work against self-promoting,
agentic women who will be perceived as violating prescriptive
expectations). Most of the evidence on the “backlash”
phenomenon has involved laboratory experiments conducted with
undergraduate participants.
There is some reason to believe, however, that women’s being
held to a higher standard for leadership positions is not limited to
laboratory or experimental settings. In their quantitative metaanalysis of forty-five studies of actual leaders drawn from
business, educational, and government organizations, Eagly,
Johannesen-Schmidt, and van Engen examined gender differences
46
Rudman & Glick, Prescriptive Gender Stereotypes and Backlash toward
Agentic Women, supra note 18, at 753.
47
D. L. Faigman & J. Monahan, Psychological Evidence at the Dawn of
the Law’s Scientific Age, 56 ANN. REV. OF PSYCHOL. 631, 640 (2005)
[hereinafter Faigman & Monahan, Psychological Evidence at the Dawn of the
Law’s Scientific Age]; Neil Vidmar, Experimental Simulations and Tort Reform:
Avoidance, Error, and Overreaching in Sunstein Et Al.’s Punitive Damages, 53
EMORY L.J. 1359, 1373 (2004).
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
625
in leadership styles.48 The study divided leadership styles into three
categories: transformational, transactional, and laissez faire.
Transformational leaders are distinguished by a more encouraging,
nurturing, and supportive interaction style with subordinates.
Notably, this style has been independently established as one of the
most effective, that is, other meta-analyses have shown positive
correlations between leaders’ effectiveness and all components of
transformational leadership.49 The less effective but more
conventional style of delegating responsibility and rewarding
satisfactory performance is defined as transactional.50 The third
and least effective style, laissez faire, is characterized by a lack of
involvement in and general failure to take responsibility for
managing.51 Investigating gender differences in the use of these
three styles, the researchers found that as compared to male
leaders, female leaders were significantly more transformational.
Additionally, men exhibited significantly higher scores for less
effective methods of leadership, namely passive management by
exception and laissez-faire.
In summary, this research suggests that, without any
information about the prior successes of an applicant, people tend
to automatically assume that male candidates are more qualified
and competent than female candidates. In order to counteract these
perceptions and their consequences, some women may selfpromote and make explicitly clear that they are exceptionally
qualified candidates and top performers in their field. However,
because agentic behaviors are viewed as contraindicative of
feminine “communal” prescriptions for behavior, these selfpromoting women are viewed as lacking in social skills and
“niceness.” Because agency does not have the same influence on
perceptions of male likeability, this may lead to discrimination in
more feminized managerial jobs that emphasize the role of
interpersonal skills. In addition, the finding that likeability is
48
Alice H. Eagly et al., Transformational, Transactional, and LaissezFaire Leadership Styles: A Meta-Analysis Comparing Women and Men, 129
PSYCHOL. BULL. 569 (2003).
49
Id. at 571.
50
Id.
51
Id.
BORGIDA MACROED 051605.DOC
626
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
associated with recommendations for salary increases and
promotions may have some interesting implications for
understanding demonstrable salary disparities between men and
women. If women want to be successful, then they must both selfpromote and emphasize their “niceness.” Unfortunately, this
suggests that female managers and executives are being held to a
different standard than their male counterparts.
Thus, the Eagly et al. meta-analysis, consistent with the
findings of this area of research, suggests that female managers
who are more likely to enact a transformational leadership style in
actual organizational settings perhaps do so because this repertoire
of effective leadership behaviors may allow them to lessen role
incongruity dilemmas in their daily organizational lives.52
CONCLUSION
To date, the courts that have accepted expert testimony on
gender stereotyping have admitted this testimony in the form of a
social framework analysis. Such an approach is established and
grounded in legal and social science scholarship,53 and its use is
not confined to employment discrimination law.54 Social
frameworks are offered to the trier of fact through expert testimony
to provide a scientifically informed context for thinking about the
matters in dispute. As Gutek and Stockdale have suggested, “[i]t is
52
Id. at 584.
E.g., Faigman & Monahan, Psychological Evidence at the Dawn of the
Law’s Scientific Age, supra note 47; John Monahan & Laurens Walker, Judicial
Use of Social Science Research, 15 L. AND HUM. BEHAV. 571 (1991); John
Monahan & Laurence Walker, Judicial Use of Social Science Research After
Daubert, 2 SHEPARD’S EXPERT & SCI. EVIDENCE Q. 327 (1994); JOHN
MONAHAN & LAURENS WALKER, SOCIAL SCIENCE IN LAW: CASES AND
MATERIALS (4th ed. Foundation Press 1998); Laurens Walker & John Monahan,
Social Frameworks: A New Use of Social Science in Law, 73 VA. L. REV. 559
(1987); Laurens Walker & John Monahan, Social Facts: Scientific Methodology
as Legal Precedent, 76 CAL. L. REV. 877 (1988).
54
Susan T. Fiske & Eugene Borgida, Social Framework Analysis As Expert
Testimony in Sexual Harassment Suits, in SEXUAL HARASSMENT IN THE
WORKPLACE: PROCEEDINGS OF NEW YORK UNIVERSITY 51ST ANNUAL
CONFERENCE ON LABOR 575 (Samuel Estreicher ed., 1999).
53
BORGIDA MACROED 051605.DOC
GENDER STEREOTYPING RESEARCH
5/23/2005 10:26 PM
627
important to point out that a social framework analysis does not
demonstrate that discrimination or harassment either does or does
not occur but provides information to help the jury determine
whether or not discrimination or harassment occurred.”55 Thus, the
focus, at least with respect to expert testimony on gender
stereotypes, is on general causation and not on establishing specific
causation.56
In pattern and practice class action sex discrimination cases,
general, but not specific, causation is the test; however, this issue is
far from resolved.57 Some of the legal opinions in cases involving
sex discrimination have referred to social framework testimony as
providing a “sound, credible theoretical framework” for thinking
about the role of gender stereotyping in a given case.58 For
example, in Beck v. Boeing Company, the federal district court
judge ruled against a motion to exclude such testimony on the
grounds that the testimony was based on sufficient facts and data,
that the testimony was the product of reliable principles and
methods, that the principles applied reliably to the facts of the case
(and that not all factors affecting gender stereotyping had to apply
to a specific case), and that general and not specific causation was
the relevant test.59 Finally, while some members of the defense bar
view expert testimony on gender stereotyping as “junk science,”
others have referred to it as “a potentially powerful theory of sex
discrimination which has particular significance for merit-based
compensation systems.”60
55
Barbara A. Gutek & Peggy Stockdale, Social Framework Analysis, in
EMPLOYMENT DISCRIMINATION LITIGATION 245 (Frank J. Landy ed., 2005).
56
Eugene Borgida, Social Framework Analysis and Employment
Discrimination (2004) (paper presented at the American Psychological Society
annual meeting, Chicago, Ill.) (on file with author).
57
E.g., Faigman & Monahan, Psychological Evidence at the Dawn of the
Law’s Scientific Age, supra note 47, at 652.
58
E.g., Jenson v. Eveleth Taconite Co., 824 F. Supp. 847, 882 (D. Minn.
1993).
59
Beck v. Boeing Company, No.C00-0301P (D. Wash. May 14, 2004)
(order denying defendant’s motion to exclude expert report, opinions, and
testimony of plaintiffs’ expert Eugene Borgida, Ph.D.).
60
Lyndsay E. Harris & James E. Boddy, Sex Discrimination Class Actions
and Merit-Based Compensation: Is Your System at Risk?, 16 EMP. L. COMMENT.
BORGIDA MACROED 051605.DOC
628
5/23/2005 10:26 PM
JOURNAL OF LAW AND POLICY
Importantly, a body of scientific knowledge, even one from the
social sciences, should not be held to a standard that requires a
complete consensus within the field of scientific inquiry for it to
play a role in educating factfinders; in fact, we would argue that
such a criterion is unrealistic and scientifically naïve in any
scientific field. Social and psychological science, like the
biological and physical sciences, are cumulative. There will always
be some inconsistencies and disagreements in a field and these
typically are empirically resolved over time and with peer review.
As discussed in this article, however, it is possible to identify areas
of comparative consensus among researchers of gender
stereotyping. Disagreements about such assessments should not
lead to the dismissal of the entire body of knowledge as
inapplicable or “junk science.” Indeed, the U.S. Supreme Court in
Daubert was correct in cautioning against the “wholesale
exclusion” of expert testimony.61 As with other areas of
psychological science, we would endorse the Court’s approach in
Daubert, which advises that “[v]igorous cross-examination,
presentation of contrary evidence, and careful instruction on the
burden of proof are the traditional and appropriate means” of
challenging admissible evidence.62
1, 1 (2004), available at http://www.mofo.com/docs/PDF/ELC0204.pdf.
61
509 U.S. 579, 596 (1993).
62
Id.
MULVEY MACROED 051805.DOC
5/23/2005 8:19 PM
ASSESSING THE LIKELIHOOD OF FUTURE
VIOLENCE IN INDIVIDUALS WITH MENTAL
ILLNESS: CURRENT KNOWLEDGE AND
FUTURE ISSUES
Edward P. Mulvey, Ph.D.∗
INTRODUCTION
The relationship between mental illness and violence is at the
nexus of numerous legal policies, ranging from involuntary civil
commitment to probation supervision, and most recently,
mandated community treatment.1 Yet it is often unclear whether
mental illness significantly increases an individual’s risk of
violence in the community and how the factor of mental illness can
be integrated into thoughtful legal policy that both protects the
community and respects individual freedoms. During the last few
decades, researchers have made considerable progress in
investigating how mental disorder might be associated with an
increased risk for violence.
In general, studies have shown a modest association between
the presence of a mental disorder and an individual’s involvement
in violence.2 In epidemiological surveys of community residents,
∗
Professor of Psychiatry and Director of the Law and Psychiatry Program,
Western Psychiatric Institute and Clinic, University of Pittsburgh School of
Medicine, Pittsburgh, Pennsylvania.
1
See generally John Monahan et al., Mandated Treatment in the
Community for People with Mental Disorders, 22 HEALTH AFF. 28, 29-31
(2003).
2
Randy K. Otto, Prediction of Dangerous Behavior: A Review and
Analysis of “Second-Generation” Research, 5 FORENSIC REP. 103, 129 (1992);
Edward P. Mulvey, Assessing the Evidence of a Link Between Mental Illness
and Violence, 45 HOSP. & COMMUNITY PSYCHIATRY 663, 663-64 (1994).
629
MULVEY MACROED 051805.DOC
630
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
individuals who reported symptoms of a type and intensity that
qualified them for a psychiatric diagnosis were also more likely to
report involvement in violence during the recall period.3
Researchers have also found in studies of discharged mental
patients that endorsement of disorder-related symptoms or
behaviors, particularly those associated with drug and alcohol use,
indicates an increased likelihood of involvement in violence.4 This
relationship between reported indicators of mental illness and
violence, however, does not mean that most people with mental
illness are violent or that most violent acts are committed by
people with mental disorders. The presence of a mental disorder
does not predict involvement in violence with a high degree of
certainty; in fact, it accounts for only about 4% of the variability
seen in reported violence. And only a small proportion of the
violence seen in a community, about 5%,5 involves individuals
with mental illness. While there is a likely association between
mental disorder and involvement in violence, determining how and
when to focus on this connection remains a daunting task.
I. THE RESEARCH ON ASSESSING THE LIKELIHOOD OF FUTURE
VIOLENCE IN INDIVIDUALS WITH MENTAL ILLNESS
Research into the relation between mental illness and future
violence has generally addressed one of three questions. First, how
do clinicians determine when someone with a mental disorder
poses a risk of violence? Knowledge of the process by which
clinicians make this determination provides necessary background
information for improving future practice. Second, how accurate
3
Bruce G. Link & Ann Stueve, Psychotic Symptoms and the Violent/Illegal
Behavior of Mental Patients Compared to Community Controls, in VIOLENCE
AND MENTAL DISORDER: DEVELOPMENTS IN RISK ASSESSMENT 137, 140-41
(John Monahan & Henry J. Steadman eds., 1994).
4
Marvin Swartz et al., Violence and Severe Mental Illness: The Effects of
Substance Abuse and Nonadherence to Medication, 155 AM. J. PSYCHIATRY
226, 230 (1998).
5
See Elizabeth Walsh et al., Violence and Schizophrenia: Examining the
Evidence, 180 BRIT. J. PSYCHIATRY 490 (2001) (discussing several studies
researching the “link between schizophrenia and violence” in the community).
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
631
are clinicians when they predict that an individual might become
violent in the near future? The amount of discretion allowed to
mental health professionals and the weight given to their clinical
determinations depend heavily on the demonstrated accuracy of
these clinical judgments. Third, what factors are related to violence
in the community by individuals with mental disorders? Clinicians
and courts can only improve their accuracy in assessing and
intervening with individuals with mental illness by looking at the
right factors for making determinations regarding the likelihood of
future violence. Each of these questions helps to lay the
groundwork for more informed and effective practice and policy in
this area.
A. How Clinicians Make Determinations of the Likelihood of
Future Violence
Surprisingly little work has been done to describe the clinical
process for assessing individuals for a likelihood of future
violence. There are a few observational studies of clinicians
making decisions in real-world settings, some studies of clinicians
making judgments about hypothesized cases, and some studies that
integrate findings from both lines of earlier investigations.6 In
general, when making determinations regarding the likelihood of
future violence by patients, clinicians appear to rely mainly on a
few straightforward factors, such as a patient’s history of violence
and a patient’s current level of disorder or hostility, rather than any
elaborate clinical formulation.7
Several researchers have proposed that clinicians follow a
conditional prediction model when assessing likely future
violence.8 In this model, the clinician uses cognitive “scripts”9 of
6
See Eric Elbogen, The Process of Violence Risk Assessment: A Review of
Descriptive Research, 7 AGGRESSION & VIOLENT BEHAV. 591, 592 (2002).
7
VERNON L. QUINSEY ET AL., VIOLENT OFFENDERS: APPRAISING AND
MANAGING RISK 141-43 (Bruce D. Sales et al. eds., 5th ed. 2003).
8
See, e.g., Edward P. Mulvey & Charles W. Lidz, Conditional Prediction:
A Model for Research on Dangerousness to Others in a New Era, 18 INT’L J.L.
& PSYCHIATRY 117, 135-38 (1995); Kirk Heilbrun, Prediction Versus
Management Models Relevant to Risk Assessment: The Importance of Legal
MULVEY MACROED 051805.DOC
632
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
how the violence might unfold and assesses the case according to
the envisioned pattern of events. For instance, an individual might
be seen as likely to be violent because he might go home, get
drunk, and beat his girlfriend. However, if this individual moves
somewhere else and attends Alcoholic Anonymous meetings, and
his girlfriend subsequently moves out of town, his likelihood of
violence might drop appreciably. Clinical determinations are rarely
straightforward approximations of the likelihood of an event
occurring; instead, they are judgments based on the perceived
likelihood of a series of supporting or inhibiting events and
conditions that might produce violence in a particular individual.
This formulation suggests that improvement in the assessment of
future violence might rest on encouraging clinicians to be explicit
about the violence they predict and to tailor treatment plans to
target those factors most likely to precipitate violence in a
particular case. Research, therefore, should be directed toward
documenting the conditions that clinicians consider when assessing
the likelihood of future violence and toward evaluating whether
these conditions are actually related to the occurrence of violence
in the community.
B. The Accuracy of Clinical Predictions of Future Violence
For years, clinicians were considered to be rather poor at
predicting future violence in individuals with mental disorders. In
general, clinicians were thought to be right a third of the time
about whether an individual with mental illness would be involved
in future violence.10 The standard conclusion was that relying on
clinical expertise was not appreciably better than flipping a coin.11
Decision-Making Context, 21 LAW & HUM. BEHAV. 347, 357-58 (1997).
9
Cf. ROGER C. SCHANK & ROBERT P. ABELSON SCRIPTS, PLANS, GOALS
AND UNDERSTANDING: AN INQUIRY INTO HUMAN KNOWLEDGE STRUCTURES
(Roger C. Schank ed., 1977).
10
See JOHN MONAHAN, THE CLINICAL PREDICTION OF VIOLENT BEHAVIOR
(1981); Otto, supra note 2, at 105-06.
11
Bruce J. Ennis & Thomas R. Litwack, Psychiatry and the Presumption of
Expertise: Flipping Coins in the Courtroom, 62 CAL. L. REV. 693, 701 (1974)
(noting that the chances of two psychiatrists agreeing on diagnosis is about 50-
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
633
However, the methodology used in the research supporting this
general conclusion was rather weak.12 For example, the groups of
patients about whom judgments were made often were not
representative of the types of patients usually seen by clinicians. In
addition, the researchers’ conclusions regarding the accuracy of
clinical predictions of future violence often were based on official
outcomes (e.g., a patient’s release from a facility or a patient’s
involuntary commitment) rather than a direct measurement of the
clinician’s evaluative process. Finally, the outcome measures of
violence in the community often were based on the official arrest
or rehospitalization of patients—a biased underrepresentation of
involvement in violence.
A rigorous field investigation conducted by Lidz, Mulvey, and
Gardner challenged and changed this general conclusion.13 In their
study, Lidz et al. asked clinicians to rate their concern about the
likelihood of future violence in a group of patients appearing in a
psychiatric emergency room.14 A group of patients who were
assessed as being at a high risk for future violence and a matched
group of patients (patients of the same age, race, gender, and
hospitalization status) were then interviewed in the community
every two months for a six-month period. Researchers asked
participants about their involvement in violence, changes in their
living situation, and their involvement in treatment. Collateral
informants, that is, persons named by the research participants as
individuals who know what is going on in their lives, were
interviewed on the same schedule. Official records were also
reviewed to capture incidents in which violence produced an arrest
or hospitalization.
The study produced several striking findings. First, patients
were involved in more violence in the community than previously
thought. Using mainly self reports, investigators found that 53% of
50).
12
Edward P. Mulvey & Charles W. Lidz, Back to Basics: A Critical
Analysis of Dangerousness Research in a New Legal Environment, 9 LAW &
HUM. BEHAV. 209, 212-14 (1985).
13
Charles W. Lidz et al., The Accuracy of Predictions of Violence to
Others, 269 JAMA 1007, 1008 (1993).
14
Id.
MULVEY MACROED 051805.DOC
634
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
the group predicted to be violent and 36% of the comparison group
reported having laid hands on another person or threatened another
individual with a weapon during the six-month follow-up period.15
Second, clinicians’ judgments about the likelihood of future
violence were more accurate than previously believed. Even after
controlling for age, gender, race, history of violence, and
disposition of the cases, clinical judgments about the likelihood of
future violence by patients generally corresponded with the
patients’ later involvement in violent incidents.16 Interestingly,
however, this accuracy only resulted when clinicians assessed the
likelihood of future violence in males, not females.17 Clinicians of
both sexes consistently underestimated the likelihood that females
in the sample would become involved in violent incidents, at least
partially because they had different conceptualizations of what
male and female violence might look like in the studied
individuals.18 Finally, these investigators found that clinicians
focused primarily on treatment-related variables (e.g., whether the
psychiatric condition of the person deteriorated and whether the
person stayed in treatment) when making determinations about the
factors that might precipitate or inhibit violent incidents.19
Clinicians also appeared to overestimate the role of medication and
illegal drug use when assessing future violence.20
These findings, coupled with reviews of the existing research,21
indicate that clinicians indeed demonstrate some appreciable
accuracy in assessing the likelihood of future violence in
15
Id. at 1008-09.
Id. at 1009.
17
Id. at 1010.
18
Jennifer Skeem et al., Gender and Risk Assessment Accuracy:
Underestimating Women’s Violence Potential, LAW & HUM. BEHAV.
(forthcoming 2005) (on file with author).
19
Edward P. Mulvey & Charles W. Lidz, The Clinical Prediction of
Violence as a Conditional Judgment, 33 SOC. PSYCHIATRY & PSYCHIATRIC
EPIDEMIOLOGY S107, S113 (1998).
20
Id. at S112.
21
See, e.g., Douglas Mossman, Assessing Predictions of Violence: Being
Accurate about Accuracy, 62 J. CONSULTING & CLINICAL PSYCHOL. 783, 78890 (1994).
16
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
635
individuals with mental illness. However, this does not mean that
clinicians are infallible or even that their conclusions are highly
accurate in most situations. Rather, it only means that clinical
judgments add a moderate amount of valid information to other
factors known about the case.
C. Factors Related to Violence in the Community in Individuals
with Mental Disorders
The value of clinical predictions of future violence may be
limited by the failure of some clinicians to systematically assess
the most relevant characteristics of their cases. If clinicians
concentrate on the “wrong” features of a person or combine
information in a biased manner, it should come as no surprise that
their predictions might be less than optimal. In order to improve
their performance, clinicians must consider sound empirical
information about the factors related to community violence and
use the most efficient methods for combining this information. A
considerable amount of recent research has been aimed at
providing this information.22
The MacArthur Risk Assessment Study is the largest recent
undertaking of this kind.23 This large-scale, multidisciplinary study
followed more than 1,000 individuals discharged from psychiatric
hospitals in three different locales (Kansas City, Missouri;
Pittsburgh, Pennsylvania; and Worcester, Massachusetts) for a
year.24 Research participants were assessed extensively during
their hospital stays and interviewed every ten weeks in the
community after their discharge. As in the Lidz et al. study,
collateral reports and official records were also collected to
provide information about changes in the research participants’
lives as well as their involvement in violence.
The research participants enrolled in this study were limited
to individuals between the ages of eighteen and forty who did not
22
See Otto, supra note 2, at 129.
See generally JOHN MONAHAN ET AL., RETHINKING RISK ASSESSMENT:
THE MACARTHUR STUDY OF MENTAL DISORDER AND VIOLENCE (2001).
24
Id. at 16-17, 147-48.
23
MULVEY MACROED 051805.DOC
636
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
have a diagnosis of retardation. Thus, the participants were
representative of a reasonable cross-section of people found in
acute care mental hospitals throughout the United States. Fiftynine percent of the sample was male, 69% was white, and 42% was
composed of individuals with a diagnosis of depression.25 Also,
more than 40% of the sampled individuals had been diagnosed
with both substance use disorders and another major mental
disorder.26
In addition to tracking individuals discharged from mental
hospitals, the researchers also conducted one-time interviews with
a sample of people in the Pittsburgh site who lived in the same
neighborhoods as the discharged patients. The researchers then
interviewed collateral informants named by these individuals and
pulled the arrest records of these community residents. Researchers
gathered this information with two purposes in mind. First, the
information would allow for a comparison between the discharged
patients and their neighbors regarding both groups’ levels of
involvement in violence. Second, it would allow for an
examination of whether the factors that predicted violence in
individuals with mental illness also predicted violence in those
without mental illness.
One of the notable findings from this study is the comparability
of the types of violence reported by individuals with mental illness
and their neighbors. The violent incidents reported by both groups
were roughly equivalent in seriousness. Moreover, the coparticipants in the violent incidents involving individuals with
mental illness closely mirrored those in violent incidents involving
community residents. About 80% to 90% of the incidents in both
groups involved family members, friends, or acquaintances; but
neither group had a high rate of engaging in violence with
strangers.27 In the sample of individuals with mental illness,
however, women were more likely than men to be involved in
25
Id. at 62, 160.
Id. at 160.
27
Henry J. Steadman et al., Violence by People Discharged from Acute
Psychiatric Inpatient Facilities and by Others in the Same Neighborhoods, 55
ARCHIVES OF GEN. PSYCHIATRY 393, 400 (1998).
26
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
637
incidents involving family and friends,28 possibly reflecting a
lifestyle more rooted in the domestic environment.
Some of the study’s most dramatic findings related to the
relationship between drug and alcohol use and violence. When the
sample of individuals with mental illness was divided into three
groups—1) those with a major mental disorder such as
schizophrenia or severe depression and a co-occurring substance
use diagnosis, 2) those with a less severe disorder and a cooccurring substance use disorder, and 3) those with a mental health
disorder, but not a substance use disorder—individuals with a
substance use disorder reported significantly more violence. The
group of individuals with just a mental health disorder reported a
level of involvement in violence that was identical to the level
reported by their neighbors.29 Substance use by both individuals
with mental illness and community residents seemed to be a strong
factor behind involvement in violent incidents. Additional
investigation into the mechanisms of substance use in individuals
with mental illness might refine treatment approaches aimed at
reducing violence.
The study also indicated that most of the violence in the group
of people with mental disorders occurred shortly after the
individuals’ discharge from the hospital.30 The number of
individuals in the sample that became involved in violent incidents
dropped off markedly after about twenty weeks in the community.
This held true even after correcting for the effects imposed by
confining certain individuals during each period. Individuals were
most at risk of being involved in a violent incident shortly after
their return to the community, indicating a need for speedy and
comprehensive community-based services for individuals likely to
become violent after their hospitalization. Involvement in
treatment also proved relevant, as individuals who attended
treatment sessions in the period after discharge demonstrated a
level of involvement in violence during the follow-up period that
28
Pamela C. Robbins et al., Mental Disorder, Violence, and Gender, 27
LAW & HUM. BEHAV. 561, 565 (2003).
29
See Steadman et al., supra note 27, at 400.
30
See MONAHAN et al., supra note 23, at 27.
MULVEY MACROED 051805.DOC
638
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
was only about a quarter that of individuals who attended no
treatment sessions during the initial period.31
Recently, the researchers connected with the MacArthur Risk
Assessment Study developed sophisticated methods for applying
their findings to the task of clinical assessment. Using what is
known as a classification tree approach,32 these researchers devised
a method for asking a series of interdependent questions to
determine an individual’s risk level for involvement in a violent
incident during the twenty weeks following discharge.33 This
methodology performs well in classifying a large proportion of the
sample (74%) into groups that have a prevalence rate of violence
either one-half or twice the expected base rate seen in the total
sample.
II. EMERGING ISSUES
Recent research has laid the groundwork for substantial
improvements in both the practice and policy of assessing the
likelihood of future violence in individuals with a mental disorder.
Work on predictive accuracy has led to calls for examination of the
conditions surrounding the type of violence envisioned by
clinicians and for more systematic efforts by clinicians to manage
the reported risks rather than simply predicting them. This
approach promotes assessments that are explicit about how
31
Jennifer Skeem et al., Psychopathy, Treatment Involvement, and
Subsequent Violence among Civil Psychiatric Patients, 26 LAW & HUM. BEHAV.
577, 581 (2002). This held true even after controlling for the factors that
influence whether an individual will seek treatment at all (e.g., age, diagnosis,
gender, prior violence).
32
See LEO BREIMAN ET AL., CLASSIFICATION AND REGRESSION TREES
(John Kimmel et al. eds., 1984) (explaining tree methodology in data analysis);
William Gardner et al., A Comparison of Actuarial Methods for Identifying
Repetitively Violent Patients with Mental Illnesses, 20 LAW & HUM. BEHAV. 35,
36-37 (1996).
33
See Henry J. Steadman et al., A Classification Tree Approach to the
Development of Actuarial Violence Risk Assessment Tools, 24 LAW & HUM.
BEHAV. 83, 84 (2000); Steven Banks et al., A Multiple Models Approach to
Violence Risk Assessment among People with Mental Disorder, 31 CRIM. JUST.
& BEHAV. 324, 326-27 (2004).
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
639
interventions might be fashioned to minimize the occurrence of
violence. The expanded basic knowledge base provided by the
MacArthur Risk Assessment Study, meanwhile, points to several
aspects of an individual’s background or functioning that are
necessary components of any informed judgment. These findings
also highlight the need to provide integrated services in a timely
manner.
Like all good research, however, these efforts also raise new
and challenging issues. One that looms on the horizon is the use of
the construct of “psychopathy” as a predictor of future violence. A
growing number of studies, including the MacArthur Risk
Assessment Study, have found the designation of this personality
disorder to be a very potent predictor of violence in both criminal
and civil psychiatric samples.34 Practitioners and policymakers will
have to grapple with what it means for an individual to be labeled a
psychopath. Another major challenge is that of integrating
actuarial predictions into the administration of justice with regard
to individuals with mental illness. Several “products” that combine
data effectively to produce risk estimates for future violence have
emerged recently, and this trend is likely to grow. How clinicians
and the courts integrate these findings and new practices into their
operations could have important implications for the operation of
law at the “borderland of justice.”35
A. What Is Psychopathy and Why Does It Matter?
There is a solid and growing body of research indicating that
ratings of an individual’s “psychopathy” are superior to many
other measures in predicting which individuals will become violent
during the follow-up period after assessment.36 A psychopath—a
term coined in 1941 by a prominent clinician, Hervey Cleckly—is
someone who operates in a guileless fashion without regard for
34
Robert D. Hare et al., Psychopathy and the Predictive Validity of the
PCL-R: An International Perspective, 18 BEHAV. SCI. & L. 623, 628 (2000).
35
FRANCIS ALLEN, THE BORDERLAND OF CRIMINAL JUSTICE viii (1964)
(referencing the use of the criminal justice system for administering social
services).
36
See Hare et al., supra note 34, at 623, 631.
MULVEY MACROED 051805.DOC
640
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
others and who lives a generally antisocial lifestyle with no
remorse. The gold standard for determining an individual’s level of
psychopathy is the Psychopathy Checklist – Revised (PCL-R).37
The PCL-R gives an individual a score by combining a set of
ratings completed by a trained professional after conducting both a
semi-structured interview and a file review. Psychopathy, as rated
by the PCL-R, is usually considered to consist of two components.
One component is an emotional detachment factor typified by a
superficial, grandiose, and deceitful attitude, and the other
component is a social deviance factor indicated by impulsiveness,
poor behavioral controls, irresponsibility, and antisocial behaviors
(such as being arrested). Individuals who score above a threshold
are considered psychopaths. These individuals are thought to have
a personality disorder that makes them view the world differently
and engage in dangerous behaviors without feeling the normal
sense of risk connected with these activities. Practitioners have
maintained for some time that psychopaths do not respond to
standard psychological treatment approaches or punishment. 38
Making sense of the findings that psychopathy predicts
violence is a trickier task than it might seem at first. Based on
common conceptions of the “bad seed,” one might easily consider
the designation of psychopaths as simply a modern-day method for
identifying character-flawed individuals whom everyone knows
are out there. Upon closer examination, however, the premises
supporting this interpretation do not hold up very well.
First, the evidence that PCL-R scores predict violence cannot
be cleanly interpreted as indicating that an individual with what
one would commonly think of as a psychopathic personality (e.g.,
someone with a remorseless, grandiose style) is more prone to
violence. The problem here is that, while the total score on the
37
ROBERT D. HARE, THE HARE PSYCHOPATHY CHECKLIST-REVISED:
MANUAL (1991) (on file with author).
38
Carl B. Gacono et al., Treating Conduct Disorder, Antisocial, and
Psychopathic Personalities, in TREATING ADULT AND JUVENILE OFFENDERS
WITH SPECIAL NEEDS 99, 111 (Jose B. Ashford et al. eds., 2001); Otto Kernberg,
The Psychotherapeutic Management of Psychopathic, Narcissistic, and
Paranoid Transferences, in PSYCHOPATHY: ANTISOCIAL, CRIMINAL, AND
VIOLENT BEHAVIOR 372, 390-91 (Theodore Millon et al., eds., 2003).
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
641
PCL-R is related to the likelihood of future violence, this
relationship appears to be driven (especially in civil psychiatric
patients) by the score of an individual on the social deviance,
rather than the emotional detachment, factor.39 This means that
irresponsibility, impulsiveness, and prior antisocial activity really
account for most of the connection between having a high PCL-R
score and engaging in violence. Importantly, high PCL-R scores do
not necessarily mean that the person fits the classic picture of a
Ted Bundie-like psychopath or sociopath. Moreover, the
stereotypic notion of a link between being a somewhat cold and
heartless person and committing violence does not seem to be
strongly supported by the data.
Second, the idea that psychopathy is a “burnt-in” character
flaw impervious to treatment is still an open question. Several
scholars who have reviewed prior studies claiming that the
treatment of psychopaths had no, or possibly a negative, effect on
subsequent violence have been skeptical about this conclusion.40 In
addition, other research using the MacArthur Risk Assessment
Study data has shown that psychopaths, although more likely to be
violent in general, are just as likely as non-psychopaths to show
reduced violence from higher levels of treatment involvement.41 A
cautious reading of the existing studies leads to the conclusion that
more systematic research on the treatment of psychopathic
individuals must be conducted before firm conclusions can be
reached.
The link between psychopathy (or at least high PCL-R scores)
and involvement in violence is overwhelming in recent research.42
39
Jennifer Skeem & Edward P. Mulvey, Psychopathy and Community
Violence among Civil Psychiatric Patients: Results from the MacArthur
Violence Risk Assessment Study, 69 J. CONSULTING & CLINICAL PSYCHOL. 358,
369 (2001).
40
See, e.g., Ronald Blackburn, Treatment of Incapacitation? Implications
of Research on Personality Disorders for the Management of Dangerous
Offenders, 5 LEGAL & CRIMINOLOGICAL PSYCHOL. 1, 15-17 (2000); Randall T.
Salekin, Psychopathy and Therapeutic Pessimism: Clinical Lore or Clinical
Reality?, 22 CLINICAL PSYCHOL. REV. 79, 95-102 (2002).
41
Skeem et al., supra note 31, at 598-99.
42
M.E. Rice et al., A Follow-Up of Rapists Assessed in a Maximum
Security Psychiatric Facility, 5 J. INT’L VIOLENCE 435, 436-37 (1990); Randall
MULVEY MACROED 051805.DOC
642
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
The challenge for clinicians and judges is to avoid the facile
interpretation of what this means. The picture is more complicated
than simply bad people doing bad things, and the next phase of risk
research must elucidate the mechanisms behind these associations
to avoid the potential harm of simplistic interpretations.
B. What Can Be Made of Findings from Actuarial Tools?
During the past decade or so, actuarial tools have become
readily available for assessing the risk of future violence in specific
populations. The best guess is that still more tools will become
available in the future. The data for constructing such tools are
being collected regularly and systematically, and the technological
advances for combining information in sophisticated ways are
becoming commonplace in the world of social science research.
People now will develop actuarial tools to predict violence because
they can and people will use them because it will be hard to justify
not using them.
There are at least three well-validated methods for assessing
the likelihood of future violence in individuals with mental illness:
the Historical Clinical Risk - 20 (HCR-20),43 the Violence Risk
Appraisal Guide (VRAG),44 and the Classification of Violence
Risk (COVR).45 There are also specialized instruments for
assessing the risk of continued domestic violence, continued sexual
offending, and violent offending in juveniles.46 Each of these
instruments combines selected bits of information about an
individual to provide an estimate of the likelihood of a certain
outcome behavior (e.g., an arrest for a violent offense) within a
T. Salekin et al., A Review and Meta-Analysis of the Psychopathy Checklist and
Psychopathy Checklist-Revised: Predictive Validity of Dangerousness, 3
CLINICAL PSYCHOL.: SCI. & PRAC. 203, 215 (1996).
43
See CHRISTOPHER D. WEBSTER ET AL., HCR-20: ASSESSING RISK FOR
VIOLENCE (version 2, 1997).
44
See QUINSEY ET AL., supra note 7, at 141-48.
45
See John Monahan et al., Prospective Test of an Actuarial Model of
Violence Risk Assessment for People with Mental Disorder, PSYCHIATRIC
SERVICES (in press).
46
See QUINSEY ET AL., supra note 7, at 155-59.
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
643
given time period after assessment. The two most common
methods for combining information about an individual are the
“regression” and the classification tree approaches.
In the regression approach, each value (or score) that an
individual has for a relevant measure is first multiplied by a preset
weight for that measure.47 The resulting “weighted” scores (the
products of the multiplication process) are summed into a total
score for the individual. This total score is then examined to see if
it falls above a cut-off score, where people above the cut-off score
are statistically more likely to engage in violence than those below
the cut-off score. The weight given to each value is determined
from prior research studies and is the weight that maximizes the
ability of that value in combination with other weighted values to
produce a total score that differentiates those who are later violent
from those who are not.
In the classification tree approach, an individual is classified as
a member of a particular group with an expected high or low
likelihood of future violence based on successive answers to
questions or scores on selected measures.48 For example, an
individual’s psychopathy score might first be considered. If the
psychopathy score exceeds a certain level, an individual might then
be asked whether she experienced physical abuse before the age of
fifteen. If the psychopathy score is below the preset level, the
individual might instead be asked about any prior arrests. At each
step of the process, an individual is asked a question, or a score on
a measure is considered, based on the score or answer at a previous
step. This process continues until an individual’s membership in a
group known to have either a high or low risk for future violence
can be established.
The regression and classification tree approaches use
different methods to achieve the same end result of assigning a risk
estimate or designation to an individual. In the regression
47
See MONAHAN, supra note 23, at 93-95. More influential measures are
weighted more heavily.
48
See generally William Gardner et al., A Comparison of Actuarial
Methods for Identifying Repetitively Violent Patients with Mental Illnesses, 20
LAW & HUM. BEHAV. 35, 36 (1996) (illustrating the classification tree
approach).
MULVEY MACROED 051805.DOC
644
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
approach, the same set of predictors is assumed to apply to all of
the individuals being classified (e.g., individuals being discharged
from a hospital), and the full set of measures is applied and scored
to get a total score indicating risk of future violence. In the
classification tree approach, certain measures are only relevant for
some subgroups of individuals (e.g., people who score high on
psychopathy are asked different questions than those who score
low on psychopathy). The assumption is that there are multiple
combinations of scores that might put a person into a high risk
category.
Despite these slightly different methods, all actuarial
approaches still strive to assign some level of likelihood for future
violence to an individual based on the consistent use of an
algorithm for combining “objective” information about that
individual. Given the consistency of the methods used for
combining information and the verifiable nature of much of the
information combined, actuarial methods are often considered
more reliable and valid than unaided clinical judgments for
determining things such as the likelihood of future violence.49 It is
important to remember, however, that actuarial methods have their
own limitations, and these affect how such instruments might be
applied effectively in the decision-making process of the courts.
First, it is important to keep in mind that actuarial instruments
show a decrement in performance when they are applied outside of
the context in which they were developed or to individuals unlike
those upon which the measures were initially based. When data are
presented about how well an actuarial instrument performs in
terms of identifying those who will later become violent, these
estimates are almost always the best figures one might ever
achieve with the actuarial instrument in question. This is because
any actuarial instrument (whether it uses a regression or
classification tree approach) is developed using an “optimization”
procedure. The algorithm behind the actuarial instrument is
49
William M. Grove & Paul E. Meal, Comparative Efficacy of Informal
(Subjective, Impressionistic) and Formal (Mechanical, Algorithmic) Prediction
Procedures: The Clinical-Statistical Controversy, 2 PSYCHOL., PUBLIC POL’Y &
L. 293, 315-16; John Swets et al., Psychological Science Can Improve
Diagnostic Decisions, 1 PSYCHOL. SCI. PUB. INT. 1, 10-11.
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
645
calculated to combine variables to get the most accurate
predictions on the data set examined. The weights given to
different measures, the cut points chosen, or even the variables
considered are all determined by their power for differentiating
violent from nonviolent individuals in the research sample. When
a different sample is examined, these assigned values might, but in
almost all cases will not, be the best ones to use for differentiating
the violent and nonviolent in the new group. This result is a
statistical regularity in that there is always “shrinkage” in
performance when an algorithm is applied to a new sample. The
greater the discrepancy between the sample on which the actuarial
instrument is applied and the sample on which it was validated, the
less one can rely on the estimates derived.
Any estimate from an actuarial instrument, therefore, must first
be examined for its relevance to the individual being examined.
One can have more confidence in the estimate produced by the
instrument if that individual being examined is in the same
situation as the subjects in the sample used to develop the
instrument were (e.g., being discharged from a mental hospital)
and looks like the subjects who constituted the sample on which
the instrument was developed (e.g., the subject is a white
Canadian). If there are wide discrepancies between the sample and
the subject, then one must question the validity of the actuarial
estimate.
Finally, it is useful to remember that actuarial instruments are
not panaceas for hard judgments. They can work effectively as
adjuncts to, rather than replacements for, clinical judgment. It is
often tempting to believe that the hard numbers or clear categories
produced by an actuarial instrument must be more precise than the
often rambling conjectures of a mental health professional. Always
choosing one over the other, however, is probably bad practice,
regardless of the direction of the ultimate decision.
What actuarial instruments provide, when applied
appropriately, is a validated estimate of how people who look like
the individual in question will usually act. This does not, however,
mean that the person assessed will act like the average person.
Clinical insights about factors possibly not considered in the
actuarial instrument are all valuable things for a decisionmaker to
MULVEY MACROED 051805.DOC
646
5/23/2005 8:19 PM
JOURNAL OF LAW AND POLICY
know.50 Actuarial instruments prove most valuable when they are
used as a starting point for, or one component of, a reasoned
clinical formulation. The best aspects of both actuarial and clinical
judgment can be obtained if these two approaches are integrated
thoughtfully rather than pitched against each other.
CONCLUSION
This article has provided a short review of the current status of
research on predicting future violence in individuals with mental
illness. Investigators have made considerable progress in recent
decades in fleshing out the association between violence and
mental disorder, but there is obviously much more work to be
done. We now know that violence is more commonplace among
individuals with mental illness than previously thought, that
clinicians have some identifiable accuracy in assessing the
likelihood of future violence, that substance use rather than mental
disorder alone seems to drive the association between mental
illness and violence, and that individuals with mental illness are at
an increased risk for involvement in violence shortly after hospital
discharge. These pieces of information can move us toward more
reasoned practice and policy regarding interventions with violent
individuals with mental illness.
At the same time, recent research raises two issues that will test
the thoughtfulness of practitioners and policymakers in the near
future. The concept of psychopathy has come to the fore as a factor
related to violence, and it is clearly an idea that can be
misinterpreted and misapplied in the legal system. Practitioners
and policymakers will have to be especially diligent in their efforts
to clarify what this construct actually means when used in the
research world before they hurry to conclusions about its
application in the delivery of justice. Similarly, actuarial
instruments will continue to appear in legal proceedings, often with
less scrutiny than they might deserve. Each of these developments
50
Examples of factors that might not be considered are why one person is
not like those usually assessed with the instrument or how an individual might
respond to the life situations confronting her.
MULVEY MACROED 051805.DOC
PREDICTING FUTURE VIOLENCE
5/23/2005 8:19 PM
647
will test the thoughtfulness of clinicians, judges, and policymakers
as they address the complicated situations confronting them daily.
Hopefully, a healthy dialogue with the research community can
enrich this process.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
VICTORIES FOR PRIVACY AND LOSSES
FOR JOURNALISM?
FIVE PRIVACY CONTROVERSIES FROM
2004 AND THEIR POLICY IMPLICATIONS
FOR THE FUTURE OF REPORTAGE
Clay Calvert∗
INTRODUCTION
A primary role of the press in a democratic society, protected
broadly in the United States under the First Amendment,1 is to
fairly, truthfully, and comprehensively report to citizens on matters
of public concern.2 The ethics code of the Radio-Television News
∗ Associate Professor of Communications & Law and Co-Director of the
Pennsylvania Center for the First Amendment at The Pennsylvania State
University. B.A., 1987, Communication, Stanford University; J.D. (Order of the
Coif), 1991, McGeorge School of Law, University of the Pacific; Ph.D., 1996,
Communication, Stanford University. Member, State Bar of California. The
author thanks David Johnson of the McGeorge School of Law for his editing
assistance and review of a draft of this manuscript.
1
The First Amendment to the United States Constitution provides in relevant
part that “Congress shall make no law . . . abridging the freedom of speech, or of
the press.” U.S. CONST. amend. I (emphasis added). The Free Speech and Free
Press Clauses have been incorporated through the Fourteenth Amendment Due
Process Clause to apply to state and local government entities and officials. See
Gitlow v. New York, 268 U.S. 652, 666 (1925).
2
The code of ethics of one major journalism organization provides:
Members of the Society of Professional Journalists believe that public
enlightenment is the forerunner of justice and the foundation of
democracy. The duty of the journalist is to further those ends by
seeking truth and providing a fair and comprehensive account of events
and issues. Conscientious journalists from all media and specialties
strive to serve the public with thoroughness and honesty.
649
CALVERT MACROED CORRECTED 060605.DOC
650
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Directors Association, for instance, advances a policy that
admonishes broadcast journalists to “pursue truth aggressively and
present the news accurately, in context, and as completely as
possible.”3 In addition to this pivotal role as truthteller, the press
plays a vital function as a watchdog or unofficial Fourth Estate,
checking and exposing government abuses of power.4
There are, however, social and legal concerns at loggerheads
with policies that permit, privilege, and promote an aggressive
press. In particular, privacy, which Professor Anita Allen recently
described as “a dominant theme in public policy in the United
States,”5 often conflicts and competes with the jobs of journalists
by denying them access to information or images that the public is
interested in receiving.6 For instance, the common law tort of
intrusion into seclusion restricts journalists’ ability to gather
information while safeguarding individual privacy of both space
and action.7 California’s anti-paparazzi law also protects privacy
Society of Professional Journalists’ Code of Ethics, available at http://www.
spj.org/ethics_code.asp (last visited Jan. 5, 2005).
3
Radio-Television News Directors Association Code of Ethics and
Professional Responsibility, available at http://www.rtnda.org/ethics/coe. shtml
(last visited Jan. 5, 2005).
4
As Columbia University Professor Herbert Gans recently wrote, the
watchdog role represents “the journalists’ finest opportunity to show that they
are working to advance democracy.” HERBERT J. GANS, DEMOCRACY AND THE
NEWS 79 (2003). Whether the press actually plays this role today, however, is
up for debate. See Marty Kaplan, The Armstrong Effect, DAILY VARIETY, Jan.
19, 2005, at 60 (contending that “[b]y and large, neither politicians nor
entertainment executives regard the press as a check on the abuse of power, or
as the representatives of the public. They regard journalists as nuisances—useful
idiots”).
5
Anita L. Allen, 2003 Daniel J. Meador Lecture: Privacy Isn’t Everything:
Accountability as a Personal and Social Good, 54 ALA. L. REV. 1375, 1375
(2003).
6
See generally Sandra F. Chance, The First Amendment in the New
Millennium: How a Shifting Paradigm Threatens the First Amendment and Free
Speech, 23 U. ARK. LITTLE ROCK L. REV. 169, 174-75 (2000) (describing a
number of legislative initiatives and policies designed to protect privacy that
negatively affect journalists’ ability to gather information).
7
See, e.g., Marich v. MGM/UA Telecommunications, Inc., 7 Cal. Rptr. 3d
60 (Cal. Ct. App. 2004) (describing the elements of the tort of intrusion into
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
651
against journalists who might engage in either physical or
constructive trespasses to obtain images of, as the statute puts it,
“personal or familial activities.”8 In brief, as the author of this
article and a colleague argued elsewhere in 2004, “the First
Amendment is not a license to destroy an individual’s privacy.”9
On the other hand, journalists often assert and claim privacy
interests of their own when gathering news, such as keeping
private and confidential the names of sources who have supplied
them with important information. As Eugene Volokh, a professor
of law at the University of California, Los Angeles (UCLA),
recently wrote, “[T]ips from confidential sources often help
journalists (print or electronic) uncover crime and misconduct. If
journalists had to reveal such sources, many of these sources
would stop talking.”10
This article examines five separate issues and controversies
that arose in 2004, each involving privacy, and analyzes their
potential impact on the practice and policies of journalism and the
free flow of information to the public in 2005 and beyond. In one
instance, as this article argues, the odds of a particular privacy
concern immediately impacting journalists are minimal. On the
other hand, the implications of the other four privacy controversies
addressed here may well have a profound and lasting effect on
reporters and the press, and on how they perform the democratic
duties mentioned at the start of this article.11
seclusion).
8
CAL. CIV. CODE § 1708.8 (2004).
9
Robert D. Richards & Clay Calvert, Suing the Media, Supporting the First
Amendment: The Paradox of Neville Johnson and the Battle for Privacy, 67
ALB. L. REV. 1097, 1109 (2004).
10
Eugene Volokh, You Can Blog, But You Can’t Hide, N.Y. TIMES, Dec. 2,
2004, at A39.
11
See supra notes 2-3 and accompanying text (describing the truth-seeking
and truth-telling obligations of journalists imposed by ethics codes).
CALVERT MACROED CORRECTED 060605.DOC
652
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Drawing from an eclectic mix of both legislative measures and
courtroom decisions, the five privacy issues addressed in this
article are:
1. The federal Video Voyeurism Prevention Act of 2004,12
signed into law on December 23, 2004 by President George
W. Bush,13 and impacting the use of miniature cameras,
camera phones, and video recorders in public places;
2. The U.S. Supreme Court’s March 2004 decision in
National Archives and Records Administration v. Favish,14
which significantly expands a privacy exemption of the
federal Freedom of Information Act (FOIA)15 in order to
protect the privacy interests of relatives of the dead;
3. The Supreme Court of Colorado’s July 2004 opinion in
Colorado v. Bryant16 protecting the privacy interest of a
complaining witness in a sexual assault case over the
media’s right to disseminate truthful and lawfully obtained
information about an event of public significance involving
basketball superstar Kobe Bryant, and the efforts of
California trial court judge Rodney S. Melville to keep
private numerous details of the sexual assault case pending
against music superstar Michael Jackson;17
12
18 U.S.C. § 1801 (2005); Pub. L. No. 108-495.
See generally Bruce Alpert, La. Victim Hails Voyeur Law, TIMESPICAYUNE (New Orleans), Dec. 30, 2004, at 1 (describing the new law and some
of the reasons for its enactment).
14
541 U.S. 157 (2004).
15
5 U.S.C. § 552 (2004).
16
94 P.3d 624 (Colo. 2004), stay denied, 125 S. Ct. 1 (2004).
17
See generally Adam Liptak, Privacy Rights, Fair Trials, Celebrities and
the Press, N.Y. TIMES, July 23, 2004, at A20 (writing that “in the Jackson case,
Judge Rodney S. Melville of Santa Barbara County Superior Court has issued a
series of orders barring the release of essentially all information concerning
evidence and potential witnesses’ identities”).
13
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
653
4. A federal district court’s August 2004 decision in
Turnbull v. American Broadcasting Companies18 refusing
to grant summary judgment for the media defendants on
multiple causes of action based on the surreptitious
recording of conversations by a producer for the television
newsmagazine 20/20; and
5. Multiple battles across the United States in 2004
involving the efforts of journalists, such as Matthew
Cooper,19 Judith Miller,20 and James Taricani,21 to keep
private the names of confidential sources after those
journalists and their news agencies were called upon in
court to reveal the sources’ identities.
When considered collectively, the five subjects analyzed in this
article, which cut across the privacy landscape, reveal a startling
and disturbing finding for working journalists—that the privacy
interests of others either prevailed or were expanded in all of the
above situations, with the lone exception being when journalists
18
32 Media L. Rep. (BNA) 2442 (C.D. Cal. 2004).
In re Special Counsel Investigation, 346 F. Supp. 2d 54 (D.D.C. 2004)
(denying the request of Time magazine reporter Matthew Cooper to quash
subpoenas issued by Special Counsel Patrick Fitzgerald as part of an ongoing
investigation into the potentially illegal disclosure of the identity of CIA covert
operative Valerie Plame).
20
In re Special Counsel Investigation, 338 F. Supp. 2d 16 (D.D.C. 2004)
(denying the motion of New York Times investigative reporter Judith Miller to
quash grand jury subpoenas served upon her as part of the ongoing investigation
into the potentially illegal disclosure of the identity of CIA covert operative
Valerie Plame).
21
In re Special Proceedings, 373 F.3d 37 (1st Cir. 2004) (affirming and
upholding a civil contempt order against Taricani for refusing to give up the
name of the individual who leaked to him a surveillance videotape). See Pam
Belluck, Reporter Who Shielded Source Will Serve Sentence at Home, N.Y.
TIMES, Dec. 10, 2004, at A28 (describing how James Taricani, an awardwinning reporter for a Rhode Island television station, was sentenced “to six
months of home confinement for refusing to reveal who gave him an F.B.I.
videotape that was evidence in a investigation of government corruption in
Providence”).
19
CALVERT MACROED CORRECTED 060605.DOC
654
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
themselves asserted privacy interests in the name of gathering and
disseminating news. Parsed differently and more bluntly, privacy
rights grew except when journalists needed them to grow. The year
2004, in brief, was not a good one for journalists when confronting
legislative and judicial concerns regarding privacy interests. But
because the issues discussed here will not disappear in 2005 and
beyond, journalists and news agencies must learn from these
negative results and adopt new policies and approaches both to
newsgathering and the judicial challenges they encounter.
I. PRIVACY IN PUBLIC PLACES?: THE VIDEO VOYEURISM
PREVENTION ACT OF 2004
In 2000, the author of this article first identified a growing
problem caused by the development of miniaturized technology
that had outpaced the current state of privacy law—the deviant and
prurient use of tiny, backpack-hidden cameras in public places to
take pictures underneath the skirts or dresses of girls and women,
and the posting of those photographs, descriptively known as
upskirts, on the World Wide Web.22 The problem from a legal
perspective was that most of the upskirting occurred in public
places—malls and parks—where, under traditional legal policy,
people simply do not possess a reasonable expectation of privacy.23
As I elaborated in a book at that time:
Backpacks and bags are the precise kind of tool employed
by many so-called upskirt voyeurs. They drop a backpack
near the feet of a woman standing in line and then hope that
the covert camera, buried within but with its lens
unobstructed and pointing upward, gets a crisp shot of the
woman’s underwear or lack thereof.24
Today, five years later, the problem of upskirt voyeurism has
22
Clay Calvert & Justin Brown, Video Voyeurism, Privacy, and the
Internet: Exposing Peeping Toms in Cyberspace, 18 CARDOZO ARTS & ENT.
L.J. 469 (2000).
23
Id. at 489.
24
CLAY CALVERT, VOYEUR NATION: MEDIA, PRIVACY, AND PEERING IN
MODERN CULTURE 126 (2000).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
655
not gone away but, in fact, has been exacerbated by the rapid
growth and development of camera phones.25 As a reporter for one
newspaper observed in August 2004, “Some people have used
camera phones as technological peepholes to take photographs up
women’s skirts and post them on a growing list of upskirt and
voyeurism Web sites.”26 This statement foreshadowed a typical
instance of alleged upskirt voyeurism with regard to which
criminal charges were filed in December 2004 in Washington
State.27 In that case, a man allegedly used “a cellular telephone
camera to attempt to take pictures up a 16-year-old girl’s skirt
while she stood in a grocery checkout line.”28 According to police
accounts, the suspect, Patrick Donald Armour,
knelt down behind the girl, ostensibly to reach for a candy
bar on a bottom shelf, and held a camera-equipped cellular
phone under the girl’s skirt. A woman standing in the
checkout line ahead of the girl said she looked back and
saw Armour place his camera-phone under the girl’s skirt
twice.29
Sadly, the incident is not rare. On the other side of the country,
in Florida, a man faced criminal charges in July 2004 for allegedly
“using his camera phone to snap a picture underneath a 14-year-old
girl’s skirt.”30 That incident took place in a mall—a public place
where, traditionally, one has no reasonable expectation of
privacy—as did an incident in Texas in 2004, in which police
25
See Pui-Wing Tam, Entreaty to Camera-Phone Photographers: Please
Print, WALL ST. J., Dec. 28, 2004, at B1 (writing that “[s]ales of camera phones
outstripped stand-alone digital cameras for the first time in 2003. This year,
research firm IDC expects 186.3 million camera phones to be sold, more than
double its projected 68.8 million for digital camera sales”).
26
Meena Thiruvengadam, Privacy Issues; The Popularity of Camera
Phones Raises Concerns About Voyeurism and the Right to Take Photos in
Public Places, SAN ANTONIO EXPRESS-NEWS, Aug. 4, 2004, at 1E.
27
John Craig, Up-Skirt Voyeurism Alleged; Newport Man Faces Charge in
Checkout Line Incident, SPOKESMAN REV. (Spokane, Wash.), Dec. 19, 2004, at
B1.
28
Id.
29
Id.
30
Virginian Charged in Florida with Voyeurism Using Camera Phone,
ASSOC. PRESS STATE & LOCAL WIRE, July 10, 2004.
CALVERT MACROED CORRECTED 060605.DOC
656
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
“arrested a 28-year-old Houston electrician using a digital camera
to film images under the skirts of girls as young as 10 at a
Woodlands Mall department store.”31 Such cases are more difficult
and troubling from a legal perspective than those that take place in
bathrooms and changing rooms, where victims have a reasonable
expectation of privacy, and thus, redress is possible.32
But shortly before Christmas in 2004, President George W.
Bush signed into law Senate Bill 1301, a measure designed to
punish and deter such upskirt voyeurism on federal property.33 The
Video Voyeurism Prevention Act of 2004 provides in relevant
part:
Whoever, in the special maritime and territorial jurisdiction
of the United States, has the intent to capture an image of a
private area of an individual without their consent, and
knowingly does so under circumstances in which the
individual has a reasonable expectation of privacy, shall be
fined under this title or imprisoned not more than one year,
or both.34
More importantly, the new law radically changes the traditional
legal tenet that a person does not possess a reasonable expectation
of privacy in a public place. In particular, the Act defines the
phrase “under circumstances in which that individual has a
reasonable expectation of privacy” to include “circumstances in
which a reasonable person would believe that a private area of the
individual would not be visible to the public, regardless of whether
that person is in a public or private place.”35
31
Charlie Bier, Digital Technology a Boon to Criminals, HOUS. CHRON.,
July 15, 2004, at This Week 1.
32
While the focus of this part of the article is on voyeurism in public
places, violations of privacy through the use of miniature cameras continued in
2004 in more private places such as bathrooms. See, e.g., Michael A. Scarcella,
Detectives Trying to ID Voyeur Victims, SARASOTA HERALD-TRIB., Dec. 24,
2004, at B1 (describing a case in Florida in which authorities allege that “[f]rom
a bathroom ceiling at a local gymnastics studio, the video camera rolled as the
teenage girls undressed”).
33
S. 1301, 108th Cong. (2004) (enacted).
34
18 U.S.C. § 1801 (2005).
35
Id. (emphasis added).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
657
It is the italicized portion of the Act cited above that breaks
with the traditional policy by granting people privacy, at least
when it comes to certain areas on their bodies,36 even if they are in
public places. The phrase “public privacy” thus is no longer an
oxymoron. In adopting a new policy of what might be considered
bodily privacy in public places, Congress has essentially borrowed
the reasoning of both the drafters of the Restatement (Second) of
Torts from a quarter-century ago and the Alabama Supreme Court
in Daily Times Democrat v. Graham37 four decades past. In
particular, a comment by the drafters of the Restatement provides
that “even in a public place . . . there may be some matters about
the plaintiff, such as his underwear or lack of it, that are not
exhibited to the public gaze; and there still may be an invasion of
privacy when there is intrusion upon these matters.”38 Similarly, in
Daily Times Democrat, the Alabama high court ruled in favor of a
woman, Flora Bell Graham, who was photographed in public
outside of an amusement fun house as air jets blew up her skirt,
exposing her “from the waist down, with the exception of that
portion covered by her ‘panties.’”39 In protecting Graham, the
court wrote:
Where the status he [the plaintiff] expects to occupy is
changed without his volition to a status that is embarrassing
to an ordinary person of reasonable sensitivity, then he
should not be deemed to have forfeited his right to be
protected from an indecent and vulgar intrusion of his right
to privacy merely because misfortune overtakes him in a
public place.40
The Video Voyeurism Prevention Act of 2004 in essence
recognizes the “misfortune” that overtakes the victims of high-tech
Peeping Toms in public places and the related embarrassment and
36
The Video Voyeurism Prevention Act defines the protected private areas
of the body to include “the naked or undergarment clad genitals, pubic area,
buttocks, or female breast of that individual.” Id.
37
162 So. 2d 474 (Ala. 1964).
38
RESTATEMENT (SECOND) OF TORTS § 652B cmt. c (1977) (emphasis
added).
39
Daily Times Democrat, 162 So. 2d at 476.
40
Id. at 478.
CALVERT MACROED CORRECTED 060605.DOC
658
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
harm that ensues from such intrusions. But how might the new law
impact journalists and their use of hidden cameras in investigative
reports when they approach people in public places? The
journalistic use of hidden cameras in public places, after all, is not
an uncommon target of invasion of privacy lawsuits.41
On its face, the Video Voyeurism Prevention Act of 2004
should have little or no effect on journalists who use hidden
cameras in public places. Why? Because the new law does not
forbid all uses of hidden cameras in public places, but only their
use to film or record “a private area of an individual without their
consent.”42 Broadcast journalists who use hidden cameras would
rarely try to capture images of the private area of a person; rather,
they would more likely attempt to capture images of deceit, graft,
and other wrongdoings by individuals or corporations in line with
the roles described in this article’s introduction.43 As Kevin M.
Goldberg, an attorney for the American Society of Newspaper
Editors, stated, “Theoretically, you never want to see a criminal
penalty imposed upon speech, but practically I believe this will
have little effect on the mainstream media.”44
The real risk for journalism, however, lies in the danger that
the new statute’s recognition of a privacy right in a public place
will be expanded by future legislation to apply to scenarios and
situations beyond those of upskirt voyeurism. If federal law now
officially recognizes a right to bodily privacy in crowded public
41
See, e.g., Deteresa v. ABC, 121 F.3d 460 (9th Cir. 1997), cert. denied,
523 U.S. 1137 (1998) (involving several privacy-based causes of action for,
among other things, the use of a hidden video camera by an ABC employee to
videotape an individual, Beverly Deteresa, without her knowledge from a public
street as she stood on the doorstep of her condominium); Wilkins v. NBC, 84
Cal. Rptr. 329 (Cal. Ct. App. 1999) (involving an unsuccessful claim for
intrusion into seclusion based upon the hidden camera videotaping by producers
for NBC’s investigative newsmagazine, Dateline, at an outdoor patio table at a
crowded public restaurant in Malibu, California).
42
18 U.S.C. § 1801(a) (2005).
43
See supra notes 2-4 (describing the aspirational roles and goals of the
press).
44
Reporters Committee for Freedom of the Press, Congress Approves
Criminal “Video Voyeurism” Law, available at http://www.rcfp.org/news/
2004/0923s1301b.html (last visited Oct. 4, 2004).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
659
places such as malls, then what is to prevent it in the future from
recognizing conversational privacy in crowded public places?
People have intimate conversations in public places, after all, that
they may not believe others can hear.
What is more, groups such as the American Civil Liberties
Union (ACLU) already claim that the current use in several major
cities of security cameras that capture images of people’s faces in
outdoor public spaces violates individual privacy rights.45 In other
words, the Video Voyeurism Prevention Act of 2004 may have
opened the floodgates for the expansion of other privacy interests
in public places. In turn, it may affect the use of surveillance
cameras outdoors in public places.46 Whether this ultimately
occurs, however, remains to be seen. In the meantime, the new law
and its expansion of privacy rights to public places should have no
impact on the work of professional journalists and their camera
people.
45
See Mark F. Bonner, Parish Gets Money for Street Cameras; ACLU’s
Concerns Fail to Dissuade Sheriff, TIMES-PICAYUNE (New Orleans), July 24,
2004, at 1 (writing that a plan to install surveillance cameras in an area near
New Orleans, Louisiana, has “drawn opposition from the American Civil
Liberties Union,” and quoting Joe Cook, executive director of the ACLU in
Louisiana, for the proposition that the cameras constitute a “reckless gamble of
privacy rights that wastes tax dollars”); Doug Donovan, Camera System
Expands in City, BALT. SUN, Dec. 2, 2004, at 1B (describing the objections of
the ACLU to a network of 24-hour surveillance cameras in the Inner Harbor
district of Baltimore, Maryland, and stating that the “American Civil Liberties
Union opposes the Baltimore network, saying the camera system infringes on
privacy rights and are [sic] ineffective in fighting crime or terrorism”); Jessica
Garrison, Cameras to Keep Watch in Hollywood, L.A. TIMES, Oct. 28, 2004, at
A1 (citing the objections of the American Civil Liberties Union to the
installation of surveillance cameras on public city streets by the Los Angeles
Police Department, and quoting Ramona Ripston, executive director of the
American Civil Liberties Union of Southern California, for the proposition that
“[t]his is creeping Big Brotherism, and it’s really disturbing. More and more, we
are losing our right to any kind of privacy”).
46
Surveillance cameras “have come under fire from the American Civil
Liberties Union, which in recent years has campaigned against them in several
cities. The organization has cited a range of objections, from skepticism about
the cameras’ ability to produce results to possible privacy violations.” Frank
Donze, Crime-Time Program, TIMES-PICAYUNE (New Orleans), Jan. 14, 2005,
at 1 (emphasis added).
CALVERT MACROED CORRECTED 060605.DOC
660
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
II. EXTENDING STATUTORY PRIVACY RIGHTS TO RELATIVES OF THE
DEAD: THE LASTING LEGACY OF VINCENT FOSTER
While the Video Voyeurism Prevention Act of 2004 created a
new federal statutory privacy right and policy with regard to
certain images of people captured in public places, the U.S.
Supreme Court’s 2004 decision in National Archives and Records
Administration v. Favish47 stretched and expanded an existing
federal statutory right of “personal privacy”48 to apply to a
“decedent’s family when the family objects to the release of
photographs showing the condition of the body at the scene of
death.”49 In brief, the case transformed a right of personal privacy
into a familial or relational privacy right, at least in relation to
death-scene photographs. As an editorial in the Plain Dealer in
Cleveland, Ohio, argued, “The high court effectively ripped out of
the Freedom of Information Act a great chunk of the public’s right
to know what its government is doing in its name.”50 The
implications for journalists of the Favish opinion, as this part of the
article argues, are both serious and far reaching. To understand
those implications, it is first necessary to understand the facts and
issues of the case.
The case centered on the efforts of Allan Favish, a California
attorney, to obtain government-taken, death-scene photographs of
Vincent Foster, Jr., former deputy counsel to President Clinton.51
Although numerous government investigations concluded that the
shooting of Foster was a suicide, Favish doubted their findings,
and he thus made a request for the photographs under the federal
Freedom of Information Act (FOIA) in order to determine for
himself what might have really happened.52 Under FOIA, any
person may request copies of records from a federal government
47
541 U.S. 157 (2004).
5 U.S.C. § 552(b)(7)(C) (2005).
49
Favish, 541 U.S. at 160.
50
A Feel-Bad Ruling; Supreme Court’s Emotions Get in the Way of Its
Judgment in Vince Foster Case, and the Public’s Rights Suffer, PLAIN DEALER
(Cleveland, Ohio), Apr. 3, 2004, at B8.
51
Favish, 541 U.S. at 160-61.
52
Id. at 161.
48
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
661
agency and the agency must supply it unless the information falls
within one of nine statutorily defined exempt areas.53 In this case,
the government agencies that held the Foster photographs at one
time or another included: 1) the National Park Service, as U.S.
Park Police had taken the photographs of Foster; 2) the Office of
Independent Counsel (OIC), which, under both Robert Fiske and
Kenneth Starr, had investigated Foster’s death and concluded it
was a suicide; and 3) the National Archives and Records
Administration, which took possession of the photographs at the
conclusion of the OIC’s investigation.54
Foster’s immediate relatives, however, objected to Favish’s
request for the death-scene photographs, asserting, as the Supreme
Court noted, “their own refuge from a sensation-seeking culture for
their own peace of mind and tranquility.”55 Specifically, Sheila
Foster Anthony, sister of Vincent Foster, wrote in an affidavit that
the release of the death-scene images of her late brother “would
constitute a painful unwarranted invasion of my privacy, my
mother’s privacy, my sister’s privacy, and the privacy of Lisa
Foster Moody (Vince’s widow), her three children, and other
members of the Foster family.”56
Foster’s relatives asserted that Exemption 7(C) of FOIA
prevented the release of the photographs. This exemption prevents
and shields the disclosure of records or information compiled for a
law enforcement purpose that “could reasonably be expected to
constitute an unwarranted invasion of personal privacy.”57 The
battle in the U.S. Supreme Court hinged initially on the meaning of
the term “personal privacy,” with Favish emphasizing the
importance of the word “personal” and asserting that the term
should be narrowly construed to mean “the right to control
information about oneself.”58 The self in this case was Foster, and,
so went the argument of Favish, since Foster was dead, he could
53
MARC A. FRANKLIN ET AL., MASS MEDIA LAW: CASES AND MATERIALS
670 (6th ed. 2000).
54
Favish, 541 U.S. at 160-64.
55
Id. at 166.
56
Id. at 167.
57
5 U.S.C. § 552(b)(7)(C) (1995).
58
Favish, 541 U.S. at 165.
CALVERT MACROED CORRECTED 060605.DOC
662
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
not assert his own privacy interest.59 Stated differently, Favish
contended that “the individual who is the subject of the
information is the only one with a privacy interest.”60
Foster’s immediate relatives, however, asked the nation’s high
court to adopt a much broader construction of the term “personal
privacy” that would include the decedent family’s privacy
interests.61 The Supreme Court sided with the Foster family,
adopting an expansive interpretation of personal privacy and
noting that “the statutory privacy right protected by Exemption
7(C) goes beyond the common law and the Constitution.”62 Citing
a very odd mix of authorities in support of its conclusion—the
Encyclopaedia Britannica and a Greek drama by Sophocles,
among others—Justice Anthony Kennedy wrote for a unanimous
Court that “[f]amily members have a personal stake in honoring
and mourning their dead and objecting to unwarranted public
exploitation that, by intruding upon their own grief, tends to
degrade the rites and respect they seek to accord to the deceased
person who was once their own.”63 Kennedy added that the Court
had “little difficulty . . . in finding in our case law and traditions
the right of family members to direct and control disposition of the
body of the deceased and to limit attempts to exploit pictures of the
deceased family member’s remains for public purposes.”64 He
noted that the “well-established cultural tradition acknowledging a
family’s control over the body and death images of the deceased
has long been recognized at common law.”65
In expanding the term “personal privacy” to sweep in family
members of the deceased and in refusing to limit the term to
individuals who are the subjects of the information or images in
question, the Supreme Court also invoked a parade-of-horrors
argument. In particular, it attempted to demonstrate the evils that
might result if it ruled against Vincent Foster’s family:
59
60
61
62
63
64
65
Id.
Id.
Id. at 166-67.
Id. at 170.
Favish, 541 U.S. at 168.
Id. at 167.
Id. at 168.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
663
We are advised by the Government that child molesters,
rapists, murderers, and other violent criminals often make
FOIA requests for autopsies, photographs, and records of
their deceased victims. Our holding ensures that the privacy
interests of surviving family members would allow the
Government to deny these gruesome requests in
appropriate cases. We find it inconceivable that Congress
could have intended a definition of “personal privacy” so
narrow that it would allow convicted felons to obtain these
materials without limitations at the expense of surviving
family members’ personal privacy.66
If this is correct, then how could the Court’s finding possibly
harm the practice of journalism? Because the holding makes it
clear that the familial right of privacy outweighs the public’s
unenumerated First Amendment right to know, at least when
images of the dead are involved.
The Court’s logic has immediate implications for press
coverage of the ongoing war in Iraq, particularly with regard to
photographs of caskets of dead soldiers as they are flown home
and arrive in the United States. Images of dead American soldiers,
gruesome though they may be for some, are important for the
public to see because they bring home the reality of war; put more
bluntly, media images depicting the loss of life, through their
power to galvanize public opinion against a war, may save lives in
the future. But the reasoning of the Court in Favish would suggest
that the privacy rights of families would prevent the public from
viewing these images. Thus, if the Court’s privacy calculus is
extended beyond the reaches of FOIA Exemption 7(C), the Favish
opinion does not bode well for journalists who seek access to
information and images related to the human costs of war; in fact,
it is perilous precedent.
These issues and implications are far more than speculative. As
journalist Hal Bernton wrote in December 2004, it was the
publication of a photograph of flag-draped coffins carrying dead
American soldiers, lined up in the fuselage of an airplane, that
“rekindled debate about a Pentagon policy—sometimes waived in
66
Id. at 170.
CALVERT MACROED CORRECTED 060605.DOC
664
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
years past but enforced by the Bush administration—to ban media
coverage of transport of military coffins.”67 The logic in Favish,
however, supports and bolsters the U.S. Department of Defense’s
thirteen-year-old policy that “has banned photographs and videos
that show the flag-draped coffins of American soldiers”68 as they
arrive at Dover Air Force Base in Delaware. In June 2004, the U.S.
Senate defeated by a 54-to-39 vote a bill that would have permitted
news photographers access to Dover Air Force Base.69 While the
ban ostensibly is designed to protect the privacy interests of the
families of the deceased soldiers,70 it has been described by at least
one critic, The New York Times’s Maureen Dowd, as “the
Pentagon’s self-serving ban.”71
The immediate implications of Favish, which took place within
the context of a FOIA case, may well come to bear on a new
FOIA-based lawsuit, filed in late 2004, that “seeks to force the
Pentagon to release photographs and videotape of coffins of
service members killed overseas and brought back to the United
States.”72 As Meredith Fuchs, one of the attorneys involved in that
suit told a reporter, “These are the kind of documents that directly
serve the core purpose of FOIA. . . . Everyone says a picture is
worth a thousand words. Well, the pictures have an impact and
help people understand what war is really about in a way that
nothing else does.”73 The question now is whether the privacy
interests of the relatives of the deceased will trump the public’s
right to know, as served by journalists. Unfortunately, the
67
Hal Bernton, Reflecting on an Image Taken, a Choice Made, a Life
Altered, SEATTLE TIMES, Dec. 26, 2004, at A1.
68
Frank Harris III, America’s War Dead Should Be Shown, HARTFORD
COURANT, Oct. 11, 2004, at A11.
69
Sheryl Gay Stolberg, Senate Backs Ban on Photos of G.I. Coffins, N.Y.
TIMES, June 22, 2004, at A17.
70
See id. (writing that “President Bush has insisted that the policy banning
the photography protects the privacy of the families of the dead”).
71
Maureen Dowd, Wolfie’s Fuzzy Math, N.Y. TIMES, May 2, 2004, at
Section 4, 11.
72
George Edmonson, Suit Seeks Military Coffin Photos, ATLANTA J.CONST., Oct. 5, 2004, at A7.
73
Id.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
665
reasoning adopted by the Supreme Court in Favish militates
against the latter interest.
The decision in Favish also is troubling to journalists for a
second reason—a reason beyond its expansive interpretation of the
term “personal privacy.” As the Christian Science Monitor noted,
“The decision makes it more difficult for media organizations,
government watchdog groups, and historians to obtain certain
types of documents held in government files.”74 Specifically, the
decision creates evidentiary barriers and hurdles for reporters
requesting information under FOIA whenever Exemption 7(C) is
raised to block the release of that information. The Court in Favish
abrogated what it called “the usual rule that the citizen need not
offer a reason for requesting the information”75 and instead held
that
the person requesting the information [must] establish a
sufficient reason for the disclosure. First, the citizen must
show that the public interest sought to be advanced is a
significant one, an interest more specific than having the
information for its own sake. Second, the citizen must show
the information is likely to advance that interest. Otherwise,
the invasion of privacy is unwarranted.76
This two-step test, it should be noted, somewhat mirrors two
aspects of the Supreme Court’s four-part commercial speech
doctrine, created in Central Hudson Gas & Electric Co. v. Public
Service Commission.77 In particular, that doctrine requires the
government to prove that it has a “substantial” interest before it
can restrict truthful advertising for lawful products and that this
interest is directly advanced by the regulation.78 What is different,
of course, is that the Central Hudson test imposes a burden on the
government before it can restrict speech, while the Favish test,
conversely, imposes a burden on private individuals, such as
74
Warren Richey, In Vincent Foster Case, Court Upholds Privacy,
CHRISTIAN SCI. MONITOR, Mar. 31, 2004, at 3.
75
541 U.S. 157, 172 (2004).
76
Id.
77
447 U.S. 557, 566 (1980).
78
Id. at 564.
CALVERT MACROED CORRECTED 060605.DOC
666
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
journalists and members of the public at large, before they can
obtain speech.
Favish, it should be stressed, sought the death-scene
photographs not because he harbored some prurient interest or
deviant desire in them, but rather because he doubted the
credibility and accuracy of multiple government investigations into
the death of a person, Vincent Foster, who was very closely
connected with the highest ranking government official in the
country, then-President of the United States Bill Clinton. The
Court held that in such instances in which “the public interest
being asserted is to show that responsible officials acted
negligently or otherwise improperly in the performance of their
duties,” the requester of information first “must produce evidence
that would warrant a belief by a reasonable person that the alleged
Government impropriety might have occurred.”79
The problem with this from a requester’s perspective is that the
photographs or information being sought might be either the only
type of physical evidence that exists or the most important piece
available. As Lucy Dalglish, head of the Reporters Committee for
Freedom of the Press, remarked, “I don’t know how you can
expect requesters to prove a negative before they are entitled to a
record under the Freedom of Information Act.”80 Favish, the high
court ultimately concluded, had not met this burden; in fact, he had
“not produced any evidence that would warrant a belief by a
reasonable person that the alleged Government impropriety might
have occurred.”81
The Supreme Court’s requirement that courts must engage in
“a meaningful evidentiary showing”82 when FOIA exemption 7(C)
privacy concerns are raised, with the burden being placed squarely
on the requester of information, clearly elevates privacy interests
above free speech interests, including the public’s right to know.
The decision thus represents a judicial blow, struck in the name of
relational or familial privacy, against journalists’—and, by
79
Favish, 541 U.S. at 173.
Linda Greenhouse, Justices Unanimously Bar Release of Photos From
the Suicide of a Top Clinton Aide, N.Y. TIMES, Mar. 31, 2004, at A16.
81
Favish, 541 U.S. at 175.
82
Id.
80
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
667
extension, the public’s—access to government records. Although
“Favish may be a helpful precedent for persons seeking privacy
protection from an intrusive media,”83 it is decidedly damaging to
a democratic society when applied to journalistic intrusions related
to the reporting of alleged misconduct by public officials. Thus,
while it may be possible for some simply to dismiss Favish as just
another conspiracy theorist, one cannot so easily dismiss the
speech-related implications of the case and precedent that now
bear his name.
This part of the article has demonstrated how one type of
privacy concern trumped access to government-held information in
2004. One question raised by this outcome is whether the case
reveals the judiciary’s implicit assumptions about what constitutes
worthy impositions on, in contrast to unworthy prying into,
individual privacy. Two additional cases from 2004, involving
high-profile celebrities from the worlds of music and sports,
implicate this question, similarly placing the concept of privacy
squarely in the balance. The next part of this article examines these
two cases—one centering on Michael Jackson, and the other, on
Kobe Bryant—in which the privacy interests of two high-profile
celebrities outweighed, as they did in Favish, the First Amendment
interest in the right to receive and publish information about
matters of public interest.
III. PRIVACY AND SEXUAL ASSAULT CASES: JOURNALISTIC LOSSES
IN THE KOBE BRYANT & MICHAEL JACKSON CASES
When jury selection finally began in February 2005 in the
sexual molestation case against Michael Jackson, the news and
entertainment media were out in full force, with cameras and boom
microphones at the ready, to capture and cover every courtroom
entrance and exit by the so-called King of Pop. For journalists, the
opportunity to witness the trial’s daily happenings, even in such a
circus-like atmosphere, was a welcome relief from the excessive
secrecy that had cloaked the case in 2004.
83
2004).
Ronald J. Riccio, Subjecting War to the Law, 177 N.J.L.J. 321 (July 26,
CALVERT MACROED CORRECTED 060605.DOC
668
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Laurie Levenson, a professor at Loyola Law School in Los
Angeles and frequent media commentator,84 summed up the
massive sealing of documents in the case against Jackson,
remarking, “I’ve never seen a case with this level of secrecy.
You’d think we were dealing with the Pentagon Papers. Everyone
is filing papers in code and we’re on the eve of trial.”85 Her
sentiment was echoed by Dalglish, who wrote in a Sacramento
Bee-published commentary that the “Michael Jackson prosecution
has been conducted under a cloak of secrecy. Unbelievably, the
judge in the Jackson case has refused to even release the
indictment against the entertainer.”86 Beyond that, as another
newspaper observed, the “[l]awyers and investigators on all sides
of the case are barred from speaking to the media.”87
In the Michael Jackson case, which centers on ten different
felony counts related to the singer’s alleged molestation of a 13year-old leukemia patient in 2003 at his secluded ranch in Southern
California,88 Judge Rodney Melville justified the need for massive
privacy and the denial of the public’s right to know by citing the
singer’s constitutional right to a fair trial.89 As Melville remarked
in June 2004 in rejecting a motion to lift orders sealing records in
the case, “The court is trying to balance the First Amendment right
against the right to a fair trial. This defendant is known around the
84
Levenson, a former prosecutor, also is the William M. Rains Fellow and
Director of Loyola Law School’s Center for Ethical Advocacy. Loyola Law
School Web site, available at http://www.lls.edu/academics/faculty/
levenson.html (last visited Jan. 6, 2005).
85
Linda Deutsch, Court Rulings Continue to Shield Evidence in Michael
Jackson Case, ASSOC. PRESS NEWSWIRES, Jan. 4, 2005.
86
Lucy Dalglish, Courts Undermine Freedom of the Press, SACRAMENTO
BEE, Aug. 22, 2004, at E3.
87
John M. Broder, From Grand Jury Leaks Comes a Clash of Rights, N.Y.
TIMES, Jan. 15, 2005, at A8.
88
See generally Eric Slater & Steve Chawkins, Opening Statements in
Jackson Trial Today, L.A. TIMES, Feb. 28, 2005, at B1.
89
See generally Martin Kasindorf, New Set of Charges Awaits Jackson,
USA TODAY, Apr. 30, 2004, at A3 (writing that “Melville, citing the perils to a
fair trial that publicity could pose, has sealed document after document in the
case”).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
669
world and that makes it very difficult to get a fair trial.”90 The right
to a fair trial is codified in the Sixth Amendment to the U.S.
Constitution, which provides in relevant part that “[i]n all criminal
prosecutions, the accused shall enjoy the right to a speedy and
public trial, by an impartial jury of the state and district wherein
the crime shall have been committed . . . .”91
The Jackson case was not the only celebrity criminal trial in
2004, however, in which privacy interests trumped the public’s
right to know. Even more significantly, in July 2004, the Supreme
Court of Colorado upheld a prior restraint on publication against
seven media entities that had lawfully obtained accurate
information about the sexual conduct and history of the woman
who accused Los Angeles Laker Kobe Bryant of sexual assault.92
What is critical here is that while the Colorado high court
acknowledged that a “[p]rior restraint of publication is an
extraordinary remedy attended by a heavy presumption against its
constitutional validity,”93 it nonetheless concluded that the privacy
interests of the accuser, as protected by a state rape shield statute,
were sufficient to overcome this presumption and the First
Amendment interests of a free press and the public’s right to
know.94 The editors of the Denver Post opined in an editorial that
the Colorado Supreme Court’s decision was “an impermissible
encroachment on the First Amendment. It was an effort to balance
First Amendment and privacy concerns, a delicate task. But the
decision was one of flawed logic that would set a bad precedent.”95
In referencing the decision’s impact on the press, University of
Colorado Law Professor Paul Campos stated, “We’re not talking
90
Michelle Caruso, Jax Case Secrets to Stay Off-Limits, DAILY NEWS
(N.Y.), June 26, 2004, at 6.
91
U.S. CONST. amend. VI. The Sixth Amendment is also applicable to the
states by incorporation. U.S. CONST. amend. XIV, § 1.
92
People v. Bryant, 94 P.3d 624 (Colo. 2004), stay denied, 125 S. Ct. 1
(2004).
93
Id. at 628.
94
Id. at 628-32.
95
Editorial, State Court Trying to Rewrite Constitution, DENVER POST, July
21, 2004, at B-06.
CALVERT MACROED CORRECTED 060605.DOC
670
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
about a chilling effect, we’re talking about a freezing effect.”96
While the criminal case against Bryant eventually was dropped,97
the prior restraint precedent in Colorado remains on the books. It
thus is important to understand what happened in that case and
how privacy triumphed over journalists’ ability to report truthful
news of public interest.
The complex prior restraint issues in the criminal case against
Kobe Bryant all began because of simple human errors and the
push of a button on a computer. In particular, a court reporter
mistakenly emailed to seven news organizations transcripts of an
in camera pretrial proceeding conducted by the trial court judge to
determine the relevancy, if any, of the prior or subsequent sexual
conduct of the woman who accused Bryant of rape.98 The notation
“IN CAMERA PROCEEDINGS” was marked on every page of
the transcripts, which were mistakenly sent out over the Internet
“because the court reporter maintained an electronic list for media
entities subscribing to transcripts of the public proceedings in the
case.”99 The news organizations thus became the fortunate
recipients of accurate information that would either confirm or
deny rumors about the complaining witness’s sexual history that
had circulated in the court of public opinion and on the World
Wide Web.
When the error was called to the attention of trial court judge
Terry Ruckriegle, however, he “ordered the recipients to delete or
destroy their copies and prohibited them from reporting the
contents.”100 The order was immediately appealed directly to the
Colorado Supreme Court by the media outlets on the grounds that
it constituted an unconstitutional prior restraint.101 The order at
issue before the high court of Colorado provided:
96
Jeff Kass, Case Sparks Debate About Key Principles, ROCKY MTN.
NEWS (Denver), Aug. 26, 2004, at 8K.
97
Kirk Johnson, As Accuser Balks, Prosecutors Drop Bryant Rape Case,
N.Y. TIMES, Sept. 2, 2004, at A1.
98
People v. Bryant, 94 P.3d 624, 626 (Colo. 2004).
99
Id. at 627.
100
Charlie Brennan, Media Appeal Judge’s Order, ROCKY MTN. NEWS
(Denver), June 29, 2004, at 17A.
101
Id.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
671
It has come to the Court’s attention that the in camera
portions of the hearings in this matter on the 21st and 22nd
were erroneously distributed. These transcripts are not for
public dissemination. Anyone who has received these
transcripts is ordered to delete and destroy any copies and
not reveal any contents thereof, or be subject to contempt
of Court.102
The Colorado Supreme Court was forced to weigh the privacy
interests of Bryant’s accuser against the First Amendment interests
of a free press. The precedent in support of the news media was
clear. The U.S. Supreme Court held a quarter-century ago that “if a
newspaper lawfully obtains truthful information about a matter of
public significance then state officials may not constitutionally
punish publication of the information, absent a need to further a
state interest of the highest order.”103 It was a rule the Court
affirmed and applied as recently as 2001.104 As applied in the
Bryant situation, the news media had lawfully obtained the truthful
transcripts about a matter of clear public concern that had attracted
massive media attention.
Weighed against this precedent, however, was the interest “in
providing a confidential evidentiary proceeding under the rape
shield statute, because such hearings protect victims’ privacy,
encourage victims to report sexual assault, and further the
prosecution and deterrence of sexual assault.”105 Under Colorado’s
rape shield statute, the prior or subsequent sexual conduct of an
alleged sexual assault victim is presumed to be irrelevant unless
the judge determines that one of several specified exceptions
applies.106 Although the trial court judge in Bryant’s case
ultimately found that the accuser’s sexual conduct during a
seventy-two-hour period prior to her medical examination at a
hospital after the alleged assault by Bryant was relevant and
102
Bryant, 94 P.3d at 626.
Smith v. Daily Mail, 443 U.S. 97, 103 (1979).
104
See Bartnicki v. Vopper, 532 U.S. 514, 527-28 (2001) (noting how the
court has “repeatedly held” this rule).
105
Bryant, 94 P.3d at 626.
106
COLO. REV. STAT. § 13-25-131 (2004).
103
CALVERT MACROED CORRECTED 060605.DOC
672
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
admissible,107 the transcript of the in camera hearing on the issue
remained sealed.
To weigh the competing interests and to determine the
constitutionality of the trial judge’s prior restraint order, a majority
of the Colorado Supreme Court fashioned a three-part test that
asked whether:
1. the state of Colorado had an interest of the highest order
that would justify a prior restraint;
2. the restraint was the narrowest available remedy to
protect the alleged interest of the highest order; and
3. the prior restraint was “necessary to protect against an
evil that is great and certain, would result from the
reportage, and cannot be mitigated by less intrusive
measures.”108
In a 4-3 decision, a majority of the Colorado Supreme Court on
July 19, 2004 applied this three-pronged approach and upheld that
part of the trial court’s order that prevented the media from
revealing the contents of the sealed documents, emphasizing that
the state’s interests of “the highest order in this case not only
involve the victim’s privacy interest, but also the reporting and
prosecution of this and other sexual assault cases.”109 The majority
reasoned that “the harms in making these in camera judicial
proceedings public would be great, certain, and devastating to the
victim and to the state. These harms justify the remedy we fashion
in this case.”110 To assure that the order was as narrow as possible,
107
See Kirk Johnson, Judge Limiting Sex-Life Shield at Bryant Trial, N.Y.
TIMES, July 24, 2004, at A1.
[T]he judge ruled that the defense could introduce direct or
circumstantial evidence about any sexual conduct on the woman’s part
in the 72 hours preceding her physical examination by doctors at a
hospital in Glenwood Springs, Colo., on July 1, 2003. The incident
with Mr. Bryant took place on the evening of June 30, more than 12
hours earlier, at a resort hotel near Vail.
Id.
108
Bryant, 94 P.3d at 628.
109
Id. at 636.
110
Id. at 637.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
673
the majority struck down that portion of Judge Ruckriegle’s order
requiring the media to delete or destroy the documents it had
received, and it noted that “[p]ublication of information the media
has obtained or obtains by its own investigative capacities is not
limited by the District Court’s order or our judgment, even though
such information may also be spoken of or referred to in the
transcripts.”111
Three members of Colorado’s high court signed off on a
vigorous dissent, written by Justice Michael Bender, contending
that
two striking facts about this case make it obvious that the
prior restraint issued by the district court is an
unconstitutional violation of the freedom of the press
guaranteed by the First Amendment. First, most of the
private details of the alleged victim’s sexual conduct
around the time of the alleged rape, which is also the
subject matter of the confidential hearings in this case, are
already available through public court documents and other
sources and have been widely reported by the media.
Second, the media did nothing wrong in obtaining the
transcripts. Under well-established prior restraint doctrine,
these two factors alone require this Court to direct the
district court to vacate its order immediately.112
The dissent’s passionate argument, however, failed to carry the
day in court, and the majority allowed the prior restraint to remain
in place. The decision shocked First Amendment scholars such as
Erwin Chemerinsky, who remarked, “This is a court order
prohibiting publication. Unless the Supreme Court dramatically
changes the law of the First Amendment, this decision can’t
stand.”113
But, unfortunately for free press advocates, it did. Although the
media quickly asked the U.S. Supreme Court to step in to prevent
the enforcement of the prior restraint, the nation’s high court
111
Id. at 638.
Id. at 639 (Bender, J., dissenting).
113
Steve Henson & Henry Weinstein, Court Bars Disclosure by Media,
L.A. TIMES, July 20, 2004, at D1.
112
CALVERT MACROED CORRECTED 060605.DOC
674
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
refused to stay the order, in part because of timing issues.114 Justice
Stephen Breyer wrote that
the trial court’s determination as to the relevancy of the
rape shield material will significantly change the
circumstances that have led to this application [for a stay of
the prior restraint]. As a result of that determination, the
trial court may decide to release the transcripts at issue here
in their entirety, or to release some portions while redacting
others. Their release . . . is imminent.115
In essence, the Supreme Court passed on the issue, but strongly
suggested that the trial court judge quickly review and release as
much of the transcripts, redacted if necessary, as possible. Justice
Breyer also wrote that the news organizations could re-file in two
days’ time for a stay with the U.S. Supreme Court if the trial court
judge had not, by that time, made his findings regarding which
portions of the transcripts could be released.116 In response to
Breyer’s rather forceful encouragement, Judge Ruckriegle ordered
the prosecution and defense “to work together to produce an edited
version of disputed transcripts that can be released to the
public.”117
The media entities involved ultimately dropped a second
appeal to the U.S. Supreme Court in August 2004 after the trial
judge unsealed large portions of the transcripts of the closed-door
hearings about the sexual history of Bryant’s accuser.118 One
reason the appeal was dropped, however, was to avoid “the
possibility that the U.S. Supreme Court might uphold Colorado’s
high court, setting a national precedent in favor of do-not-publish
orders.”119 Thus, while the transcripts were made public in the
Bryant case, “the greater battle over prior restraints remains.”120
114
Associated Press v. District Court, 125 S. Ct. 1 (2004).
Id. at 2.
116
Id.
117
Steve Henson, Judge Seeks Edits of Transcripts, L.A. TIMES, July 28,
2004, at D3.
118
Karen Abbott, Media Groups Drop Plans to Appeal Publishing Ban,
ROCKY MTN. NEWS (Denver), Aug. 4, 2004, at 5A.
119
Id.
120
Steve Lipsher & Felisa Cardona, Media Drop Bryant Lawsuit, DENVER
115
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
675
In summary, in 2004, when it came to sexual assault cases
involving high-profile public figures such as Kobe Bryant and
Michael Jackson, privacy and secrecy trumped the public’s right to
know and the First Amendment interest in a free press. This does
not bode well for the press in 2005 and beyond. As media reporter
Tim Rutten observed in the Los Angeles Times, “[T]he precedent
established in the Bryant case ominously stands in Colorado.”121
Why are trial court judges such as Terry Ruckriegle and
Rodney Melville now coming down squarely on the side of privacy
and against the First Amendment interests of the public and
journalists? Could there be factors at play here besides legal rules
and principles (such as the right to a fair trial and rape shield
statutes) that might be influencing their opinions?122 One
extrajudicial, contextual variable that might be playing an unseen
role is a possible growing sentiment that the prying and peering
behavior of the news media that has, in part, given rise to our
voyeuristic culture has simply gone too far.123 In fact, Michael
Jackson’s attorneys, in arguing for certain information to be kept
sealed, wrote that media coverage of the case was “voyeuristic and
entertainment-related”124 and that the press was simply seeking
POST, Aug. 4, 2004, at B-02.
121
Tim Rutten, Regarding Media; Secrecy Proves Costly, L.A. TIMES,
Sept. 3, 2004, at E1.
122
The law often is influenced by variables that may have nothing to do
with legal rules. Benjamin N. Cardozo, the late U.S. Supreme Court Justice,
wrote more than eighty years ago that the forces that influence judges in their
opinions
are seldom fully in consciousness. They lie so near the surface,
however, that their existence and influence are not likely to be
disclaimed. . . . Deep below consciousness are other forces, the likes
and dislikes, the predilections and prejudices, the complex of instincts
and emotions and habits and convictions, which make the man, whether
he be litigant or judge.
Benjamin N. Cardozo, The Nature of the Judicial Process, in SELECTED
WRITINGS OF BENJAMIN NATHAN CARDOZO 178 (M.E. Hall ed., 1947).
123
See generally CLAY CALVERT, VOYEUR NATION: MEDIA, PRIVACY, AND
PEERING IN MODERN CULTURE 133-37 (2000) (detailing the media’s voyeuristic
news practices and the public’s appetite for such content).
124
Liptak, supra note 17, at A20.
CALVERT MACROED CORRECTED 060605.DOC
676
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
“material that might sell magazines or provide higher ratings
during sweeps week on television.”125 What the author of this
article has elsewhere identified as the voyeurism value in First
Amendment jurisprudence126 may finally have met its match in the
form of celebrity cases involving sexual assault.
Parsed differently, what occurred in 2004 may well represent
the start of an unfortunate wave of judicial backlash in response to
prior journalistic indiscretions or out of fear of future foibles. A
recent article in the ABA Journal on celebrity cases and the sealing
of records, including those involving Kobe Bryant and Michael
Jackson, noted how some people “say increasing numbers of
competing news organizations and the rise of tabloid journalism
have overwhelmed courts and forced judges’ hands.”127
Or could it be that the judges’ decisions against the news media
reflect a much broader and pervasive lack of public trust in the
news media?128 Put differently, if the public does not trust the news
media, then why should Judge Melville—a member of that same
public—trust news organizations with the sensitive sexual
information at issue in the Michael Jackson case? After all, a
survey conducted in 2004 on behalf of the Project for Excellence
in Journalism found that the “public believes that news
organizations are operating largely to make money, and that the
journalists who work for these organizations are primarily
motivated by professional ambition and self-interest.”129 It would
be distressing for journalists, of course, to believe that they in part
brought this situation on themselves, but that may be the case.
Ultimately, regardless of the actual reasons, the Michael
Jackson and Kobe Bryant cases in 2004 represent triumphs of
privacy over a free press and the public’s right to know. The next
125
Steve Chawkins, Jackson Lawyers Scold Media for Seeking Records,
L.A. TIMES, July 20, 2004, at B6.
126
Clay Calvert, The Voyeurism Value in First Amendment Jurisprudence,
17 CARDOZO ARTS & ENT. L.J. 273 (1999).
127
John Gibeaut, Celebrity Justice, ABA J., Jan. 2005, at 42, 47.
128
Cf. Mark Jurkowitz, Public’s Cynicism About Media Has Become A
Pressing Concern, BOSTON GLOBE, Apr. 14, 2004, at C1 (writing that “public
distrust of the news media appears to be at a dangerously high level”).
129
Id.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
677
part of this article illustrates, using the opinion of another court in
2004 in a decidedly non-celebrity setting, some of the specific,
privacy-intrusive journalistic practices that may result in legal
liability. Despite the different setting and players in this next case,
privacy again prevailed in court. In contrast to the Jackson and
Bryant cases, in which cameras were a pervasive presence outside
of the public courthouses, the case discussed in the next part of the
article poses ramifications for the covert use of cameras and
microphones by journalists inside of private places.
IV. UNDERCOVER JOURNALISM AND PRIVACY: SOME LESSONS
FROM 20/20 IN 2004
Los Angeles-based attorney Neville Johnson has made a name
for himself as a litigator by suing news media organizations on
behalf of people who claim that their privacy interests were
invaded by duplicitous and invasive newsgathering techniques.130
He successfully posited such an argument before the Supreme
Court of California in Sanders v. American Broadcasting
Companies, Inc.131 In that case, which examined the use of hidden
cameras and microphones, Johnson coaxed from the court a
decision holding that a plaintiff need not prove a complete
expectation of privacy to recover under the tort of intrusion into
seclusion.132
In 2004, Johnson was at it once again, this time in federal
court, in a case called Turnbull v. American Broadcasting
Companies, Inc.133 The case pivoted on the surreptitious recording
of both images and voices by an undercover ABC producer for a
20/20 newsmagazine segment called “Pay to Play” that aired in
November 2002. The lawsuit focused “on the alleged intrusion of
130
See generally Richards & Calvert, supra note 9 (profiling Johnson and
providing the transcript of an in-depth, first-person interview with him).
131
20 Cal. 4th 907 (1999).
132
Id. at 916 (holding that “[t]here are degrees and nuances to societal
recognition of our expectations of privacy: the fact that the privacy one expects
in a given setting is not complete or absolute does not render the expectation
unreasonable as a matter of law”).
133
32 Media L. Rep. (BNA) 2442 (C.D. Cal. 2004).
CALVERT MACROED CORRECTED 060605.DOC
678
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
privacy rather than the broadcast of the program,”134 as the
plaintiffs did not assert a cause of action for defamation.135
In August 2004, U.S. District Court Judge S. James Otero
issued an order rejecting ABC’s motion for summary judgment on
a number of privacy-based causes of action filed by Johnson and
his partner, Brian Rishwain, thus allowing much of the case to
proceed to a jury trial.136 Although the plaintiffs ultimately lost at
trial on October 28, 2004,137 there are several significant aspects of
Judge Otero’s summary judgment ruling that bode well for privacy
advocates and that may impact and limit future hidden-camera and
hidden-microphone investigations by journalists. In particular,
Judge Otero’s decision permitting causes of action based on
eavesdropping,138 intrusion into seclusion,139 trespass,140 and
134
Id. at 2446.
See CAL. CIV. CODE § 44 (Deering 2004) (defining defamation in
California to include both libel and slander).
136
Turnbull, 32 Media L. Rep. 2442 (C.D. Cal. 2004).
137
See E-mail from Jim Ryan, associate attorney for Johnson & Rishwain,
LLP, to Clay Calvert, Associate Professor of Communications and Law at The
Pennsylvania State University (Jan. 18, 2005, 14:39:59 PST) (on file with
author) (setting forth the date of the jury verdict, and noting that a motion for a
new trial had been filed and was, at that time, under consideration).
138
CAL. PENAL CODE § 632 (Deering 2004). This section, which applies to
the secretive recording of confidential communications, provides in relevant
part:
Every person who, intentionally and without the consent of all parties
to a confidential communication, by means of any electronic
amplifying or recording device, eavesdrops upon or records the
confidential communication, whether the communication is carried on
among the parties in the presence of one another or by means of a
telegraph, telephone, or other device, except a radio, shall be punished
by a fine not exceeding two thousand five hundred dollars ($ 2,500), or
imprisonment in the county jail not exceeding one year, or in the state
prison, or by both that fine and imprisonment.
Id.
139
See generally Reporters Committee for Freedom of the Press, Invasion
of Privacy: Intrusion, First Amendment Handbook, available at
http://www.rcfp.org/handbook/c02p02.html (last visited Jan. 18, 2005)
(describing the intrusion tort).
140
See Miller v. NBC, 187 Cal. App. 3d 1463, 1480 (Cal. Ct. App. 2004)
135
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
679
physical and constructive invasion of privacy under California’s
anti-paparazzi statute141 is significant because it:
• identifies a number of specific, objective indicators or
signals that a reasonable expectation of privacy exists in
certain scenarios when hidden recording devices are used;
• reflects an expansive interpretation of both the California
Supreme Court’s decision in Sanders regarding privacy
expectations142 and the state’s anti-paparazzi law; and
(defining the tort of trespass under California law).
141
CAL. CIV. CODE § 1708.8 (Deering 2004). The law was enacted after the
death of Princess Diana and amid a public furor about the actions of so-called
paparazzi. See generally Clay Calvert & Robert D. Richards, The Irony of News
Coverage: How the Media Harm Their Own First Amendment Rights, 24
HASTINGS COMM. & ENT. L.J. 215 (2002) (discussing the evolution of antipaparazzi legislation). California’s anti-paparazzi statute has two key
components—one for physical invasions of privacy, the other for constructive
invasions of privacy, with the former providing:
A person is liable for physical invasion of privacy when the defendant
knowingly enters onto the land of another without permission or
otherwise committed a trespass, in order to physically invade the
privacy of the plaintiff with the intent to capture any type of visual
image, sound recording, or other physical impression of the plaintiff
engaging in a personal or familial activity and the physical invasion
occurs in a manner that is offensive to a reasonable person.
CAL. CIV. CODE § 1708.8 (a) (Deering 2004).
In addition to targeting physical invasions of privacy on personal or familial
activities, the statute also restricts constructive invasions of privacy by
providing:
A person is liable for constructive invasion of privacy when the
defendant attempts to capture, in a manner that is offensive to a
reasonable person, any type of visual image, sound recording, or other
physical impression of the plaintiff engaging in a personal or familial
activity under circumstances in which the plaintiff had a reasonable
expectation of privacy, through the use of a visual or auditory
enhancing device, regardless of whether there is a physical trespass, if
this image, sound recording, or other physical impression could not
have been achieved without a trespass unless the visual or auditory
enhancing device was used.
CAL. CIV. CODE § 1708.8 (b) (Deering 2004).
142
See supra notes 131-32 and accompanying text.
CALVERT MACROED CORRECTED 060605.DOC
680
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
• contains dicta suggesting that First Amendment protection
for the use of hidden cameras is limited, especially when
journalist-defendants freely admit during discovery that
such surveillance techniques were not necessary to report
the story in question.
To better understand the potential implications of the Turnbull
decision, however, it is first important to briefly review the
underlying facts of the case. The 20/20 segment at issue focused on
the activities of certain “casting workshops” in the Los Angeles
area.143 The gist of the story was that these workshops for aspiring
actors had very little educational or learning component to them,
but instead, were merely paid opportunities for actors to meet and
appear before casting directors. As such, the plaintiffs, most of
whom were aspiring actors, contended that the 20/20 segment
“made them look like ‘whores,’ or desperate losers on the fringe of
the acting community in Los Angeles.”144 But because the causes
of action focused on how the information for the segment was
gathered rather than on the segment itself, the gravamen of the
complaint was that the “[p]laintiffs object[ed] to the very fact that
their presence at the workshop was recorded”145 and that private
and embarrassing conversations were recorded.
To obtain footage and audio at these workshops, ABC producer
Yoruba Richen went undercover and attended several workshops,
paying an admission fee to enter as if she too were an aspiring
actress.146 It is undisputed that “Richen’s primary purpose in
attending the workshops was to do a story on the workshops, not to
practice her acting.”147 It also was undisputed that “Richen did not
tell anyone at the workshops that she was wearing a hidden camera
or planned to wear a hidden camera.”148
While at the workshops, Richen recorded performances by the
actors doing scenes for the casting directors and, more importantly,
143
Turnbull v. ABC, 32 Media L. Rep. (BNA) 2442, 2445 (C.D. Cal.
2004).
144
145
146
147
148
Id. at 2445-46.
Id. at 2447.
Id. at 2446.
Turnbull, 32 Media L. Rep. at 2447.
Id.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
681
secretly recorded “conversations between actors while they were
waiting for performances to begin,”149 including “personal
conversations between workshop participants to which Richen was
not a party.”150 In one instance, she recorded a plaintiff-actor
“making an offensive and overtly sexual comment to”151 another
plaintiff-actor, while in another situation she captured one
plaintiff-actor expressing something that she did not want the
casting directors to know. Beyond this, Richen “even filmed a
journey into the womens’ [sic] lavatory.”152 The tape also captured
conversations “overheard from across the room as two people
talk[ed] in a corner, or while their backs [were] turned to Ms.
Richen, apparently unaware that an ABC News reporter [was]
recording their every word.”153
With these undisputed facts in mind, one can better understand
the three significant aspects of Judge Otero’s opinion identified
earlier in this section of the article. First, in holding that the
plaintiffs had reasonable expectations of confidentiality and
privacy in their communications at the workshops, the judge
articulated a number of objective indicators of privacy that were
manifested in the setting, the situation, and the behavior of the
plaintiffs. This was all part of what Judge Otero called “a common
sense approach”154 to privacy. These factors, which, if heeded,
should help journalists avoid future lawsuits for privacy invasions,
include:
• Plaintiffs’ Body Language: In particular, in finding a
conversational privacy expectation, Judge Otero noted that
two of the plaintiffs “had their back[s] turned to”155 ABC’s
producer, Richen, while the plaintiffs were talking among
themselves.
• Plaintiffs’ Distance and Location from Defendant: Judge
149
150
151
152
153
154
155
Id.
Id.
Id.
Turnbull, 32 Media L. Rep. at 2447.
Id.
Id. at 2453 n.8.
Id. at 2451.
CALVERT MACROED CORRECTED 060605.DOC
682
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Otero specifically observed that defendant “Richen was
standing across the room”156 from the plaintiffs at the time
she recorded one of their conversations. It also appeared to
make a difference to the judge that the plaintiffs, during
some conversations, were “in a corner”157 of a room.
• Content of the Communications: Judge Otero’s opinion
suggests that if the plaintiffs are engaged in a conversation
that includes potentially embarrassing or offensive remarks,
it may signal that they did not expect their conversation to
be recorded. In particular, he emphasized that, in one
instance, a plaintiff made an overtly sexual comment that
“probably was not for Ms. Richen’s benefit.”158
In addition, in another conversation recorded by Richen, a
different plaintiff, Sharon Johnston, specifically said, “But
they don’t have to know that.”159 For Judge Otero, this
statement made it “clear”160 that the plaintiff “did not want
her conversation disseminated outside of the intended
audience.”161 As the judge wrote, “By stating ‘they don’t
have to know that,’ Johnston was signaling her expectation
of privacy.”162
• Number of People in the Setting: In finding a privacy
expectation, Judge Otero observed that the workshops
“were small, consisting of 10 to 20 people,” and in many
instances when conversations were recorded, “there were
only two or three people in a room.”163
• Custom of the Activities in the Setting: The very nature of
the educational workshop at issue in the case also appeared
to play an important role in the judge’s privacy calculus. In
particular, Judge Otero wrote that “[i]t is not difficult to
156
157
158
159
160
161
162
163
Id.
Turnbull, 32 Media L. Rep. at 2451.
Id.
Id. (emphasis in original).
Id.
Id.
Turnbull, 32 Media L. Rep. at 2451.
Id.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
683
imagine a litany of classroom or workshop settings where
the students might reasonably expect privacy.”164 In
important dicta, Otero observed:
Customarily, in law school and undergraduate
university lectures, students must ask for the
instructor’s permission prior to recording university
lectures. Closer to the point, if a group of aspiring
authors decided to attend a seminar with a writer in
residence at a local university in order to obtain
feedback and criticism regarding unfinished work, it
would probably be reasonable for them to assume their
activities, readings, and the instructor’s comments,
were not being overheard by a person who was not
similarly situated; let alone being recorded by a
journalist.165
What is interesting here is the suggestion that, in certain
learning environments in which people voluntarily expose
themselves to the risk of criticism from others (instructors
or classmates) for the ultimate purpose of improving
themselves based on feedback, they do not voluntarily
expose themselves to a risk of recordation of their activities
and conversations.
• Admission and Entrance to the Setting: Judge Otero
pointed out in his analysis of privacy expectations on the
tort of intrusion that “the workshops were closed to the
general public. To gain entry, a prospective participant had
to audition, pay an entry fee and check-in.”166 He added
that “[t]he workshops took place in a private room of a
private building few actors know about.”167 In the judge’s
view, the restrictions on the program’s accessibility
seemingly added to the degree of privacy expected by
workshop participants.
164
165
166
167
Id.
Id.
Id. at 2453 n.9.
Id.
CALVERT MACROED CORRECTED 060605.DOC
684
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Viewed collectively, the abovementioned laundry list of
privacy factors that can be distilled from Judge Otero’s opinion
should prove useful for journalists considering whether and when
to use hidden cameras and microphones. Indeed, journalists should
seriously consider incorporating these variables into their
newsgathering policies and practices.
In addition to these privacy variables, Judge Otero’s opinion is
significant in its expansive reading of the Supreme Court of
California’s holding in Sanders v. American Broadcasting
Companies, Inc.168 In Sanders, the California high court held that
“[a] person who lacks a reasonable expectation of complete
privacy in a conversation because it could be seen and overheard
by coworkers (but not the general public) may nevertheless have a
claim for invasion of privacy by intrusion based on a television
reporter’s covert videotaping of that conversation.”169
Judge Otero extended this logic from the workplace setting of
Sanders, which involved the tele-psychic industry, to the
educational and classroom setting of Turnbull. This extension
allowed Otero to conclude, in part, that the plaintiffs “could not
have expected, as they talked amongst themselves in the corners or
against the wall of the classroom, in their chairs awaiting class to
begin, much less the ladies [sic] room, that a reporter was covertly
recording their conversations.”170
Judge Otero adopted a similarly expansive construction of
California’s anti-paparazzi law and, in particular, its requirement
that the alleged privacy invasion must relate to “the plaintiff
engaging in a personal or familial activity.”171 Clearly the factual
situation at issue in Turnbull did not involve “familial activity”;
indeed, the workshops were all about acting and meeting casting
directors. Thus, to receive the protection of California’s antipaparazzi law, the plaintiffs’ conduct at the casting workshops
would have to be characterized as “personal activity.” The
defendants contended in their summary judgment motion that ABC
168
169
170
171
20 Cal. 4th 907 (1999).
Id. at 923.
Turnbull, 32 Media L. Rep. at 2454.
CAL. CIV. CODE § 1708.8(a) (Deering 2004) (emphasis added).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
685
producer Richen “did not record personal activity.”172 The judge,
however, rejected this contention. In allowing the plaintiffs to
proceed to trial on this statutory cause of action, Judge Otero ruled
that the defendants “recorded personal conversations and other
matters without permission.”173 This expansive reading of the antipaparazzi statute stretches the term “activity”174 to include
conversations. The judge’s view thus rejects the existence of a
conduct-versus-speech dichotomy that separates and distinguishes
an activity from a conversation. Under this interpretation, personal
conversations, not just personal activities, fall within the ambit of
California Civil Code Section 1708.8. This interpretation
represents an important victory for privacy advocates.
Finally, the third significant aspect of Judge Otero’s summary
judgment ruling in Turnbull is the following statement made by the
judge: “[T]here is no point in according First Amendment
protection in the instant case because Defendants freely admit that
they would have gone ahead with the same story even if secret
camera footage was unavailable.”175
If this proposition really is true, as Judge Otero believes it is,
then Turnbull’s implications for hidden-camera journalists and
producers are profound: if the same story can be told regardless of
whether hidden cameras are used, then journalists should not
expect the First Amendment to come to their rescue if they are
sued for invasions of privacy based on the use of hidden cameras.
Likewise, journalists and producers should never admit in
depositions or affidavits that they could have told the same story or
would have done the same story without the hidden surveillance
devices. Indeed, Judge Otero cited the deposition testimony of
Brian Ross, the chief investigative correspondent at ABC and “one
of the individuals who decided to do the story and decided to use
hidden cameras for the program,”176 as proof that “Ross would not
have scrapped the story if he could not have used hidden
172
173
174
175
176
Turnbull, 32 Media L. Rep. at 2456.
Id. (emphasis added).
CAL. CIV. CODE § 1708.8 (Deering 2004) (emphasis added).
Turnbull, 32 Media L. Rep. at 2458 (emphasis added).
Id. at 2448.
CALVERT MACROED CORRECTED 060605.DOC
686
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
cameras.”177 To some extent, then, ABC was done in by its own
words.
Judge Otero’s statement further suggests that the First
Amendment will come to journalists’ defense in such situations if
hidden cameras were the only way to tell the story. This forces
news producers and in-house media counsel to make very tough
choices about how to cover stories and whether to risk the use of
hidden cameras. Turnbull advises that a true journalistic and legal
cost-benefit analysis is in order for those in the newsrooms and
executive suites of the broadcast networks.
In summary, Judge Otero’s summary judgment ruling in
Turnbull, although of precedential value today in only one federal
district court, contains reasoning, logic, and analysis that, if
adopted by other courts, may have significant ramifications for
journalism policies and practices in the future. In the interim, the
laundry list of privacy-expectation signals identified by the judge
should prove useful for journalists in guiding their own conduct in
future investigative-report scenarios.
V. PRIVACY IN SOURCE-REPORTER RELATIONSHIPS:
THE DIFFICULTY OF KEEPING CONFIDENCES IN 2004
This article so far has illustrated how privacy concerns often
prevailed in 2004 against the interests of both journalists and the
public’s right to know. It thus is more than a little bit ironic that,
when journalists in 2004 asserted their own privacy interests—in
particular, the right to keep private and secret the names of their
confidential sources—they were thoroughly rebuffed and rebuked
by the judiciary. In fact, as the author of this article and a colleague
wrote in a newspaper commentary in November 2004, “[t]he list of
reporters now facing jail time for refusing to disclose a source’s
identity grows longer each day.”178
Chief among those journalists was Jim Taricani, an
investigative television journalist for NBC-affiliate WJAR,
177
Id. at 2458.
Robert D. Richards & Clay Calvert, Keeping Public Confidences; It’s
Time for a Federal Shield Law Protecting Journalists from Source Revelation,
PITTSBURGH POST-GAZETTE, Nov. 30, 2004, at A-21.
178
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
687
Channel 10, in Providence, Rhode Island, who was held in both
civil and criminal contempt in November 2004.179 In particular,
Taricani was convicted of criminal contempt for refusing to reveal
the identity of the person who leaked to him a copy of an FBI
surveillance videotape showing a bribe being accepted by a
Providence city official.180 The tape, which Taricani’s station aired
on February 1, 2001,181 had been under seal by a court, and thus,
the person who leaked it to the reporter violated a court order.182
U.S. District Court Judge Ernest C. Torres appointed Marc DeSisto
as a special prosecutor to try “to find out who gave Taricani the
secret videotape.”183
Taricani’s November criminal conviction followed a decision
earlier that same year by the U.S. Court of Appeals for the First
Circuit affirming a civil contempt ruling and holding that Taricani
did not have a First Amendment privilege or right to refuse to
reveal his source to DeSisto.184 The decision was not surprising.
Although thirty-one states now have shield laws that grant
journalists varying degrees of protection against testifying about
certain confidential information in their possession,185 there is no
179
See generally Lynne Tuohy, Reporter Convicted; Shielded Source,
HARTFORD COURANT, Nov. 19, 2004, at A1 (providing an excellent overview of
the battles Taricani faced for protecting his source).
180
Pam Belluck, Reporter Is Found Guilty for Refusal to Name Source,
N.Y. TIMES, Nov. 19, 2004, at A24.
181
Tuohy, supra note 179, at A1.
182
See Eileen McNamara, Journalists Under Attack, BOSTON GLOBE, Nov.
21, 2004, at B1 (writing that “[w]hoever leaked the videotape to Taricani,
though, did so in violation of a court order that all such materials were to be
sealed”).
183
Tracy Breton, Taricani Told to Reveal Source or Risk Prison,
PROVIDENCE J., Nov. 5, 2004, at A-01.
184
In re Special Proceedings, 373 F.3d 37 (1st Cir. 2004).
185
See, e.g., CAL. EVID. CODE § 1070 (setting forth California’s shield
law); see generally DON R. PEMBER & CLAY CALVERT, 2005-2006 EDITION
MASS MEDIA LAW 390-93 (2005) (discussing state shield laws); Reporters
Committee for Freedom of the Press, The Reporter’s Privilege Compendium: An
Introduction, available at http://www.rcfp.org/cgi-local/privilege/item. cgi?i=
intro (last visited Jan. 16, 2005).
[Thirty-one] states and the District of Columbia have enacted statutes—
CALVERT MACROED CORRECTED 060605.DOC
688
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
federal shield law to protect source privacy; in fact, in the only
instance in which the U.S. Supreme Court has considered a
possible constitutional privilege, the Court rejected a First
Amendment privilege for journalists to refuse to testify before
grand juries.186
After the appellate court’s ruling, Judge Torres began fining
Taricani $1,000 per day, hoping that the civil contempt remedy
would persuade Taricani to give up his source.187 Some $85,000 in
paid fines later, Torres switched tactics from civil to criminal
contempt.188
Why did Taricani refuse to reveal his source to the special
prosecutor? As the Emmy Award-winning journalist explained
outside the courthouse after his conviction,
I wish all my sources could be on the record, but when
people are afraid, a promise of confidentiality may be the
only way to get the information to the public, and in some
cases, to protect the well-being of the source. I made a
shield laws—that give journalists some form of privilege against
compelled production of confidential or unpublished information. The
laws vary in detail and scope from state to state, but generally give
greater protection to journalists than the state or federal constitution,
according to many courts.
Id.
186
Branzburg v. Hayes, 408 U.S. 665 (1972). See In re Special
Proceedings, 373 F.3d 37, 44 (1st Cir. 2004) (writing that “[i]n Branzburg, the
Supreme Court flatly rejected any notion of a general-purpose reporter’s
privilege for confidential sources, whether by virtue of the First Amendment or
of a newly hewn common law privilege”).
187
See Tracy Breton, Taricani Could Face Harsher Sanctions,
PROVIDENCE J., Sept. 30, 2004, at A-01 (describing the court-imposed, $1,000per-day fine on Taricani, and how it was being paid everyday by a check
“delivered to the clerk of the U.S. District Court—written from a bank account
of Channel 10 investigative reporter Jim Taricani”).
188
See Belluck, supra note 180, at A24 (writing that “Taricani was fined
$1,000 for each day he continued to refuse to name his source” and, when he
refused to relent “after he had paid $85,000—for which he was reimbursed by
his employer—Judge Torres changed the civil contempt case into a criminal
contempt case”).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
689
promise to my source, which I intend to keep.189
Taricani’s promise of privacy to his source, however,
ultimately resulted in six months of home confinement—the
sentence Judge Torres meted out in December 2004 for the
criminal contempt conviction.190
Taricani was not the only journalist under a determined judicial
assault in 2004 aimed at compelling the revelation of his sources.
In fact, as a reporter for the Boston Globe summarized the situation
in late 2004:
This past summer, five reporters were found in contempt
for refusing to disclose sources used in reporting on Wen
Ho Lee, the former nuclear scientist who was the suspect in
an espionage case.191 And leaked information from the
BALCO steroid grand jury investigation could leave some
Bay Area reporters facing penalties for not revealing their
sources.192
In the BALCO situation, which centered on alleged steroid use
by individuals such as baseball superstars Barry Bonds and Jason
Giambi, U.S. Attorney Kevin V. Ryan asked journalists from the
San Francisco Chronicle to reveal their sources for leaked grand
jury testimony.193 By early 2005, the Chronicle’s editor, Phil
Bronstein, maintained that the newspaper would not give up its
confidential sources, stating that “[t]he press has certain
responsibilities in society, but one of them is not to enforce the
provisions of the federal grand jury system. Obviously, there are
people who disagree with that, including the Justice Department.
189
Id.
W. Zachary Malinowski, Taricani Won’t Appeal Punishment,
PROVIDENCE J., Dec. 22, 2004, at B-03.
191
Lee v. U.S. Dept. of Justice, 327 F. Supp. 2d 26 (D.D.C. 2004). The five
journalists in that case were Bob Drogin of the Los Angeles Times, H. Josef
Hebert of the Associated Press, Jeff Gerth and James Risen of The New York
Times, and Pierre Thomas, a former CNN reporter who now works for ABC
News. Id. at 27 n.1.
192
Mark Jurkowitz, Journalists Push for a State Shield Law, BOSTON
GLOBE, Dec. 21, 2004, at D1.
193
John M. Broder, From Grand Jury Leaks Comes a Clash of Rights, N.Y.
TIMES, Jan. 15, 2005, at A8.
190
CALVERT MACROED CORRECTED 060605.DOC
690
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
But that’s not the view from here.”194
Two other major cases from 2004 involving the journalistic
desire to keep private the identities of sources both pivot on the
question of who leaked and disclosed the name of covert CIA
operative Valerie Plame in July 2003 to several members of the
media, including Robert Novak.195 Novak later blew Plame’s cover
by printing her name in his syndicated column that same month.
Novak cited his sources for the scoop on Plame’s employment as
“two senior administration officials,”196 neither of whom he
identified. It is a violation of federal law to reveal the names of
covert CIA agents, and the Justice Department named a special
prosecutor, U.S. Attorney Patrick J. Fitzgerald, to conduct a grand
jury investigation into who leaked Plame’s name to Novak. With
Novak refusing to tell anyone whether he had even spoken with the
special prosecutor or was cooperating with the government
investigation,197 Fitzgerald soon began “aggressively taking on
other journalists who reported on the story.”198 Among those
journalists was Time magazine’s Matthew Cooper and, although
she never wrote a story on the matter, The New York Times’s
Judith Miller.199 In November 2004, U.S. District Court Judge
Thomas F. Hogan refused to quash a subpoena served on Cooper
194
Id.
See generally Lorne Manly & Adam Liptak, At Leak Inquiry’s Center, a
Circumspect Columnist, N.Y. TIMES, Dec. 31, 2004, at A18 (providing a
thorough overview of the facts and legal issues in the dispute).
196
See Adam Liptak, Judges Skeptical of First Amendment Protection for
Reporters in C.I.A. Leak Inquiry, N.Y. TIMES, Dec. 9, 2004, at A28 (“Robert
Novak, the syndicated columnist, was the first to disclose Ms. Plame’s identity
publicly, in a column published on July 14, 2003. He had been told, he wrote, by
‘two senior administration officials’ seeking to cast doubt on an opinion column
by Ms. Plame’s husband, Joseph C. Wilson IV, a former diplomat.”).
197
See Charles Duhigg, Media Law; Robert Novak: How Does He Stay Out
of Jail, L.A. TIMES, Dec. 12, 2004, at Opinion M6 (describing how Novak has
managed to escape the same wrath that has faced Matthew Cooper and Judith
Miller).
198
Richard B. Schmitt, The Nation; Prosecutor’s Lips Still Sealed in Probe
of Leaked Information, L.A. TIMES, Oct. 23, 2004, at A10.
199
See Punishing the Press, N.Y. TIMES, Dec. 20, 2004, at A28 (writing
that Miller “never wrote a single article about the Plame controversy”).
195
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
691
and Time, writing:
Mr. Cooper and Time have no privilege based in the First
Amendment or common law, qualified or otherwise,
excusing them from providing documents to or testifying
before the grand jury in this matter. Therefore, Mr. Cooper
and Time must fulfill their obligations to answer valid
subpoenas issued to them by a grand jury acting in good
faith.200
This decision followed an earlier August 2004 order holding
Cooper in civil contempt and fining Time $1,000 per day until it
handed over the subpoenaed documents.201 Judith Miller’s motion
to quash the subpoena of Fitzgerald also was rejected by Judge
Hogan.202 Similar to his finding with Matthew Cooper, Judge
Hogan opined that Miller “has no privilege, based in the First
Amendment or common law, qualified or otherwise, excusing her
from testifying before the grand jury in this matter. . . . Ms. Miller
must fulfill her obligation, shared by all citizens, to answer a valid
subpoena issued to her by a grand jury acting in good faith.”203
With both Cooper and Miller facing up to eighteen months in jail
for refusing to disclose their sources, the reporters and their news
organizations took their case to a federal appellate court in
December 2004.204 The three-judge panel seemed skeptical during
oral argument of granting a privilege to Cooper and Miller,205 and
in February of 2005, it ruled against the journalistic duo.206 The
200
In re Special Counsel Investigation, 346 F. Supp. 2d 54, 56 (D.D.C.
2004).
201
In re Special Counsel Investigation, 332 F. Supp. 2d 33 (D.D.C. 2004).
In re Special Counsel Investigation, 338 F. Supp. 2d 16 (D.D.C. 2004).
203
Id. at 19.
204
See Richard B. Schmitt, A Sign of Hope for Reporters in CIA Leak Case,
L.A. TIMES, Dec. 9, 2004, at A22 (describing the arguments before the U.S.
Court of Appeals for the District of Columbia).
205
See Carol D. Leonnig, Judges Weigh Press Freedoms, WASH. POST,
Dec. 9, 2004, at A11 (describing how “Judge David B. Sentelle grew visibly
irritated as he repeatedly asked longtime First Amendment lawyer Floyd
Abrams to explain how Cooper and Miller’s circumstances differed from those
of the Kentucky reporter”).
206
In re Grand Jury Subpoena, No. 04-3138, 2005 U.S. App. LEXIS 2494
202
CALVERT MACROED CORRECTED 060605.DOC
692
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
appellate court wrote that “there is no First Amendment privilege
protecting the evidence sought, but no decision had been reached
by the end of the year,”207 and it added that “if any such common
law privilege exists, it is not absolute, and in this case has been
overcome by the filings of the Special Counsel with the District
Court.”208
In situations such as those involving Taricani, Cooper, and
Miller, the First Amendment interest in privacy of information—
privacy of source identity, in particular—is clear. As veteran media
defense attorney James C. Goodale observed, what journalists such
as “Taricani are fighting for is the right to do their job. They
cannot do it without confidential sources.”209 And what is that job?
New York Times columnist William Safire summed it up well in a
recent commentary calling for a privilege for journalistic source
confidentiality when he wrote that “it is the publication’s
obligation to the public to publish what it considers newsworthy—
and not to assist the government in punishing the provider of that
news.”210
Indeed, a promise of privacy to a source is sometimes the only
way that a journalist such as Taricani can obtain what Safire terms
“newsworthy” information. A journalist who burns such a source
by breaching that promise harms not only himself and the source,
but all journalists and, more importantly, the public in general.211
As Eileen McNamara of the Boston Globe wrote in 2004,
breaching a promise of confidentiality “undermine[s] the work of
(D.C. Cir. Feb. 15, 2005).
207
Id. at *2.
208
Id.
209
James C. Goodale, Communications and Media Law; Why Reporters Go
to Jail, N.Y. L.J., Dec. 3, 2004, at 3.
210
William Safire, Judges as Plumbers, N.Y. TIMES, Dec. 13, 2004, at
A27.
211
Cf. Maggie Mulvihill, As You Were Saying . . . This Journalist Stands
Tall Rather Than Give Up a Source, BOSTON HERALD, July 3, 2004, at 16
(writing that “no governmental whistleblower would confide in a reporter if he
thought the reporter were in cahoots with prosecutors or would blow his cover”
and pointing out that “so much that the public should know would remain secret
if reporters didn’t keep their promises and refrain from ratting out their
confidential sources to the government”).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
693
all journalists by discouraging people in sensitive situations from
sharing information about wrongdoing with reporters. Journalists
would always prefer to put their sources on the record, but there
are instances in which such candor could cost a source his job or
his physical safety.”212
The situation was so bad that, in November 2004, U.S. Senator
Christopher Dodd (D–Conn.) introduced a bill titled “The Free
Speech Protection Act of 2004,” which was designed to create a
federal shield law to protect individuals and organizations involved
in gathering and disseminating news from being hauled into
federal court and forced to disclose their sources or other
unpublished information.213 In proposing the measure, Dodd
contended that “[w]hen the public’s right to know is threatened,
and when the rights of free speech and free press are at risk, all of
the other liberties we hold dear are endangered.”214 In a January
2005 opinion piece published in the Atlanta Journal-Constitution,
Dodd elaborated on this argument, writing:
If reporters are unable to promise confidentiality to their
sources, many conscientious citizens will choose not to
come forward with information out of fear for their jobs,
their reputations, even their lives. The public’s ability to
hold those in power accountable—whether in the
government or in the private sector—will be severely
compromised. In a real sense, when the public’s right to
know is threatened, so are all of the other liberties we hold
dear.215
There was good reason to think that the public would support
the measure; a national survey of more than 650 adults conducted
in October 2004 on behalf the First Amendment Center in
Nashville, Tennessee, found that seventy-two percent of
respondents either strongly or mildly agreed with the statement
212
McNamara, supra note 182, at B1.
S. 3020, 108th Cong. (2004).
214
Andy Thibault, Good Time to be Enemy of the State, CONN. L. TRIB.,
Jan. 10, 2005, at 20.
215
Christopher J. Dodd, Public’s Right to Know on Endangered List,
ATLANTA J.-CONST., Jan. 15, 2005, at 11A.
213
CALVERT MACROED CORRECTED 060605.DOC
694
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
that “journalists should be allowed to keep a news source
confidential.”216
Dodd’s proposal for congressional action clearly had the
support of the mainstream news media. In an editorial in
November 2004, the Washington Post opined:
Mr. Taricani’s case unfortunately is not unusual at all. It is
part of a rash of recent cases in which judges are seeking to
force journalists to renege on promises of confidentiality,
using the threat of jail as leverage. Without such promises,
much good journalism wouldn’t happen. If the federal
courts will not recognize a privilege for reporters such as
Mr. Taricani, as most states do, Congress needs to step in
and do it for them.217
If Congress does indeed act in 2005, it will represent a battle
between the legislative and judicial branches of government, with
the latter coming down squarely against the privilege in 2004 in
the cases of Jim Taricani, Matthew Cooper, and Judith Miller. This
clear preference for disclosure over source protection raises the
question: Why is there such reluctance on the part of judges to
extend a privacy privilege to journalists to protect their sources?
Why was there in 2004, as William Safire puts it, a “sudden wave
of judicial repression”218 of reporters? Mark Jurkowitz of the
Boston Globe observes that while “First Amendment advocates say
that privilege is vital to the free flow of information . . . some of
the public seems more skeptical, viewing journalists as putting
themselves above the law.”219 Might such skepticism be present
among judges who might see journalists as too often intruding on
others’ privacy in order to get information to sell newspapers? In
other words, if there is a perception among judges that journalists
push the boundaries of other people’s privacy rights in the name of
216
First Amendment Center, 2004 Confidential-Sources Survey, available
at http://www.firstamendmentcenter.org/about.aspx?item=2004_confidential_
sources (last visited Jan. 16, 2004). See Richards & Calvert, supra note 178, at
A-21 (citing the survey finding).
217
Jailing Reporters, WASH. POST, Nov. 28, 2004, at B06.
218
Safire, supra note 210, at A27.
219
Mark Jurkowitz, Departures Anchored the Year’s Top Media Stories,
BOSTON GLOBE, Dec. 26, 2004, at N10.
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
695
newsgathering and reporting, then why should judges not show
them how it feels to have their private information revealed?
CONCLUSION
If the judicial opinions and statutes described and analyzed in
this article were tallied up on a mythical legal scoreboard in a
contest pitting privacy advocates against working journalists, it
would show a decisive victory for privacy in 2004 and an
overwhelming defeat for the press. As discussed in this article:
• A right to personal privacy was extended by Congress,
albeit in limited circumstances, to people in public places
under the Video Voyeurism Prevention Act of 2004;
• A right to personal privacy also was extended, in certain
federal FOIA actions, by the U.S. Supreme Court to the
family members and close relatives of the dead who, for
obvious reasons, could not assert their own privacy claims;
• The press was not allowed to print, because of privacy
concerns, the contents of truthful documents that it had
lawfully obtained in the Kobe Bryant sexual assault case;
• The press was prohibited from obtaining access, also
because of privacy concerns accompanied by right-to-fairtrial issues, to basic and fundamental information about a
criminal case pending against one of the world’s most wellknown celebrities, Michael Jackson;
• The use of journalistic hidden cameras and microphones
that intrude on personal privacy was rebuked by a federal
court judge who took, as was noted earlier, what he called a
“common sense”220 approach to privacy that rejected a
media summary judgment motion and allowed numerous
privacy-based causes of action to proceed to trial.
When journalists, however, asserted their own privacy rights—
in particular, the right to keep private the identity of their
confidential sources—they lost in several high-profile cases, such
as those involving Jim Taricani, Matthew Cooper, and Judith
220
See supra note 154 and accompanying text.
CALVERT MACROED CORRECTED 060605.DOC
696
6/20/2005 7:10 PM
JOURNAL OF LAW AND POLICY
Miller.
What does all of this mean? For the news media, the
preponderance of judicial opinions and legislation against it in
2004 may reflect the findings of an independent survey conducted
that same year revealing a growing belief among the public that, as
compared to their counterparts from years past, modern
“journalists are sloppier, less professional, less moral, less caring,
more biased, less honest about their mistakes, and generally more
harmful to democracy.”221 Judges and legislators may harbor these
very same beliefs about the press—legislators, of course, often
pander to public sentiment to win elections, whether or not they
agree with those sentiments—and this may be influencing their
actions. There just might be then a disturbing correlation here for
journalists: less trust in the press may lead to fewer favorable
judicial rulings and less favorable legislation. The less-trust side of
the equation is clear,222 and the negative legal side has been
illustrated amply with multiple examples in this article.
While the news media devote a great amount of time to handwringing about whether there is a liberal news media bias,223 as do
others involved in the media,224 perhaps the media’s time would be
221
Mark Jurkowitz, Media Distrust May Be Libel-Case Key, BOSTON
GLOBE, Jan. 9, 2005, at B1 (quoting from a “major study released in March
2004 by the Project for Excellence in Journalism”).
222
See David Weddle, Swagland, L.A. TIMES MAG., Jan. 16, 2005, at 14
(writing that “[a] recent Gallup poll found that only 21% of those surveyed rated
newspaper reporters’ ethical standards as high or very high. Journalists ranked
lower than bankers, auto mechanics, elected officials and nursing home
operators”).
223
See, e.g., Joe Strupp et al., The Liberal Media: Myth or Reality?,
EDITOR & PUBLISHER, Aug. 1, 2004 (providing a comprehensive analysis of
both data and opinions regarding a potential liberal bias in the news media).
224
See, e.g., ERIC ALTERMAN, WHAT LIBERAL MEDIA? THE TRUTH ABOUT
BIAS AND THE NEWS (2003) (attempting to refute allegations that there is a
liberal bias in the media); BERNARD GOLDBERG, ARROGANCE: RESCUING
AMERICA FROM THE MEDIA ELITE (2003) (setting forth multiple instances of
what the former reporter for CBS News believes is a liberal bias in the
mainstream news media, including, most notably, The New York Times);
BERNARD GOLDBERG, BIAS: A CBS INSIDER EXPOSES HOW THE MEDIA DISTORT
THE NEWS (2002) (providing examples that the author contends illustrate a
liberal bias in the media).
CALVERT MACROED CORRECTED 060605.DOC
6/20/2005 7:10 PM
JOURNALISM AND THE RIGHT TO PRIVACY
697
better spent now focusing on the specific conduct and actions that
tend to erode respect for the news media while simultaneously
elevating judicial and legislative respect for privacy rights.
Journalists, in other words, cannot sit back and simply blame
pandering politicians and judges for their current state of woe.
Instead, a self-examination of their own actions may point them
out of this mess and toward a reasonable solution. If journalists
expect a right of privacy in their own relationships with sources
(think Jim Taricani and Matthew Cooper), then they may need to
be more careful about intruding on the privacy interests of others.
The solution, of course, must strike a balance that respects
privacy rights, but that allows journalists to perform their roles in a
democratic society. To achieve this balance, journalists must
educate the public (judges and legislators included) through their
actions, and not simply their pontifications in self-serving
editorials and commentaries, about the importance of their roles as
both watchdogs of government abuses of power225 and conveyors
of truthful and accurate news.226 The proper location of the
fulcrum in this delicate privacy-versus-reporter balance is, of
course, difficult to precisely pinpoint; however, it is clear that in
2004 more judicial and legislative weight was placed on the side of
personal privacy than on the side of journalists and reporters. The
press must now convince judges and legislators that the policy
interest in protecting a free press in a democratic society requires
shifting that balance back to a point that affords journalists greater
access to information and greater freedom to report the material
that they lawfully obtain.
225
See Leathers v. Medlock, 499 U.S. 439, 447 (1991) (observing that
“[t]he press plays a unique role as a check on government abuse” and “as a
watchdog of government activity”).
226
See supra notes 2-3 and accompanying text (describing this role).
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING: IF OMISSION IS THE
MISSION, WHERE IS THE FEDERAL TRADE
COMMISSION?
Brooke E. Crescenti*
INTRODUCTION
In 1915, the Federal Trade Commission (FTC) was organized
as an independent federal administrative agency pursuant to the
congressional mandate set forth in the Federal Trade Commission
Act of 1914 (FTC Act).1 The FTC is granted the authority to
investigate, prevent, and prosecute unfair or deceptive acts or
practices in or affecting commerce, including cases of false
advertising and unsubstantiated product claims.2 However, in spite
*
Brooklyn Law School Class of 2006; B.S. New York University, 2003.
The author wishes to thank her parents for their constant love and faith. She
would also like to thank the staff of the Journal of Law and Policy for their
assistance and hard work. Special thanks to her family and J.P., R.Z. and S.M.
for their encouragement and to J.G. for his patience and support.
1
15 U.S.C. § 41 (2003).
2
15 U.S.C. § 45(b) (1938). With regard to the administration and
enforcement of the FTC Act, if the FTC has reason to believe that a violation
has occurred of any of the proscriptions in the Act against unfair or deceptive
acts or practices in or affecting commerce or unfair methods of competition in
and affecting commerce, it may issue a complaint setting forth the
Commission’s charges. Once the complaint is served on the individual or
business entity, the FTC will hold a hearing. Id. If, after the hearing, the FTC
believes that the individual or business entity indeed engaged in deceptive acts
or practices, it may issue a cease and desist order against the practice. Id. The
federal courts of appeals have exclusive jurisdiction over challenges to FTC
cease and desist orders. Id. § 45(d). The findings of fact, if supported by
evidence, are conclusive on appeal. Id. § 45(c). Violations of final orders,
whether adjudged by the FTC or the courts of appeals, result in civil penalty
699
BROOKE MACROED CORRECTED 053105.DOC
700
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
of the FTC’s authority to investigate deceptive acts and its
requirement that advertisers disclose material connections between
product endorsers and the companies selling the endorsed
products,3 the FTC thus far has failed to scrutinize undercover
marketing, a growing, innovative, and unconventional form of
marketing communication.4
Undercover marketing is “a type of advertisement using actors
in real-life settings to make them appear to be average people.”5 It
utilizes “people hired by a company to surreptitiously promote a
product in public establishments.”6 In a typical undercover
marketing campaign, a marketer hires an actor to go into public
places and approach consumers with one objective in mind: to
pitch a product without revealing to consumers that they are the
subjects of a marketing campaign and that the actor is being paid to
promote a product.7 A successful undercover marketing operative
appears to be just another satisfied customer.8 The target consumer
believes that she is simply engaged in a spontaneous conversation
with an average Joe or Jane.9 Marketing firms believe that
actions, brought by the Attorney General, and mandatory injunctions. Id. § 45(l).
In addition, the FTC is granted rulemaking power to supplement the FTC Act in
order to curb deceptive practices or unfair competition occurring on an industrywide basis. 15 U.S.C. § 57a (1914).
3
Guides Concerning Use of Endorsements and Testimonials in
Advertising, 16 C.F.R. §§ 255.0-255.5 (1980) [hereinafter FTC Guides].
4
As of Apr. 19, 2005, all articles and sources cited in and researched in
preparation for this note failed to report any FTC proceedings against
undercover marketers.
5
Undercover Marketing, WEBSTER’S NEW MILLENNIUM DICTIONARY OF
ENGLISH (2003), available at http://dictionary.reference.com/search?q=
undercover+marketing.
6
Id.
7
Walking, Talking Stealth Ads, THE JOURNAL RECORD, Sept. 12, 2002, at
2002 WL 4937133; 60 Minutes: Undercover Marketing Uncovered (CBS
television broadcast, July 25, 2004), available at http://www.cbsnews.com/
stories/2003/10/23/60minutes/main579657.shtml.
8
The Evening Standard: If This Woman Offers You a Drink, Don’t Think
It’s Your Night (UK television broadcast, July 31, 2001) (noting that undercover
marketing operatives are not genuine, disinterested consumers, but rather, the
“secret agents of capitalism”).
9
Id.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
701
undercover marketing works to capture the coveted 18- to 34-yearold demographic, which historically has been unmoved by
traditional print and broadcast advertising.10 Sam Ewen, CEO of
Interference, Inc., a major New York undercover marketing firm,
says of undercover marketing, “[W]e can target customers at those
times when they’re open to being talked to. It’s not as passive as
television or radio. That gives us an advantage.”11
FTC regulations, however, mandate that advertisers disclose
any “material connection” between a person endorsing a product
and the company selling the product.12 The FTC Guides
Concerning Use of Endorsements and Testimonials in Advertising
(FTC Guides) define a “material connection” as a relationship that
might affect the weight or credibility of the endorsement.13 This
note argues that undercover marketing is necessarily deceptive and
10
Vickie Maye, You’re Roach Bait to Marketing Spies, SUN HERALD, Aug.
19, 2001, at 42 (explaining that undercover marketers believe that going
undercover is the only way to reach consumers ages 12 to 34 who are “too savvy
to fall for traditional advertising methods”).
11
Ryan Naraine, Questions for Sam Ewen, CEO of Interference, Inc.,
atnewyork.com, July 25, 2001, at http://www.atnewyork.com/news/article.php
/8511_808381.
12
FTC Guides, supra note 3, at § 255.5. The regulation states:
When there exists a connection between the endorser and the seller of
the advertised product which might materially affect the weight or
credibility of the endorsement (i.e., the connection is not reasonably
expected by the audience) such connection must be fully disclosed. An
example of a connection that is ordinarily expected by viewers and
need not be disclosed is the payment or promise of payment to an
endorser who is an expert or well known personality, as long as the
advertiser does not represent that the endorsement was given without
compensation. However, when the endorser is neither represented in
the advertisement as an expert nor is known to a significant portion of
the viewing public, then the advertiser should clearly and
conspicuously disclose either the payment or promise of compensation
prior to and in exchange for the endorsement or the fact that the
endorser knew or had reasons to know or to believe that if the
endorsement favors the advertised product some benefit, such as an
appearance on TV, would be extended to the endorser.
Id. (emphasis added).
13
Id.
BROOKE MACROED CORRECTED 053105.DOC
702
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
therefore violates FTC regulations because the principal goal of
any undercover marketing campaign is to assure that target
customers are unaware that they are being pitched to by interested
product marketers.14 Jonathan Ressler, the pioneer of undercover
marketing in the United States and the founder of Big Fat
Promotions, Inc., one of the most prominent undercover marketing
firms in New York,15 boasts, “You can never, ever tell we’re doing
it. And we’ll never admit to it. If people ever know they’re being
marketed to, we’re not doing our job properly.”16 Although the
FTC prohibits marketers from engaging in this type of conduct
without revealing to consumers the endorsers’ financial
connections to the company,17 in practice, the agency has declined
to bring enforcement actions against undercover marketers
engaged in deceptive practices, despite the statutory mandate
directing the FTC to prevent deceptive acts.18
Part I of this note explores the role of the FTC in regulating,
enforcing, and defining the parameters of permissible marketing
techniques. Part II argues that undercover marketing is a deceptive
practice subject to FTC jurisdiction and that the FTC should
investigate the practice in order to comply more fully with its
14
Brian Steinberg, Undercover Marketing Is Gaining Ground, WALL ST. J.,
Dec. 18, 2000, at B17D (explaining that “the ploy has advertisers plant
seemingly average Joes in a demographically desirable crowd—without tipping
consumers off that the people touting the goods are hired to do so”).
15
JOEL BAKAN, THE CORPORATION: THE PATHOLOGICAL PURSUIT OF
PROFIT AND POWER 132 (2004) (explaining that Ressler is credited with the
invention of the undercover marketing technique).
16
Undercover Agencies, THE AUSTRALIAN, Sept. 27, 2001, at M03.
17
FTC Guides, supra note 3, at § 255.5. See regulation text cited supra
note 12.
18
15 U.S.C. § 45(a)(2) states that “[t]he Commission is hereby empowered
and directed to prevent persons, partnerships or corporations . . . from using . . .
unfair or deceptive acts or practices in or affecting commerce” (emphasis
added). Undercover marketing falls within this ambit because undercover
marketers do not disclose material connections between their companies and
their endorsers, as required by the FTC Guides. FTC Guides, supra note 3, at §
255.5 (commanding that “when there exists a connection between the endorser
and the seller of the advertised product which might materially affect the weight
or credibility of the endorsement [i.e., the connection is not reasonably expected
by the audience] such connection must be fully disclosed”).
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
703
mandate to enforce the FTC Act. Part III discusses the striking
similarities between undercover marketing and infomercials,19
which have been heavily regulated by the FTC in recent years, and
argues that undercover marketing should command similar FTC
regulation. Finally, Part IV evaluates the potential consequences of
the FTC’s inaction with regard to undercover marketing practices,
including further tarnished corporate credibility and widespread
consumer distrust. This note concludes with a call for action by the
FTC.
I. THE FTC’S REGULATION OF DECEPTIVE PRACTICES
The FTC was initially established to enforce antitrust
regulations.20 Thus, the FTC’s current position as an advertising
and marketing enforcer was a “fortuitous by-product” of the terms
of the FTC Act.21 In 1972, the U.S. Supreme Court, in Federal
Trade Commission v. Sperry & Hutchinson Co., construed the Act
to give the FTC the power to regulate deceptive advertising and
marketing affecting consumers.22 The Court remanded to the Fifth
Circuit a judgment setting aside the FTC’s cease-and-desist order
against Sperry & Hutchinson Co., a trading stamp company, for
violating federal antitrust law by suppressing customer trading
stamp exchanges.23 Finding that the FTC had inadequately linked
19
According to the FTC, infomercials are advertisements “presented in the
guise of a talk-show format.” FTC v. California Pacific Research, Inc., No. CVN-88-602BRT, 1991 WL 208470, at *3 (D. Nev. Aug. 27, 1991). Similarly,
undercover marketing schemes are face-to-face marketing interactions with
customers presented in the guise of everyday conversations with averagelooking strangers. See, e.g., The Evening Standard, supra note 8.
20
15 U.S.C. § 41 (effective Sept. 26, 1914).
21
W.H. Ramsay Lewis, Infomercials, Deceptive Advertising and the
Federal Trade Commission, 19 FORDHAM URB. L.J. 853, 854 (1992) (citing
EARL W. KITNER, A PRIMER ON THE LAW OF DECEPTIVE PRACTICES 56 (1978)).
While the FTC Act was originally intended to prevent instances of unfair
competition between companies, the FTC’s current regulatory scheme now
focuses on protecting consumers from marketing companies, in addition to
protecting companies from one another. Id.
22
FTC v. Sperry & Hutchinson Co., 405 U.S. 233 (1972).
23
Id. at 250.
BROOKE MACROED CORRECTED 053105.DOC
704
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
Sperry & Hutchinson Co.’s conduct to a violation of antitrust law,
the Court considered whether the FTC was limited to regulating
only those deceptive acts or practices that violate the spirit of the
antitrust laws, that is, only those practices that are unfair to
consumers and also affect competition, as the Fifth Circuit held.24
The Court held that the FTC has broad regulatory power to protect
both companies and consumers, noting that Congress, when
creating the statutory FTC powers, explicitly considered and
rejected the inclusion of a rigid list of unfair practices to be
regulated.25
The Federal Trade Commission’s authority to regulate
deceptive advertising and marketing is contained within Section 5
of the FTC Act (Section 5).26 Section 5 provides that unfair or
deceptive acts or practices in and affecting commerce are
unlawful.27 In Federal Trade Commission v. Motion Picture
Advertising Service Company, the Supreme Court noted the
generality of Section 5, explaining that the statute’s proscriptions
are flexible to permit their later definition “with particularity by the
myriad of cases from the field of business.”28 Further, in a
conference report, Congress explained that Section 5 is necessarily
24
Id. at 239, 248.
Id. at 239-240. Congress acknowledged that the potential for human
inventiveness in the field of consumer manufacturing made it unwise to strip the
FTC of regulatory discretion and latitude. Id.
26
As amended in 1938. Kathyleen A. O’Brien, Strategies for Successfully
Defending Against Federal Trade Commission Investigations of False and
Deceptive Advertising, 775 PLI/COMM 269 (1997) (explaining that while Section
12 of the Act relates specifically to false advertising concerning food, drugs,
devices, and cosmetics, Section 5 of the Act grants the FTC broad jurisdiction
over other types of deceptive acts and practices in and affecting commerce that
relate to the public interest).
27
15 U.S.C. § 45(a)(1) (1938) (stating “[u]nfair methods of competition, in
and affecting commerce, and unfair or deceptive acts or practices in or affecting
commerce, are hereby declared unlawful”).
28
FTC v. Motion Picture Advertising Service Co., 344 U.S. 392, 394
(1953). The Court, after explaining that Section 5 of the FTC Act gives the FTC
broad power to regulate and define unfair or deceptive acts and practices, held
that the FTC exercised proper discretion in finding that a film distributor’s
exclusive screening agreements with theater operators unfairly restrained
competition. Id. at 394-95.
25
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
705
broad because there is “no limit to human inventiveness in this
field” and an attempt to pinpoint particular prohibited practices
would prove futile.29
The FTC has interpreted its expansive mandate to permit the
agency’s investigation into
(1) any acts, practices, conduct, or circumstances which the
Commission has been authorized by law to investigate; (2)
suspected violations of the laws and regulations enforced
by the Commission; (3) industry practices to determine
whether a trade regulation, rule, legislation, or other means
of corrective action would be appropriate; or (4) possible
violations of a Commission order to cease and desist.30
FTC investigations may take two forms: nonpublic or public.31
In a nonpublic investigation, the FTC keeps confidential the details
of its investigation of individuals or business entities to protect
against premature adverse publicity.32 Conversely, the FTC may
conduct public investigations of the practices of an entire industry
or group of industries when it feels that the interests of the public
will be best served by an open investigation.33 In either case, with
29
H.R. CONF. REP. NO. 63-1142, at 19 (1914). See also FTC v. ColgatePalmolive Co., 380 U.S. 374, 385 (1965) (explaining that the generality of
Section 5 “necessarily gives the Commission an influential role in interpreting §
5 and in applying it to the facts of particular cases arising out of unprecedented
situations”).
30
FTC Operating Manual, ch. 3.1.2.1, available at http://www.ftc.gov/
foia/adminstaffmanuals.htm.
31
Id. at ch. 3.3.3.1, 3.3.3.2.
32
Id. at ch. 3.3.3.1.
33
Id. at ch. 3.3.3.2. The FTC publicly announces open investigations
through news releases outlining the FTC’s charges against an individual or
company and the allegedly offending act or practice. The news release may
also provide a synopsis of prior FTC action against the individual or
company. Each release directs readers to the FTC’s webpage or a mailing
address to obtain a copy of the formal complaint. See, e.g., News Release,
Federal Trade Commission, Ads for Various Diet Supplements and Topical
Gels Don’t Cut the Fat, Says the FTC (June 16, 2004) (announcing the
FTC’s administrative complaint against Basic Research, L.L.C. and others
for making false and unsubstantiated claims about weight-loss and fat-loss
gels and supplements), available at http://www.ftc.gov/opa/2004/06/
BROOKE MACROED CORRECTED 053105.DOC
706
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
the objective of compelling enforcement or corrective action, the
agency makes detailed inquiries into possible violations of the laws
and regulations it enforces.34
Enforcement mechanisms and other corrective measures
employed by the FTC include injunctions, investigational consent
agreements, and recommendations for formal FTC complaints,
trade regulation rules, industry guides, and policy statements.35
The wide spectrum of enforcement mechanisms available to the
FTC allows the agency to choose from among various remedies,
including the implementation of prophylactic measures, such as the
issuance of a new industry guide or enforcement policy
statement.36 These measures may guard against the potential evils
dietsupp.htm; News Release, Federal Trade Commission, Marketer of
Electronic Abdominal Exercise Belt Charged With Making False Claims
(Oct. 1, 2003) (announcing the FTC’s administrative complaint against
Telebrands Corp. for using deceptive practices, specifically
unsubstantiated product claims, in the marketing and selling of the “Ab
Force” exercise device), available at http://www.ftc.gov/opa/2003/10/
abforce.htm.
34
FTC Operating Manual, supra note 30, at ch. 3.3.4.1.
35
Id. at ch. 3.1.3.4.
36
Issuance of a new industry guide may be appropriate “[w]here there is an
indication that a large number of persons are engaged in a similar type of
violation” or when widespread violations are threatened by the existence of
“competitive considerations” that may “cause many individual persons and
firms to adopt and be reluctant to abandon a particular practice until other
members of the industry have begun to do so.” Id. at ch. 8.3.3(1)-(2). The
Operating Manual instructs that, “[w]hen promulgation of a guide may provide
the impetus for members of an industry to voluntarily correct their business
practices and thereby eliminate violations, it can be an effective means of
achieving compliance.” Id. at ch. 8.3.3(2). Industry guides may be more
instructive than official FTC regulations because the Operating Manual
encourages industry guide drafters to include “meaningful factual criteria” for
determining when a violation exists rather than making blanket legal
conclusions, such as stating that a practice is illegal when it has the capacity to
deceive. Id. at ch. 8.3.3(3). For example, in 2001, the FTC promulgated the
Guides for the Jewelry, Precious Metals, and Pewter Industries, which
addressed such industry-wide issues as the proper usage of certain terms and
representations regarding gemstones, metals, and product quality. Federal Trade
Commission, For Business, Jewelry Guides, available at http://www.ftc.gov/
bcp/guides/jewel-gd.htm.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
707
inherent in certain industries before the FTC specifically targets
individuals or business entities by issuing complaints and holding
formal hearings.37 The FTC thus enjoys great discretion in crafting
remedies to combat deceptive practices.38 To assist agency
officials in determining whether particular practices are deceptive,
the FTC has developed a Policy Statement on Deception.39
A. The FTC’s Policy Statement on Deception: Factors for
Identifying Deceptive Practices
In recognition of the potential for confusion regarding the
reach of the FTC’s broad mandate, the agency sought to provide a
single definitive statement on what constitutes a deceptive act or
practice by issuing a Policy Statement on Deception (Policy
Statement) in 1983.40 Although FTC policy statements are not
binding law,41 they set forth the circumstances in which the
Issuance of an enforcement policy statement may be based on the
“accumulated expertise of the Commission acquired from numerous
investigations and proceedings concerning a particular industry or practice,” or
may be “the outgrowth of an independent Commission determination that
pronouncement of an interpretive statement or enforcement policy upon a
particular subject will further the public interest.” FTC Operating Manual, supra
note 30, at ch. 8.5.4. For example, in 1994, the FTC issued the Enforcement
Policy Statement on Food Advertising, explaining its joint enforcement
jurisdiction over food advertising with the Food and Drug Administration and
the FTC’s method of evaluating certain nutritional claims. Federal Trade
Commission, For Business, Advertising Guidance, Food Advertising, available
at http://www.ftc.gov/bcp/policystmt/ad-food.htm.
37
See supra text accompanying note 2 (describing FTC enforcement
procedures).
38
FTC Operating Manual, supra note 30, at ch. 3.1.2.2 (explaining that
“[t]he Commission possesses broad jurisdiction to deal with unfair or deceptive
acts and practices and unfair methods of competition under the FTC [Act] and
the various special statutes enforced by the Commission”).
39
FTC Policy Statement on Deception (Oct. 14, 1983) [hereinafter FTC
Policy Statement], available at http://www.ftc.gov/bcp/policystmt/ad-decept.
htm.
40
Id.
41
FTC Operating Manual, supra note 30, at ch. 8.6.1 (stating that while
policy statements are intended to clarify the rules that they describe, they do not
BROOKE MACROED CORRECTED 053105.DOC
708
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
Commission intends to take corrective action, or provide
interpretive statements on laws or substantive rules.42 The FTC’s
1983 Policy Statement outlined three criteria that the FTC should
consider in determining whether a particular act or practice is
deceptive; each of the three factors must be present for the FTC to
intervene.43 These factors include the likelihood that the practice
will mislead the consumer, the reasonableness of the consumer’s
reaction to the practice, and the materiality of the practice from the
consumer’s perspective.
The FTC officially adopted the Policy Statement in In the
Matter of Cliffdale Associates.44 In that case, the administrative
law judge concluded that an unfair or deceptive practice was “any
advertising representation that ha[d] the tendency and capacity to
mislead or deceive a prospective purchaser.”45 The FTC, however,
rejected that approach as “circular and therefore inadequate to
provide guidance on how a deception claim should be analyzed.”46
Rather, the FTC explained that it must evaluate deceptive practices
have binding force).
42
Id. at ch. 8.5.2. The principle function of an enforcement policy
statement is to deter violations of the law by clarifying any ambiguities or
uncertainties that may arise concerning FTC enforcement policies. Id. at
ch. 8.5.3.
43
Id.
44
103 F.T.C. 110 (1984) (appending the Policy Statement to the decision).
The FTC held that Cliffdale engaged in deceptive acts and practices by making
unsubstantiated claims about its Ball-Matic Valve (an automobile retrofit
device). Id. Further, the FTC held that the claims were deceptive because
Cliffdale had failed to reveal its relationship with Ball-Matic endorsers, many of
whom were Cliffdale business associates being passed off as disinterested
customers. Id. In support of its decision, the FTC cited the FTC Guides, supra
note 3, at §§ 255.0-55.5, which require companies to disclose material
connections between the company and its product’s endorsers. Id. See infra Part
I.B for further discussion of the FTC Guides.
45
Cliffdale, 103 F.T.C. 110.
46
Id. The FTC recognized that a clearer, more articulable standard was
necessary for evaluating deceptive practices. Simply asking whether the
particular alleged deceptive practice had a tendency to deceive an undefined
class of consumers provided no framework for future evaluations under the FTC
Act. Id.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
709
based on the factors laid out in its Policy Statement.47
1. The “Likely to Mislead” Factor
The Policy Statement provides that a practice will be deemed
deceptive only in the case of a “representation, omission or
practice that is likely to mislead the consumer.”48 Because Section
5 is more prophylactic than punitive in nature, a mere capacity to
deceive may garner FTC attention.49 Thus, a finding of deception
does not require actual deception so long as the practice is likely to
mislead.50 If an individual or entity induces contact with a
consumer through “a representation, practice, or omission that is
likely to mislead the consumer,”51 a violation of the Act will result,
even if the consumer is later informed about the marketer’s use of
deceptive tactics.52 Further, if a message’s overall impression is
misleading, it is no defense that some elements of the
communication are true because “words and sentences may be
literally and technically true and yet be framed in such a setting as
to mislead or deceive.”53 For example, in In the Matter of
47
Id.
FTC Policy Statement, supra note 39.
49
FTC v. Sterling Drug, 317 F.2d 669, 674 (2d Cir. 1963). After noting the
broad scope of FTC power to regulate deceptive practices, the court held that the
FTC was not erroneously denied a temporary injunction against Sterling Drug.
Id. The FTC alleged that Sterling Drug deceived consumers by improperly
linking its drugs to an American Medical Association endorsement, but the court
found that the FTC failed to present sufficient evidence for such a finding at the
preliminary injunction stage. Id.
50
Id.; Resort Rental Car Sys. Inc. v. FTC, 518 F.2d 962, 964 (1975)
(explaining that “advertising capable of being interpreted in a misleading way
should be construed against the advertiser” and that “actual deception need not
be shown”).
51
Cliffdale, 103 F.T.C. 110.
52
Resort, 518 F.2d at 964 (explaining that “[t]he Federal Trade Act is
violated if [the act or practice in question] induces the first contact through
deception, even if the buyer later becomes fully informed”).
53
Bockenstette v. FTC, 134 F.2d 369, 371 (10th Cir. 1943). In
Bockenstette, the court upheld an FTC cease-and-desist order based on findings
that defendant hatchery owners deceptively advertised that they were linked to
the National Poultry Improvement Plan, which represented a choice group of
48
BROOKE MACROED CORRECTED 053105.DOC
710
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
Encyclopedia Britannica, the FTC found a deceptive omission in
violation of the FTC Act when sales representatives
misrepresented the purpose of their initial contact with
customers.54 Encyclopedia Britannica salespersons went door-todoor claiming to be conducting brand recognition surveys when, in
fact, the main purpose of their contact with the homeowners was to
sell encyclopedias.55 The FTC held such a failure to disclose
impermissible and deceptive because, although surveys were in
fact conducted, they were merely the gateway to encyclopedia
sales.56
2. The Reasonableness of Interpretation Factor
The Policy Statement provides that the second factor for
determining the existence of a deceptive act or practice is the
reasonableness of the consumer’s reaction to or interpretation of
the act or practice.57 The statement provides that “to be deceptive
the representation, omission or practice must be likely to mislead
reasonable consumers under the circumstances.”58 Reasonable,
however, does not necessarily mean intelligent or discerning.59
When reviewing FTC orders, courts recognize that in order to
receive protection, consumers should not be expected to be
especially bright or discerning.60 Indeed, as noted by the Supreme
female hens, when in fact they were not. Id. While the hatchery’s hens were
previously approved by the National Poultry Improvement Plan, once they came
under the defendants’ ownership they lost that title and could no longer be
advertised as such. Id.
54
87 F.T.C. 421 (1976).
55
Id.
56
Id.
57
FTC Policy Statement, supra note 39.
58
Id.
59
Beneficial Corp. v. FTC, 542 F.2d 611, 618 (3d Cir. 1976). In
Beneficial, the Court upheld an FTC order, holding that Beneficial’s
advertisements for loans to those entitled to income tax refunds were deceptive
because they failed to indicate that potential borrowers had to meet certain credit
standards set by Beneficial. Id.
60
Id. (quoting Callman, Unfair Competition and Trademarks § 19.2(a)(1),
which, inter alia, described the general consumer public as a vast multitude that
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
711
Court in Federal Trade Commission v. Standard Education
Society,
[l]aws are made to protect the trusting as well as the
suspicious. The best element of business has long since
decided that honesty should govern competitive enterprises,
and that the rule of caveat emptor should not be relied on to
reward fraud and deception.61
As to the act or practice in question, the Policy Statement cites
to Beneficial Corp. v. Federal Trade Commission, in which the
FTC found that “[i]n determining the meaning of . . . a sales
presentation, the important criterion is the net impression that it is
likely to make on the general populace.”62 Expanding on this
notion, in American Home Products v. Federal Trade Commission,
the Third Circuit held that when considering the reasonableness of
the consumer’s reaction, the FTC has the right to examine the total
impression made by an advertising tactic.63 Otherwise, “the
Commission would have limited recourse against crafty advertisers
whose deceptive messages were conveyed by means other than, or
in addition to, spoken words.”64 Therefore, when considering a
consumer’s reaction to an act or practice, the FTC must
acknowledge that “[t]he ultimate impression upon the mind of the
[consumer] arises from the sum total of not only what is said but
includes “the ignorant, and unthinking and the credulous, who, in making
purchases, do not stop to analyze but too often are governed by appearances and
general impressions”); Aronberg v. FTC, 132 F.2d 165, 167 (7th Cir. 1942)
(quoting Callman as cited in Florence Mfg. Co. v. Dowd, 178 F. 73 2d Cir.
1910). The Aronberg court affirmed an FTC cease-and-desist order against the
defendant, charging that Aronberg engaged in deceptive practices by advertising
his over-the-counter medicinal remedy without revealing its potential dangers to
consumers’ health. Id.
61
302 U.S. 112, 116 (1937).
62
Beneficial, 542 F.2d at 618 (citing Grolier, 91 F.T.C. 315, 430 (1978),
remanded on other grounds, 615 F.2d 1215 (9th Cir. 1980), modified on other
grounds, 98 F.T.C. 882 (1981), reissued, 99 F.T.C. 379 (1982)).
63
695 F.2d 681 (3d Cir. 1982). In American Home Products, the court held
that substantial evidence supported an FTC finding that the defendant
deceptively represented that its nonprescription analgesics were proven superior
to competitors.
64
Id. at 688.
BROOKE MACROED CORRECTED 053105.DOC
712
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
also of all that is reasonably implied.”65
3. Materiality Factor
The Policy Statement also requires that the representation,
omission, or practice be “material” in order to give rise to a finding
of deception.66 The Policy Statement defines a material
representation as an act or practice likely to affect the consumer’s
conduct or decision with regard to the challenged product or
service.67 So long as materiality is found, it is irrelevant whether a
consumer was actually deceived by the practice.68 In Federal
Trade Commission v. Colgate-Palmolive Co., the Supreme Court
made clear that material representations or omissions are not
limited to those related to the substantive aspects of the product or
service being promoted.69 Indeed, the Court held that any
representations or omissions that materially induce consumer
action are subject to the proscriptions of the Act.70
65
Aronberg, 132 F.2d at 167 (noting that the public is often influenced by
appearances and general impressions and not simply the specific language used
in advertisements).
66
FTC Policy Statement, supra note 39. The Policy Statement cites the
definition of materiality in the RESTATEMENT OF TORTS, SECOND § 538(2)
(1977), which states that a material misrepresentation or omission is one that
“the reasonable person would regard as important in deciding how to act, or one
which the maker knows the recipient, because of his or her own peculiarities, is
likely to consider important.” The RESTATEMENT § 538(2)(a) cmt. d (1977)
states that the material fact does not necessarily have to affect the money paid in
a transaction and explains that “there are many more-or-less sentimental
considerations that the ordinary man regards as important.” Id.
67
FTC Policy Statement, supra note 39.
68
Id.
69
Colgate-Palmolive Co., 380 U.S. at 386-87 (rejecting respondent’s
argument that “the only material facts are those which deal with the substantive
qualities of a product,” in favor of the FTC’s position that “the
misrepresentation of any fact so long as it materially induces a purchaser’s
decision to buy is a deception prohibited by § 5”). In Colgate-Palmolive, the
Court reinstated an FTC order that held that the undisclosed use of a prop made
of plexiglass and sand in a commercial to simulate a razor shaving a piece of
sandpaper clean was a material deceptive practice. Id.
70
Id. Accordingly, deception may stem from marketing or advertising
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
713
B. FTC Guides Concerning Use of Endorsements and
Testimonials in Advertising
In addition to its Policy Statement on Deception, the FTC has
provided guidance regarding the use of advertising endorsements
in Section 255 of Title 16 of the Code of Federal Regulations.71
Promulgated by the FTC, this section, known as the FTC Guides
Concerning Use of Endorsements and Testimonials in
Advertising,72 defines endorsement as
[a]ny advertising message (including verbal statements,
demonstrations, or depictions of the name, signature,
likeness or other identifying personal characteristics of an
individual or the name or seal of an organization) which
message consumers are likely to believe reflects the
opinions, beliefs, findings, or experience of a party other
than the sponsoring advertiser. The party whose opinions,
beliefs, findings, or experience the message appears to
reflect will be called the endorser and may be an individual,
group or institution.73
Section 255.5 provides that when there is a material connection
between an endorser and the seller of an advertised product that
might “materially affect the weight or credibility of the
endorsement,74 such connection must be fully disclosed.”75
Specifically, the FTC mandates that if actors are employed to pose
as actual customers, this fact must be disclosed to the public.76 The
methods, not just from unsubstantiated product claims.
71
FTC Guides, supra note 3, at §§ 255.0-255.5.
72
15 U.S.C. § 57a (1914) (giving the FTC authority to prescribe rules with
respect to deceptive acts or practices in and affecting commerce).
73
FTC Guides, supra note 3, at § 255.0(b) (emphasis added).
74
FTC Guides, supra note 3, at § 255.5. In other words, the connection is
not reasonably expected by the audience.
75
Id. (emphasis added).
76
FTC Guides, supra note 3, at § 255.2(b). The regulations state,
“[a]dvertisements presenting endorsements by what are represented, directly or
by implication, to be ‘actual customers’ should utilize actual customers, both in
the audio and video or clearly and conspicuously disclose that the persons in
such advertisements are not actual customers of the advertised product.” Id.
BROOKE MACROED CORRECTED 053105.DOC
714
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
FTC clarified the appropriate application of the regulation with the
following hypothetical example:
An advertisement purports to portray a “hidden camera”
situation in a crowded cafeteria at breakfast time. A
spokesperson for the advertiser asks a series of actual
patrons of the cafeteria for their spontaneous, honest
opinions of the advertiser’s recently introduced breakfast
cereal. Even though the words “hidden camera” are not
displayed on the screen, and even though none of the actual
patrons is specifically identified during the advertisement,
the net impression conveyed to consumers may well be that
these are actual consumers, and not actors. If the actors
have been employed, that fact should be disclosed.77
II. UNDERCOVER MARKETING IS A DECEPTIVE PRACTICE BASED ON
FTC STANDARDS
A marketing industry publication aimed at chief marketing
officers defines undercover marketing as “the use of actors or shills
to pitch a product in a public place without it being revealed that
they are salespeople.”78 Advertising and marketing industry leaders
and numerous consumer advocacy groups are increasingly
questioning and criticizing the use of this practice.79 The executive
director for the Center for Digital Democracy calls the undercover
77
FTC Guides, supra note 3, at § 255.2, Example 3 (emphasis added). Part
III of this note further discusses the FTC’s application of the FTC Guides
to infomercials and argues that undercover marketing is as deceptive as certain
infomercial practices, which are heavily regulated by the FTC.
78
Lingo Lab: A Marketing Glossary: Undercover Marketing, CMO
Magazine, available at http://www.cmomagazine.com/glossary/term.html?
CID=55.
79
Suzanne Vranica, That Guy Showing Off His Hot New Phone May Be a
Shill, WALL ST. J., July 31, 2002, at B1 (noting that the executive director of
Commercial Alert, a nonprofit consumer activist organization, criticizes
undercover marketing as deceptive); Walking, Talking Stealth Ads, supra note 7
(explaining that many people in the advertising industry agree with consumer
protection advocates that undercover marketers go too far and quoting a
managing director of Atlanta’s Titan Advertising who argues that undercover
marketers are lying to and deceiving the public).
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
715
marketing phenomenon the “brand-washing of America.”80
Nevertheless, the FTC thus far has taken no initiative to challenge
the practice, despite its signature element of deceptiveness.81
Although the FTC has no specific regulations addressing the
practice of undercover marketing, its authority to regulate
undercover marketing falls within the sweeping coverage of
Section 5 of the FTC Act.82
As a threshold matter, the FTC has jurisdiction to regulate
undercover marketing as a form of advertising in and affecting
commerce.83 Advertising has been defined to include “[a]ny action
intended to draw the attention of the public or of a segment thereof
to merchandise, a service, a person or organization, or to a line of
conduct.”84 Undercover marketing easily falls within this
categorization. Operatives take action to draw unsuspecting
consumers’ attention to a certain product.85 Typical undercover
marketing techniques include placing attractive young men and
80
Daniel Eisenberg & Laura Bradford, It’s an Ad, Ad, Ad, Ad World: As
Conventional Methods Lose Their Punch, More Marketers are Going
Undercover to Reach Customers, TIME (Canadian Edition), Sept. 23, 2002
(explaining that critics believe that undercover marketing “tinker[s]” with
consumers’ minds). The Center for Digital Democracy seeks to enhance public
understanding of the U.S. digital media system and make the media industry
more accountable to the public interest. See CDD Mission Statement, at
http://www.democraticmedia.org/cddmissionstatement.html.
81
As of Apr. 19, 2005, all articles and sources cited in and researched in
preparation for this note fail to report any FTC proceedings against undercover
marketers.
82
15 U.S.C. § 45 (a)(2) (1938) (stating “[t]he Commission is hereby
empowered and directed to prevent persons, partnerships, or corporations . . .
from using unfair methods of competition in or affecting commerce and unfair
or deceptive acts or practices in or affecting commerce”).
83
FTC Policy Statement, supra note 39 (explaining that marketing
practices, such as bait-and-switch schemes or providing incomplete information
to the consumer, are covered by the Act in addition to traditional advertising);
see also 15 U.S.C. § 45(2) (empowering and directing the FTC to “prevent
persons, partnerships, or corporations . . . from using . . . unfair or deceptive acts
or practices in or affecting commerce”).
84
O’Brien, supra note 26, at 276 (quoting George E. Rosden & Peter E.
Rosden, The Law of Advertising, 17.02[2] at 17-36 (1995).
85
60 Minutes, supra note 7.
BROOKE MACROED CORRECTED 053105.DOC
716
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
women in bars to conspicuously talk about how great their vodka
tastes and recommend that other patrons order the same,86 paying
mothers to extol the benefits of their new laundry detergent at their
children’s Little League games,87 and compensating celebrities to
appear on talk shows and praise prescription drugs without
indicating that they are spokespersons for the manufacturer.88
Although FTC spokespersons have indicated that undercover
marketing raises concerns, in their view, undercover marketing
may simply be unethical, not illegal.89 An assistant director of the
FTC’s Advertising Practices Division has said of undercover
marketing, “It’s troubling, but whether it rises to the level of being
illegal is not clear. At a minimum it’s not clear that there’s enough
harm done to make it a priority for the FTC.”90 However, an
analysis of a typical undercover marketing encounter reveals that
the success of undercover marketing depends on meeting all three
of the FTC’s criteria for a deceptive practice,91 none of which
references injury or “harm done” to the consumer.92 Therefore, the
FTC is empowered to regulate undercover marketing and should
make regulation of this deceptive practice a priority.
An example of a typical undercover marketing scheme assists
in illustrating the deceptive nature of this practice. In 2002, Sony
Ericsson hired the marketing firm Interference, Inc. and launched
86
Jim Rutenberg, Phenomenon (Buy Me), N.Y. TIMES, July 15, 2001, at 21.
Catherine Donaldson-Evans, Advertisers Go Undercover to Push
Products, FOX NEWS, Aug. 17, 2001, available at http://www.foxnews.com/
story/0%2C2933%2C32179%2C00.html.
88
Eisenberg, supra note 80.
89
Jana Ritter, Buyer Beware: Uncovering Undercover Advertising, THE
GALT GLOBAL REVIEW, Mar. 18, 2003, at http://www.galtglobalreview.com/
business/ buyer_beware.html (explaining that the Federal Trade Commission
has yet to charge Big Fat Promotions, Inc. with anything other than bad taste);
Donaldson-Evans, supra note 87 (quoting Mary Engle, assistant director of the
FTC’s advertising practices division).
90
Donaldson-Evans, supra note 87.
91
See Part II.A-C (noting the three factors necessary in order to find the
existence of a deceptive practice, including “likely to mislead,” “reasonableness
of interpretation” and “materiality”).
92
See Part I.A. 3 (noting that actual consumer deception is not necessary
for a practice to be deemed deceptive so long as there is a capacity to deceive).
87
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
717
its “fake tourist” campaign.93 Sony sent actors to crowded tourist
destinations, such as New York’s Times Square and Seattle’s
Space Needle, to pose as tourists.94 The operatives went out in
pairs, posing as boyfriend and girlfriend, and asked passersby if
they would mind taking pictures of the couple.95 The operatives
handed willing parties a new Sony camera cell phone, acting as if
it were their own, and initiated conversations about the phone and
its various features.96 At no point during the interaction did the
operatives reveal that they were employed by Sony to market the
new camera phone.97
Sony’s campaign is representative of many undercover
marketing schemes and, as such, it may serve to highlight the
applicability of the FTC’s criteria for identifying deceptive
practices to undercover marketing.98 Indeed, Sony’s campaign
satisfies each of the FTC’s criteria for deception, thus indicating
that undercover marketing is ripe for FTC regulation and
enforcement action. First, Sony misleads consumers by omitting
the fact that the “tourists” are actually paid endorsers.99 Second,
the beliefs of consumers that they are not part of a paid marketing
campaign when they are spontaneously stopped on the street are
93
Vranica, supra note 79; 60 Minutes, supra note 7. See also Case Study:
Sony Ericsson, at http://www.interferenceinc.com/sony.html. The case study
outlines the marketing elements utilized in the Sony campaign, including fake
tourists and “leaners,” the term used for attractive actors placed in bars and clubs
to promote the camera.
94
Vranica, supra note 79; 60 Minutes, supra note 7 (noting that in Times
Square, the “ancient home of American marketing, people on the street once
“picked pockets, but now, they mess with your mind”).
95
Vranica, supra note 79 (noting that the operatives have no intention of
identifying that they work for Sony); 60 Minutes, supra note 7.
96
Vranica, supra note 79 (explaining that the idea is to have consumers
believe that they’ve “stumbled onto a hot new product”); 60 Minutes, supra note
7.
97
Vranica, supra note 79; 60 Minutes, supra note 7 (explaining that the
Sony operatives were “irresistibly innocent looking” and that the “Good
Samaritan” picture-taking consumers had no idea that they were “being had”).
98
See supra Part I.A (outlining the three FTC criteria for finding a practice
deceptive).
99
See infra Part II.A.
BROOKE MACROED CORRECTED 053105.DOC
718
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
entirely reasonable.100 Finally, Sony’s failure to disclose the true
nature of the interaction is a material omission.101
A. Failure to Disclose That Undercover Operatives Are Paid
Marketers Is An Omission Likely to Mislead the
Consumer
In order to qualify as a deceptive practice, the Policy Statement
first requires that there be an omission likely to mislead the
customer.102 Undercover marketing easily satisfies this criterion
because the omission in undercover marketing is the essence of the
sales pitch. Instead of deceiving the consumer about a particular
attribute of the product or service, a successful undercover
marketing operative deceives the consumer into believing that he is
not being marketed to at all.103 In fact, the thought should never
enter the consumer’s mind and, if it does, the undercover
marketing interaction is viewed as a failure.104 If an undercover
marketing operative fails to disclose that she is paid by a company
to promote its product and gives no information to that effect, the
customer is almost certain to be deceived by this omission.105
With undercover marketing, omission is the mission.106 In fact,
100
See infra Part II.B.
See infra Part II.C.
102
FTC Policy Statement, supra note 39. See Part I.A.1.
103
See Part I.A.1 (discussing deceptive representations and omissions).
104
Undercover Agencies, supra note 16 (noting undercover marketer
Jonathan Ressler’s insistence that his operatives’ true identities remain secret in
order to preserve the success of the campaign).
105
A useful comparison may be made to infomercials, a marketing practice
heavily targeted by the FTC in recent years. See Lewis, supra note 21; O’Brien,
supra note 26. Infomercials are troubling to the FTC because it is rarely clear
whether the audience members or endorsers in the infomercials are paid for their
participation, thereby creating a tendency to deceive viewers. See Lewis, supra
note 21. Undercover marketing is much the same. The deception lies not in the
claims about the product, but in the representation that the endorser has no
financial connection to the product he is extolling. See infra Part III for a further
discussion of this comparison.
106
John Heinzl, Beware Tourists With Talking Cameras, THE GLOBE AND
MAIL, Aug. 1, 2002, at A1 (commenting on the Sony “fake tourist” operatives,
101
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
719
“[t]he whole key to undercover marketing is never knowing that
it’s going on.”107 Once a consumer finds out that the person he was
chatting with is actually a shill, the cover is blown and the
transaction loses its effectiveness.108 In the Sony cell phone
campaign, for example, the passerby is not told that the “tourist” is
actually a Sony employee.109 The seemingly innocent acts of being
asked to take a photograph and engaging in friendly banter about
the camera are not likely to tip off consumers that they are really
the targets of a marketing pitch. A skillful operative cleverly
disguises her mission.110 Consequently, there is an omission that is
likely to deceive.111 Further, in addition to the Policy Statement,
the endorsement-specific regulations contained in the FTC Guides
Concerning Use of Endorsements and Testimonials in
Advertising112 provide that when marketers do not reveal that their
endorsers are paid, customers will be deemed to have been
unlawfully deceived.113 The scenario engineered by Sony’s
Jon Maron, director of marketing for Sony Ericsson, explained, “they’re not
identified as Sony Ericsson employees because it takes the spontaneity of the
conversation away”).
107
BAKAN, supra note 15, at 132-34 (quoting Jonathan Ressler of Big Fat
Promotions, Inc.).
108
Id.
109
Heinzl, supra note 106.
110
Undercover Agencies, supra note 16.
111
Encyclopedia Britannica, 87 F.T.C. 421.
112
FTC Guides, supra note 3, §§ 255.0-55.5. See Part I.B (discussing the
FTC Guides).
113
See, e.g., In the Matter of Creative Health, Inc., 2004 FTC LEXIS 51
(2004) (holding that Creative Health Inc. engaged in a deceptive practice in
violation of the FTC Act and the FTC Guides by failing to reveal that some
endorsers were principals in a public relations company that earned
commissions on sales related to its promotions and that other endorsers were
product distributors who earned profits based on their sales of the product); In
the Matter of TrendMark, Inc., 126 F.T.C. 375 (1998) (holding that TrendMark,
Inc.’s failure to reveal that its endorsers were distributors of its diet products
who earned profits on their sales was a deceptive practice in violation of the
FTC Act and the FTC Guides); In the Matter of Bodywise International, Inc.,
120 F.T.C. 704 (1995) (holding that Bodywise International, Inc.’s failure to
reveal that its endorsers were healthcare professionals and physicians who
derived income from Bodywise product sales was a deceptive practice in
BROOKE MACROED CORRECTED 053105.DOC
720
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
operatives ensures this result.
B. Consumers in Undercover Marketing Campaigns Will Not
Reasonably Suspect That They Are the Subjects of a
Sales Pitch
The second criterion in evaluating a potentially deceptive
practice requires an examination of the act or practice from the
perspective of a consumer acting reasonably in the
circumstances.114 The relevant question is whether the consumer’s
reaction to or interpretation of a practice is reasonable; in this case,
whether a consumer’s impression that he was not part of a paid
marketing interaction is reasonable.115
Undercover marketers strive to penetrate consumers’ lives and
buying processes in a seamless and undetectable manner.116 The
goal of undercover marketers is to make sure that the deception is
as authentic as possible.117 The target audience is the 18- to 34year-old market, a market that increasingly rejects the traditional,
obvious sales pitch.118 Operatives will congregate in places where
consumers usually gather, such as bars and parks.119 They have
violation of the FTC Act and the FTC Guides).
114
FTC Policy Statement, supra note 39. See also Part I.A. 2.
115
The FTC Policy Statement explains that the appropriate inquiry is
whether the consumer’s interpretation of a practice is reasonable. See supra note
39. Therefore, in an undercover marketing situation, the appropriate inquiry
would be whether the consumer’s belief that he was conversing with an average
Joe or Jane is reasonable.
116
Undercover Agencies, supra note 16 (quoting John Palumbo, Big Fat
Promotions, Inc.’s chief strategy officer). Palumbo explained, “[P]eople have to
see [the product], they have to understand it in a real way. The only way for
them to understand it in a real way is for it to be in their world. And that’s what
we do. We put it in their life.” Id.
117
Rutenberg, supra note 86.
118
Maye, supra note 10 (explaining that Jonathan Ressler believes that
undercover marketing is the only way to reach young people who are “too savvy
to fall for traditional advertising methods”); The Evening Standard, supra note 8
(noting that 12- to 34-year-old consumers are especially hard to reach because
they have “grown up with the heavy-sell and are now inured to it”).
119
Rutenberg, supra note 86.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
721
similar physical characteristics as the consumers, enabling them to
fit in with the target audience, or are especially attractive in order
to attract the attention of the target audience.120 When the
marketing communication is over, the consumers should have no
inclination that they were the subjects of a marketing pitch.121
Undercover marketing actors who do their jobs well create the
reasonable impression that they are not being paid to push a
product.122 Therefore, it is perfectly reasonable for consumers to
believe that they are not part of a paid marketing interaction. The
“sum total” of what is said and what is reasonably implied would
lead a reasonable consumer to believe that he was innocently
conversing with a friendly passerby, not that he was the subject of
an undercover marketing scheme.123
By way of example, Essential Reality, a gaming company,
launched an undercover campaign to promote its new “P-5 Glove,”
a gaming accessory that enables users to fly planes and fire
weapons on their computers.124 Innocent-looking actors were hired
to place themselves in Starbucks coffee shops, use the glove, and
120
Consuelo Lauda Kertz & Roobina Ohanian, Recent Trends in the Law of
Endorsement Advertising: Infomercials, Celebrity Endorsers and Nontraditional
Defendants in Deceptive Advertising Cases, 19 HOFSTRA L. REV. 603 (1991)
(explaining that endorsement marketing is most effective when the consumer
can identify with the endorser); The Evening Standard, supra note 8 (explaining
that undercover marketing companies choose their operatives with care to find
just the right look for the target consumer).
121
Rutenberg, supra note 86.
122
Id.
123
Aronberg v. FTC, 132 F.2d 165, 167 (7th Cir. 1942) (explaining that
“[t]he ultimate impression upon the mind of the [consumer] arises from the
sum total of not only what is said but also of all that is reasonably
implied”); see, e.g., 60 Minutes, supra note 7 (quoting Malcolm Gladwell,
author of THE TIPPING POINT). Gladwell explained of undercover marketing:
[T]here’s a set of rules that govern a lot of advertising and we’re aware
of the rules. We’re aware that the woman in the advertising for Ivory
Soap is prettier than most women in our lives. A line is crossed, I think,
when you go outside of those normal boundaries and start to deceive
people in ways that they are . . . totally unwitting to what’s going on.
Id.
124
60 Minutes, supra note 7.
BROOKE MACROED CORRECTED 053105.DOC
722
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
initiate conversation about the device.125 John Flaherty, an
unsuspecting coffee drinker, conversed with one of the actors
about the glove.126 When later told by 60 Minutes broadcast
producers that he was a subject in an undercover campaign, he
admitted that there was something disturbing about the scheme
once he learned the true purpose of the interaction.127 The Sony
“fake tourist” example is similar.128
The FTC might seek to regulate undercover marketing by
proscribing practices that deceive only the most savvy of
consumers;129 however, the FTC’s protection is intended for all
consumers, regardless of varied levels of intelligence, naivety, or
skepticism.130 Judging by the efforts expended by undercover
marketers to maintain their cover as average consumers, it is
highly unlikely that consumers will take away the reasonable
impression that they are conversing with paid actors.131
125
Id.
Id.
127
Id. Flaherty explained, “[i]t just seemed to me like a nice, friendly
encounter, and it kind of restores your faith in your fellow New Yorkers. And
then, to find out it was all fake, it was just kind of, I don’t know—I don’t like, I
don’t like the ring of it.” Id.
128
Vranica, supra note 79 (quoting Gary Ruskin, executive director of
consumer activist organization Commercial Alert). When told of the Sony
campaign, Ruskin argued, “[i]t’s deceptive. People will be fooled into thinking
this is honest buzz.” Id.
129
Lewis, supra note 21, at 860 (explaining that the FTC’s protection is
universal and intended for the gamut of American consumers). Lewis explains
that the FTC is “the guardian of the ignorant, unthinking and credulous, the
defender of ‘Mortimer Snerds’ and the protector of ‘wayfaring men, though
fools.’ This victim of deception may not read all that he should, and may merely
grab a general impression.” Id.
130
Id.
131
Vranica, supra note 79 (explaining that Sony has “gone to great lengths
to train its actors to avoid detection”). Peter Groome, president of Omnicom
Group Inc.’s Fathom Communications, explained, “[i]f you put [the actors] in a
Sony Ericsson shirt, then people are going to be less likely to listen to them in a
bar.” Id.
126
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
723
C. The Failure to Reveal That Undercover Operatives Are Paid
Marketers Is a Material Omission
The final requirement for finding a deceptive practice is that
the omission be material.132 A material omission is one that is
likely to affect the customer’s disposition toward the product being
pitched.133 Undercover marketing clearly satisfies this final
criterion, given that its ultimate goal is to favorably influence the
customer’s attitude toward the product.134 If Sony did not believe
that its stealth tactics would help to sell more camera phones or
generate greater buzz about its product, it would have relied
exclusively upon traditional advertising mechanisms to increase
sales.135 However, companies such as Sony and Essential Reality
are increasingly cognizant of the potential to increase profits
through the use of interpersonal deception.136 As a result of the
“fake tourist” campaign, Interference, Inc. reports that Sony
realized a fifty-percent increase in sales in the markets in which the
undercover marketing campaign was conducted.137 Marketers
recognize the importance of protecting their operatives’ identities,
and undercover marketing executives rarely divulge their client
132
FTC Policy Statement, supra note 39.
Id. The Policy Statement explains that “[w]here the seller knew, or
should have known, that an ordinary consumer would need omitted information
to evaluate the product or service . . . materiality will be presumed because the
manufacturer intended the . . . omission to have an effect.” Id.
134
Steinberg, supra note 14.
135
Maye, supra note 10 (noting that undercover marketers believe that
undercover marketing tactics are necessary to reach savvy, young consumers
who are unaffected by traditional advertising methods).
136
BAKAN, supra note 15 (arguing that corporations have no problem
deceiving consumers so long as they can benefit financially from the
interaction); Steinberg, supra note 14 (quoting marketing executives who say
that the trend in advertising must be to interpersonal communication and away
from traditional media); Alyson Ward, Marketers Finding Sneaky New Ways to
Pitch, SEATTLE POST-INTELLIGENCER, Sept. 17, 2002, at E1 (noting that
marketers are learning that they have to reach out to consumers where they
congregate, not just through a television set).
137
Case study: Sony Ericsson, Interference Inc. Home, Case Studies, Sony
Ericsson, at http://www.interferenceinc.com/sony.html (last visited Apr. 19,
2005).
133
BROOKE MACROED CORRECTED 053105.DOC
724
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
lists for fear of confidentiality breaches that could blow open
ongoing campaigns.138
Undercover marketing operatives are not merely satisfied
customers who wish to spread the joy they have found from using
the product they are pitching. They instead are being paid to make
the product look and sound as appealing as possible while posing
as a typical consumer.139 The use of this tactic makes undercover
marketing operatives endorsers by FTC standards because they are
advancing an advertising message that consumers are likely to
think is independent of the sponsoring marketer.140 When the
omission is related to an endorsement, the FTC specifically
mandates that any material connection between an endorser and
the seller of the product be disclosed.141 Given that omissions
relating to endorsements are considered material by the FTC,
undercover marketing-specific endorsement omissions should
receive due attention by the FTC. Marketers are required to
disclose to consumers material connections between endorsers and
sellers, and naturally, undercover marketing would be of no value
if this connection were disclosed.142 Indeed, companies require that
undercover operatives sign confidentiality agreements to ensure
that the marketers do not reveal that they are being paid to promote
the product.143 Customers deserve protection from these tactics
because such material omissions offend contemporary notions of
fairness,144 and more importantly, expressly violate FTC
138
Steinberg, supra note 14 (explaining that marketers are wary of naming
names for fear of rendering the campaigns ineffective).
139
Jim McBeth, Buying into the Virus, THE SCOTSMAN, Aug. 17, 2001, at 4
(explaining that undercover marketing operatives should be approachable and
not too good looking or obvious).
140
FTC Guides, supra note 3, at § 255.0 (providing that “[t]he party
whose opinions, beliefs, findings, or experience the [advertising] message
appears to reflect will be called the endorser and may be an individual,
group or institution”).
141
FTC Guides, supra note 3, at § 255.5.
142
Id.
143
BAKAN, supra note 15, at 134 (noting that undercover operatives are
contractually bound to conceal their relationships by the companies that employ
them).
144
Arthur Best, Controlling False Advertising: A Comparative Study of
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
725
regulations.145
In In the Matter of Cliffdale Associates, Inc.,146 the FTC,
interpreting the FTC Guides Concerning Use of Endorsements and
Testimonials in Advertising,147 opined that consumers are far more
likely to rely on endorsements they believe to be “independent and
unbiased” than those they know to be compensated.148 The FTC
held that the company’s failure to disclose the relationship between
itself and the endorsers materially affected the weight consumers
gave to a particular endorsement.149 Because making a false claim
of impartiality is material to consumers, such an omission is a
deceptive practice under the Act and the FTC Guides.150 For the
FTC, therefore, what matters is whether the consumer would have
thought differently of the message had he known it was
communicated by a paid employee.151 The agency’s holding in
Public Regulation, Industry Self-Policing, and Private Litigation, 20 GA. L.
REV. 1 (1985) (arguing that fairness is the goal of advertising regulation).
145
FTC Policy Statement, supra note 39; FTC Guides, supra note 3, at §
255.
146
103 F.T.C. 110 (1984).
147
FTC Guides, supra note 3, at §§ 255.0-55.5.
148
Cliffdale, 103 F.T.C. 110.
149
Id.
150
Id.
151
O’Brien, supra note 26. See also FTC Policy Statement, supra note 39
(citing RESTATEMENT (SECOND) OF TORTS § 538 cmt. d (1977)). The
Restatement explains that a material fact does not necessarily have to hinge on
the finances of the transaction, stating that “there are many more-or-less
sentimental considerations that the ordinary man regards as important.” The
inclusion of this Comment in the FTC Policy Statement indicates that the FTC
realizes that consumers are not solely concerned with the money they lose
because of deceptive practices. Undercover marketing affects emotions as well
in that one should be able to trust those with whom one interacts to be
forthcoming about their purposes for the interaction. See also 60 Minutes, supra
note 7 (quoting Malcolm Gladwell, author of THE TIPPING POINT). Of
undercover marketing, Gladwell noted:
Part of what makes real word-of-mouth so powerful is the
understanding that we have . . . that the person telling us about it is
telling us about it for . . . disinterested reasons. They’re not being paid
by somebody. They have our interests at heart. That is worlds apart
from a situation where the person telling us something is telling us that
BROOKE MACROED CORRECTED 053105.DOC
726
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
Cliffdale has been consistently applied to subsequent FTC
enforcement actions related to the failure of endorsers to reveal to
consumers their material connections with marketers, and applies
easily to undercover marketing.152
III. UNDERCOVER MARKETING SHOULD BE REGULATED BY THE
FTC, GIVEN ITS MARKED SIMILARITY TO INFOMERCIALS, WHICH
ARE THE SUBJECTS OF HEAVY FTC SCRUTINY
Since the 1990s, the FTC has strictly regulated deceptive
practices in the “infomercial” industry.153 Infomercials take the
form of full-length talk show programs (often enhanced by studio
audiences) devoted exclusively to demonstrating the particular
product being marketed.154 The product demonstrators and
audience members endorsing the product in these programs appear
to be objective and independent of the company selling the
product.155 In light of the potential of these programs to confuse
consumers, the FTC has found that infomercials violate the FTC
Act’s proscriptions on deceptive practices when they are designed
because they have some private agenda. They’re getting paid. They’re
being planted.
Id.
152
See, e.g., In the Matter of Melinda R. Sneed, 128 F.T.C. 322 (1999)
(holding that the Sneeds’ failure to disclose that John Sneed, as an endorser,
received a financial benefit from product sales was a deceptive practice in
violation of the FTC Act because such a fact would materially affect the weight
and credibility given by customers to the endorsement and would be material to
customers in their decision to purchase the product); In the Matter of Taleigh
Corp., 119 F.T.C. 835 (1995) (holding that Taleigh’s failure to disclose that
endorsers of its diet pills were compensated was a deceptive practice in violation
of the FTC Act because such a fact would be material to customers in making
purchase decisions regarding the products).
153
O’Brien, supra note 26 (citing several enforcement actions and
explaining that “‘[i]nfomercials,’ which are program-length advertisements
which frequently masquerade as talk shows, are quickly becoming a favorite
target of the FTC”). According to the FTC, infomercials are advertisements
“presented in the guise of a talk-show format.” FTC v. California Pacific
Research, Inc., No. CV-N-88-602BRT, 1991 WL 208470, at *3 (D. Nev. 1991).
154
Lewis, supra note 21, at 853.
155
Id. at 869.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
727
to look like independent television programs rather than paid
commercial advertising, and thus, are likely to mislead
consumers.156 Additionally, the FTC has found that the failure of
infomercial sponsors to disclose that demonstrators and
participants receive compensation from the sponsors constitutes a
violation of the FTC Guides on Endorsements and Testimonials in
Advertising.157 The guides mandate disclosure of any material
connection between an endorser and the sponsor company.158
The FTC has instituted enforcement actions against several
companies for broadcasting infomercials that are presented as
independent television programs rather than paid commercial
advertisements.159 In In re Twin Star Products,160 a highly
publicized enforcement action, the FTC enjoined Twin Star from
scripting its infomercials in a way that misled consumers into
156
See, e.g., In the Matter of Bogdana Corporation, 126 F.T.C. 37 (1998)
(finding that Bogdana employed a deceptive format for its infomercials in
violation of the FTC Act by expressly or impliedly representing that the
infomercials were independent radio programs, not paid commercial
advertising); In the Matter of Mega Systems International, Inc., 1998 FTC
LEXIS 207 (finding that Mega Systems engaged in deceptive acts and practices
in violation of the FTC Act by formatting its infomercials so as to resemble
independent television programming, not paid commercial advertising); In the
Matter of Nutrivida, Inc., 126 F.T.C. 339 (1998) (finding that Nutrivida made
false and misleading representations in violation of the FTC Act by expressly or
impliedly representing that its infomercials were independent television
programs, not paid commercial advertising); In the Matter of Wyatt Marketing
Corp. Inc., 118 F.T.C. 117 (1993) (finding that Wyatt’s infomercial format
constituted a deceptive practice in violation of the FTC Act because it expressly
or impliedly represented itself as an independent television program, not a paid
commercial advertiser).
157
In re Twin Star Prods. Inc., No. C 3307 (FTC Oct. 2, 1990) (decision
and order) (enjoining Twin Star from further broadcasting its infomercials and
from making deceptive claims about its paid endorsers); FTC Guides, supra note
3, at § 255.5.
158
FTC Policy Statement, supra note 39. See supra Part I.B (explaining the
coverage of the FTC Guides).
159
See supra note 156 for several examples of typical infomercial
injunctions.
160
In re Twin Star Prods. Inc., No. C 3307 (FTC Oct. 2, 1990) (decision
and order).
BROOKE MACROED CORRECTED 053105.DOC
728
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
believing that the broadcasted pieces were independent consumer
programs.161 In particular, Twin Star deceptively represented that
endorsers appearing on the program reflected “[t]ypical members
of the public who . . . were independent from all entities marketing
the product, when, in fact, the endorsers were its paid
employees.”162 The FTC’s action against Twin Star indicates that
the agency recognizes the danger inherent in fooling consumers
into believing that they are hearing endorsements from purportedly
objective consumers.163
Undercover marketing is similarly problematic. Infomercials
first attracted the attention of the FTC because of their deceptive
format.164 Product demonstrators and audience members appeared
to be extolling the benefits of products of their own accord.165
Undercover marketing campaigns employ the same tactics, using
seemingly objective and disinterested street operatives who are, in
actuality, paid actors.166 Both infomercial sponsors and undercover
marketers intend and strive to deceive customers to capture their
audiences’ attention and market their products.167 In light of the
parallel nature of these tactics, the FTC should similarly target
undercover marketing for enforcement action, given that
undercover marketing campaigns, much like infomercials, are
deceptively structured so as to resemble otherwise objective
interactions.
The infomercial format developed when advertisers realized
that while very few people would be inclined to sit and watch a
161
Id.
Id.
163
Id.
164
O’Brien, supra note 26 (discussing the FTC’s penchant for strictly
regulating infomercials). O’Brien explains, “So great are infomercials’ capacity
to deceive that they have been described as the ‘seamy underside of the
advertising business.’” Id. (citing Joanne Lipman, Infomercial Makers Try to
Clean Up Their Act, WALL ST. J., Mar. 4, 1991, at B3).
165
See supra note 156 (citing examples of typical FTC enforcement actions
based on the false independence of the demonstrators and audience members).
166
Walking, Talking Stealth Ads, supra note 7.
167
Lewis, supra note 21, at 871 (arguing that infomercial producers never
intend clarity in their advertisements because of the need to make customers
believe that they are watching an objective television program).
162
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
729
thirty-minute commercial, many more people would watch a
thirty-minute commercial disguised as an objective talk show.168
Similarly, undercover marketing developed when marketers
realized that their target audience, resistant to traditional
advertising, would be much more willing to listen to an average
Joe or Jane than a paid spokesperson.169 Based on the FTC’s active
regulation of infomercials,170 logic dictates that the FTC should be
just as concerned with undercover marketing because, like
infomercials, undercover marketing campaigns are premised on
attempts to deceive consumers through the use of orchestrated
endorsements.171
IV. A CALL FOR ACTION BY THE FTC
The FTC thus far has failed to seriously scrutinize undercover
marketing practices. The public record is devoid of pending
168
Id. at 865.
Maye, supra note 10 (explaining that undercover marketers believe that
going undercover is the only way to reach consumers ages 12 to 34 who are “too
savvy to fall for traditional advertising methods”).
170
See supra note 156 for several examples of typical infomercial
injunctions.
171
FTC Guides, supra note 3, at § 255.5 (stating that any material
connection between an endorser and a seller of a product must be disclosed).
This blanket rule applies to any marketing practice, regardless of form, so long
as an endorsement is involved. On its official website, the FTC addresses
infomercials in a section titled Frequently Asked Advertising Questions, warning
potential infomercial sponsors that they should make sure not to “deceptively
mimic the format of news reports, talk shows or other independent
programming.” Further, recognizing that many infomercials contain
endorsements, the FTC refers interested parties to the FTC Guides on
Testimonials and Endorsements in Advertising, at http://www.ftc.gov/bcp/
conline/pubs/buspubs/ad-faqs.htm. Although the FTC Guides do not specifically
mention infomercials, it is clear that the rule is intended to cover all
endorsements. Undercover marketing, like infomercials, utilizes endorsements
and, as such, is subject to the FTC Guides governing endorsements. FTC
Guides, supra note 3, at § 255.0-55.5. The FTC should, therefore, give
undercover marketing the same attention it has given to infomercials, given the
analogous nature of the practices. See Part III (discussing the similarities
between infomercials and undercover marketing).
169
BROOKE MACROED CORRECTED 053105.DOC
730
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
investigations of the practice, filings of individualized FTC
complaints, or entries of enforcement orders against specific
undercover marketers.172 However, undercover marketing
undeniably is a deceptive practice under FTC standards,173 and
therefore, the FTC should take affirmative measures to regulate
this practice.
Several policy reasons support the FTC’s regulation of
undercover marketing. First, given the FTC’s broad power to
regulate deceptive practices, the agency should not shy away from
regulating undercover marketing simply because the practice is an
innovative one.174 If the FTC fails to regulate such new
technologies, the effectiveness of its regulation of deceptive
practices will be severely diminished.175 Additionally, the FTC
should act to regulate undercover marketers based on the
increasing threat to consumer protection posed by the nowblossoming use of undercover marketing campaigns by top
corporations.176 Moreover, should undercover marketing tactics
come to the attention of the public, consumer backlash might
result; the FTC would be wise to anticipate this response and
proactively protect the consumer market.177 Finally, undercover
marketing creates a culture of deception that is antithetical to the
mandate of the FTC, an agency created for the express purpose of
ensuring fair and honest commercial practices.178
A. The FTC’s Authority and Responsibility to Widen Its Scope
of Investigation and Enforcement to Cover New
Technologies, Including Undercover Marketing
The FTC has a duty to act in the interest of the public and
172
Based on Westlaw and Lexis Nexis searches of FTC proceedings as of
Apr. 19, 2005.
173
FTC Policy Statement, supra note 39; FTC Guides, supra note 3, at §
255.
174
See infra Part IV.A.
175
Id.
176
See infra Part IV.B.
177
See infra Part IV.C.
178
See infra Part IV.D.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
731
investigate deceptive acts and practices.179 Given the expanding
budgets devoted to and the scope of undercover marketing
campaigns, the FTC should act expeditiously to regulate this
marketing technique.180 Currently, undercover marketing is
primarily of interest only to journalists. Notably, articles on this
marketing practice are devoid of any mention of FTC action.181
Perhaps more telling, when interviewed on the subject, Jonathan
Ressler, of Big Fat Promotions, Inc., reported that he has yet to
hear from the FTC and does not expect to.182
Although undercover marketing is significantly different in
form from traditional print and broadcast advertising, the FTC
should not retreat from regulating this marketing practice,
particularly in light of the widespread criticism that undercover
marketing has garnered from both consumer groups and
advertising executives.183 In American Home Products v. Federal
Trade Commission, the Supreme Court explained that the FTC
must be able to expand the parameters of its enforcement to ensure
that narrow investigatory or regulatory interpretations do not leave
loopholes through which marketers may accomplish the prohibited
179
15 U.S.C. § 45(b) (1914).
There are emerging signs of awareness of the practice at the FTC. In a
New York Times Magazine interview, the assistant director of advertising
practices at the FTC said of undercover marketing, “If [sic] testimonial is
affiliated with you in some way, you have to disclose that.” However, the
director would not comment on whether the practice was being investigated.
Rutenberg, supra note 86. Another associate director for advertising practices at
the FTC told the Wall Street Journal that undercover marketing “certainly raises
ethical questions. At some point it raises legal questions. If a person in a bar
makes claims about a product without revealing the fact that he is being paid to
promote, you could well have FTC problems.” Steinberg, supra note 14.
181
As of Apr. 19, 2005, all articles cited in and researched in preparation
for this note fail to report any FTC proceedings against undercover marketers.
182
John Heinzl, Advertising Slinks Undercover, THE GLOBE AND MAIL,
July 20, 2001.
183
Heinzl, supra, note 106 (quoting Gary Ruskin, executive director of
Commercial Alert, who explained that undercover marketing is the
commercialization of human relationships); Rutenberg, supra note 86 (quoting
Jeff Chester, a board member at the Center for Media Education, who called
undercover marketers deceptive “commercial kamikazes”).
180
BROOKE MACROED CORRECTED 053105.DOC
732
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
goal of deception.184
Following this mandate, the FTC has continually widened its
lens of investigation to cover new technologies.185 In In the Matter
of All-State Industries of North Carolina, then-FTC Commissioner
Philip Elman explained that the FTC retains jurisdiction over
practices that may not have been thought of previously and has the
duty of developing a body of unfair trade practices law that is
“adapted to the diverse and changing needs of a complex and
evolving competitive system.”186 Undercover marketing, if left
unchecked, could severely diminish the effectiveness and bite of
the FTC’s regulation of deceptive practices.187 Section 5 of the Act
was drafted broadly for the specific purpose of giving the FTC the
necessary flexibility to combat “new and improved” deceptive
practices.188 Therefore, because the FTC is empowered to regulate
this deceptive practice, the agency should investigate undercover
marketing with vigor and resist justifying agency inaction by
referencing the innovative nature of the practice.
184
American Home Products, 695 F.2d at 704. The Court explained, “[i]f
the Commission is to attain the objectives Congress envisioned . . . it must be
allowed effectively to close all roads to the prohibited goal, so that its order may
not be bypassed with impunity.” Id.
185
Lewis Rose, Stealth Marketing (Interactive Marketing Under
Government Scrutiny), MARKETING COMPUTERS, Apr. 1, 1995, at 20 (noting that
the FTC has applied its traditional rules to developing technologies over the
years, including print ads in 1914, radio ads in the 1920s, television ads in the
1950s, infomercials in the 1980s, and 900 numbers and Internet scams in the
1990s).
186
In the Matter of All-State Industries of North Carolina, Inc., 75 F.T.C.
465 (1969) (noting that the changing characteristics of the American
marketplace forced the FTC to focus its attention upon deceptive practices
associated with credit transactions).
187
American Home Products, 695 F.2d at 704. As the Court recognized in
American Home Products, the potential for innovation and craftiness in the
marketing industry makes it essential that the FTC enforce its provisions from
all angles. The Court pointed out, “[i]f the Commission is to attain the objectives
Congress envisioned . . . it must be allowed effectively to close all roads to the
prohibited goal, so that its order may not be bypassed with impunity.” Id.
188
See supra Part I (noting the generality of Section 5).
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
733
B. Undercover Marketing’s Rapid Growth and Popularity with
Major Corporations Frustrates the Protection of
Consumers
Undercover marketing is no longer a grassroots practice.189
Large, big-budget corporations, such as Microsoft, Kellogg, BMW,
Apple, and Cadbury Schweppes PLC, are putting their advertising
dollars into undercover marketing campaigns.190 Further, Big Fat
Promotions Inc.’s profits increased fivefold in the first two years of
the company’s existence.191 Clearly, undercover marketing is no
longer a niche endeavor; however, one would be hard pressed to
compile reliable estimates of corporate spending on undercover
marketing, in part because companies resist admitting that they use
stealth methods.192 Presumably, the more money spent on
undercover marketing campaigns, the more consumers will be
subjected to polished schemes, inevitably leading to widespread
deception.193 Consumers are not the only ones falling prey to
undercover marketers. Business owners frequently have no idea
that their establishments are being used by undercover marketing
operatives to push products; further, they are given no
compensation in exchange for the use of their venues.194 As
undercover marketers secretly pour greater amounts of money into
189
John Arlidge, The Way We Live Now: Too Good to be True, THE
OBSERVER (London Edition), Aug. 12, 2001, at 6; Vranica, supra note 79.
190
Arlidge, supra note 189; Vranica, supra note 79.
191
Morgan Campbell, You Won’t Feel a Thing, TORONTO STAR, Aug. 7,
2001, at D01.
192
Eisenberg, supra note 80.
193
Id. (explaining that “industry experts say that outlays for alternative
campaigns are growing rapidly–and that Madison Avenue has little choice but to
seek new ways to push products”).
194
Jane Standley, Undercover Advertising Targets Consumers, BBC NEWS,
August 17, 2001, available at http://news.bbc.co.uk/1/hi/business/ 1496213.stm
(explaining that “the conversations struck up or the recommendations you
overhear are selling not just talking and sometimes even the business owner
doesn’t even know what is going on”).
BROOKE MACROED CORRECTED 053105.DOC
734
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
more elaborate campaigns, they will soon have both unwitting
consumers and business owners literally eating, drinking, and
snapping photos out of their proverbial hands.
C. Consumer Backlash: An Inevitable Consequence of
Undercover Marketing in a Time of Tarnished
Corporate Credibility
Despite some undercover marketers’ hopeful outlook for the
potential of undercover campaigns, as undercover marketing
tactics are revealed to the general public, marketers run the risk of
even further tarnishing corporate credibility in this age of corporate
scandal.195 The FTC might determine that it is in the best interest
of society and a well-functioning marketplace to preempt the
spread of distrust that may balloon as undercover marketing tactics
come to light.196 The fact that successful advertising industry
leaders themselves have decried the tactics of undercover
marketers should make clear to the FTC that undercover marketing
is not only a questionable practice, but also a dangerous one for
corporate goodwill.197 Some in the industry warn that backlash
195
Ritter, supra note 89 (arguing that undercover marketing could be bad
for business overall, in that if advertising techniques become so deceptive,
corporate credibility could take a hit and result in an overriding state of distrust
of companies).
196
See Harmon Leon, Secret Agents of Capitalism: Is That Hottie Really
Flirting With You, or Is It Undercover Marketing?, METROACTIVE, May 12,
2004, at http://www.metroactive.com/papers/metro/05.12.04/marketing-0420.
html (imagining an “Orwellian . . . paranoid environment” resulting from
undercover marketing, where interactions with others are made to be
questioned).
197
Vranica, supra note 79 (quoting David Lubars, president and creative
director at Fallon Worldwide, a prominent advertising agency, who cautioned
that marketers should be honest with customers in order to avoid a backlash in
the future once undercover marketing strategies come to light). See also
Eisenberg, supra note 80 (quoting Keith Reinhard, chairman of DDB
Worldwide advertising agency, criticizing undercover marketing as “bad
business”); Walking, Talking Stealth Ads, supra note 7 (quoting Scott Marticke,
managing director of Atlanta’s Titan Advertising, who decried undercover
marketing, saying, “[a]s it goes on, it will create a kind of disconnect, and
people won’t know whom to trust”).
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
735
should be a real concern of undercover marketers.198 However,
judging by the burgeoning number of corporations employing
undercover marketers and the money being spent on these
campaigns, it does not appear that advertisers are taking such
warnings to heart.199
A 2004 Marketing and Ethics News Poll conducted by CMO
Magazine, a publication for chief marketing officers, indicates that
marketing executives worry about the effects of unethical practices
in business, especially in light of the recent scandals involving
Enron and Martha Stewart.200 Seventy-three percent of those
polled believe that increasing penalties for offenders is the best
way to deter future unethical practices.201 The marketing
executives also stated that deceptive sales and marketing practices
were the top ethical issue facing their industry today, followed
closely by dishonesty with customers.202 Furthermore, those polled
viewed deceptive marketing practices as one of the top three
198
See, e.g., 60 Minutes, supra note 7 (quoting Malcolm Gladwell, author
of THE TIPPING POINT). Of undercover marketing, Gladwell warned:
My problem with undercover marketing is not what happens in the
moment. It’s what happens in a week, or two weeks, or a month down
the road, when we discover we’ve been duped. And I think that the
moment when we discover we’ve been duped causes a backlash.
Companies who engage in this practice are courting that backlash. And
that’s a very, very dangerous thing to play with.
Id. See, e.g., Rogier van Bakel, A Letter From the Editor: Big Fat Liars,
ADVERTISING AGE’S CREATIVITY, Aug. 1, 2001, at 6. Van Bakel, editor of
Advertising Age’s Creativity, warns that “[u]nlike flavored water poured over
vodka, credibility and deceit never mix. You can’t build a brand without first
building trust, and the fastest way to squander trust is to play people for
suckers. . . .” Id.
199
See supra Part IV.B (discussing the rapid growth of corporate budgets
for undercover marketing campaigns).
200
IDG’s CMO Magazine Reports the Untold Challenges, Evolving
Pressures that Chief Marketing Officers Face, PRIMEZONE MEDIA NETWORK,
Aug. 30, 2004, available at 2004 WL 91766902.
201
CMO Marketing and Ethics News Poll, available at http://www.
cmomagazine.com/ethics_poll/.
202
Id. at 5. See, e.g., Vranica, supra note 79 (quoting a marketing executive
who characterized undercover marketing as “reprehensible” and “desperate”).
BROOKE MACROED CORRECTED 053105.DOC
736
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
ethical issues facing U.S. businesses generally.203 Perhaps this is
because 44.6 percent of those polled admitted that they had
witnessed someone in their company engage in or develop
deceptive or misleading sales or promotion tactics.204 These
numbers are telling. Marketing executives recognize that deceptive
marketing practices pose a disturbing ethical dilemma and concede
that regulatory enforcement and penalties would effectively deter
such conduct.205 The FTC should take advantage of this corporate
attitude by enacting regulations to rein in the use of these troubling
and deceptive practices.
D. Undercover Marketing Breeds an Atmosphere of Distrust
and Skepticism
Undercover marketing adds a new twist to advertising.
Although traditional advertising makes use of persuasion or charm,
it does not typically involve explicit trickery.206 Undercover
marketers flagrantly deceive consumers, and they are not shy about
it.207 One undercover marketer told the New York Times Magazine
that his goal for each mission was simple: to subtly impart the
sponsor’s message to consumers and “implant things about the
product into their head.”208 One scholar argues that undercover
marketing is distinguishable from most other types of advertising
because it is based in artificiality.209 Consumers expect
203
CMO Marketing and Ethics News Poll, supra note 201.
Id.
205
Id.
206
Yvonne Cartwright, Undercover Marketing: Pitching You on the Sly,
BELLINGHAM BUSINESS JOURNAL, Nov. 1, 2003, at B13 (discussing the opinion
advanced by Malcolm Gladwell, author of THE TIPPING POINT, that undercover
marketing is a bit of a con game). Gladwell’s statement highlights the point that
traditional advertising tries to coax customers towards a certain product, while
undercover marketing tries to hoodwink them. Id.
207
Undercover Agencies, supra note 16 (quoting Jonathan Ressler, founder
of Big Fat Promotions, Inc., who boasted that his operatives will never reveal
their true agendas).
208
Rutenberg, supra note 86.
209
Thomas Nord, Stealth Marketing – Is it the Next Big Thing or Just a Big
Fat Flop?, THE COURIER-JOURNAL, Aug. 3, 2001, at 1C (quoting a professor of
204
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
737
commercials on television, but not in their local bars and hangouts.
Companies such as Big Fat Promotions, Inc. and Interference, Inc.
intentionally situate endorsers in places where people’s guards are
down.210
Jonathan Ressler of Big Fat Promotions, Inc. attempts to
defend his practices by stating that the operatives are telling the
truth about the positive attributes of the products they are
promoting.211 However, courts have held that truthfulness within
marketing will not save a message that is misleading overall.212
Although the FTC mandates that endorsers reveal their connections
to sponsor companies, undercover marketers avoid doing so at all
costs.213 So long as the FTC continues to turn a blind eye to these
deceptive practices, people may eventually become increasingly
skeptical of one another’s motives. They may begin to ask
themselves, “does this girl really want to have a drink with me or is
she being paid by the gin company?” or “does my neighbor really
prefer that brand of detergent or has she been hired to chat me up
about it?” Undercover marketing creates a “sort of Truman Show
situation where the world is full of ‘real people’ acting as your best
mates when, in fact, they are paid brand spokesmen.”214 Such a
state of distrust cannot be in the public’s best interest, which the
marketing at Indiana’s Kelley School of Business).
210
Id.
211
BAKAN, supra note 15, at 134 (explaining that Ressler is proud of his
commitment to honesty in telling the truth about the products). See also
Campbell, supra note 191 (quoting John Palumbo, chief strategy officer at Big
Fat Promotions, Inc., who said that the practice is honest because the company
only sends out operatives who use the products being hocked).
212
Bockenstette, 134 F.2d at 371 (explaining that “words and sentences
may be literally and technically true and yet be framed in such a setting as to
mislead or deceive”).
213
FTC Guides, supra note 3, at § 255.5; Campbell, supra note 191
(quoting Jonathan Ressler of Big Fat Promotions, Inc., who explained that the
company will not reveal its clients’ names). Ressler explained, “by naming them
we render the whole promotion useless. The key here is confidentiality.” Id.
214
Arlidge, supra note 189 (quoting Sean Pillot de Chenecey, a trends
forecasting consultant who works for Saatchi & Saatchi, Weiden & Kennedy,
and McCann Erickson).
BROOKE MACROED CORRECTED 053105.DOC
738
6/20/2005 7:11 PM
JOURNAL OF LAW AND POLICY
FTC is empowered to promote.215 Kalle Lasn, the editor of
Adbusters magazine, lamented the potential effect of undercover
marketing on communities as a whole. She explained, “It is a form
of cultural corruption at a time when advertising already pervades
the landscape. It’s much more insidious because marketers are
creating culture at the grassroots level, on the streets and where we
live.”216
Even children are being exploited by undercover marketers.
Hasbro, a leading children’s toy company, recently recruited 1,600
boys from Chicago, Illinois, ages eight to ten, and paid them to
play a new handheld video game called “Pox” and to tell their
friends about it.217 Sales of the game skyrocketed and Hasbro spent
a fraction of what it would have on a traditional advertising
campaign.218 Undercover marketing is invading the playground,
the local watering hole, Times Square, and the laundromat, and can
easily spread anywhere. In its Policy Statement, the FTC warned
that the Commission “intends to enforce the FTC Act
vigorously . . . [and] will investigate, and prosecute where
appropriate, acts or practices that are deceptive.”219 The FTC must
wholeheartedly combat undercover marketing if it truly intends to
investigate and prosecute deceptive acts.
CONCLUSION
Although undercover marketing is unconventional and
somewhat enigmatic,220 the FTC retains the power to continually
215
15 U.S.C. § 45(b) (1938) (empowering the FTC to initiate proceedings
against individuals and companies using any unfair or deceptive practice in or
affecting commerce if such proceeding is in the public interest).
216
McBeth, supra note 139 (noting the observation of some critics that
undercover marketing is sinister in nature).
217
Id. (outlining the Hasbro campaign).
218
Id.
219
FTC Policy Statement, supra note 39.
220
As compared to a television commercial, for example. While
undercover marketing can take many forms in many different locales,
commercial advertising is considerably standard. Television commercials do not
command consumer interaction, while undercover marketers seek to engage
consumers both physically and verbally.
BROOKE MACROED CORRECTED 053105.DOC
6/20/2005 7:11 PM
UNDERCOVER MARKETING AND THE FTC
739
expand its focus to keep pace with the evolution of the marketplace
and to develop new enforcement priorities as times change.221 The
FTC should heed the charge of Judge Learned Hand, who observed
that the FTC’s duty is to “discover and make explicit those
unexpressed standards of fair dealing which the conscience of the
community may progressively develop.”222 FTC investigations
often serve as a vital means of gathering information for the
issuance of new FTC policy statements or industry guides.223
Therefore, even if the FTC is unsure at this stage whether
undercover marketing practices demand individual prosecutions,
the agency still can be proactive by providing undercover
marketers with specific guidance as to the permissible bounds of
their practices and by alerting them to the imposition of a more
regulatory stance in the future.224 As illustrated by the Sony
example, undercover marketing encapsulates all of the FTC criteria
for deceptive practices and, as such, should be addressed by the
FTC. However the FTC sees fit to best serve the public interest, it
must act now before undercover marketing becomes so seamless
that it is completely undetectable.
221
All-State Industries of North Carolina, Inc., 75 F.T.C. 465 (noting the
expectation that the FTC will continually adapt its regulations to changing
competitive systems).
222
FTC v. Standard Education Society, 86 F.2d 692, 696 (2d Cir. 1936),
rev’d on other grounds, 302 U.S. 112 (1937).
223
FTC Operating Manual, supra note 30, at ch. 3.1.3.4.
224
Id. at ch. 8.3.3 (explaining that the FTC may issue an industry guide to
address a specific practice in lieu of taking individualized enforcement action if
such a guide might spur companies to proactively curb the practice in order to
avoid FTC prosecution). However, the Operating Manual does caution that an
industry guide would be inappropriate if there are indications that the violations
are willful or wanton or if compliance with the standard of lawful conduct is
unlikely to be attained without an enforcement action or creation of a
substantive rule having binding effect. Id. at ch. 8.3.4. While an industry guide
on undercover marketing may indicate in black and white that undercover
marketing contravenes the FTC Act and the FTC Guides, undercover marketers
seem intent on doing exactly what the FTC Guides prohibit, namely, refusing to
disclose material connections between companies and endorsers. FTC Guides,
supra note 3, at §255.5. Therefore, an industry guide may do little more than put
a name to a practice that the FTC Act already prohibits.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
IN SEARCH OF CONGRESSIONAL INTENT:
DOES LIHPRHA RESTRICT STATE AND
LOCAL GOVERNMENTS FROM
PRESERVING AFFORDABLE HOUSING?
Michael Freedman∗
INTRODUCTION
During the past twenty-five years, the federal government has
reduced its role in maintaining the nation’s affordable housing
supply,1 shifting the burden of housing America’s poor to state and
local governments.2 From the New-Deal era through the 1970s, the
∗
Brooklyn Law School Class of 2006; B.A., University of California, Los
Angeles, 2000. The author would like to thank his family and friends from
coast-to-coast for their love, encouragement and unyielding support. Special
thanks to Professor David Reiss for his guidance and insight. Many thanks to the
staff of the Journal of Law and Policy for their assistance and patience.
1
Affordable housing refers to housing intended for “low-income” or “very
low-income” people. Families with incomes less than eighty percent of the area
median income are considered “low-income.” 42 U.S.C. § 1437(a)(b)(2) (2002).
Families with incomes less than fifty percent of the area median income are
considered “very low-income.” Id.
2
Roger K. Lewis, Free-Market Housing Programs Come Up Short as
Affordability Gap Widens With Age; Shaping the City, WASH. POST, Oct. 30,
1999, at G10; Wayne Washington, White House Starts Shift of Housing Burden
to States HUD Spokesmen Cite Delays in Using U.S. Aid, BOSTON GLOBE, Feb.
14, 2003, at A3; Glenn Thrush & Graham Rayman, Housing Sees Cuts in
Budget; Bush Administration Proposals Would Mean as Much as $200M Less
for Low-Income Vouchers, Subsidies, NEWSDAY (New York), Feb. 9, 2005, at
A07; Tim May, Low-Income Rentals Threatened as Vote Looms in U.S. Senate;
Housing: Affordability Rents and Efforts by Tenants to Own Their Buildings
May Hinge on Fate of Some Federal Subsidies. Los Angeles Could Be Hit
Particularly Hard, L.A. TIMES, Sept. 25, 1995, at B1; David W. Chen, HUD Aid
741
FREEDMAN MACROED CORRECTED 5-31-05.DOC
742
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
federal government played a major role in providing affordable
housing to the nation’s poor and low-income communities through
the creation of large, project-based subsidy programs that provided
incentives to owners to build and maintain affordable housing
projects.3 During the 1960s and 1970s, more than two million units
were built under project-based programs.4 Starting with the Reagan
Administration, however, federal housing policies began to reflect
a preference for letting the market address housing demands by
providing subsidies directly to tenants through housing vouchers,
while preserving and rehabilitating the existing stock of affordable
projects.5 Housing policies shifted further throughout the Clinton
Administration, during which the federal government eliminated
funding for the development of affordable housing projects and for
preservation incentives.6 The current administration is attempting
Short by $50 Million, City Reports, N.Y. TIMES, Jan. 27, 2005, at B5; Hillary
Stout, Housing Subsidies in Doubt, N.Y. TIMES, May 8, 1988, at sec. 8, 1; Ann
Mariano, Housing Policy Faces New Challenges; Low-Income Programs Seen
Threatened, WASH. POST, Nov. 24, 1985, at G1 (“Five years ago, housing the
poor was high on the list of national priorities, the fastest-growing federal
subsidy program of the decade. . . . But little more than a year ago, with the
election of Ronald Reagan, all that began to change.”).
3
Alfred M. Clark, III, Can America Afford to Abandon a National Housing
Policy?, 6 J. AFFORDABLE HOUS. & COMM. DEV. L. 185, 185 (1997) (citing the
Housing Act of 1949, 42 U.S.C. § 1452b (repealed 1990)); see also Peter
Salinas, Toward a Permanent Housing Problem, No. 95 Public Interest 22-23
(1986), reprinted in CHARLES E. DAYE ET AL., HOUSING AND COMMUNITY
DEVELOPMENT 14 (3d ed. 1999). Under these programs, private owners
contracted with the federal government to receive subsidies in return for keeping
rents affordable for low-income income tenants. Id. The term “project” is
sometimes used in reference to government owned and operated public housing.
“Project” in this note generically refers to any affordable housing development.
In addition to federal housing efforts, virtually every state and most larger
cities have agencies dedicated to the housing the poor. Id. State and local efforts
act as both conduits for federal assistance and supplement local efforts. Id.
4
Clark, supra note 3, at 187.
5
Michael Grunwald, Further Cuts Feared in Housing for the Poor;
Menino Says City Needs Game Plan, BOSTON GLOBE, Jan. 5, 1995, at 1 (“The
federal government has not built much affordable housing since 1980, but it has
left intact and often rehabilitated the existing stock.”); see also Mariano, supra
note 2.
6
HUD’s Flawed Blueprint, BOSTON GLOBE, Jan. 9, 1995, at 10; Grunwald,
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
743
to completely sever any ties between the federal government and
low-income tenants by providing housing funds directly to state
governments through block grants, along with the responsibility
for administering federal housing programs.7
The transfer to the states of responsibility for the creation and
maintenance of affordable housing has occurred largely with
Congress’s consent.8 State and local governments, however, are
supra note 5 (“Clinton’s plan to overhaul HUD to free cities and states from its
oversight and to increase choice for low-income tenants could chip away at that
stock, HUD officials conceded.”) (quoting Sue Marsh, executive director of the
Massachusetts Coalition for the Homeless).
7
Thrush & Rayman, supra note 2; Steve Twedt, The Fraying Safety Net,
PITTSBURGH POST-GAZETTE (Pennsylvania), Mar. 27, 2005, at A10; Ron
Nissimov, City Moving Quickly to Spend Block Grant Money; Funds Not Used
by May 1 Are at Risk of Being Lost, HOUS. CHRON., Mar. 25, 2005, at B5;
Jocelyn Y. Stewart, U.S. Blamed in Cuts to Rental Aid for Poor; Changes to
Housing Program Will Lead to Higher Costs, Canceled Contracts, Critics Say,
L.A. TIMES, June 3, 2004, at B1. Unfortunately, federal funding of block grants
for housing falls short of the amount promised to participants in current voucher
programs and is insufficient to maintain the programs’ effectiveness. David W.
Chen, U.S. Is Asked to Increase Housing Aid, N.Y. TIMES, Jan. 19, 2005, at B1
(“New York City officials say their projected share of federal funds for lowincome housing vouchers is more than $61 million short of what is needed . . . .
In all, 492 out of an estimated 2,500 housing agencies that issue vouchers have
asked for more money.”).
8
See Paulette J. Williams, Special Series: Developing Sustainable Urban
Communities: The Continuing Crisis in Affordable Housing: Systemic Issues
Requiring Systemic Solutions, 31 FORDHAM URB. L.J. 413, 455 (2004).
The story of affordable housing development during the 1980s is a
story of disinvestment by the federal government, devolution of the
responsibility for housing upon state and local government, and an
increasing sense that the private enterprise with government subsidies
could do a better job of addressing the continuing crisis than
government did at any level.
Id.; Michael H. Schill, Privatizing Federal Low Income Housing Assistance:
The Case of Public Housing, 75 CORNELL L. REV. 878, 907-16 (1990)
(advocating a market-based approach to the availability of public housing based
on the availability of maximal tenant choice and the use of vouchers to obtain
private housing).
The federal government has abandoned commitments in a variety of social
welfare contexts, creating new challenges for state and local governments, and
also new opportunities for the private market, primarily non-for-profit
FREEDMAN MACROED CORRECTED 5-31-05.DOC
744
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
not prepared, nor are they able, to take over where the federal
government left off. Budgetary challenges limit the ability of state
and local governments to develop new affordable housing projects
and force states to turn to their most controversial housing
regulations, including rent control and eminent domain, to preserve
a small but important stock of permanent affordable housing.9
organizations. See generally John J. Ammann & Peter W. Salsich, Jr.,
Symposium: Nonprofit Housing Providers: Can They Survive the “Devolution
Revolution”?, 16 ST. LOUIS U. PUB. L. REV. 321 (1997) (discussing a widening
role for nonprofits in the development of affordable housing policy).
The “devolution revolution,” as exemplified by the 1996 welfare
reform legislation, has created major uncertainties for housing and
homeless services providers. How the states will respond to new
responsibilities that are accompanied by fewer resources is a matter of
conjecture at the moment. As welfare reform begins to be implemented,
it is increasingly clear that it will have major impact on housing policy.
Low-income families may be able to accept “a new social contract that
expects and rewards work and responsible behavior” in return for help
in finding jobs, protecting children and escaping poverty.
Id. at 352.
9
David W. Chen, Bill on Affordable Housing Sets Up Sides for a Debate,
N.Y. TIMES, Oct. 29, 2004, at B5. See generally William A. Fishel & Perry
Shapiro, Takings, Insurance, and Michelman: Comments on Economic
Interpretations of ‘Just Compensation’ Law, 17 J. LEGAL STUD. 269 (1988),
reprinted in ROBERT C. ELLICKSON ET AL., PERSPECTIVES ON PROPERTY LAW
514 (2d ed. 1995).
State and local governments have used alternative means of developing
efficient, low-cost solutions to the growing affordable housing crisis. New York
City’s Tenant Empowerment Act: Hearing on Intro. No. 186 and Proposed Res.
No. 388-A Before the Committee on Housing and Buildings of the Council of the
City of New York (Oct. 28, 2004) (statement of James Grow, Esq., National
Housing Law Project) [hereinafter Grow]. These measures seek to reinstitute the
key “component of the federal preservation policy for HUD-subsidized
properties that the federal government abandoned since 1995—that preservation
is of sufficient importance to warrant restrictions on owner conversion, so long
as those restrictions are supported with market-value compensation.” Id. at 6.
While the over the direction of American housing policy continues, this
note assumes that, at the present time, the free-market is unable solve the
immediate needs of the nearly two million families that will be forced to
relocate if current privately-owned subsidized affordable housing projects
remove affordability restrictions. See, e.g., Peter Dreier, The New Politics of
Housing, 63 J. OF THE AM. PLANNING ASS’N 5 (1997), reprinted in DAYE, supra
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
745
Further, in their efforts to maintain the rapidly diminishing
affordable housing stock, state and local legislatures have
encountered difficulty in determining the scope of their authority
with respect to housing, given the uncertain application of a single
provision within a much larger federal statute—the Low Income
Housing Preservation and Resident Homeownership Act of 1990
(LIHPRHA or the Preservation Act).10
Programs for privately owned, federally subsidized housing
projects, under forty-year subsidized mortgages, originally
permitted developers to free themselves of affordability restrictions
after twenty years.11 LIHPRHA set forth a number of federal
preservation policies intended to protect low-income tenants from
profit-motivated owners seeking to convert their federally
subsidized housing projects to market-rate units.12 The
Preservation Act and its predecessor, the Emergency Low-Income
Housing Preservation Act of 1987,13 created a burdensome process
for converting federal affordable housing projects to market-rate
units (“opting out”), including a requirement that owners provide
findings that low-income tenants would not be adversely impacted
by the conversion of the regulated properties.14 This high threshold
note 3, at 7; RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 480 (4th ed.
1992).
10
Low-Income Housing Preservation and Resident Homeownership Act of
1990, 12 U.S.C. §§ 4101-4125 (1996) [hereinafter LIHPRHA]. See also New
York City’s Tenant Empowerment Act: Hearing on Intro. No. 186 and Proposed
Res. No. 388-A Before the Committee on Housing and Buildings of the Council
of the City of New York (Oct. 28, 2004) (statement of Prof. Paula Franzese, Esq.,
Seton Hall School of Law) (examining federal preemption of a proposed New
York City preservation law) [hereinafter Franzese] (on file with author).
11
See infra notes 37-41 and accompanying text.
12
12 U.S.C. §§ 4101-4125. David Etezadi & Peter Manzo, LIHPRHA:
Great Opportunity for Community Control, 4 J. AFFORDABLE HOUS. & COMM.
DEV. L. 67 (1994-1995).
13
Emergency Low-Income Housing Preservation Act of 1987, 12 U.S.C. §
1715l (1989 & Supp. 1995).
14
See LIHPRHA, 12 U.S.C. §§ 4101, 4102(a), 4107, 4108. Under
LIHPRHA’s prepayment procedures, an owner first had to file with HUD,
tenants, and others a “notice of intent” to prepay. Id. § 4102(a). The owner then
was required to submit a plan of action (POA) setting forth information relating
to the proposed prepayment. Id. § 4107. HUD would then evaluate the owner’s
FREEDMAN MACROED CORRECTED 5-31-05.DOC
746
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
essentially prohibited owners from prepaying forty-year mortgages
and from opting out of their respective federal housing programs
after the twenty years.15 Owners were protected under LIHPRHA,
however, from additional burdens imposed by state or local
governments through a preemption provision that prohibited states
from further restricting the prepayment of mortgages subsidized by
the U.S. Department of Housing and Urban Development
(HUD).16 This provision is codified in section 4122 of Title 12 of
the United States Code (Section 4122).
Facing budget concerns, Congress began to remove the
preservation restrictions in 1996 through various appropriations
acts. The first of these, the Housing Opportunity Program
Extension Act (HOPE),17 provided owners seeking to opt out an
alternative to LIHPRHA.18 Under HOPE, owners were able to opt
out of federal housing programs without HUD’s consent so long as
they agreed to not increase rents for sixty days.19 The question
remained, however, whether LIHPRHA’s preemption provision
would protect buildings opting out under HOPE’s terms from state
or local preservation initiatives. HOPE contains no preemption
provision, nor does it refer to LIHPRHA’s preemption provision.20
Indeed, there is no language in either of the respective statutes
POA for approval. Id. § 4101(a). Under the Act, HUD could issue such approval
for prepayment only after making certain written findings that the prepayment
would not adversely affect the low-income housing supply or involuntarily
displace current tenants. Id. §§ 4101(a), 4108.
15
Arthur R. Hessel & Susan Sturman, Heard From HUD: LIHPRHA
Comes Full Circle, 5 J. AFFORDABLE HOUS. & COMM. DEV. L. 316, 318 (1996).
16
12 U.S.C. § 4122.
17
Housing Opportunity Program Extension Act of 1996, Pub. L. No. 104120, § 2(b), 110 Stat. 834 (1996) (codified at 12 U.S.C.A. § 4101 note (1996))
[hereinafter HOPE].
18
See Kenneth Arms Tenant Ass’n v. Martinez, No. CIV. S-01-832, 2001
U.S. Dist. LEXIS 11470, at *13 (E.D. Cal. 2001) (“The Housing Opportunity
Program Extension Act of 1996 . . . permitted mortgage prepayment without
HUD’s approval.”).
19
See infra note 74.
20
Kenneth Arms, 2001 U.S. Dis. LEXIS 11470, at *26 (holding that HOPE
does not contain a preemption provision and when owners are involved with the
HOPE prepayment scheme, LIHPRHA’s preemption provision does not apply).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
747
directly discussing the relationship between and interplay of HOPE
and LIHPRHA.
Two recent cases considering the applicability of LIHPRHA’s
preemption provision to HOPE’s prepayment provisions
demonstrate the confusion surrounding this issue. The Eighth
Circuit, in Forest Park II v. Hadley,21 and the Ninth Circuit, in
Topa Equities v. City of Los Angeles,22 examined two principal
questions: (1) whether the LIHPRHA preemption provision applies
to buildings opting out under HOPE’s prepayment schedule; and
(2) whether the LIHPRHA provision preempts the respective
Minnesota and California preservation statute challenged in the
cases.23 Both courts agreed that, despite the defunding of
LIHPRHA, the Act’s preemption provision is applicable to
HOPE’s prepayment schedule because of the plain meaning of
LIHPRHA’s preemption provision and the state laws’ apparent
effect of regulating HUD.24 The Eighth and Ninth Circuits reached
conflicting holdings with respect to the second issue.25 The Eighth
Circuit employed a “practical effects” analysis, under which all
state or local preservation laws that have the effect of limiting or
delaying owners’ expectations of converting their affordable
housing projects to market rates are preempted.26 By contrast, the
Ninth Circuit employed a “legal consequences” test, under which
laws that “restrict or inhibit” the prepayment of federally
subsidized mortgages are preempted.27 The result is that state or
local laws (e.g., rent control regulations) in the Eighth Circuit that
restrict owners from realizing the potential gains from market-rate
rents following opt out are preempted, while those in the Ninth
Circuit, according to the court in Topa Equities, are not.
21
Forest Park II v. Hadley, 336 F.3d 724 (8th Cir. 2003), reh’g and reh’g
en banc denied (2003).
22
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
23
Id. at 1067; Forest Park, 336 F.3d at 727.
24
Topa Equities, 342 F.3d at 1069; Forest Park, 336 F.3d at 732.
25
Topa Equities, 342 F.3d at 1070; Forest Park, 336 F.3d at 733. See also
Independence Park Apts. v. United States, 61 Fed. Cl. 692, 704 (2004).
26
Forest Park, 336 F.3d at 733.
27
Topa Equities, 342 F.3d at 1070.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
748
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
The courts’ holdings in Forest Park and Topa Equities will
have direct consequences for whether current programs, which
increasingly are being administered by states, can continue to meet
“the critical and growing needs of lower-income Americans for
decent and affordable housing.”28 Nationwide, more than four and
a half million seniors, people with disabilities, and families with
low incomes live in federally subsidized affordable housing rental
units.29
In examining the application of LIHPRHA’s preemption
provision to buildings opting out under HOPE and other alternative
prepayment provisions, this note argues that the circuit courts
incorrectly adopted a plain meaning approach to the interpretation
of LIHPRHA’s preemption provision. In this case, preemption
jurisprudence dictates that ambiguous statutory language should be
interpreted in light of the relevant legislative history and the
underlying purpose and structure of the statutes. Indeed, a review
of the legislative purpose and history of LIHPRHA and the past
practice of HUD lead to the conclusion that LIHPRHA’s
preemption provision should not apply to buildings opting out
under HOPE.
Part I of this note examines the background of federal policies
regarding the development and preservation of affordable housing
projects. Part II discusses current state and local affordable housing
preservation policies. Part III examines Supreme Court
jurisprudence regarding constitutional preemption standards. Part
IV analyzes the holdings of Forest Park30 and Topa Equities.31
Part V.A examines the debate about whether LIHPRHA’s
preemption provision should be applied to alternative prepayment
processes such as HOPE. This part highlights the circuit courts’
28
Grow, supra note 9, at 3. These units represent more than one-third of
our country’s subsidized housing inventory. Id.
29
U.S. DEP’T OF HOUS. AND URBAN DEV., RENTAL HOUSING DYNAMICS: IS
AFFORDABLE HOUSING FOR THE POOR AND ENDANGERED SPECIES? (2003),
available at http://www.huduser.org/datasets/ahs/ahsReports.html#2. U.S. DEP’T
OF HOUS. AND URBAN DEV., THE WIDENING GAP: NEW FINDINGS ON HOUSING
AFFORDABILITY IN AMERICA (1997).
30
Forest Park, 336 F.3d at 733.
31
Topa Equities, 342 F.3d at 1070.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
AFFORDABLE HOUSING PRESERVATION
749
reliance on a plain meaning approach to the interpretation of the
LIHPRHA preemption provision, as compared to the broad
approach suggested by affordable housing advocates, who argue
that the interpretation of LIHPRHA requires the use of a range of
sources, including legislative history, and agency policy and
practice. Part V.B proposes that if courts continue to apply the
LIHPRHA preemption provision to HOPE or other prepayment
statutes, then constitutional preemption jurisprudence and
congressional history support the use of the Ninth Circuit’s legal
consequences test rather than the Eighth Circuit’s practical effects
approach. Part VI recommends that state and local governments
interested in preserving their privately owned, federally subsidized
affordable housing should enact preservation laws that enable and
aid the transfer of such properties to non-speculative preservation
owners following the current owners’ opting-out of federal
affordable housing programs.
I. BACKGROUND OF FEDERAL HOUSING DEVELOPMENT AND
PRESERVATION POLICIES
Since 1949, the priority of U.S. housing policy has been to
achieve “a decent home and suitable living environment for every
American family.”32 The National Housing Act of 194933
employed an approach that was unique among New Deal-era
policies in enticing private industry to aid the implementation of a
government initiative.34 Through below-market interest rate loans
32
Salinas, supra note 3, at 22-23. In pursuit of the objectives of the
Housing Act of 1949, the national government subsidized construction of 1.2
million new low-income apartments, 800,000 new apartments for moderateincome families, and 700,000 new apartments for the elderly. Id. Adding more
than 1.5 million rent supplements per year and other subsidy programs, the
federal government has subsidized more than 5 million households with new or
rehabilitated housing units since 1950. Id.
33
42 U.S.C. § 1441 (1949) [hereinafter National Housing Act of 1949].
34
Peter W. Salsich, Jr., Will the “Free Market” Solve the Affordable
Housing Crisis?, J. OF POVERTY L. AND POL’Y, Jan.–Feb. 2002, at 573
(explaining how the U.S. Housing Act of 1937, which created the federal public
housing program, created programs designed to construct and manage housing
projects in order to provide homes but also in major part to help the country
FREEDMAN MACROED CORRECTED 5-31-05.DOC
750
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
and general subsidies, federal housing programs provided
incentives to private developers to build inexpensive housing.35
In 1961, Congress enacted a below-market interest rate
program (Section 221(d)(3)),36 which was later replaced in 1968 by
an interest subsidy program (Section 236).37 Under Section
221(d)(3) and Section 236, HUD was authorized to insure loans
made by private lending institutions and to subsidize interest
payments on loans extended to profit-motivated developers.38
Through subsidies, HUD reduced the interest rates of private
market mortgages (usually about eight to ten percent) to between
one and three percent, and offered developers the option of
extending mortgage loans for the construction period plus forty
years.39 Owners were expected to share the benefits of the program
with tenants through lower rents.40
In exchange for favorable financing terms, owners of Section
221(d)(3) and 236 housing projects were required to comply with
certain minimum property standards and to maintain the housing
for occupancy by low-income families.41 Mortgage documents for
projects under these sections prohibited project owners from
prepaying their forty-year mortgages for a period of twenty years
without HUD’s prior consent.42 This restriction prevented owners
come out of the Great Depression by creating new construction jobs); Angela D.
Chatman, Housing Cuts; Restructuring HUD Will Shred the Safety Net for the
Poor. But the Pain Will Be Shared by All, PLAIN DEALER (Cleveland, Ohio),
Mar. 26, 1995, at 1C.
35
Salsich, supra note 34, at 576.
36
National Housing Act, 12 U.S.C. § 17151(d)(3) (1961).
37
Id. § 1715z-1 (1968).
38
Sheldon Winkelman, Low Income Housing Preservation and Resident
Homeownership Act of 1990, 73 MICH. B.J. 1160, 1160 (1994).
39
Id.
40
See Cienega Gardens v. United States, 194 F.3d 1231, 1234 (Fed. Cir.
1998), cert. denied sub nom. Sherman Park Apartments v. United States, 528
U.S. 820 (1999).
41
Winkelman, supra note 38, at 1160.
42
Cienega Gardens, 194 F.3d at 1234-35.
Generally, when obtaining a HUD-insured mortgage under either of
the above programs, an owner executed a deed of trust note payable to
a private lending institution. The note evidenced a loan made to the
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
751
from escaping the low-income housing limitations and use
restrictions during this period.43
Despite the government’s efforts to generate and maintain
affordable housing through the Section 221(d)(3), Section 236, and
Section 8 project-based programs, America’s affordable housing
needs remained substantially unmet.44 Many of the projects built
owner pursuant to a loan agreement between the owner and the lending
institution that contemplated advances to the owner. Payment of the
indebtedness evidenced by the note was secured by a deed of trust, or a
mortgage, on the subject property. The note and deed of trust were
printed on forms approved by HUD, and HUD endorsed the note as
part of its mortgage insurance. The repayment term of the loan was
generally forty years. Simultaneously, in exchange for HUD’s
endorsement for insurance (pursuant to a commitment for insurance),
the owner entered into a regulatory agreement with HUD, under which
the owner agreed, among other things, to certain “affordability
restrictions,” including restrictions on the income levels of tenants,
restrictions on allowable rental rates, and restrictions on the rate of
return the owner could receive from the housing project. The regulatory
agreement and the mortgage insurance provided by HUD were to
remain in effect so long as the loan remained outstanding.
While the regulatory agreement made no mention of the right to
prepay the outstanding loan, a rider to the deed of trust note permitted
the owner to prepay the loan in full, without HUD approval, after
twenty years. Developers could not prepay their loans prior to twenty
years, except under certain conditions, including HUD approval. The
prepayment rules in the riders reflected contemporaneous HUD
regulations governing the Section 221(d)(3) and Section 236 programs.
By prepaying the outstanding loan, an owner could terminate HUD’s
affordability restrictions on the property. The owner then could convert
the property into a conventional rental property and charge market
rental rates, thereby obtaining a greater return on the investment.
Id. (internal quotations and citations omitted).
43
Id. at 1235.
44
Grow, supra note 9, at 2. Throughout the 1970s, projects under Sections
221(d)(3) and 236 suffered large operating losses, as rents remained static
despite increasing maintenance costs. Id. In response, Congress supplied
additional subsidies in the form of Section 8 Loan Management Set-Aside
contracts, commonly known as Section 8 Housing Assistance Payment (HAP)
contracts. See Henry A. Herrman, Comment, Privity: How HUD Avoided
Contract Liability under ELIHPA and LIHPRHA, 30 SW. U. L. REV. 323, 329
(2001). Section 8 subsidies originally provided direct payments to owners of
FREEDMAN MACROED CORRECTED 5-31-05.DOC
752
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
under these programs quickly fell into disrepair due to the projects’
failure to generate sufficient “revenue to keep pace with rising
operating, management, and maintenance costs.”45 Attempting to
address the deficiencies of its project-based tenant subsidy
programs, Congress subsequently developed a tenant-based
Section 8 voucher program that largely replaced project-based
subsidy programs.46 As a result, appropriations under Sections
221(d)(3) and 236 stopped.47 However, the approximately 800,000
units built under these programs continued in operation pursuant to
the twenty-year prepayment restriction.48
In the mid-1980s, more than 800,000 units built under Sections
221(d)(3) and 236 were still held under federal rent prohibitions
and remained an important part of the nation’s affordable housing
stock. Congress became concerned that a significant portion of this
pool could be lost through conversion to market-rate units because
the twenty-year restriction on prepayment for most of the
properties would soon expire.49 Upon prepayment, “the units
would no longer be restricted to low-income occupancy.”50
Because in most cases local rents for comparable properties
exceeded the rents earned by assisted housing units, the owners of
these units could increase their profits by converting properties to
market-rate rents or condominium status.51 Conversions of both
types threatened the same result—the removal from the market of
affordable housing for low-income tenants.
section 236 projects to supplement the rent payable by low-income tenants and
to offset operating costs for the projects. Clark, supra note 3, at 187-88.
45
Herrman, supra note 44, at 329.
46
Id. at 187-88. For a discussion of section 8 subsidies, see DAYE, supra
note 3, at 210-20; see also Brian Maney & Sheila Cowley, Scarcity and Success:
Perspectives on Assisted Housing, 9 J. AFFORDABLE HOUS. & COMM. DEV. L.
319 (2000).
47
Clark, supra note 3, at 187.
48
Id.
49
C. Theodore Koebel & Cara L. Bailey, State Policies and Program to
Preserve Federally Assisted Low-Income Housing, 3 HOUSING POL’Y DEBATE,
995, 995-96 (1990) (citing John Moore, Expiring Uses, NAT’L J., Aug. 2, 1986,
at 1184-88).
50
Id.
51
Id. See also Grow, supra note 9, at 4.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
753
The prospect that nearly one million units of low-income
housing built during the 1960s would soon become eligible to be
released from rent restrictions prompted Congress to enact the
Emergency Low-Income Housing Preservation Act of 1987
(Emergency Act).52 The Emergency Act prohibited owners of
projects eligible for prepayment from converting these properties
to market-rate units unless they first complied with cumbersome
provisions.53 The Emergency Act effectively placed a moratorium
on prepayment by owners of their Section 221(d)(3) and 236
housing development loans at the end of the original twenty-year
period.54 Owners, outraged at the sudden abrogation of their
contractual rights, filed lawsuits challenging the constitutionality
of the Emergency Act.55
52
Emergency Low-Income Housing Preservation Act of 1987, 12 U.S.C. §
1715l (1989 & Supp. 1995) [hereinafter ELIHPA]. See also Johnson v. United
States Dep’t of Housing & Urban Dev., 911 F.2d 1302, 1305 (8th Cir. 1990)
(citing H.R.CONF.REP. No. 426, 100th Cong., reprinted in 1987 U.S. Code
Cong. & Admin. News 3458, 3489-3790).
The Preservation Act was prompted by the prospect that nearly one
million units of low income housing built in the 1960s would soon be
eligible to be released from rent restrictions because of the expiration of
the twenty-year moratorium on unilateral mortgage prepayment and the
expiration of Section 8 rental assistance contracts. To prevent a national
crisis in the availability of low income housing, Congress passed a
temporary measure making it more difficult for owners of § 221(d)(3)
housing to release themselves from the terms of the regulatory
agreement.
Id. (internal citations omitted).
53
Koebel & Bailey, supra note 49, at 996.
54
Winkelman, supra note 38, 1160. See also Clark, supra note 3, at 189
(explaining that Congress passed ELIHPA as emergency legislation, intending
government and private industry to work together in developing permanent
solutions to the impending disaster of losing nearly 800,000 affordable housing
units).
55
ELIHPA was controversial because many owners believed that the terms
of the Act breached the federal government’s agreement to free the owners from
rent and land use restrictions. Howard Cohen & Taylor Mattis, Prepayment
Rights: Abrogation by the Low Income Housing Preservation and Resident
Homeownership Act of 1990, 28 REAL PROP. & TR. J. 1 (1993). Under
federal
project-based subsidy programs of the National Housing Act, the private
FREEDMAN MACROED CORRECTED 5-31-05.DOC
754
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
While these suits were pending, Congress repealed the
Emergency Act and enacted the National Affordable Housing Act,
including the Low-Income Housing Preservation and Resident
Homeownership Act of 1990 (LIHPRHA or Preservation Act).56
LIHPRHA established a comprehensive program to recapitalize
privately-owned assisted housing and to commit these properties
for use as low-income housing for their “remaining useful life” of
fifty years.57 To effectuate this goal, LIHPRHA encouraged
owners to refinance their units under the program58 or to sell their
developer secured a loan from a private lender, evidenced by a deed of trust note
payable to a private lending institution (the “Note”). Cienega Gardens v. United
States, 33 Fed. Cl. 196, 196-203 (1995). HUD endorsed the Note as part of its
process of insuring the mortgage. Id. The Note contained a rider, which
expressly permitted the owner to prepay the loan in full, without HUD approval,
after twenty years. Cienega Gardens, 194 F.3d at 1235. The rider was the only
document that expressly mentioned the borrower’s right to prepay. Id. Pursuant
to the rider, developers were not permitted to prepay their HUD-insured loans
prior to twenty years except under certain conditions. Id. Owners argued that
Congress had interfered with their contractual right of prepayment. See
Herrman, supra note 44, at 330 n.66 (citing Orrego v. 833 W. Buena Joint
Ventura, 943 F.2d 730 (7th Cir. 1991) (holding ELIHPA not retroactive without
reaching the constitutional question); Thetford Properties v. U.S. Dep’t of Hous.
& Urban Dev., 907 F.2d 445, 450 (4th Cir. 1990) (dismissing appellant’s
constitutional claim for failure to exhaust administrative remedies); Johnson v.
U.S. Dept. of Hous. & Urban Dev., 911 F.2d 1302 (8th Cir. 1990) (dismissing
plaintiffs’ claim for failure to exhaust administrative remedies); Orrego v. U.S.
Dep’t of Hous. & Urban Dev., 701 F. Supp. 1384 (N.D. Ill. 1988)).
56
LIHPRHA, 12 U.S.C. §§ 4101-4147. LIHPRHA was part of a more
comprehensive housing law called the Cranston-Gonzalez National Affordable
Housing Act, 42 U.S.C. §§ 12701-12898 (1996). See generally Winkelman,
supra note 38, at 1161.
57
LIHPRHA, 12 U.S.C. § 4112(c). “[T]he term ‘remaining useful life’
means. . . the period during which the physical characteristics of the housing
remain in a condition suitable for occupancy, assuming normal maintenance and
repairs are made and major systems and capital components are replaced as
becomes necessary.” Id. Fifty years after the commencement of the POA, the
owner may petition HUD for a hearing to determine whether the useful life of
the project has expired. Id.
58
Koebel & Bailey, supra note 49, at 997 (“LIHPRHA preservation
incentives include insured or direct capital improvement financing, an equity
takeout loan, an 8 percent return on preservation equity, access to reserves,
increased Section 8 and non-Section 8 rents, and insured acquisition loans and
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
755
properties to “qualified purchasers.”59 With the hope of attracting
willing owners, LIHPRHA offered a range of incentives, including
increased rent ceilings, increased allowable rates of return on
investments, and equity loan funds for capital improvements.60
LIHPRHA specifically addressed the contested contractual
issues raised by the Emergency Act by permitting owners to
prepay their mortgage loans; however, the Act discouraged the
widespread exercise of this option through the creation of a
burdensome approval process.61 Under the Act’s prepayment
procedures, an owner first had to file with HUD, tenants, and
others a “notice of intent” to prepay.62 The owner then was
required to submit a “plan of action” (POA) setting forth
information relating to the proposed prepayment.63 HUD would
then evaluate the owner’s POA for approval.64 Under the
Preservation Act, HUD could issue approval for prepayment only
after making certain written findings that the prepayment would
not adversely affect the low-income housing supply or
involuntarily displace current tenants.65 In the event that HUD
could not make the necessary findings, LIHPRHA required that the
agency disapprove the owner’s POA and deny the owner’s request
for prepayment approval.66
grants to qualified purchasers.”).
59
Id.
“Qualified purchaser” is a technical term in the act identifying either a
priority purchaser as previously defined or another purchaser
committing to continuation of the units as assisted housing. Priority
purchasers include a local or state government agency, a nonprofit
housing corporation, or a tenant organization. For-profit buyers are
eligible as qualified purchasers if they are willing to commit to the use
restriction.
Id.
60
Cienega Gardens, 194 F.3d at 1235.
61
See LIHPRHA, 12 U.S.C. §§ 4101-4147.
62
Id. § 4102(a).
63
Id. § 4107.
64
Id. § 4101(a).
65
Id. §§ 4101(a), 4108. See also Forest Park, 203 F. Supp. 2d at 1074.
66
Id. § 4108(c). Under LIHPRHA’s stringent procedures, the prepayment
option was effectively limited to situations in which (1) there was no bona fide
FREEDMAN MACROED CORRECTED 5-31-05.DOC
756
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
Under the market conditions of the mid-1990s, few owners
could reasonably satisfy the Preservation Act’s requirements for
prepayment.67 Further, the properties of many owners were
insufficiently valuable to qualify for additional financial incentives
under the Act. Therefore, most “eligible” owners did not to
participate in LIHPRHA at all.68 Of the owners that did participate,
virtually all filed POAs seeking preservation incentives to remain
in the affordable housing program.69
While Congress intended to burden owners with additional
federal restrictions through the Preservation Act, it also sought to
protect them from additional burdens imposed by state or local
governments. Section 4122 of LIHPRHA expressly preempted
state and local governments from establishing or enforcing laws or
regulations that would “restrict or prohibit” the prepayment of
loans on LIHPRHA-eligible housing projects.70 By its terms,
purchaser for the owner’s property, (2) there was insufficient public funding to
implement preservation, or (3) the owner’s POA was supported by sufficient
negative impact findings by HUD. Winkelman, supra note 38, at 1161-62.
An owner who wishes to prepay must file a notice of intent to do so,
simultaneously with HUD as well as with the appropriate state or local
government officer, the holder of the existing mortgage and the tenants.
HUD’s permission to prepay is only granted if the project fits into one
of two categories, namely: The project must be located in an area where
there is no need for low-income housing and where there will be no
substantial economic effect on the tenants; or the project must have a
value so high that the amount of federal incentives which must be
offered (“Federal Cost Limits”) cannot support the appraised value of
the property. It is generally felt that it would require an extremely
unique set of facts and circumstances to lead to HUD’s granting
permission for prepayment; that is, all areas arguably need low-income
housing and very few projects have a value greater than the Federal
Cost Limits. Therefore, the option of prepayment is probably a fiction.
Id.
67
Winkelman, supra note 38 (citing Thetford Properties, 907 F.2d at 450)
(“[T]he option of prepayment is probably a fiction . . . .”).
68
Id. See also Grow, supra note 9, at 3.
69
LIHPRHA, 12 U.S.C. § 4109.
70
Id. § 4122. In relevant part, the statute reads:
(a) In general. No State or political subdivision of a State may
establish, continue in effect, or enforce any law or regulation that . . .
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
757
Section 4122 did not apply to local laws of general applicability.71
The Preservation Act, along with its express preemption provision,
ensured affordable housing for the most at-risk low-income
Americans, while providing equitable treatment of federally
funded incentives among the states. 72
In the late 1990s, an emerging interest in balancing the national
budget, coupled with HUD’s promotion of its “Reinvention
Blueprint”—a radical proposal to substitute vouchers for all
project-based assistance—led to Congress’s defunding of the
Preservation Act and the creation of a series of new programs to
address the low-income housing shortage.73 Congress mandated
(1) restricts or inhibits the prepayment of any mortgage described in
section 4119(1) of this title . . . on eligible low income housing; (2)
restricts or inhibits an owner of such housing from receiving the
authorized annual return provided under section 4101 of this title; (3) is
inconsistent with any provision of this subchapter, including any law,
regulation, or other restriction that limits or impairs the ability of any
owner of eligible low income housing to receive incentives authorized
under this subchapter . . . or (4) in its applicability to low income
housing is limited only to eligible low income housing for which the
owner has prepaid the mortgage or terminated the insurance
contract . . .
(b) Effect. This section shall not prevent the establishment, continuing
in effect, or enforcement of any law or regulation of any State or
political subdivision of a State not inconsistent with the provisions of
this subchapter, such as any law or regulation relating to building
standards, zoning limitations, health, safety, or habitability standards
for housing, rent control, or conversion or rental housing to
condominium or cooperative ownership, to the extent such law or
regulation is of general applicability to both housing receiving Federal
assistance and nonassisted housing . . . .
Id. (emphasis added).
71
Id.
72
H.R.CONF.REP. No. 101-943, at 458, 460 (1990), reprinted in 1990
U.S.C.C.A.N. 6070, 6165 [hereinafter House LIHPRHA Report] (“Preemption:
The [LIHPRHA] solution would recognize that a fair Federal preservation
policy must apply uniformly to all affected properties regardless of location. For
that reason, the solution would preempt State and Local laws that target only
prepayment projects for special treatment.”).
73
Forest Park II, 336 F.3d at 729. See Grow, supra note 9, at 3. While not
endorsing HUD’s proposal, in 1996, Congress reduced funding for the
FREEDMAN MACROED CORRECTED 5-31-05.DOC
758
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
that, effective October 1, 1996, HUD suspend processing of any
unapproved POAs under LIHPRHA.74 As a practical matter,
LIHPRHA’s restrictive prepayment requirements have not been
LIHPRHA preservation program. Id. See Pub. L. No. 104-204 (FY 97
Appropriations Act); see also H.R.Cong.Rep. No. 812, 104th Cong. (1996).
The preservation program has been redesigned to reduce excessive
costs. . . . To assist the Congress in making a determination of whether
this program is the most cost-effective way to provide affordable
housing opportunities to low-income families, the conferees request the
General Accounting Office (GAO) to evaluate and review the program.
Id., available at http://thomas.loc.gov.
74
HOPE authorized the HUD secretary to limit LIHPRHA funding to
certain developments meeting specified criteria and permitted prepayment so
long as the owners held off on rent increases for sixty days. HOPE § 2(b). The
Quality Housing and Work Responsibility Act § 219, permanently defunded
LIHPRHA and authorized prepayments notwithstanding its terms. Pub. L. No.
105-276, 112 Stat. 2461 (1998).
Sec. 219. (a) Prepayment Right.—Notwithstanding prior acts:
(1) the owner of the project may prepay, and the mortgagee may accept
prepayment of, the mortgage on the project, and
(2) the owner may request voluntary termination of a mortgage
insurance contract with respect to such project and the contract may be
terminated notwithstanding any requirements under sections 229 and
250 of the National Housing Act.
(b) Conditions.—Any prepayment of a mortgage or termination of an
insurance contract authorized under subsection (a) may be made—
(1) only to the extent that such prepayment or termination is consistent
with the terms and conditions of the mortgage on or mortgage
insurance contract for the project;
(2) only if the owner of the project involved agrees not to increase the
rent charges for any dwelling unit in the project during the 60-day
period beginning upon such prepayment or termination; and
(3) only if the owner of the project provides notice of intent to prepay
or terminate, in such form as the Secretary of Housing and Urban
Development may prescribe, to each tenant of the housing, the
Secretary, and the chief executive officer of the appropriate State or
local government for the jurisdiction within which the housing is
located, not less than 150 days, but not more than 270 days, before such
prepayment or termination.
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
759
applied to any HUD-subsidized mortgage prepayments since 1996,
although Congress has never explicitly repealed LIHPRHA.
As part of its new housing policy, Congress passed the
Housing Opportunity Program Extension Act of 1996 (HOPE),75
an appropriations bill that limited LIHPRHA funding and shifted
resources to tenant-based subsidy programs, such as Section 8
housing vouchers. Further, Section 219 of HOPE permitted
prepayment of mortgages for subsidized projects without HUD
approval, provided that owners delayed rent increases for at least
sixty days.76 HOPE essentially reinstated most of the owners’
original rights to prepay Section 221(d)(3) and 236 mortgages,
including the right to prepay their mortgages without HUD
approval after twenty years.77
Following the passage of HOPE, a series of congressional
enactments, as part of the Quality Housing and Work
Responsibility Act, permanently defunded the Preservation Act’s
incentive programs and authorized the prepayment of federally
subsidized mortgages.78 Neither Section 219 nor any other
provision in HOPE or the subsequent Acts contained a preemption
clause such as that in LIHPRHA or references to LIHPRHA’s
express preemption provision.79 Consequently, the interplay
between LIHPRHA’s preemption provision and subsequent federal
housing acts has been the subject of various preservation battles
between low-income tenants and profit-motivated owners. Only
two circuit courts have addressed the relationship between
LIHPRHA and HOPE—the Eighth Circuit in Forest Park II v.
Hadley and the Ninth Circuit in Topa Equities v. City of Los
Angeles. Both courts have found that the LIHPRHA preemption
provision is still applicable law, regardless of the provision under
which an owner purports to be opting out.
75
HOPE § 2(b).
Id. (“[O]nly if the owner of the project involved agrees not to increase
the rent charges for any dwelling unit in the project during the 60-day period
beginning upon such prepayment or termination. . . .”).
77
Quality Housing and Work Responsibility Act § 219; see also Cienega
Gardens, 38 Fed. Cl. at 70.
78
Quality Housing and Work Responsibility Act § 219.
79
Topa Equities, 342 F.3d at 1069; Forest Park, 336 F.3d at 732.
76
FREEDMAN MACROED CORRECTED 5-31-05.DOC
760
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
Since the 1950s, HUD has subsidized approximately 1.7
million rental units in more than 23,000 privately-owned
properties.80 More than 80,000 low-income apartment units were
preserved through LIHPRHA and the Emergency Act.81 However,
between 1995 and 2003, following the enactment of HOPE and the
return of the twenty-year prepayment option, more than 300,000
units have been removed from the affordable housing stock.82 In
2004, 1.4 million affordable units remained; more than forty
percent of the tenants are elderly.83 Ever-increasing rents in most
urban centers and the trend of owners opting-out of federal housing
programs for larger profits have forced greater involvement by
state and local governments in the provision of decent, affordable
homes to American families.84
80
Hearing on H.R. 4679 Before the House Subcommittee on Housing and
Community Opportunity, House Financial Services Comm., 108th Cong. (2004)
(Testimony of Michael Bodaken, President, National Housing) (July 20, 2004)
[hereinafter
National
Housing
Trust
Testimony],
available
at
http://www.nhtinc.org/documents/072004_NHT_Testimony.pdf.
81
SO. CAL. ASS’N OF GOV’T, THE AFFORDABLE HOUSING PRESERVATION
CRISIS: A RISK ASSESSMENT OF THE POTENTIAL HUD-ASSISTED MULTIFAMILY
HOUSING IN THE SOUTHERN CALIFORNIA REGION (Dec. 2000) [hereinafter RISK
ASSESSMENT STUDY], available at http://api.ucla.edu/workshop/Housing
Preservation/LinksResources/HousingPreservationReportFinal.htm.
82
NAT’L HOUS. TRUST, CHANGES TO PROJECT-BASED MULTIFAMILY UNITS
IN HUD’S INVENTORY BETWEEN 1995 AND 2003 6 (2004), available at
http://www.nhtinc.org/documents/PB_Inventory.pdf.
Analysis of HUD data finds that between 1995 and 2003, the number of
such units dropped from 1.7 million to 1.4 million. The loss of 300,000
affordable rental units is substantially larger than previous estimates.
This loss is especially concerning in light of the shortage of other
available affordable housing for extremely low-income households
across the country.
Id.
83
National Housing Trust Testimony, supra note 80.
84
On average, annual rent increases among American’s major metropolitan
centers were three percent per year from 1988-1997. JACK GOODMAN, NAT’L
MULTI HOUS. COUNCIL, PERFORMANCE ACROSS LOCAL APARTMENT MARKETS
(1999), available at http://www.nmhc.org. See also KALIMA ROSE ET AL., PRATT
INST. CTR. FOR CMTY. & ENV. DEVELOPMENT, INCREASING HOUSING
OPPORTUNITY IN NEW YORK CITY: THE CASE FOR INCLUSIONARY ZONING
(2004) (explaining that the average income for New York renter households
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
AFFORDABLE HOUSING PRESERVATION
761
II. STATE AND LOCAL GOVERNMENT PRESERVATION POLICIES
Prior to the enactment of the Emergency Act and LIHPRHA,
state and local governments provided limited protections for
federal affordable housing projects. In the wake of LIHPRHA’s
defunding, federal law no longer provides preservation guarantees;
consequently, states and localities have assumed an even larger
role in preserving affordable housing projects.85 Low-income
housing advocates push for increased local government
involvement in preservation because preservation, as compared to
programs hinging on tenant vouchers, prevents the displacement of
residents from their homes and communities, maintains affordable
units for future tenants in need, and provides superior housing
security for tenants and communities.86 In light of recent economic
downturns, preservation also has proved more cost effective for
local governments than the development of new affordable
projects.87 Aggressive state and local preservation policies have
achieved marked success in limiting the conversion of low-income
housing to market-rate units.88 These protections come in a variety
of forms, including procedural requirements for opting out,
limitations on property owners’ returns, and the transfer of
properties to preservation owners.89
grew just three percent from 1975 to 1999, but the average rent went up thirtythree
percent),
available
at
http://www.policylink.org/
Research/NYIZ/default.html.
85
Brian Galle, National Housing Law Project, Preserving Federally
Assisted Housing at the State and Local Level: A Legislative Tool Kit, 29 HOUS.
L. BULL. 183 (1999) (survey of state and local preservation initiatives), available
at http://www.nhlp.org/html/hlb/1099/1099toolkit.htm.
86
Grow, supra note 9, at 5. See, e.g., S. REP. NO. 106-161(1999)
(expressing that tenant-based Section 8 vouchers “do not always provide real
rental choice for assisted families” noting that “in a number of cases, families
with vouchers are unable to use their vouchers to obtain affordable housing”),
available at http://thomas.loc.gov.
87
Grow, supra note 9, at 5.
88
See generally Risk Assessment Study, supra note 81.
89
Galle, supra note 85. Numerous state and localities have also developed
alternative programs aimed primarily at preserving projects through preservation
incentives to current owners. Id. In exchange for new or extended agreements,
FREEDMAN MACROED CORRECTED 5-31-05.DOC
762
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
A. Procedural Requirements for Opting Out
Procedural requirements provide information to tenants or local
governments in order to prepare them for the impending removal
of affordability restrictions on subsidized housing projects.90
Current federal law guarantees no less than six months’ notice to
HUD and tenants of an impending market-rate conversion.91 Many
states have statutes requiring that notice be provided to tenants or
local governments in advance of the current 150-day federal notice
period.92
some of these programs provide: refinancing or cash-out current equity; equity
takeout loans for other purposes; partial access to residual receipts or excess
income accounts; and increased dividends. See also National Council of State
Housing Agencies National Preservation Survey (1998), available at
www.nhlp.org/html/pres/state/index.htm (citing programs in California,
Colorado, Maine, Maryland, Massachusetts, Michigan, Minnesota,
Pennsylvania, and Wisconsin). These state and local incentive-based
preservation programs are beyond the scope of this note and are not likely in
conflict with LIHPRHA preemption. Id.
90
Galle, supra note 85, at 4. Notice requirements may additionally provide
sufficient time for arranging a transfer of the project to a non-profit owner or
deter some owners from prepaying. Id.
91
Quality Housing and Work Responsibility Act § 219 (requiring any
owner who anticipates a termination of the Section 221(d)(3) or Section 236
mortgage to provide no less than 150 days and no more than 270 days notice to
tenants and to HUD); see also Galle, supra note 85.
92
See CAL. GOV’T. CODE § 65863.10, et seq. (2001) (requiring, upon any
action that would terminate subsidy for all HUD-subsidized housing, one year’s
notice to tenants, state and local housing authorities, and local governments
prior to termination or prepayment); 1988 Conn. Pub. Acts 88-262 (requiring
one year notice); 310 ILL. COMP. STAT. § 60/3 (2004) (requiring, upon the
intended sale or disposition of property for all HUD-subsidized housing, six
months’ notice to tenants and to the state housing authority); ME. REV. STAT.
ANN. tit. 30-A, § 4973 (1993) (requiring ninety days notice to state and local
housing authority triggered upon any action that would terminate subsidy for all
HUD-subsidized housing); MD. CODE ANN., art. 83B, § 9-101, et seq. (1989)
(requiring, upon intended sale or disposition of property for all HUD-subsidized
housing, no less than one year or more than two years notice to locality, tenant
association, state – notice is triggered by any action that may terminate subsidy);
MINN. STAT. § 566.17 (1998) (requiring one year notice); R.I. GEN. LAWS § 3445-4 et seq. (1988) (providing two years notice to tenant association, state, city;
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
763
For example, a Minnesota statute requires that a landlord who
seeks to opt out of federally subsidized rental housing must
provide tenants with one year’s written notice.93 A second
Minnesota statute requires that, at least twelve months prior to
termination, owners seeking to opt out or prepay their mortgage
loans must submit to the state housing agency, local government,
and affected residents a “tenant impact statement” outlining the
potential impact of the termination on residents.94 These statutes
were the subjects of a challenge by owners in Forest Park II v.
Hadley, a case before the Eighth Circuit.95
Under a similar California law, an owner of an assisted housing
development who seeks to terminate a project-based contract must
and tenant access to information triggered upon sale, conversion, prepayment for
all HUD-subsidized housing); TEX. GOV’T. CODE ANN. § 2306.185(f), et seq.
(2005) (requiring one year’s notice to state housing authority triggered upon any
action that would terminate subsidy for all HUD-subsidized housing); WASH.
REV. CODE § 59.28 (2005) (requiring one year notice); DENVER, CO., MUN.
CODE § 12-106, et seq. (2000) (requiring, triggered upon opt out or sale for all
HUD-subsidized housing, one year’s notice to city, tenants for Section 8
contract expirations; 210 days for long-term contract expirations; and 150 days
for one year extensions); Portland, Or., City Code § 30.01.030, et seq. (2005)
(requiring one year notice to city, tenants for § 8 contract expirations; 210 days
for long-term contract expirations; and 150 days for 1 year extensions triggered
upon opt out or sale for all HUD-subsidized housing); SAN FRANCISCO, CA.,
ADMIN. CODE § 60.4, et seq. (1990) (requiring, upon the intended sale or
disposition of property Section 8 contracts, eighteen months’ notice to city and
tenants for prepayment; twelve months’ notice for Section 8 contract
expirations).
93
MINN. STAT. ANN. § 504B.255 (West 2002).
The landlord of federally subsidized rental housing must give
residential tenants of federally subsidized rental housing a one-year
written notice under the following conditions: (1) a federal section 8
contract will expire; (2) the landlord will exercise the option to
terminate or not renew a federal section 8 contract and mortgage; (3)
the landlord will prepay a mortgage and the prepayment will result in
the termination of any federal use restrictions that apply to the housing;
or (4) the landlord will terminate a housing subsidy program.
Id.
94
MINN. STAT. ANN. § 471.9997 (West 2001). See Forest Park, 336 F.3d at
730.
95
Forest Park II v. Hadley, 336 F.3d 724 (8th Cir. 2003).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
764
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
provide at least nine months’ notice of the proposed change to each
affected tenant household in the assisted housing development.96 In
addition, California law provides that an owner’s notice to tenants
shall simultaneously be filed with a number of public entities,
including the Board of Supervisors of the county, and the
96
Cal. Gov’t. Code § 65863.10(b) mandates that the notices contain
specific information for the purpose of explaining to the tenants the process and
ramifications of the owners’ decision to opt out. Cal. Gov’t. Code §
65863.10(b). Cal Gov’t Code § 65863.10 provides:
(b) At least nine months prior to the anticipated date of termination of a
subsidy contract or prepayment on an assisted housing development,
the owner proposing the termination or prepayment of governmental
assistance shall provide a notice of the proposed change to each
affected tenant household residing in the assisted housing development
at the time the notice is provided.
Id. § 65863.10. The amended Cal. Gov’t Code § 65863.10(b)(1) requires twelve
months notice. Cal. Gov’t. Code § 65863.10(b)(1)-(6) provides:
(1) The anticipated date of the termination or prepayment of the federal
program, and the identity of the federal program . . . .
(2) The current rent and anticipated new rent for the unit on the date of
the prepayment or termination of the federal program . . . .
(3) A statement that a copy of the notice will be sent to the city or
county, or city and county, where the assisted development is located,
to the appropriate local public housing authority, if any, and to the
Department of Housing and Community Redevelopment.
(4) A statement of the possibility that the housing may remain in the
federal program after the proposed date of subsidy termination or
prepayment if the owner elects to do so.
(5) A statement of the owners’ intention to participate in any current
replacement federal subsidy program made available to affected
tenants.
(6) The name and telephone number of the city, county, or city and
county, the appropriate local public housing authority, if any, the
Department of Housing and Community Development, and a legal
services organization, that can be contacted to request additional
written information about an owner’s responsibilities and the rights and
options of an affected tenant.
Id. § 65863.10(b)(1)-(6).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
765
Department of Housing and Community Development.97
B. Limiting Property Owners’ Returns
The primary purpose of market-rate conversion is to increase
an owner’s rents and profits. With this in mind, states and localities
have also attempted to prevent owners from opting out through
regulations that increase the costs of conversion.98 Governments
have achieved this through either direct regulation of rent levels
(i.e., rent control) or other requirements, such as so-called
“statutory leases” or prepayment fees.99
The Los Angeles Rent Stabilization Ordinance (LARSO),100
for example, prohibits owners of buildings previously subsidized
by HUD from moving their rents to market level.101 LARSO
97
CAL. GOV’T CODE § 65063.10(c)(1). These entities shall send additional
notices containing supplemental information regarding the number of tenants
affected, the number of units that are government assisted and the types of
assistance they receive, the number of the units that are not government assisted,
the number of bedrooms in each unit that is government assisted, and the ages
and income of the affected tenants. Id. § 65863.10(c)(2).
98
Galle, supra note 85, at 2.
99
Id. at 3 (“Typically, [rent control] in the area involves setting the ‘base
rent’ for any property exiting the federal subsidy or assistance program at the
last rent level in effect under that program and then subsequently applying the
generally applicable rent regulations on general and individual rent
adjustments.”). See, e.g., SAN FRANCISCO, CA. CODE § 37 (1998); LOS
ANGELES, CA MUNICIPAL CODE § 151.02 (1995). Massachusetts has a similar
statute that applies to all projects that terminated their government
“involvement” in either insurance, interest subsidies, or rental assistance. MASS.
GEN. LAWS ch. 40, § 14 (2001).
States and localities have enacted legislation imposing tenant relocation
costs on owners who convert their affordable housing project to market-rate. See
R.I. GEN. LAWS § 34-45-11 (imposing moving costs); MD. ANN. CODE ART.
83B § 9-105(a) (imposing moving costs up to $975); WASH. REV. CODE § 59.28
(imposing relocation payment level of $2,000, half payable by the owner and
half payable by the city). Seattle enacted a local relocation ordinance that
applies to any displacement caused by demolition, change of use, substantial
rehabilitation, or removal of use restrictions on federally assisted housing
developments. SEATTLE, WA MUNICIPAL CODE §22.210 (enacted 1990).
100
LOS ANGELES MUNICIPAL CODE §§ 151.01-04. (1995).
101
Id. See also Topa Equities, 342 F.3d at 1067. This includes buildings
FREEDMAN MACROED CORRECTED 5-31-05.DOC
766
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
requires that, instead of entering the open market, properties
exiting federal subsidy or assistance programs must enter the rent
stabilization program at the rent previously charged when the
buildings were under federal rent restrictions.102 The ordinance
also provides that rents may not be adjusted “if a rental unit is
vacated as a result of the termination of the regulation of the rental
unit under any local, state, or federal program,” requiring the unit
to remain available to low-income tenants even though vacancy
decontrol would normally free the unit.103 This regulation was the
subject of a challenge by owners in Topa Equities v. City of Los
Angeles.104
Another alternative involves so-called “statutory leases,”
through which tenants in converted buildings receive mandatory
temporary lease renewals under terms “specified by law at rent
levels roughly equal to those in effect under the federal program”
prior to conversion.105 Both Rhode Island and Maryland have
statutes that employ this concept.106
C. Transfer of Properties to Preservation Owners
Recently, several states introduced and passed laws designed to
permanently preserve at-risk properties by transferring ownership
of the buildings to non-speculative or not-for-profit owners.107
that participated under Section 221(d)(3) and 236 programs. Id.
102
Originally LARSO was Los Angeles’s general rent control regulation
and applied only to standard, “free-market” rental units. Topa Equities, 342 F.3d
at 1067. In 1990, Los Angeles amended LARSO to apply to properties that
participate in federal programs under the NHA. Id. LARSO also contains a
vacancy decontrol provision that permits owners to raise the rent for a unit to
market level after the tenancy for that unit has terminated. Id.
103
Id.
104
342 F.3d 1065, 1067 (9th Cir. 2003).
105
Galle, supra note 85, at 3.
106
R.I. GEN. LAWS § 34-45-11; MD. ANN. CODE ART 83B §§ 9-101–9-114;
ME. REV. STAT. ANN. tit.30-A § 4976.
107
CAL. GOV’T CODE § 65863.10, et seq. (2001) (providing an offer to
purchase to tenant associations, non-profit organizations, some for-profit
organizations, and public agencies upon any action that would terminate the
subsidy regarding all HUD-subsidized housing); 20 ILL. COMP. STAT. § 3805/8.1
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
767
Many affordable housing advocates argue that such transfers are
the best way to preserve at-risk housing because they are more
likely to keep tenants in their homes and preserve properties as
future housing sources.108 Generally, these laws provide non(2004) (providing the tenant association with a true right of first refusal for all
HUD-subsidized housing that is triggered upon the intended sale or disposition
of property); ME. REV. STAT. ANN. tit. 30-A, § 4973 (1993) (providing the state
housing authority with a true right of first refusal that is triggered upon any
action that would terminate the subsidy for all HUD-subsidized housing); MD.
CODE ANN., art. 83B, § 9-101, et seq. (1989) (providing the local housing
authority, groups representing tenants, and non-profit organizations with a true
right of first refusal that is triggered upon the intended sale or disposition of
property for all HUD-subsidized housing); R.I. GEN. LAWS § 34-45-4 et seq.
(1988) (providing the tenant association, state housing authority, local housing
authority, and locality with a true right of first refusal that is triggered for all
HUD-subsidized housing upon sale, conversion, prepayment); TEX. GOV’T.
CODE ANN. § 2306.185(f), et seq. (2005) (providing time for the state to locate a
potential buyer, triggered upon any action that would terminate the subsidy for
all HUD-subsidized housing); DENVER, CO., MUN. CODE § 12-106, et seq.
(2000) (providing time for the city to negotiate for purchase, triggered upon opt
out or sale for all HUD-subsidized housing); PORTLAND, OR., CITY CODE §
30.01.030, et seq. (2005) (providing time for the city to negotiate for purchase,
triggered upon opt-out or sale for all HUD-subsidized housing); SAN
FRANCISCO, CA., ADMIN. CODE § 60.4, et seq. (1990) (providing the city, tenant
association, and non-profit organizations with a true right of first refusal
triggered upon the intended sale or disposition of Section 8 properties).
108
Galle, supra note 85, at 6.
If cost were no object, permanent preservation of at-risk properties via
transfer to non-speculative ownership would probably be the best way
to preserve at-risk housing. Moving projects into the hands of entities
whose purpose is providing housing rather than generating profit—such
as tenant-endorsed or controlled non-profits— is more likely to keep
tenants in their homes and preserve the property as a future housing
resource. Obtaining both the necessary funds for transfer and site
control from private owners remain difficult challenges in the everchanging policy and budget picture. In addition, local government
activities that enable nonprofit purchasers to be competitive with other
options available to owners interested in converting to market-rate use
(e.g., identifying potential conversion candidates, contacting owners to
explore transfer options, providing predevelopment support for
purchasers) will continue to be especially important in preserving units.
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
768
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
speculative or not-for-profit entities either a true “right of first
refusal,” which permits a designated purchaser to match another
sale offer and thereby acquire title, or a “right to make an offer,”
with or without an obligation on the owner’s part to sell.109
A California law requires that owners of affordable housing
projects must, at least twelve months prior to prepayment, offer the
property for sale to everyone on a state-maintained list of
prospective purchasers who have indicated their willingness to
assure the long-term affordability of the housing.110 For 180 days,
these prospective purchasers have an exclusive right to make an
offer. For an additional 180 days thereafter, all prospective buyers
who made offers but were rejected still maintain a right of first
refusal. This regulation was the subject of a challenge by owners in
the Eastern District of California in Kenneth Arms v. Martinez,111
which upheld the law in the face of a preemption challenge. The
federal circuit courts have yet to rule on the validity of these laws
109
See, e.g., National Housing Law Program, Right of First Refusal in
Preservation Properties: Worth a Second Look, 32 HOUS. L. BULL. 1 (Jan.
2002); National Housing Law Program, Illinois Establishes Tenant Purchase
Option for Properties Terminating Federal Programs, 34 HOUS. L. BULL. 150
(July 2004) (discussing the Illinois law that provides an opportunity to purchase
the property and preserve it as low-income housing before an owner converts it
to market rent). In combination with mechanisms to obtain site control, some
localities have adopted formula(s) specifying the sale price of preservation
properties. See also ME. REV. STAT. ANN. tit. 30-A, § 4973 (creating
“preemptive options,” triggered by the act of prepayment or opt-out, that
combine notice requirements with a right of first refusal for the state housing
agency whenever the owner takes an action that would terminate a project’s
subsidies); SAN FRANCISCO, CA ADMIN. CODE § 60.7(a) (setting a “Fair Return
Price” based upon certain appraisal assumptions that the owner must accept as a
sale price).
110
CAL. GOV’T CODE § 65863.11 (2001). Under Cal. Gov’t Code §
65863.11(b)-(c), an owner may not sell or otherwise dispose of his development
in a manner that would result in either discontinuance of the development’s
status as an assisted housing development or the termination of any low-income
use restrictions that apply to the development, unless the owner provides an
opportunity to purchase the developments to specified public and private
entities. Kenneth Arms, 2001 U.S. Dist. LEXIS 11470, at *33 (citing CAL.
GOV’T CODE § 65863.11(b)-(c) (amended 2001)).
111
Kenneth Arms, 2001 U.S. Dist. LEXIS 11470, at *36.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
769
in light of Section 4122 of LIHPRHA. Discussions regarding the
effect of federal preemption on these regulations frequently arise,
however, in state and city legislatures at hearings and debates
about preservation policies.112
III. CONSTITUTIONAL STANDARDS FOR PREEMPTION
Preemption is the power of the federal government to supplant
state law with respect to matters the federal government has the
power to regulate under the U.S. Constitution.113 The Supreme
Court has repeatedly held that congressional intent determines
when a congressional act preempts state or local law.114
Preemption of a state law by federal law may be either explicitly
stated in the language of a federal statute, such as the LIHPRHA
preemption provision, or read by the courts to be implicitly
contained in the statute’s structure or purpose; that is, preemption
may be express or implied.115 If a federal law expressly or
112
Grow, supra note 9, at 2; see also Galle, supra note 85, at 2; Chen,
supra note 10.
113
Preemption power is generally viewed as arising from the Supremacy
Clause of the United States Constitution.
This Constitution, and the Laws of the United States which shall be
made in Pursuance thereof; and all Treaties made, or which shall be
made, under the Authority of the United States, shall be the supreme
Law of the Land; and the Judges in every State shall be bound thereby,
any Thing in the Constitution or Laws of any State to the Contrary
notwithstanding.
U.S. CONST. art. VI, cl. 2. See Stabile, infra note 115, at 2, n.2; see, e.g.,
Cipollone v. Liggett Group, Inc., 505 U.S. 504, 516 (1992) (noting “that state
law conflicting with federal law is without effect”).
114
Gade v. National Solid Wastes Management Ass’n, 505 U.S. 88, 96
1992 (“The question whether a certain state action is pre-empted by federal law
is one of congressional intent. The purpose of Congress is the ultimate
touchstone. To discern Congress’s intent we examine the explicit statutory
language and the structure and purpose of the statute.”) (internal citations and
quotations omitted).
115
Susan J. Stabile, Preemption of State Law by Federal Law: A Task for
Congress of the Courts?, 40 VILL. L. REV. 1, 4-5 (1995); see also Jones v. Rath
Packing Co., 430 U.S. 519, 525 (1977); Fidelity Fed. Sav. & Loan Ass’n v. de la
Cuesta, 458 U.S. 141, 152-54 (1982) (explaining how federal regulatory or
FREEDMAN MACROED CORRECTED 5-31-05.DOC
770
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
impliedly preempts a state law, then “the state law may not be used
by a plaintiff to impose liability on a defendant.”116 There is a legal
presumption against the preemption of state or local housing laws;
however, Congress may preempt both explicitly.117
Express preemption “occurs when a statute contains an explicit
statement addressing the preemptive effect of the statute on state
law claims, rather than leaving it to the courts to decide, in any
given dispute, whether the federal statute preempts state law.”118
Courts are usually called upon to interpret the precise scope of an
express preemption provision, the application of which is clear, as
Congress added the provision to the statute.119 Forest Park, Topa
Equities, and their progeny are distinct in requiring the court to
apply LIHPRHA’s preemption provision to HOPE, a distinct
statutory program. The standard express preemption inquiry does
not require this additional step because a preemption provision is
generally part of the statute at issue.120
In the absence of an express preemption provision, preemption
may be implied.121 This implication may arise from a pervasive
scheme of federal regulation, in which case federal law is said to
“occupy the field.”122 Alternatively, implied preemption may arise
statutory law may expressly or impliedly preempt state law).
116
Stabile, supra note 115, at 3.
117
For discussion on the presumption against federal preemption of local
police powers, see infra note 133 through 136 and accompanying text.
118
Stabile, supra note 115, at 2.
119
Id.
120
See, e.g., Employee Retirement Income Security Act of 1974 § 1144(a),
29 U.S.C. §§ 1001-1461 (1988 & Supp. V 1993) [hereinafter ERISA]
(preempting all state laws that relate to employee benefit plans covered by
ERISA); Hazardous Material Transportation Act, 29 U.S.C. app. 1811(a)
(1988).
121
Forest Park II v. Hadley, 336 F.3d 724, 732 (8th Cir. 2003); Topa
Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065, 1069 (9th Cir. 2003).
122
See, e.g., City of Milwaukee v. Illinois, 451 U.S. 304, 317 (1981)
(concluding that the 1972 Amendments to the Federal Water Pollution Control
Act “occup[y] the field through the establishment of a comprehensive regulatory
program supervised by an expert administrative agency”); Rice v. Santa Fe
Elevator Corp., 331 U.S. 218, 230 (1947), rev’d sub nom. Rice v. Board of
Trade of Chicago, 331 U.S. 247 (1947) (stating that federal regulatory scheme
may be so pervasive or federal interest so dominant that enforcement of state
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
771
from a conflict between state law and federal law.123 Such a
conflict can be actual, such as where it is impossible to satisfy both
federal and state law simultaneously, or indirect, when a state law
stands as an obstacle to the accomplishment of Congress’s
objectives.124
In both express and implied preemption contexts, congressional
intent is “the ultimate touchstone” in determining the extent of
federal preemption.125 Thus, in determining whether preemption is
expressly addressed in a statute, “courts ask the question: did
Congress intend its law to preempt a challenged state law?”126
Since courts generally prefer to give effect to the plain and
ordinary reading of statutory language, in express preemption
situations, congressional intent is sought primarily in the language
of the preemption provision.127 That is, where a statute contains an
laws on same subject is precluded).
123
See Fidelity Fed. Sav. & Loan Ass’n, 458 U.S. at 153. Preemption of
state law where an actual conflict exists between a federal enactment and state
law is compelled by the Supremacy Clause. For further discussion of the
Supremacy Clause, see supra note 113.
124
See Pacific Gas & Elec. Co. v. Energy Resources Conservation & Dev.
Comm’n, 461 U.S. 190, 204 (1983) (noting that state law is preempted where it
is physically impossible to comply with both federal and state law); Florida
Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-43 (1961) (“A
holding of federal exclusion of state law is inescapable and requires no inquiry
into congressional design where compliance with both federal and state
regulations is a physical impossibility . . . .”).
125
Stabile, supra note 115, at 7 (citing Gade, 505 U.S. at 96).
126
Id.
127
Cipollone, 505 U.S. at 516. In using the plain meaning approach, it is
assumed that the legislature probably used the words, grammar, and punctuation
in a normal way to communicate its intent, so the words, grammar, and
punctuation are to be given the meaning that they would ordinarily produce
when trying to determine the legislature’s intent. The plain meaning statutory
analysis begins with “the assumption that the ordinary meaning of [the]
language accurately expresses the legislative purpose.” Park ‘N Fly, Inc. v.
Dollar Park and Fly, Inc., 469 U.S. 189, 194 (1985). The plain meaning rule
instructs a court to give the words of a statutory provision their “natural
meaning,” unless doing so “would lead to a clearly unreasonable, absurd
interpretation or there is otherwise clear evidence that Congress intended
something other than the plain meaning of the statute.” Stabile, supra note 115,
at 3, n.12 (citing Patterson v. Shumate, 504 U.S. 753, 758-62 (1992) (noting that
FREEDMAN MACROED CORRECTED 5-31-05.DOC
772
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
express preemption provision, the issue of whether a state law is
preempted is viewed as a question of statutory interpretation.128
Prior to 1992, many courts engaged in implied preemption
analysis after finding that a preemption provision did not invalidate
state law.129 In Cipollone v. Liggett Group, Inc.,130 a case
examining the preemptive scope of the Public Health Cigarette
Smoking Act of 1969, the Supreme Court stated that there is no
need to examine further the substantive provisions of legislation to
infer congressional intent to preempt state law when Congress has
included in the legislation a provision explicitly addressing
preemption and when that provision provides a “reliable indicium
the opponent of the plain meaning bears an exceptionally heavy burden of
persuasion in proving that Congress intended an alternative reading); Garcia v.
United States, 469 U.S. 70, 75 (1984) (stating that only an extraordinary
showing of contrary intentions from the statute’s legislative history would
justify a limitation on the unambiguous plain meaning of the statute’s language);
see also Robinson v. Shell Oil Co., 519 U.S. 337, 340 (1997) (explaining that if
the language in question has a plain and unambiguous meaning, then the court
should not inquire further except in rare and exceptional cases).
The Supreme Court has used the plain meaning approach in interpreting
various preemption provisions. See Shaw v. Delta Airlines, 463 U.S. 85, 96-97
(1983) (“A law ‘relates to’ an employee benefit plan, in the normal sense of the
phrase, if it has a connection with or reference to such a plan.”) (citing BLACK’S
LAW DICTIONARY 1158 (5th ed. 1979)); Morales v. Trans World Airlines, Inc.,
504 U.S. 374, 383 (1988) (analogously applying Black’s Law Dictionary
“relates to” definition to the interpretation of ADA); Cippolone, 505 U.S. at
520-24 (citing Shaw, 463 U.S. at 97) (finding that the plain meaning of “no
requirement or prohibition” sweeps broadly enough to encompass common-law
obligations; thus the Court must give effect to the plain meaning unless there is
clear congressional intent otherwise).
128
Stabile, supra note 115, at 7.
129
Id. at 57 n.209 (citing Taylor v. General Motors Corp., 875 F.2d 816,
825-27 (11th Cir. 1989)) (holding that a common law tort action for failure to
install airbags was not expressly preempted by the National Traffic and Motor
Vehicle Safety Act, but was impliedly preempted by the provisions of Act), cert.
denied, 494 U.S. 1065 (1990); Pennington v. Vistron Corp., 876 F.2d 414, 42021 (5th Cir. 1989) (holding that claims were impliedly preempted by the Public
Health Cigarette Smoking Act of 1969, but not expressly preempted)).
130
505 U.S. 504 (1992). For a discussion on Cipollone, see Stabile, supra
note 115, at 59.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
773
of congressional intent with respect to state authority.”131 The
Court reasoned that Congress’s enactment of an express
preemption provision implies that matters beyond the reach of that
provision are not preempted.132
There is a presumption against federal preemption of laws
concerning “spheres traditionally occupied by the states.”133
Housing regulations, land use restrictions, and zoning ordinances
are within the traditional spheres of state and local police
powers.134 Historically, preemption of a state or local law is not
131
Cipollone, 505 U.S. at 517.
Id.
133
Medtronic, Inc. v. Lohr, 518 U.S. 470, 475 (1996) (“States traditionally
have had great latitude under the police powers to legislate as to the protection
of the lives, limbs, health, comfort, and quiet of all persons.”); N.Y. State
Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U.S.
645, 655 (1995) (citing and quoting Rice, 331 U.S. at 230) (noting that the
exercise by a local authority of its historic police power is not to be superseded
by federal statutes unless this was the clear and manifest purpose of Congress);
Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 440 (1982)
(explaining that preemption is not appropriate in areas in which states
traditionally have enjoyed broad power to regulate).
134
Lucas v. S.C. Coastal Council, 505 U.S. 1003, 1034 (1992) (noting that
property is bought and sold and investments are made subject to the State’s
power to regulate); Mugler v. Kansas, 123 U.S. 623, 669 (1887) (holding that
the Takings Clause, while conferring substantial protection on property owners,
does not eliminate the police power of the State to enact limitations on the use of
their property); Nollan v. California Coastal Comm’n, 483 U.S. 825, 836 (1987).
The government’s power to forbid particular land uses in order to
advance some legitimate police-power purpose includes the power to
condition such use upon some concession by the owner, even a
concession of property rights, so long as the condition furthers the same
governmental purpose advanced as justification for prohibiting the use.
Id. See also MacDonald, Sommer & Frates v. County of Yolo, 477 U.S. 340,
361 (1986).
[P]olice power regulations such as zoning ordinances and other landuse restrictions can destroy the use and enjoyment of property in order
to promote the public good just as effectively as formal condemnation
or physical invasion of property. From the property owner’s point of
view, it may matter little whether his land is condemned or flooded, or
whether it is restricted by regulation to use in its natural state, if the
effect in both cases is to deprive him of all beneficial use of it.
132
FREEDMAN MACROED CORRECTED 5-31-05.DOC
774
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
lightly presumed, and courts exercise great restraint when spheres
traditionally occupied by the states are the subjects of a preemption
challenge.135 Therefore, even with express preemption, it is
difficult to argue that a federal law was intended to broadly
supplant the power of state or local governments with regard to
housing or land use issues.136
IV. FOREST PARK AND TOPA EQUITIES
In Forest Park137 and Topa Equities,138 the Eighth and Ninth
Circuits, respectively, held that the LIHPRHA preemption
provision applies to housing projects opting out of federal
programs under HOPE.139 The courts reached opposite conclusions
regarding federal preemption of the state preservation laws at issue
because of the diverging preemption tests each court embraced.140
Id. Village of Belle Terre v. Boraas, 416 U.S. 1 (1974); Berman v. Parker, 348
U.S. 26 (1954) (noting that land use regulation is within the inherent police
powers of the States and their political subdivisions).
135
For a discussion on the presumption against preemption, see supra note
133. See also Chester v. Panicucci, 281 A.2d 811 (1971) (explaining that federal
preemption of local police powers must be explicit).
136
See Jones v. United States, 529 U.S. 848, 860 (2000) (Stevens, J.,
concurring) (citing U.S. v. Bass, 404 U.S. 336, 349 (1971)). Principles of
federalism dictate that in the absence of a clear intent to supersede the historic
police powers of the States, Congress cannot be deemed to have significantly
changed the federal-state balance. Franklin Tower, 157 N.J. at 615; see also N.Y.
State Conference of Blue Cross & Blue Shield Plans, 514 U.S. at 654. Because
the party claiming preemption bears the heavy burden of supporting that claim
by “clear and manifest evidence,” the starting point for any preemption analysis
dealing with housing issues is that Congress does not generally intend to
supplant state law. N.Y. State Conference of Blue Cross & Blue Shield Plans,
514 U.S. at 654.
137
Forest Park II v. Hadley, 336 F.3d 724 (8th Cir. 2003).
138
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
139
Topa Equities, 342 F.3d at 1069; Forest Park, 336 F.3d at 732.
140
Compare Forest Park, 336 F.3d at 732 (holding that the Minnesota
preservation statutes at issue are expressly and impliedly preempted by the
applicable federal statutes) with Topa Equities, 342 F.3d at 1070 (holding that
the Los Angeles preservation ordinance at issue is not expressly or impliedly
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
775
In Topa Equities, the Ninth Circuit attempted to distinguish the
Eighth Circuit’s holding in Forest Park on the grounds that the
preservation laws were fundamentally different.141 However, at
least one federal court has observed that the Ninth Circuit’s
holding concerning the applicable scope of the LIHPRHA
preemption provision conflicts with the Eighth Circuit’s practical
effects test.142
A. Applying the Preservation Act’s Preemption Provision to
HOPE’s Alternative Prepayment Schedule: The Plain
Reading
In Forest Park II v. Hadley, the Eighth Circuit reviewed a
tenant’s association’s attempt to delay the prepayment of a Section
236 mortgage by the owner of a low-income housing
development.143 The Minnesota statutes at issue require that an
owner provide one year’s notice in advance of prepayment, while
current federal law under HOPE requires notice of no less than 150
days.144 The owner complied with federal requirements, but failed
to comply with the state notice requirement.145 The district court
found that since the building owner was opting out of the federal
affordable housing program under HOPE’s prepayment provisions,
LIHPRHA did not apply.146 The district court granted the tenant’s
preempted by the applicable federal statutes). Topa Equities, 342 F.3d at 1069;
Forest Park II, 336 F.3d at 724.
141
Topa Equities, 342 F.3d at 1069.
142
Independence Park Apts. v. United States, 61 Fed. Cl. 692, 704 (Fed. Cl.
2004).
143
Forest Park II, 336 F.3d at 727.
144
Id. at 729-30.
145
Id. at 727.
146
Forest Park II v. Hadley, 203 F.Supp.2d 1071, 1077-78 (D. Minn.
2002).
As indicated by the language of Section 4122, the statute only preempts
state law that is “inconsistent with the provisions of this subchapter”; in
other words LIHPRHA. 12 U.S.C. § 4122(b) (2005). Section 4122
therefore has no applicability to state laws that may conflict with the
notice provisions enacted apart from LIHPRHA. In this case, Forest
FREEDMAN MACROED CORRECTED 5-31-05.DOC
776
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
association injunctive relief until the owner complied with both
state and federal law.147
On appeal, the Eighth Circuit observed that, “unlike cases
involving a field traditionally regulated by the states, there [was]
no presumption against preemption in this case” because the case
did not involve a field traditionally regulated by the states.148 The
court determined that the central issue was not whether the statutes
involved a field subject to the state’s traditional police power, but
rather, whether Minnesota law was restricting HUD from
administering the “entity it regulates”—the HUD-subsidized
project.149 The court rejected traditional preemption analysis, in
which congressional intent controls, because, under the Supremacy
Clause,150 “state statutes may not interfere with the implementation
of a federal program by a federal agency.”151 Therefore, the court
regarded congressional intent and legislative history as irrelevant
because in the regulatory realm at issue, federal law reigned
supreme.152 The court effectively short-circuited the preemption
analysis by determining that the issue was settled by the state
statute’s frustration of a federal administrative agency’s acts.
Unlike the district court below, the Eighth Circuit was not
Park II asserts that the relevant state laws conflict with Section 219 of
the 1999 HUD Appropriations Act, Pub. L. 105-276, §
219(b)(3)(1999).
Id. at 1075.
147
Id.
148
Forest Park II v. Hadley, 336 F.3d 724, 731 (8th Cir. 2003) (citing
Buckman Co. v. Plaintiff’s Legal Comm., 531 U.S. 341, 347 (2001)) (“The
relationship between a federal agency and the entity it regulates is inherently
federal in character because the relationship originates from, is governed by, and
terminates according to federal law.”).
149
Forest Park II, 336 F.3d at 731-32.
150
See Supremacy Clause, supra note 113.
151
Forest Park II, 336 F.3d at 731-32 (citing Gade, 505 U.S. at 96) (noting
that in traditional preemption analysis, the “ultimate touchstone” is
congressional intent, but that the unique federal laws and programs involved in
the case make it difficult to apply a traditional preemption analysis). The court
also noted that there was no presumption against preemption in this case. Id.
(citing Buckman, 531 U.S. at 347).
152
Forest Park II, 336 F.3d at 732.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
777
persuaded by the tenants’ argument that LIHPRHA was implicitly
repealed because Congress ceased funding its incentive programs.
The circuit court found that LIHPRHA had not been explicitly
repealed, and it was still applicable because the loans at issue were
described in the eligibility provision.153 The LIHPRHA preemption
provision, according to the court, therefore remains enforceable
against state or local laws that “restrict or inhibit” prepayment.154
Under this rationale, federal law, not state law, provides the
appropriate notice requirements.
The Ninth Circuit heard similar arguments in the case of Topa
Equities v. City of Los Angeles,155 in which an owner of an
apartment building challenged a Los Angeles ordinance
prohibiting owners of low-income housing who had previously
opted-out of federal housing programs from raising rents until
existing low-income tenancies had terminated.156 The building
owner had prepaid his Section 236 mortgage in hopes of raising
rents to market levels and claimed that Section 4122 preempted the
city ordinance.157 The city argued, and the district court agreed,
that HOPE impliedly repealed the LIHPRHA preemption clause,
given that the Act’s new opt-out requirements post-dated the
defunding of LIHPRHA.158
On appeal, the Ninth Circuit reversed the district court and
concurred with the Eighth Circuit’s holding that LIHPRHA’s
153
Id. at 733.
Congress used very broad language in defining the types of mortgages
covered by the preemption provision. To the extent that it intended
preemption to apply only to laws affecting mortgages subject to
LIHPRHA, it could have stated as much. The fact that Congress no
longer funds the incentive programs established by LIHPRHA does not
mean that the prepayment provisions contained therein are irrelevant or
that the statute is no longer the law.
Id.
154
155
Id.
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
156
157
158
Id. at 1070.
Id.
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
778
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
preemption provision applies to HOPE’s prepayment schedule.159
Specifically, the Ninth Circuit determined that LIHPRHA’s
preemption provision was still effective, despite the termination of
funding for LIHPRHA-based programs.160 The court concluded
that congressional inaction had no effect on LIHPRHA’s express
preemption provision.161 Thus, the court expressly joined the
Eighth Circuit in limiting the inquiry to the unrepealed language of
the dormant LIHPRHA program.162
B. Determining the Scope of the Preservation Act’s Preemption
Provision
In Forest Park, the Eighth Circuit determined that Congress
intended for Section 4122 to be applied broadly.163 The court
supported its holding by citing the broadness of the “restrict or
inhibit” language of LIHPRHA’s preemption provision. Further,
the court emphasized that Congress had originally intended to offer
prepayment as an incentive for owners to participate in the HUD
program.164 According to the court, the broad language used in
defining the types of mortgages covered by the preemption
provision implied that the provision’s application was not limited
to mortgages subject to a POA for LIHPRHA prepayment.165 The
court therefore interpreted the Preservation Act’s preemption
provision to apply to projects opting out under alternative
159
Id.
Id. at 1069 (citing and quoting Firebaugh Canal Co. v. United States,
203 F.3d 568, 575 (9th Cir. 2000)) (“The intention of the legislature to repeal
must be clear and manifest, and in the absence of some affirmative showing of
an intention to repeal, the only permissible justification for a repeal by
implication is when the earlier and later statutes are irreconcilable.”).
161
Topa Equities, 342 F.3d at 1069 (“While it is true that Congress has,
since enacting HOPE, ceased funding LIHPRHA’s incentive programs, that
inaction has no effect on LIHPRHA’s express preemption provision. That
provision is extant.”).
162
Id.
163
Forest Park II, 336 F.3d at 732-34.
164
Id.
165
Id.
160
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
779
prepayment processes, such as HOPE.166
In order to determine whether the Minnesota statutes “restrict
or inhibit” the prepayment of mortgages, the court examined the
“practical effect” of the state restrictions.167 The court recognized
that the statutes did not, on their face, directly “restrict or inhibit”
the prepayment of mortgages.168 Nonetheless, the Eighth Circuit
reasoned that the effect of the state statutes was to “restrict or
inhibit” prepayment of federal mortgages since an owner could be
in compliance with the federal notice requirement when prepaying
the mortgage, but would still need to wait to prepay in order to
comply with the state’s notice requirements.169 The court explained
that because compliance with state regulations is required, “the
statutes have the direct effect of impeding, burdening, and
inhibiting the prepayment of federal mortgages even if the
additional requirements may be minimal.”170
Moreover, the court found that the federal laws regarding
prepayment “impliedly preempt the [Minnesota] statutes because
the state statutes conflict with federal law.”171 Under the practical
effects test, any statute that diminishes the realization of federally
166
Id. at 729.
Id. at 733.
168
Forest Park II, 336 F.3d at 733.
169
Id. The court stated:
The effect is that the state law forces the federal government to
continue to provide financial assistance to the participant when both the
federal government and the participant have chosen to end their
relationship. In this way, the state law not only regulates the conduct of
the citizen-owner, requiring him to take additional actions in order to
withdraw, but also regulates or restricts the actions of the federal
government under its own federal program.
167
Id.
170
Id. The court described the state statute as an additional requirement
“that forces owners to remain in a federally subsidized program from which
Congress has authorized withdrawal.” Id. at 733-34. The court further explained
that “[s]ince the Minnesota law stands as an obstacle to the accomplishment and
execution of [HUD’s] full purposes and objectives as defined by Congress, it
must give way.” Id.
171
Id. at 733.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
780
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
granted expectations is preempted by Section 4122.172
The Ninth Circuit in Topa Equities v. City of Los Angeles173
determined that the preemption provision should be narrowly
applied, thereby implicitly rejecting the “practical effects”
approach endorsed by the Eighth Circuit.174 The Ninth Circuit
articulated a different test—the “legal consequences” approach—
for determining whether the state statute at issue was preempted by
LIHPRHA.175
Under the legal consequences test, the court found that the Los
Angeles rent stabilization law was not preempted as applied to a
building opting out of a Section 236 HUD-subsidized mortgage.176
The Ninth Circuit examined whether LARSO directly prohibited or
limited the ability of federal housing project owners to prepay their
mortgages by imposing some legal bar or impediment to their
doing so. Unlike the Eighth Circuit in Forest Park, the Topa
Equities court did not address whether LARSO made prepayment
impracticable as an economic matter.177
In reviewing the case, the court noted that there was a
presumption against preemption stemming from the historical
understanding of housing and land use issues as part of the states’
police powers.178 The court cited paragraph (b) of the LIHPRHA
172
173
Id.
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
174
Id.
Id.
176
Id.
177
Id.
178
Topa Equities, 342 F.3d at 1071 (quoting from Kargman v. Sullivan,
552 F.2d 2, 6 (1st Cir. 1977)). The court in Kargman addressed whether a local
rent control ordinance was preempted by HUD regulations. The First Circuit
concluded that the ordinance was not preempted because it operated
independently from the federal subsidized housing program. Kargman, 552 F.2d
at 6. A different result was reached by the First Circuit in City of Boston v.
Harris, 619 F.2d 87 (1st Cir. 1980), in which the Court of Appeals held that
Boston’s rent control ordinance directly conflicted with HUD regulations and
was accordingly preempted. The Court in Topa Equities looked to whether the
city’s “traditionally strong interest in local rent control must yield.” Topa
Equities, 342 F.3d at 1070.
175
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
781
preemption provision, which preserves certain state or local laws,
and commented that “nothing in the HUD regulations purports to
limit states from enacting their own rent control laws of general
applicability which apply equally to apartment owners who exit the
federal program as well as other apartment owners.”179
Additionally, the court cited a First Circuit decision that held that
“federal legislation creating the network of subsidized housing
laws is superimposed upon and consciously interdependent from
the substructure of local law relating to housing.”180
The Ninth Circuit reasoned that the local ordinance at issue
was not preempted by Section 4122 because it did not “restrict or
prohibit” an owner’s prepayment options.181 Specifically, the Ninth
Circuit held that subsection (b) of Section 4122 insulated the city
ordinance from challenge because the ordinance applied across the
board, preventing all owners of low-income housing from
increasing rents, regardless of whether an had owner prepaid or
opted out of federal housing programs.182 The court distinguished
the Eighth Circuit’s findings in Forest Park by holding that the
Los Angeles ordinance, unlike the Minnesota statutes, restricted
rental increases in all apartment buildings, regardless of an owner’s
past or present participation in federal housing programs.183
The Eighth Circuit in Forest Park184 and Ninth Circuit in Topa
179
Topa Equities, 342 F.3d at 1072.
Id. (citing to Kargman, 552 F.2d at 1).
181
Id. at 1067.
182
Id. at 1070.
183
Id. (finding that the Minnesota law prohibited prepayment of federal
programs if the participant did not provide a longer period of notice than the
federal notice required upon opting out of the federal program; therefore, while
the Los Angeles ordinance affected all low-income housing owners, regardless
of participation in federal low-income housing programs, the Minnesota statute
was specifically limited to those owners who sought to prepay federalsubsidized mortgages) (citing Kenneth Arms, 2001 U.S. Dist. LEXIS 11470)
(finding state statutes that required owners seeking to prepay federally
subsidized mortgages to comply with state regulations that required longer
notice than federal law and provided tenants with first refusal rights were not
preempted by the preemption language contained in LIHPRHA).
184
Forest Park II, 336 F.3d at 724.
180
FREEDMAN MACROED CORRECTED 5-31-05.DOC
782
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
Equities185 are the only circuit courts to have ruled on whether the
LIHPRHA preemption provision applies to housing projects opting
out under alternative federal statutes.186 The courts’ decisions to
apply the preemption provision in such cases were based on a plain
and ordinary reading of the statute. While the cases are
distinguishable based on the types of statutes or regulations
involved (a procedural requirement to opting out as compared to a
regulation limiting property owners’ returns upon opting out), they
appear to conflict regarding the applicable scope of LIHPRHA’s
preemption provision.187
V. ANALYSIS
The holdings of Forest Park188 and Topa Equities189 have
significant implications for the low- and very-low income tenants
residing in the remaining 1.4 million rental units in more than
23,000 privately owned, HUD-subsidized properties.190 These
holdings restrict the ability of state and local governments to
address distinctly local housing issues in the face of the federal
government’s progressive burdening of state and local
governments with responsibility for administering federal
programs to house the poor.191 In their attempts to address
affordable housing shortages, several states and localities have
enacted preservation laws that work to maintain the quickly
diminishing permanent stock of affordable housing.192 Despite the
185
Topa Equities, 342 F.3d at 1065.
Id. at 1069; Forest Park, 336 F.3d at 732.
187
Independence Park Apts. v. United States, 61 Fed. Cl. 692, 704 (Fed.
Cl., 2004). Compare Forest Park, 336 F.3d at 732 (holding that the Minnesota
preservation statutes are expressly and impliedly preempted by the applicable
federal statutes) with Topa Equities, 342 F.3d at 1070 (holding that the Los
Angeles preservation ordinance is not expressly or impliedly preempted by the
applicable federal statutes).
188
Forest Park II v. Hadley, 336 F.3d 724 (8th Cir. 2003).
189
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
190
National Housing Trust Testimony, supra note 80.
191
Topa Equities, 342 F.3d at 1069; Forest Park II, 336 F.3d at 724.
192
For discussion on state and local preservation policies, see supra Part
186
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
783
agreement of the Eighth and Ninth Circuits regarding the first issue
in the preemption analysis—whether LIHPRHA’s preemption
provision applies to HOPE’s prepayment provisions—courts
should continue to review this issue, particularly in light of the
ambiguous and incongruous results of the Eighth and Ninth
Circuit’s plain reading interpretation of an implied relationship
between LIHPRHA and HOPE.
With regard to the circuit split on the second issue—the
applicable scope of federal preemption—the Ninth Circuit’s
reasoning is more consistent with past preemption jurisprudence.
First, housing regulations, land use restrictions, and zoning
ordinances are within the zone of state police powers.193 In light of
the presumption against preemption of laws and regulations
enacted in accordance with a state’s police powers, the provision
must be construed narrowly.194 Second, LIHPRHA’s legislative
history illustrates congressional intent not to limit the ability of
state and local governments to regulate privately owned,
subsidized affordable housing projects after opt out.195 The courts’
interpretations of these issues determine whether owners are
permitted to opt out without complying with the state procedures
intended to preserve affordable housing.
A. In Search of Congressional Intent
Although the Eighth and Ninth Circuits ruled on the issue of
whether the LIHPRHA preemption provision applies to projects
opting out of federal affordability restrictions under HOPE, this
remains an open question. State courts and federal courts in other
circuits are not bound by the decisions of these circuits; indeed, a
state court in the Ninth Circuit reached the opposite conclusion
subsequent to Ninth Circuit’s decision in Topa Equities.196 The
III.
193
See supra note 133 and accompanying text.
Cipollone v. Liggett Groups, Inc., 505 U.S. 504, 516 (1992).
195
See infra notes 256 through 260 and accompanying text.
196
College Gardens Preservation Committee v. Eugene Burger, No. 03
AS02608, slip op., at 3 (Sac. Superior Court, Cal. Nov. 19, 2003) (ruling that
California’s law requiring notice to tenants in prepaying rental projects was not
194
FREEDMAN MACROED CORRECTED 5-31-05.DOC
784
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
debate over whether to apply the LIHPRHA preemption provision
to HOPE is shaped by differing methods of statutory interpretation:
the circuit courts have applied a plain meaning approach, while
affordable housing advocates have sought to direct the courts’
attention to the statutory purpose and congressional history of
LIHPRHA.197 The standard for determining the effect of the
LIHPRHA preemption provision is congressional intent; therefore,
the method of statutory interpretation applied by the courts dictates
the scope of the examination.198
The plain meaning approach used by the Eighth and Ninth
Circuits gives effect to the clear, ordinary language of the Acts.199
The LIHPRHA preemption provision, by its terms, applies to any
laws or regulations that restrict prepayments of “eligible lowincome housing.”200 In 1990, when Congress enacted LIHPRHA,
buildings in all federal housing programs were LIHPRHA eligible
and regulated by HUD.201 The Eighth Circuit determined that
preempted by the long-dormant LIHPRHA when the owners were not seeking to
prepay under LIHPRHA), available at http://www.ruralhome.org/manager/
uploads/college.pdf.
197
Forest Park, 336 F.3d at 729 (“Appellees argue that [the preemption]
provision is not applicable to Forest Park’s Section 236 mortgage because of
Congress’s subsequent actions.”); see generally Topa Equities, 342 F.3d at
1070.
198
Ultimately, rules of statutory interpretation attempt to determine the
“intent of the legislature.” 2A SUTHERLAND STATUTORY CONSTRUCTION § 45.5
(6th ed.). Differing theories of statutory interpretation are in essence means of
giving a particular law the meaning intended by Congress. For a discussion on
the plain meaning method of statutory interpretation, see supra note 127.
199
Id.
200
H.R. Rep. No. 101-559, at 78 (1990) [hereinafter House Comm. Report].
[Section] 4122 would preempt and declare null and void any state or
local law, ordinance or regulation that limits an owner’s right to pay off
a mortgage on eligible low- income housing or, limits the occupancy,
type of tenure, use or rental charges of such a property. The committee
wishes to emphasize that the pre-emption provision only applies to
eligible housing, defined in the bill as projects with mortgages that are
insured or assigned under Section 221 (d)(3)(BMIR) or Section 236
program.
Id. (emphasis added).
201
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
785
Minnesota laws mandating additional procedural requirements for
projects attempting to opt out of federal affordable housing
programs infringed upon HUD’s administrative mandate to
regulate federal housing policies and programs.202 The court
concluded that legislative history was irrelevant because the
Supremacy Clause of the U.S. Constitution prohibits the states
from interfering with the implementation of a federal program by a
federal agency.203 The court therefore held that the Minnesota laws
were preempted based on a plain reading of HUD’s federal
prerogatives and the state law’s impact on housing under federal
programs.
Through its cursory analysis, the Eighth Circuit bypassed the
opportunity to examine other evidence of congressional intent,
including relevant legislative history and HUD’s own policies,
which do not support the application of the LIHPRHA preemption
provision to projects opting out through HOPE.204 The Ninth
Circuit expressly adopted the Eighth Circuit’s holding without
significant discussion. The court appeared content with the plain
reading conclusion that because LIHPRHA had not been explicitly
repealed, the preemption provision continued to apply to HUDregulated housing projects and programs.
The circuit courts’ plain meaning interpretation of LIHPRHA
leads to ambiguous and incongruous results. As such, the circuit
courts should have considered other means of interpretation.205 In
effect, the courts have applied the terms of LIHPRHA to HOPE, a
202
See supra notes 148 through 153 and accompanying text.
Id.
204
This analysis would lead to the examination of HUD policies regarding
the application of Section 4122 to projects opting out through HOPE. As
mentioned above, the Court mentioned HUD’s policies, but did not consider
them to govern the decision. Id. This analysis appears inconsistent. Choosing to
free HUD from state regulations that it does not deem restrictive would appear
beyond what the court was called upon to do.
205
United States v. Kay, 359 F.3d 738, 743 (2004) “If, after application of
these principles of statutory construction, we conclude that the statute is
ambiguous, we may turn to legislative history. For the language to be considered
ambiguous, however, it must be susceptible to more than one reasonable
interpretation or more than one accepted meaning.” Id. (internal quotations and
citations omitted).
203
FREEDMAN MACROED CORRECTED 5-31-05.DOC
786
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
separate statute, absent statutory language or congressional records
explicitly addressing the relationship between the statutes.206 The
plain meaning approach is not appropriate when the court is
examining two separate federal statutory programs.
Nonetheless, the courts held that because LIHPRHA’s terms
incorporate all buildings opting out of federal affordability
programs today, these buildings also benefit from the Act’s
preemption provision. LIHPRHA provided HUD with guidance for
regulating all federal housing programs; therefore, all buildings
built before 1990 were LIHPRHA eligible.207 Congress, however,
subsequently ordered HUD to suspend further processing of
preservation applications that did not have approved action plans,
thereby rendering LIHPRHA dormant.208 Although LIHPRHA was
never expressly repealed, as evidenced by the fact that buildings
participating in preservation programs continue to receive funding
pursuant to agreements executed while the program was still
accepting new applications, technically no buildings remain
prospectively LIHPRHA eligible.209 For this reason, the courts’
206
See supra note 78 through 80 and accompanying text.
12 U.S.C. § 4119(1) (2005).
§ 4119(1) The term “eligible low-income housing” means any housing
financed by a loan or mortgage— (A) that is (i) insured or held by the
Secretary under section 221(d)(3) of the National Housing Act [12
USCS § 1715l(d)(3)] and receiving loan management assistance under
section 8 of the United States Housing Act of 1937 [42 USCS § 1437f]
due to a conversion from section 101 of the Housing and Urban
Development Act of 1965; (ii) insured or held by the Secretary and
bears interest at a rate determined under the proviso of section
221(d)(5) of the National Housing Act [12 USCS § 1715l(d)(5)]; (iii)
insured, assisted, or held by the Secretary or a State or State agency
under section 236 of the National Housing Act [12 USCS § 1715z-1];
or (iv) held by the Secretary and formerly insured under a program
referred to in clause (i), (ii), or (iii); and (B) that, under regulation or
contract in effect before February 5, 1988, is or will within 24 months
become eligible for prepayment without prior approval of the
Secretary.
207
Id.
208
See supra note 74 and accompanying text.
See Kenneth Arms, 2001 U.S. Dist. LEXIS 11470, at *13. HUD’s
inability to accept new preservation applications, execute new POAs or enforce
209
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
787
application of the LIHPRHA preemption provision to buildings
opting out under HOPE is inappropriate, as there is no active
program for which the buildings may be deemed eligible. Since
1996, HUD has lacked the authority to accept new LIHPRHA
applications.210 It is illogical to suggest that, at LIHPRHA’s
creation, Congress intended that LIHPRHA should apply to
properties that had never participated in LIHPRHA because, in
1990, Congress had no reason to address non-participating
properties, which did not exist as a class until Congress ceased
funding mandatory LIHPRHA preservation in 1996. HUD’s own
policy before these cases was consistent with this approach.211
As the Supreme Court held in Cipollone, “[w]here Congress
explicitly preempts state law, Congress’s enactment of a provision
defining the preemptive reach of a statute implies that matters
beyond that reach are not preempted.”212 As non-LIHPRHA
properties are not referenced in LIHPRHA’s express preemption
provision, such properties are beyond its reach. If the LIHPRHA
preemption provision were applied to non-LIHPRHA properties, it
is reasonable to assume that all LIHPRHA prepayment provisions
should apply as well. This includes requiring owners to issue a
LIHPRHA notice of intent and to submit and receive HUD
approval of a POA.213 It is counterintuitive to suggest that one
provision of LIHPRHA applies to owners when owners are not
required to meet any of the statute’s other prepayment
LIHPRHA’s provisions except as to owners already participating in LIHPRHA
prior to October 1, 1996 as sufficient reasoning for the holding that since “[t]he
Owners were never involved in the LIHPRHA Preservation Program, and never
operated under the LIHPRHA plan of action. Rather, the prepayment scheme
followed by the Owners is that embodied in [Section 219], permitting mortgage
prepayment without HUD approval, rather than LIHPRHA with its restrictions.”
Id. The court held that the preemption provision of LIHPRHA, 12 U.S.C. §
4122, does not govern and thus does not preempt the California notice and right
of first refusal statutes. Id.
210
See supra note 74 and accompanying text.
211
Letter from Clare Harringan, Office of HUD General Counsel, to Judge
Lawrence K. Karlton, June 21, 2001 [hereinafter HUD Letter] (on file with
author).
212
Cipollone v. Liggett Groups, Inc. 505 U.S. 504, 517 (1992).
213
See supra note 67 and accompanying text.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
788
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
requirements. Indeed, if the buildings that were the subjects of
review by the Eighth and Ninth Circuits were LIHPRHA eligible,
without explicit congressional language on point, LIHPRHA
would dictate that the owners of these buildings would be required
to submit findings of minimal tenant impact as a condition of
receiving HUD authorization to opt out.214 To argue, as the Eighth
and Ninth Circuits did, that a project is entitled to the benefits of
LIHPRHA (express preemption), but need not shoulder its burdens
(for example, detailed assurances that low-income tenants will not
be harmed) turns a reasonable interpretation of congressional intent
on its head. Indeed, as a slip opinion by a California Superior
Court referencing the holdings of both Forest Park and Topa
Equities recently noted, “Although LIHPRHA continues to apply
to properties participating prior to 1996, in effect the heart of
LIHPRHA has been eviscerated.”215 A cursory analysis of these
results seem incongruous or at the least unfair, thus requiring
further examination of the legislative history and purpose of both
statutes.216
214
215
Id.
College Gardens Preservation Committee v. Eugene Burger, slip op., at
3.
216
This is similar to holdings in various circuit courts in cases dealing with
the American with Disabilities Act of 1990 (42 U.S.C. § 1201). While out of the
preemption context, it does provide a workable solution about when and how to
read congressional history and legislative purpose when dealing with absurd
results of a plain meaning interpretation. The following circuit courts held
against the plain language of the ADA, which states that an employer is required
to accommodate an individual who is “regarded as” disabled. See Kaplan v. City
of North Las Vegas, 323 F.3d 1226, 1231-33 (9th Cir. 2003).
On the face of the ADA, failure to provide reasonable accommodation
to “an otherwise qualified individual with a disability” constitutes
discrimination. And, on its face, the ADA’s definition of “qualified
individual with a disability” does not differentiate between the three
alternative prongs of the “disability” definition. The absence of a stated
distinction, however, is not tantamount to an explicit instruction by
Congress that “regarded as” individuals are entitled to reasonable
accommodations. Moreover, because a formalistic reading of the ADA
in this context has been considered by some courts to lead to bizarre
results, we must look beyond the literal language of the ADA.
Id. at 1232 (citing Royal Foods Co., Inc. v. RJR Holdings, Inc., 252 F.3d 1102,
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
AFFORDABLE HOUSING PRESERVATION
789
The ambiguity created by a plain meaning approach to
interpreting the LIHPRHA preemption provision suggests that
courts should apply a clearer and more thorough interpretative
framework to the analysis of this provision by looking to statutory
purpose and legislative history to determine whether to apply the
LIHPRHA preemption provision to HOPE’s prepayment
provisions. The lack of congressional funding for LIHPRHA
should give rise to an analysis of available congressional history
and legislative purpose, which would serve to highlight the error of
applying the preemption provision to properties opting out under
HOPE. LIHPRHA’s legislative purpose and congressional intent in
this area dictate that the LIHPRHA preemption provision should
not be applied to HOPE’s provisions.217 This analysis has been
supported by a formal opinion of HUD, which, in lieu of relevant
congressional language, should be granted deference.
1. Analyzing the Legislative Purpose of LIHPRHA
Congressional hearings on LIHPRHA suggest that LIHPRHA
preemption is only applicable to LIHPRHA-eligible properties.218
Furthermore, there is no preemption language contained in HOPE,
the National Housing Act of 1934, the Emergency Low Income
Housing Preservation Act of 1987, or the regulation promulgated
by HUD that grants owners the right to prepay their mortgages.219
Through LIHPRHA, Congress demonstrated its understanding that
1108 (9th Cir. 2001)) (internal citations omitted) (recognizing that a court must
look beyond the plain language of a statute when the literal interpretation would
lead to an absurd result). See also Weber v. Strippit, Inc., 186 F.3d 907, 916-17
(8th Cir. 1999); Workman v. Frito-Lay, Inc., 165 F.3d 460, 467 (6th Cir. 1999);
Newberry v. E. Texas State Univ., 161 F.3d 276, 280 (5th Cir. 1998).
217
These arguments are expanded versions of arguments presented to the
Sacramento, California Superior Court on behalf of the plaintiff, College
Gardens Preservation Committee, in the case of College Gardens Preservation
Committee v. Eugene Burger. See Letter Brief on the Applicability of LIHPRHA,
College Gardens Preservation Committee v. Eugene Burger, No. 03 AS02608
(Sac. Superior Court, Cal. Nov. 19, 2003) (on file with author).
218
See House Comm. Report, supra note 200.
219
For a discussion on the relationship between various federal housing
statutes, see supra Part II.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
790
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
it was free to add a preemption provision when it deemed one
necessary. Congress determined that a preemption provision was
required in LIHPRHA because the Act added substantial burdens
and cumbersome regulations to subsidized properties.220
Prior to its defunding, LIHPRHA was a comprehensive federal
preservation program that substantially restricted prepayments of
“eligible low-income housing.”221 Only a few properties could
satisfy LIHPRHA’s restrictive criteria for prepayment and
conversion to market-rate housing.222 In fact, the LIHPRHA
program functioned primarily to provide federally funded
incentives to preserve eligible properties.223 Preemption was an
integral part of the comprehensive LIHPRHA statutory scheme.
The purpose of Section 4122’s preemption provision was to
prevent states and localities from singling out LIHPRHA
participants for special and disadvantageous treatment that would
reduce the federal preservation incentives otherwise available
under the program.224
The congressional record, in providing for express preemption
of prepayment restrictions, demonstrates that Congress was
concerned with equitable treatment of private affordable housing
projects throughout the states and had no intention of abrogating
state procedural requirements as they applied to non-participating
properties.225 Restrictions by individual states would have proven
unnecessary and have created non-uniform procedures nationwide.
Further, investors seeking to develop federally subsidized
properties would have been able to choose where to develop based
not on need, but rather, on the number and form of state opt-out
provisions. Because the federal government abandoned projectsubsidized programs in 1996 when it reinstituted owners’
prepayment rights, it is reasonable to assume that the federalism
interests that were of concern at the time of LIHPRHA’s enactment
220
See House LIHPRHA Report, supra note 72; College Gardens
Preservation Committee v. Eugene Burger, slip op., at 3.
221
See supra note 204 and accompanying text.
222
See Winkelman, supra note 66.
223
Id.
224
See House LIHPRHA Report, supra note 72.
225
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
791
are no longer relevant. As such, courts should not second-guess
Congress’s decision not to add a preemption provision to HOPE.226
2. Reviewing HUD’s Interpretation of the LIHPRHA Preemption
Provision’s Applicability to Non-LIHPRHA Properties
Upon the enactment of LIHPRHA, Congress charged HUD
with administering the LIHPRHA program.227 In exercising this
mandate, HUD determined that the LIHPRHA preemption
provision does not apply to non-LIHPRHA properties.228 In the
agency’s opinion letter addressing this issue and related federal
preemption questions concerning California’s prepayment notice
statutes,229 HUD’s General Counsel explained that because
Congress ceased funding for new LIHPRHA action plans in 1996,
LIHPRHA presently applies only to those projects that received
LIHPRHA preservation incentives prior to 1996.230 Therefore,
state laws can be inconsistent with LIHPRHA only with respect to
projects that have already received preservation incentives and
226
United States v. Van Lufkins, 676 F.2d 1189, 1194 (1982).
LIHPRHA, 12 U.S.C. § 4101. (“General prepayment limitation (a)
Prepayment and termination. An owner of eligible low-income housing may
prepay, and a mortgagee may accept prepayment of, a mortgage on such housing
only in accordance with a plan of action approved by the [HUD]
Secretary. . . .”).
228
HUD Letter, supra note 211.
229
Kenneth Arms, 2001 U.S. Dist. LEXIS 11470 (holding that the
LIHPRHA preemption provision does not apply to four non-LIHPRHA
apartment developments).
230
HUD Letter, supra note 211.
The preemption provision in LIHPRHA at Section 232, 12 U.S.C. Sec.
4122, was intended to afford protection to owners of properties that
were, or are, operating under the LIHPRHA Preservation Program
(emphasis added). Section 4122(b) states that the section “shall not
prevent the establishment, continuing in effect, or enforcement of any
law or regulation of any State or political subdivision of a State not
inconsistent with the provision of this subchapter.” Thus, a state law
could not be inconsistent with the provision of LIHPRHA for an owner
who was never involved in the LIHPRHA Preservation Program and
never operated under a LIHPRHA plan of action.
Id. (emphasis in original).
227
FREEDMAN MACROED CORRECTED 5-31-05.DOC
792
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
undertaken additional affordability restrictions.231 The HUD Letter
further stated that the express language of the LIHPRHA
preemption provision limits preemption to local laws that are
inconsistent with LIHPRHA.232 In addition to relying on the
limiting language of the LIHPRHA preemption provision, HUD
further reasoned that because “HUD does not have authority to
accept new preservation applications or to enter into new plans of
action, it has continued to implement and enforce the provisions of
LIHPRHA only as to those owners who were in the program prior
to the passage of HOPE [the successor statutory scheme] in
1996.”233
In United States v. Mead,234 the Supreme Court held that an
agency’s “permissible construction of [a] statute” that it is charged
with administering is entitled to “some deference,” even if the
agency’s interpretation of the statute is not rendered within a
public notice and comment rulemaking framework.235 Indeed, as
explained in Skidmore v. Swift,236 such agency interpretations are
“entitled to respect” to the extent they have the “power to
persuade.”237 Further, in Auer v. Robbins238 the Supreme Court
held that an agency’s interpretation of its own regulations is
controlling unless “plainly erroneous or inconsistent with the
regulation.”239 Like the agency interpretation at issue in Auer,
which was submitted in an amicus brief, HUD’s interpretation
regarding preemption in the Kenneth Arms case was submitted by
231
Id.
Id.
233
HUD Letter, supra note 211.
234
533 U.S. 218, 235 (2001) (finding that Customs letter “classification
ruling may at least seek a respect proportional to its power to persuade, and may
claim the merit of its writer’s thoroughness, logic and expertness, its fit with
prior interpretations, and any other sources of weight”).
235
Reno v. Koray, 515 U.S. 50, 61 (1995) (citing Chevron U.S.A. Inc. v.
Natural Resources Defense Council, 467 U.S. 837 (1984)).
236
323 U.S. 134, 140 (1944).
237
Id.
238
519 U.S. 452, 461 (1997).
239
Id.
232
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
AFFORDABLE HOUSING PRESERVATION
793
letter brief at the court’s invitation.240 The agency’s interpretation
was not at issue in Kenneth Arms. Thus, like the agency
interpretation in Auer, “[t]here is simply no reason to suspect that
the interpretation does not reflect the agency’s fair and considered
judgment on the matter in question.”241
For the foregoing reasons, future courts hearing preemption
challenges should question the continued application of
LIHPRHA’s preemption provision to owners opting out under
Section 219 of HOPE or other similar prepayment provisions.
B. Future Preemption Analysis and the Ninth Circuit’s
Approach
If courts continue to apply LIHPRHA’s preemption provision
to owners opting out under HOPE, courts should follow the Ninth
Circuit’s narrow “legal consequences” analysis.242 This approach is
240
Forest Park II v. Hadley, 203 F. Supp. 2d 1071, 1076 (2002) (citing
HUD Letter Brief in Kenneth Arms, 2001 U.S. Dist. LEXIS 11470, at *33).
241
Auer, 519 U.S. at 461. The Eighth Circuit in Forest Park, in a footnote,
cast aside Supreme Court precedent and found that the Court owes no deference
to the letter cited in Kenneth Arms. Forest Park II, 336 F.3d at 733, n.6. The
Court distinguished the letter at issue from an official agency interpretation
resulting from official agency rulemaking procedures. Id. The Court cited the
Supreme Court case of Christensen v. Harris County, in which the Supreme
court found that a Department of Labor opinion letter taking the position that an
employer may compel the use of compensatory time only if the employee has
agreed in advance to such a practice was not due deference for the application of
a clear and unambiguous statute. Christensen v. Harris County, 529 U.S. 576,
587 (2000).
In Christensen, however, the Supreme Court also found that “[a]n agency’s
opinion letter interpreting its own regulation is entitled to deference when
language of regulation is ambiguous.” Christensen, 529 U.S. at 588. The only
thing that is clear regarding LIHPRHA, its provisions, Congress’s refusal to
continue funding LIHPRHA programs, Congress’s enactment of an alternative,
and currently only prepayment option through budget Appropriation, is that the
application of HOPE and LIHPRHA is ambiguous. For a discussion on the
ambiguities created through a plain reading interpretation of the relationship
between LIHPRHA and HOPE, see supra Part V.
242
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
794
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
consistent with established preemption jurisprudence and
congressional intent. The Eighth Circuit’s expansive “practical
effects” approach erroneously applies administrative law principles
to a preemption analysis and is inconsistent with congressional
intent.243
As past preemption cases dictate, there is a presumption against
preemption within traditional spheres of local police powers,
unless it can be shown that it is the clear and manifest purpose of
Congress to preempt state authority.244 Housing regulations, land
use restrictions, and zoning ordinances fall within the traditional
spheres of state and local police powers.245 While the Supremacy
Clause does not require a narrow or broad construction in response
to the presumption against preemption,246 the Supreme Court, in
Cipollone, recently held that the preemption provision at issue
required a narrow interpretation.247
The Eighth Circuit’s holding and its doctrine of preempting
regulations that have the practical effect of restricting or inhibiting
the prepayment of HUD-subsidized mortgages is inconsistent with
the presumption against preemption. Furthermore, absent any
guidance, the application of this approach may impede the ability
of state and local governments to preserve affordable housing
within the existing framework of the federal prepayment
scheme.248 A modest application of this holding would in effect
243
Forest Park II v. Hadley, 336 F.3d 724, 732 (8th Cir. 2003), reh’g and
reh’g en banc denied (2003).
244
See discussion supra Part III and accompanying notes.
245
Id.
246
Cipollone v. Liggett Group, Inc., 505 U.S. 504, 544 (1992) (Scalia, J.,
concurring in judgment in part and dissenting in part).
247
Cipollone, 505 U.S. at 517.
248
Forest Park II, 336 F.3d at 734. The Eighth Circuit may have attempted
to limit this expansive interpretation of LIHPRHA’s preemptive scope, but its
language fell short of such a goal.
[N]ot . . . all state attempts at preserving existing federally subsidized,
low-income housing are preempted. Nothing in the federal statutes,
their legislative history, or their stated objectives indicates that states
are prohibited from instituting their own incentive plans or other
programs to preserve low-income housing within the framework of the
federal prepayment scheme. When, however, these state programs
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
795
preempt all state and local preservation laws that inhibit the
realization of expectations by owners opting out or prepaying
Section 221(d)(3) or 236 mortgages.249 This language includes
limitations on the property owners’ returns and forced transfers of
properties to preservation owners, in addition to the procedural
requirements involved in the case.250 As the court succinctly stated,
a “further requirement imposed by a state statute would directly
interfere with Congress’s original intent of offering prepayment as
an incentive.”251 Therefore, even though the Eighth Circuit found
that the Minnesota notice requirements do not explicitly bar
prepayment, it nonetheless found that they had the effect of
restricting opt out and were consequently preempted.252
The Ninth Circuit held that LIHPRHA’s structure supports the
proposition that the “restrict or inhibit” language of Section 4122
was not intended to preempt state preservation laws that do not
explicitly place barriers on the prepayment of Section 221(d)(3)
and 236 mortgages.253 The court found that LIHPRHA’s
preemption provision requires a narrow construction and expressly
states that local laws are preempted “only to the extent that [they]
violate the provisions of this subsection.” 254 Further, the statute
limits preemption to local laws “inconsistent” with LIHPRHA.255
Congressional history suggests that the Ninth Circuit’s holding
is more in line with what Congress intended and that Congress in
place additional requirements on federal program participants, restrict
the exercise of the participants’ federally granted prepayment rights, or
create delays in the prepayment process, they are preempted.
Id.
249
Id. at 732 (noting that the effect of the state statute was to limit a right to
move to the market rates that Congress provided).
250
For discussion on various preservation laws and regulations, see supra
Part III.
251
Forest Park II, 336 F.3d at 733. Nonetheless, the law that provided the
incentive is not the law that provided the preemption provision—an
inconsistency the court failed to address. Id.
252
Forest Park II, 336 F.3d at 732.
253
Topa Equities, Ltd. v. City of Los Angeles, 342 F.3d 1065 (9th Cir.
2003).
254
LIHPRHA, 12 U.S.C. § 4122.
255
Id.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
796
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
fact considered state preservation laws and did not intend to
preempt such laws.256 Statements from various House committees
illustrate that Congress intended LIHPRHA to work in conjunction
with state preservation laws; indeed, a House conference
committee specifically cited to a Maryland preservation law.257
The Maryland law noted by the conference committee was adopted
in 1989, the year before Congress enacted LIHPRHA.258 This state
law requires owners to provide notice to the local government and
to tenants at least one year before prepayment, which is more
comprehensive than the notice required by the Minnesota impact
statement law at issue in Forest Park.259 If the congressional
committee did not intend to preempt the Maryland law, then it
follows that Congress did not intend to preempt the Minnesota or
California laws, which were less far reaching.260
Further, LIHPRHA’s reporting requirements demonstrate that
Congress understood state and local efforts as a means of ensuring
256
Congressman Hoagland, sponsoring the amendment that became section
4122(b), described it as “narrowing the State and local law preemption language
in the bill so that the state and local laws that contradict this statute will be
preempted.” 136 CONG. REC. H6053-01, H6183 (daily ed. July 31, 1990). See
also S. REP. NO. 316, 101 Cong., 2nd Sess. 1, 106 (June 8, 1990), reprinted in
1990 U.S.C.C.A.N. 5763, 5868 (“Local task forces have also considered the
range of municipal responses (rent control, tax abatement, etc.) that can be taken
alone or in conjunction with a federal preservation solution . . . .”) (emphasis
added).
257
See H.R.CONF.REP. 101-943, reprinted in 1990 U.S.C.C.A.N. 6070,
6171. (“In the event of prepayment, HUD would have several tools to protect
the existing tenants and assist the affected community in replacing the stock.
The tenant protections build upon provisions contained in the House bill as well
as in State laws such as the Maryland Assisted Housing Preservation Act.”).
258
MD. ANN. CODE art. 83B, tit. 9, §§ 101-114 (2003); see statutes cited
supra notes 93-94.
259
MD. ANN. CODE art. 83B, tit. 9-103(a)(1)-(5) (2005); see statutes cited
supra notes 93-94.
260
MD. ANN. CODE art. 83B, tit. 9, §§ 101-114 (2003); see statutes cited
supra notes 93-94. See also Cal. Fed. Sav. and Loan Ass’n v. Guerra, 479 U.S.
272, 287-88 (1987) (finding “significant” the fact that Congress was aware of
and acknowledged in debates existing similar state laws when enacting federal
law and “failed to evince the requisite ‘clear and manifest’ purpose to supersede
them”).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
797
an effective flow of information regarding prepayment and viewed
such local efforts as complementary to LIHPRHA. Throughout
LIHPRHA are requirements that the owner and HUD notify both
tenants and state or local government entities of all activities
leading up to prepayment.261 LIHPRHA thus encourages state and
local entities to use this information to assist tenants in preserving
affordable housing.262 These provisions evince Congress’s desire
to empower tenants and state and local governments to effect the
preservation purposes of the Act. In a 1999 Hearing Notice
provided to all affordable housing projects, HUD clearly supported
261
LIHPRHA, 12 U.S.C. § 4102(b) (2005). In relevant part, the statute
provides:
The owner, upon filing a notice of intent under this section, shall
simultaneously file the notice of intent with the chief executive officer
of the appropriate State or local government for the jurisdiction within
which the housing is located and with the mortgagee, and shall inform
the tenants of the housing of the filing.
Id. § 4106(c). “The Secretary shall make any information provided to the owner
under subsections (a) and (b) of this section available to the tenants of the
housing together with other information relating to the rights and opportunities
of the tenants.” Id. § 4107(a)(2).
Each owner submitting a plan of action under this section to the
Secretary shall also submit a copy to the tenants of the housing. The
owner shall simultaneously submit the plan of action to the office of the
chief executive officer of the appropriate State or local government for
the jurisdiction within which the housing is located.
Id. § 4118:
The Secretary shall confer with any appropriate State or local
government agency to confirm any State or local assistance that is
available to achieve the purposes of this title and shall give
consideration to the views of any such agency when agency when
making determinations under this subchapter. The Secretary shall also
confer with appropriate interested parties that the Secretary believes
could assist in the development of a plan of action that best achieves
the purposes of this subchapter.
Id.
262
LIHPRHA, 12 U.S.C. § 4107(a)(2) (“An appropriate agency of such
State of local government shall review the plan and advise the tenants of the
housing of any programs that are available to assist the tenants in carrying out
the purposes of this title.”).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
798
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
this position, explaining that “besides meeting the Federal
notification requirement, project Owners must also comply with
any State or local notification requirements.”263
There is a distinct legal difference between imposing a barrier
to prepayment and imposing rent control on formerly subsidized
buildings.264 Speculative owners may argue that differentiating
between the two destroys the intent of the preemption provision
because applying rent control post-opt out, from an owner’s
perspective, compels the same result as a prepayment barrier.265 By
opting out of a federal program, however, an owner changes the
essence of the particular building. For example, the owner is no
longer required to comply with certain federally mandated property
standards, the owner is no longer liable under certain federal laws,
and the tax structure of the investment is changed.266 The
legislative history of LIHPRHA suggests that, while Congress
intended to provide property owners with an opportunity to free
themselves from the increased oversight and potential liability
associated with participation in a federal housing program, it did
not intend to permit owners to free themselves from the historic
police powers of states and localities to regulate housing and land
use issues.
Therefore, LIHPRHA’s structure, notice requirement, and
legislative history do not demonstrate the clear and manifest
congressional intent required for Section 4122 to preempt state
notice laws.267 Conversely, they suggest congressional approval of
supplementary state preservation efforts, such as the Maryland
263
Kenneth Arms, 2001 U.S. Dist. LEXIS 11470, at *10 (citing HUD
Housing Notice 99-36, Dec. 29, 1999).
264
Forest Park II v. Hadley, 336 F.3d 724, 732 (8th Cir. 2003), reh’g and
reh’g en banc denied (2003). See also Lifgren v. Yeutter, 767 F. Supp. 1473 (D.
Minn. 1991) (holding that the Preservation Act and regulations relating thereto
are not inconsistent with the borrower’s option to prepay at any time, but rather,
the Preservation Act and its regulations simply provide procedures that must be
followed in the event that a borrower evidences an intent to prepay).
265
Forest Park, 336 F.3d at 732.
266
Winkelman, supra note 38, at 1160.
267
See Cipollone, 505 U.S. at 516.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
6/6/2005 1:26 PM
AFFORDABLE HOUSING PRESERVATION
799
preservation statute.268
VI. RECOMMENDATION: STATE AND LOCAL GOVERNMENTS
SHOULD LOOK TO TRANSFER PROPERTIES TO PRESERVATION
OWNERS
Preservation of affordable housing is an economical and
efficient means of safeguarding housing for current and future lowincome residents.269 As the federal government continues to
abandon affordable housing programs, state and local governments
must look to preservation laws to maintain affordable housing or
face waves of homeless or displaced low-income tenants.270
Preservation policies offer security to current tenants facing
drastically increasing rents in units that have benefited from
subsidized rents for the past twenty years. Laws that encourage and
provide methods of transferring federally subsidized affordable
housing projects to non-speculative owners, such as the Illinois
Assisted Housing Preservation Act and the New York Tenant
Empowerment Act (Intro. No. 186), provide an optimal means of
preserving at-risk housing.271 The transfer of projects to tenantendorsed or tenant-controlled non-profit organizations, whose
purpose is to provide housing, not generate profit, will keep tenants
in their homes and preserve properties as future housing sources.272
268
MD. ANN. CODE art. 83B, tit. 9-103(a)(1)-(5) (2005); see statutes cited
supra note 93-94. MD. ANN. CODE art. 83B, tit. 9, §§ 101-114 (2003).
269
Grow, supra note 9, at 3. Given current funding concerns, the
development of new affordable housing units appears unlikely. Id.
270
Id.
271
Rather than providing notice of impending opt-out, these laws provide
long-term security to the tenants and their communities. Grow, supra note 9, at
3.
272
Galle, supra note 85, at 5. Obtaining both the necessary funds for
transfer and site control from private owners remains a difficult challenge for
state and local governments. Id. Some 150 states or localities have adopted
housing trust funds providing revenue to assist the preservation of affordable
homes. Id. Obtaining sufficient capital funds will usually require state or local
financial contributions, which may include formerly “federal” funds such as
Low-Income Housing Tax Credits, HOME Investment Partnership Program or
Community Development Block Grant funds. Id. Low-Income Housing Tax
FREEDMAN MACROED CORRECTED 5-31-05.DOC
800
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
Legal and practical issues regarding these laws are untested, but
following the preemption analysis set forth in the cases regarding
other preservation laws, these laws should not be subject to federal
preemption because they do not “restrict or inhibit” prepayment, as
they apply after opt out.273 Illinois’s statute and the proposed New
Credit (LIHTC) is an indirect method of subsidizing low-income housing. See
132 Cong. Rec. S8146-8158 (daily ed. June 23, 1986) (explaining the legislative
intent of the LIHTC). See also DAYE, supra note 3, at 247-49. HOME
Investment Partnerships Program, U.S. Dep’t of Housing and Urban Dev.,
Home Investment Partnerships Program (2004), at http://www.hud.gov/
offices/cpd/affordablehousing/programs/home/index.cfm. For a discussion on
the HOME Investment Partnership Act of 1990, see DAYE, supra note 3, at 24042. For a discussion on Community Development Block Grant program
(CDBG), see supra note 7.
Some states have dedicated some general revenues to maintaining
affordability in HUD-subsidized housing. Galle, supra note 85, at 6 (“For
example, in California the enacted budget for FY 2000 includes $6 million for a
broad purpose multifamily acquisition and rehabilitation program, with the first
priority for funding being the preservation of currently affordable units.”). On
the local level, San Francisco, for example, established the San Francisco
Redevelopment Agency (SFRA) to administer several million dollars of
redevelopment agency tax increment funds for a variety of uses, including grants
and below-market loans for nonprofit purchasers, pre-development assistance,
and tenant outreach, organizing, and technical assistance. Galle, supra note 85,
at 6. More information regarding SFRA available at http://www.sfgov.org/
site/sfra_index.asp.
Federal Low-Income Housing Tax Credits and tax-exempt bond allocations
may provide an important source of funds for nonprofit acquisitions. Galle,
supra note 85, at 6 (“California reserves 10 percent of its Low-Income Housing
Tax Credits for preservation. For several years in Massachusetts . . . 60 percent
of the state’s credits are allocated to large-scale projects with significant federal
resources.”) (internal quotations and citations omitted). See generally Ammann,
supra note 8. Both state and municipal governments provide other capital or
debt subsidy measures. Galle, supra note 85, at 7 n.36, 37 (“Many state housing
finance agencies, including those in Maine, Missouri, Pennsylvania, and
Vermont, use their own budgets for low- or no-interest loans to promote
preservation purchases. These funds are used for costs of purchase, to
rehabilitate properties, to expand affordability, [and] to cover predevelopment
costs. . . . ”).
273
Grow, supra note 9, at 6; Galle, supra note 85, at 6; See Franzese, supra
note 10.
A constitutional taking issue exists with the application of these
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
801
York City housing regulations are good examples of laws that
work to provide increased protections for current tenants.274
preservation laws. Grow, supra note 9, at 6. Although there have so far been no
such challenges to state or local purchase opportunity laws, the takings issue
appears fairly straightforward so long as just compensation in provided. Grow,
supra note 9, at 6 (citing Williamson County Regional Planning Comm’n v.
Hamilton Bank of Johnson City, 473 U.S. 172, 194 (1985) (“The Fifth
Amendment does not proscribe the taking of property; it proscribes taking
without just compensation.”)). The federal Constitution prohibits takings for
public use without just compensation. U.S. CONST. amend. V (“[N]or shall
private property be taken for public use, without just compensation. . . .”) Id.
Federal, state, and local regulatory action can result in a constitutional
taking of private property so long as it is for a valid public use and just
compensation is provided. Grow, supra note 9, at 6. So long as preservation
purchase opportunity laws are not mandatory or provide market value
compensation by setting transfer prices at appraised market value, they appear to
fall safely outside of the takings clause. Id. (citing Palazzolo v. Rhode Island,
533 U.S. 606, 625 (2001)) (“When a taking has occurred, under accepted
condemnation principles the owner’s damages will be based upon the property’s
fair market value . . . an inquiry which will turn, in part, on restrictions on use
imposed by legitimate zoning or other regulatory limitations.”).
A constitutional takings challenge to LIHPRHA was examined along these
lines. In Chancellor Manor v. United States, the Federal Circuit held that since
the Owners of federally subsidized affordable housing projects should have
known that HUD’s regulations regarding prepayment could be changed at any
time, and that their right to prepay their mortgages was not a property interest
protected by the Fifth Amendment’s takings clause. Chancellor Manor v. United
States, 331 F.3d 891, 903 (Fed. Cir. 2003). The court remanded the case for a
more thorough factual inquiry in order to determine whether the plaintiffs could
“establish the existence of a regulatory taking under the Penn Central
standards.” Id. at 906. The court directed the lower court to address all Penn
Central factors including: (1) the extent to which HUD’s regulatory change
interfered with the Owners’ objective reasonable expectations; (2) the economic
impact of HUD’s regulatory change on the Owners; and (3) the nature of HUD’s
regulatory change. Id. at 906.
274
New York City Council Int. No. 186-2004 (“A Local Law to amend the
administrative code of the city of New York, in relation to creating a right of
first refusal and an opportunity to purchase.”). See Chen, supra note 10. These
laws still create the challenge for purchasers and public agencies to find or
provide the necessary funds. For a discussion on funding options, see Galle,
supra note 85. Ideas for such funding in New York include the creation of a
Housing Trust Fund, similar to the Federal Housing Trust Fund discussed. See
Chen, supra note 9.
FREEDMAN MACROED CORRECTED 5-31-05.DOC
802
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
Additionally, these laws are good illustrations of why state and
local governments should feel secure in the face of a preemption
challenge.
Transfer laws seek to reinstitute a key “component of the
federal preservation policy for HUD-subsidized properties that the
federal government [has] abandoned since 1995—that preservation
is of sufficient importance to warrant restrictions on owner
conversion, so long as those restrictions are supported with marketvalue compensation.”275 Both the Illinois law and the New York
City regulation cover all properties with HUD-subsidized
mortgages when the owners threaten conversion, sale, or
disposition of properties.276 Both statutes create purchase rights
when an owner proposes to sell a property or terminate the existing
federal subsidy programs or restrictions.277 These rights provide
tenant associations or their designees the right to purchase,
recognizing that tenant associations are not always in the best
position to purchase.278 A valuation of the properties in question is
performed through multiple appraisals, and a specified resolution
procedure is available if an agreement on the property’s value
cannot be reached.279 Thus, these laws relieve the government of
275
See Grow, supra note 9, at 6. This principle still governs the federal
Rural Housing Services-subsidized multifamily inventory. 42. U.S.C. § 1742(c)
et. seq. (2004).
276
Illinois law now covers properties with HUD-subsidized mortgages,
properties with certain state-provided mortgages, properties with expiring or
terminative project-based Section 8 contracts, and properties with expiring rent
restrictions under the federally funded but state-administered Low-Income
Housing Tax Credit program. 310 Ill. Comp. Stat. § 60/3 (as amended, July
2004). Proposed New York City Council Int. No. 186-2004, Chap. 9, §26-801
et. seq.
277
310 Ill. Comp. Stat. § 60/5 (2004) (as amended by SB 2329, enacted
July 2004); Proposed New York City Council Int. No. 186-2004, §§ 26-805,
807.
278
310 Ill. Comp. Stat. § 60/5 (2004) (as amended by SB 2329, enacted
July 2004); Proposed New York City Council Int. No. 186-2004, §§ 26-805,
807.
279
310 Ill. Comp. Stat. § 60/7(b) (2004) (as amended by SB 2329, enacted
July 2004); Proposed New York City Council Int. No. 186-2004, § 26-804
(Appraisal Determination).
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
803
the burden of maintaining housing projects, but encourage and
provide means for stabilizing losses of affordable housing units.
These preservation laws can survive a LIHPRHA preemption
challenge because they apply to buildings “following opt-out” of
federal housing programs.280 Notwithstanding anti-preemption
arguments, current tenants should look for ways to preserve current
affordable housing under the assumption that courts will continue
to apply the preemption provision. The Eighth Circuit’s expansive
doctrine is problematic, as a practical effect of forced transfer laws
may be to limit conversions.281 These transfer laws may limit the
ability of owners to realize speculative gains since the goal of such
preservation laws is to encourage owners to sell their properties to
preservation owners. However, nowhere does the language or
legislative history of LIHPRHA support the position that Congress
intended to displace all state and local power on housing and land
use issues.282 Under the Ninth Circuit’s analysis, these laws will
not be preempted, as they do not expressly “restrict or inhibit”
prepayment.283 State and localities are currently using procedural
requirements in conjunction with other forms of preservation laws;
however, in order to comply with possible federal preemption
issues, it may be wise to limit this practice and repeal current
procedural requirements.284 Even under the Ninth Circuit’s
analysis, procedural requirements may be interpreted as placing a
barrier to prepayment.285
The ironic conclusion is that preservation laws that limit
property owners’ returns on opting out (i.e., rent control) and aid or
force the transfer to preservation owners (under eminent domain
power) are more controversial than procedural requirements.286
280
See Franzese, supra note 10.
The practical effects test preempts laws that have the effect of restricting
owners from realizing profits from moving their building’s units to market rates.
Forest Park II, 336 F.3d 724. According to the Eighth Circuit, this would be an
attempt to circumvent the “restrict or inhibit” language of Section 4122. Id.
282
Topa Equities, 342 F.3d 1065.
283
Id.
284
Forest Park II, 336 F.3d 724.
285
For further discussion, see supra Part V.B.
286
See generally RICHARD A. EPSTEIN, TAKINGS: PRIVATE PROPERTY AND
281
FREEDMAN MACROED CORRECTED 5-31-05.DOC
804
6/6/2005 1:26 PM
JOURNAL OF LAW AND POLICY
This is one of the more absurd results of express preemption
provisions and the reason many scholars have called for the
cessation of congressional use of preemption provisions.287
Nonetheless, preservation laws that apply post-opt out will have
significant advantages in overcoming LIHPRHA preemption issues
and helping to provide “a decent home and suitable living
environment for every American family.”288
CONCLUSION
The preservation of project-based affordable housing is a vital
part of any housing program for low- and very-low-income
individuals. In the recent past, the federal government has
maintained preservation requirements in order to continue to
provide housing for America’s poorest residents. In the mid-1990s,
Congress, at the whim of changing national priorities, determined
that funds should be redirected toward the goal of balancing the
national budget.289 Congress subsequently enacted HOPE and
discontinued federal preservation requirements.290
State and local governments both before and after these federal
efforts have used preservation policies to provide homes for lowand very-low income tenants. As witnessed throughout the nation,
these policies continue to provide security to those fearing
impending notice of a drastic increase to market-rate rents.
Congress, in drafting LIHPRHA, recognized the crucial role of
states and localities in providing affordable housing. In LIHPRHA,
Congress provided that state and local preservation laws should not
be superseded by the Act. Nonetheless, because the burdens
imposed by LIHPRHA were heavy, Congress restricted the ability
of states and localities to place further burdens on properties with a
LIHPRHA “plan of action” by enacting the LIHPRHA preemption
POWER OF EMINENT DOMAIN (1985); see also Jordan Rose, Eminent
Domain Abuse: The Growing Threat to Private Property, 174 ARIZ. ISSUE
ANALYSIS, Aug. 16, 2002.
287
See generally Stabile, supra note 115.
288
National Housing Act of 1949, 42 U.S.C. § 1441.
289
See supra note 73 through 74 and accompanying text.
290
See supra note 75 through 77 and accompanying text.
THE
FREEDMAN MACROED CORRECTED 5-31-05.DOC
AFFORDABLE HOUSING PRESERVATION
6/6/2005 1:26 PM
805
provision.
Legislative history dictates that Congress intended for
LIHPRHA’s benefits, including its preemption provision, to be
extended to those properties accepting the burdens of LIHPRHA.
Therefore, buildings opting out under alternative prepayment
schemes, such as HOPE, were not intended be the beneficial
recipients of LIHPRHA incentives, including Section 4122. The
buildings at issue in Forest Park291 and Topa Equities292 opted out
of Section 236 affordability restrictions through HOPE’s
prepayment provisions, and therefore, should not have been
afforded the benefit of LIHPRHA preemption. The analysis of both
the Eighth and Ninth Circuits overlooked a basic manifestation of
congressional intent as determined through rules of statutory
construction endorsed by the Supreme Court.
In the event that courts continue to apply LIHPRHA’s
preemption provision to affordable housing properties opting out
under alternative prepayment schemes, they should apply the Ninth
Circuit’s narrow approach. The Ninth Circuit’s interpretation and
analysis accord with established preemption jurisprudence and are
more aligned with congressional intent, as determined through
LIHPRHA’s structure and legislative history. The Eighth Circuit’s
expansive “practical effects” approach erroneously applies
administrative law principles to a preemption analysis and is not in
line with explicit congressional intent. These divergent
interpretations possess the potential to further reduce the
availability of housing to the detriment of elderly and low-income
tenants. Even though the federal government has given up on these
tenants, the federal courts should not inhibit local governments
from working for their benefit.
291
292
Forest Park, 336 F.3d 724.
Topa Equities, 342 F.3d 1065.
BETH MACROED CORRECTED FINAL 060605.DOC
6/6/2005 1:37 PM
BROADNAX V. GONZALEZ: QUESTIONING
THE IMPACT OF EXPANDING FETAL
RIGHTS ON LITIGATION AND
HEALTHCARE IN NEW YORK
Elizabeth Lemanowicz∗
INTRODUCTION
It is difficult to imagine the anguish Marta Tebbutt faced on
September 6, 1980, as she gave birth to a child she knew had died
inside her.1 Postmortem examination of the fetus revealed that a
negligently-performed amniocentesis, a test in which Marta’s
doctor inserted a syringe into her lower abdomen in order to draw
fluid from the amniotic fluid around the fetus, had possibly caused
the fetus’s subsequent death.2 Marta turned to the courts for justice,
suing her doctor and seeking to recover for her “pain, severe
disappointment, anxiety, despondency, bitterness, and suffering.”3
However, for Marta Tebbutt, relief was never granted.4 The trial
court granted a motion by the defendant doctor for summary
judgment, dismissing the complaint as insufficient as a matter of
∗
Brooklyn Law School, Class of 2006; B.A. in Justice, American
University, 2002. The author would like to thank her family and friends for their
love and support, especially Damon Osborne, who was infinitely helpful and
patient during the entire writing process. She would also like to thank the staff
and editors of the Journal of Law and Policy for all their hard work and help.
1
Tebbutt v. Virostek, 477 N.Y.S.2d 776, 777 (N.Y. App. Div. 1984).
2
Id. See also Tebbutt v. Virostek, 483 N.E.2d 1142, 1144 (N.Y. 1985)
(describing the amniocentesis that Dr. Virostek administered to Marta Tebbutt)
(Jasen, J., dissenting).
3
Tebbutt, 477 N.Y.S.2d at 777.
4
Tebbutt, 483 N.E.2d at 1143 (affirming the trial court’s order granting
defendants’ motion for summary judgment dismissing the complaint).
807
BETH MACROED CORRECTED FINAL 060605.DOC
808
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
law.5 The appellate court affirmed the motion to dismiss,6 and
finally, the New York Court of Appeals affirmed the motion as
well.7 The courts reasoned that Marta had not suffered any
physical injury that would give rise to a claim for emotional
distress as a result of the stillborn birth.8 Without an independent
physical injury, Marta had no legal right to recovery for her
emotional injuries, and this would be the case for similarly situated
women for years to come.
The legal impediments faced by women such as Marta Tebbutt
were lifted in 2004 when the New York Court of Appeals decided
the landmark case of Broadnax v. Gonzalez, holding that, “even in
the absence of an independent injury, medical malpractice
resulting in miscarriage or stillbirth should be construed as a
violation of a duty of care to the expectant mother, entitling her to
damages for emotional distress.”9 Broadnax marked the end of
nearly twenty years of precedent that denied mothers damages for
emotional distress suffered from negligently caused miscarriages
or stillbirths unless they had experienced independent injuries.10
5
Tebbutt, 477 N.Y.S.2d at 777.
Id. at 779.
7
Tebbutt, 483 N.E.2d at 1143.
8
Tebbutt, 477 N.Y.S.2d at 777-78; Tebbutt, 483 N.E.2d at 1143-44. The
court noted that Marta Tebbutt alleged no physical injury distinct from that
suffered by the fetus. Id. Having suffered no physical injury, the court held that
Marta Tebbutt’s claim was governed by Vaccaro v. Squibb, 418 N.E.2d 386
(N.Y. 1980), in which the mother sought to recover for emotional injuries
resulting from the harm done to her child in the womb. Tebbutt, 483 N.E.2d at
1143. Because the mother in Vaccaro did not learn of the harm done to the fetus
until the birth, which occurred some time after the harm occurred, the court
rejected the contention that the defendants owed a duty to the mother. Id.
Similarly, in Tebbutt, the court rejected the mother’s claim for damages for
emotional distress. Id.
9
Broadnax v. Gonzalez, 809 N.E.2d 645, 649 (N.Y. 2004).
10
Id. at 648. Accord Tebbutt v. Virostek, 483 N.E.2d 1142 (N.Y. 1985). In
general, when there is a duty owed to a plaintiff by a defendant, a breach of that
duty that results in emotional injury is compensable even though no physical
injury occurred, but only if the breach “unreasonably endangered plaintiff’s
physical safety.” 1 N.Y. P.J.I.3d § 2:284, at 1476 (2005). However, in
Broadnax, the Court of Appeals held that an expectant mother may recover
damages for emotional distress resulting from a stillbirth or miscarriage that was
6
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
809
This comment analyzes the impact of Broadnax in New York.
The focus of the analysis is two-pronged. The first prong focuses
on the potential impact of the Broadnax decision on the future of
wrongful death suits for fetuses in New York.11 The second prong
focuses on the potential effect the decision will have on the
availability and cost of gynecological services in New York and,
consequently, on the rate of malpractice liability for obstetricians
and gynecologists. This comment argues that Broadnax could open
the door for a cause of action for wrongful fetal death, which is
presently prohibited as a cause of action in New York courts. By
reconsidering and expanding the rights of the mother and the fetus
in cases of prenatal malpractice, it is conceivable that mothers and
fetuses in the post-Broadnax era will be able to further push the
boundaries of tort law and claim new causes of action grounded in
wrongful death. This may be a substantial step in tort law, and
perhaps it is theoretically just; however, it is yet unknown whether
the liability that medical practitioners face will increase if a
previously unrecognized class of plaintiffs—unborn fetuses and
expectant mothers—are afforded significant legal rights.12 Indeed,
the Broadnax decision may prove detrimental to society if, as a
result of increasing liability for physicians, the provision of
healthcare becomes sufficiently expensive to compel the exit of
physicians from the fields of obstetrics and gynecology due to high
caused by medical malpractice, regardless of whether the mother suffered an
independent physical injury or whether her physical safety was unreasonably
endangered. Id. The Broadnax decision appears to have overruled decisions in
which recovery was denied for emotional distress resulting from a stillbirth or
miscarriage that was caused by medical malpractice in which the “independent
physical injury” was limited to the physical pain and suffering that naturally
accompanies the birthing process. 1 N.Y. P.J.I.3d § 2:150, at 802 (2005).
11
An action for “wrongful death” is a “lawsuit brought on behalf of a
decedent’s survivors for their damages resulting from a tortious injury that
caused the decedent’s death.” BLACK’S LAW DICTIONARY 1607 (7th ed. 1999).
New York’s wrongful death statute is set forth in N.Y. EST. POWERS & TRUSTS §
5-4.1 (2000). New York does not have a wrongful life statute.
12
Broadnax, 809 N.E.2d at 650. “[T]here is no way for us to predict or
assess the potential effect of this expansion of liability . . . on the cost and
availability of gynecological and obstetrical services in New York State.” Id.
(Read, J., dissenting).
BETH MACROED CORRECTED FINAL 060605.DOC
810
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
insurance premiums and a fear of being sued.13
Part I.A of this comment provides an overview of case law
regarding tort-based causes of action for emotional or
psychological injuries. Part I.B chronicles the history in New York
of causes of action for emotional distress related to prenatal care,
including a discussion of Tebbutt v. Virostek, the precursor to the
Broadnax decision. Part I.C provides an analysis of the court’s
holding and rationale in Broadnax. Part II.A focuses on the
potential impact of Broadnax on wrongful death lawsuits in New
York. Specifically, it contends that the reasons previously cited by
the New York Court of Appeals for banning actions for the
wrongful death of a fetus have been effectively undercut by the
court’s decision in Broadnax. Part II.B briefly addresses the impact
of Broadnax on the malpractice jurisprudence of the past year. Part
III discusses the potential ramifications of Broadnax for the
provision of obstetrical and gynecological care in New York.
Finally, this comment concludes that the state legislature, not the
judiciary, will need to take the lead if clarity and consistency is
ever to come to the area of tort jurisprudence that encompasses
fetal rights.
I. BROADNAX V. GONZALEZ: PAST AND PRESENT
In Broadnax v. Gonzalez,14 the New York Court of Appeals
overruled Tebbutt v. Virostek,15 which held that unless an
expectant mother suffered an independent physical injury, she had
no right to recover damages for emotional distress resulting from a
miscarriage or stillbirth.16 The Broadnax decision recognized that
medical malpractice resulting in a miscarriage or stillbirth
constituted a breach of duty to the expectant mother, and damages
for emotional distress arising out of that breach should be
13
Medical Liability: Hearing on H.R.5 and H.R.4280 Before the House
Committee on Small Business, 109th Cong. (2005) [hereinafter Hearings]
(statement of Dr. Donald J. Palmisano, M.D., J.D., Immediate Past-President of
the American Medical Association) available at 2005 WL 408414 (F.D.C.H.).
14
Broadnax v. Gonzalez, 809 N.E.2d 645 (N.Y. 2004).
15
Tebbutt v. Virostek, 483 N.E.2d 1142 (N.Y. 1985).
16
Id. See also 1 N.Y. P.J.I.3d §2:280, at 1462-1463 (2005).
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
811
recoverable, even absent physical injury.17 In overruling Tebbutt, a
case governed by the court’s earlier decision in Vaccaro v.
Squibb,18 the majority in Broadnax relied heavily on the language
and logic of the dissent in Tebbutt.19
A. A Brief History of Case Law Regarding Causes of Action for
Psychological Injuries
The issue of whether to permit causes of action for emotional
or psychic injuries absent independent physical injuries has been
treated differently by New York courts throughout history. For a
greater part of the twentieth century, New York courts insisted that
a plaintiff could not recover for emotional injuries absent a
physical injury.20 In 1961, the Court of Appeals fashioned a new
rule that permitted recovery for emotional injuries absent
immediate personal injury, but only if there was immediate fear or
threat of bodily harm to the plaintiff directly.21 In Battalla v. State,
an infant-plaintiff was placed in a chair lift at a state-run ski resort
by an employee who failed to properly secure the infant and lock
17
Broadnax, 809 N.E.2d at 649; 1 N.Y. P.J.I.3d § 2:280, at 1463 (2005).
Vaccaro v. Squibb, 418 N.E.2d 286 (N.Y. 1980) (holding that a mother
who was prescribed a toxic drug that rendered her child limbless at birth could
not recover for emotional and psychic harm absent an independent injury).
19
Broadnax, 809 N.E.2d at 648 (citing Tebbutt, 483 N.E.2d at 1144 (Jasen,
J., dissenting)); Tebbutt, 483 N.E.2d at 1149 (Kaye, J., dissenting). “On its own
terms, Tebbutt may make formal sense, but it created a logical gap in which the
fetus is consigned to a state of ‘juridical limbo.’ It is time to fill the gap. If the
fetus cannot bring suit, ‘it must follow in the eyes of the law that any injury here
was done to the mother.’” Id. (quoting language from the dissenting opinions in
Tebbutt).
20
Mitchell v. Rochester Railway Co., 45 N.E. 354 (N.Y. 1896) (holding
that a woman who miscarried as a result of being frightened by a team of horses
owned by railroad company did not have a cause of action, there being no
recovery available for mere fright absent immediate personal injury).
21
Battalla v. New York, 176 N.E.2d 729 (N.Y. 1961) (holding that a cause
of action exists when a claimant alleges that she was negligently caused to suffer
emotional and psychological injuries with consequential physical injuries).
18
BETH MACROED CORRECTED FINAL 060605.DOC
812
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
the equipment.22 As a result, the infant became frightened and
hysterical while riding on the chair lift and suffered “severe
emotional and neurological disturbances with residual physical
manifestations.”23 Under the principle that “a wrong-doer is
responsible for the natural and proximate consequences of his
misconduct,”24 the court held that the claimant should have the
opportunity to prove that her emotional injuries, leading to her
subsequent physical injuries, were the proximate result of the
defendant’s negligence. Thus, after Battalla, contemporaneous or
consequential physical harm, coupled with psychological injuries,
was thought to provide an “index of reliability otherwise absent in
a claim for psychological trauma with only psychological
consequences.”25
From the Battalla decision developed the rule that “one may
have a cause of action for injuries sustained although precipitated
by a negligently induced mental trauma without physical
impact.”26 In 1969, the Court of Appeals refused to apply the
Battalla rule to cases in which the tortfeasor’s duty not to cause
physical injury did not apply to the claimant seeking damages for
emotional and subsequent physical distress.27 In Tobin v.
Grossman, a mother was in a neighbor’s home when she heard
brakes screeching outside.28 She ran outside to find her injured
child lying on the ground at the site of the accident.29 The court
held that the plaintiff-mother was barred from bringing a cause of
action for her mental and consequential physical injuries caused by
22
Id. at 729.
Id. The “residual physical manifestations” were not explained or clarified
in either the trial or appellate level decisions. See Battalla v. State, 184 N.Y.S.2d
1016 (N.Y. Ct. Cl. 1959); Battalla v. State, 200 N.Y.S.2d 852 (N.Y. App. Div.
1960).
24
Battalla, 176 N.E.2d. at 730 (quoting Ehrgott v. Mayor of City of N.Y.,
96 N.Y. 264, 281 (1884)).
25
Johnson v. New York, 334 N.E.2d 590, 592 (N.Y. 1975).
26
Tobin v. Grossman, 249 N.E.2d 419, 420-21 (N.Y. 1969) (addressing the
issue of the possibility of recovery for physical injuries resulting from a purely
mental or psychological impact).
27
Id. at 419-20.
28
Id. at 419.
29
Id.
23
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
813
shock and fear for her child.30 The principles espoused in Battalla
made clear that an individual could bring a cause of action for
injuries that caused psychological trauma and consequential
physical injuries, even absent contemporaneous physical impact.31
Yet in Tobin, the court held that a cause of action did not exist for
psychological harm sustained by a person as a result of someone
else’s injuries, regardless of whether a special relationship existed
between the two individuals or whether the emotionally injured
person was an eyewitness to the incident that resulted in harm to
the other.32
However, in 1975, New York extended the rule in Battalla to a
situation in which the defendant’s negligence caused neither
contemporaneous nor consequential physical harm to the
plaintiff.33 In Johnson v. State, the plaintiff suffered emotional
harm as a direct result of the negligence of a state hospital, which
falsely notified the plaintiff that her mother, a patient at the
hospital, had died.34 The plaintiff’s emotional injuries were
unaccompanied by any physical injury.35 The Court of Appeals
held that it was the hospital’s duty to responsibly advise the proper
next of kin of a patient’s death and that recovery for emotional
harm would be permitted by an individual subjected directly to a
tortious act, such as the negligent mishandling of a corpse or the
negligent false notification of death.36 Johnson clarified that
individuals may recover for emotional harm, even in the absence
of fear of physical injury, when they are subjected directly to the
negligence of a tortfeasor.37 For such recovery, however,
individuals must prove that any suffered psychological injuries are
genuine and substantial, and that these injuries were proximately
caused by the defendant’s conduct.38
30
31
32
33
34
35
36
37
38
Id. at 420, 424.
Id. at 420-21.
Id. at 423-24.
Johnson v. New York, 334 N.E.2d 590 (N.Y. 1975).
Id. at 591.
Id.
Id. at 593.
Id.
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
814
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
B. Causes of Action for Psychological Injuries Caused by
Negligent Prenatal Care
The status of recovery for emotional suffering can be broken
down into four main rules: (1) where a tortfeasor causes physical
injury to another, the injured party can recover for the actual
physical injury and concurrent mental and emotional suffering
resulting from the wrongful act;39 (2) where a tortfeasor directly
causes the injured party to experience fear of physical injury as a
direct result of the tortious conduct, the party can recover for
psychic injuries absent physical impact;40 (3) where a tortfeasor
physically injures one party, recovery is denied for mental and
emotional injuries experienced by a third party as a result of the
physical injuries sustained by the first party; 41 but (4) where a
tortfeasor genuinely, substantially, and proximately causes
psychological injuries to the injured party, the injured party can
recover for the emotional harm, even in the absence of fear of
potential physical injury.42 These rules can be applied to cases of
medical malpractice in which a doctor’s negligence causes
physical injuries to a fetus, resulting in the miscarriage, stillbirth,
or permanent impairment of the child. Courts previously have
addressed such cases from the vantage point of the mother and
have examined whether a mother’s right to collect damages for
emotional distress resulting from the physical injuries sustained to
the fetus inside her is a situation consistent with any of the four
main rules.
In 1977, the New York Court of Appeals decided the case of
Howard v. Lecher, based on the third rule above, holding that a
parent who suffers psychological injuries as a result of a doctor’s
medical malpractice in treating a fetus cannot recover for such
damages.43 In Howard, the plaintiffs were the parents of a child
who died from Tay-Sachs disease, a progressive degenerative
39
40
41
42
43
Mitchell v. Rochester Railway Co., 45 N.E. 354 (N.Y. 1896).
Battalla v. State, 176 N.E.2d 729 (N.Y. 1961).
Tobin v. Grossman, 249 N.E.2d 419 (N.Y. 1969).
Johnson v. New York, 334 NE.2d 590 (N.Y. 1975).
Howard v. Lecher, 366 N.E.2d 64 (N.Y. 1977).
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
815
disease that affects the nervous system and tends to appear more
often in children with parents of Eastern European Jewish
descent.44 The parents alleged that the doctor-defendant was
negligent in failing to properly perform or evaluate a genealogical
history of the parents, given his knowledge that the Howards were
both Eastern European Jews and that there was a high risk that the
fetus would be born with Tay-Sachs.45 The Howards claimed that,
had the doctor informed them of this risk, they would have chosen
to abort the fetus.46 The parents brought a cause of action to
recover from the defendant-doctor for the emotional and mental
distress they experienced from witnessing their child suffer and die
from such a devastating disease.47 In this case, the court held that
the parents were not made to suffer any physical or mental injury,
except for the pain in watching their child suffer from Tay-Sachs,
and that the doctor’s negligence was not the direct cause of the
child’s suffering from the disease.48 The court suggested that, even
in a case in which the negligent conduct of a doctor directly injured
a fetus but in no manner physically injured the parents, there could
be no recovery for the mental and emotional pain and suffering of
the parents.49 Thus, there could be no recovery for the mental and
emotional injuries experienced by the parents in Howard.50
In 1978, the court decided the case of Becker v. Schwartz based
on the principles espoused in Howard.51 In Becker, two cases were
combined in which the plaintiffs sought damages for emotional
distress alleged to have occurred as a consequence of the birth of
their infants in an impaired state, the birth of those infants having
occurred through the negligence of the defendant-doctors.52 In
Becker, the plaintiffs had received prenatal care from the
44
45
46
47
48
49
50
51
52
Howard, 366 N.E.2d at 64-65, 67.
Id. at 65.
Id.
Id. at 64-65, 66.
Id. at 66.
Id.
Id.
Becker v. Schwartz, 386 N.E.2d 807 (N.Y. 1978).
Id. at 809.
BETH MACROED CORRECTED FINAL 060605.DOC
816
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
defendant-doctors.53 Plaintiff-mother Delores Becker was thirtyseven years old at the time of her pregnancy and at no point during
the prenatal care provided by the defendant-doctors was Delores
informed that, based on her age, she was at an increased risk of
bearing a child with Down’s Syndrome.54 Becker subsequently
gave birth to an infant with Down’s Syndrome.55 Becker and her
husband claimed damages for the pecuniary expenses they bore
and would continue to bear for the care and treatment of their
infant, and for the emotional and physical injuries suffered by
Delores as a result of her child’s having been born with Down’s
Syndrome.56
In a companion case, Park v. Chessin, Hetty Park and her
husband consulted the defendant-doctors to determine the
likelihood that they would bear a child afflicted with a genetic
kidney disease.57 Having already experienced the birth of a child
who had died from a genetic kidney disease five hours after being
born, the plaintiffs were concerned with the possibility that they
might bear another child so afflicted.58 In response to the plaintiffs’
inquiry, the defendant-doctors told the Parks that the chances of
having another baby afflicted with the kidney disease were
“practically nil.”59 As a result of this information, the Parks
renewed their efforts to conceive a child and Hetty subsequently
gave birth to a baby born with a genetic kidney disease.60 The
infant survived for only two and a half years before dying from the
disease.61 Plaintiffs brought a claim seeking damages for the
pecuniary expenses they bore for the care and treatment of their
53
Id. at 808.
Id. at 808-09.
55
Id. at 808.
56
Id. at 809.
57
Id. Hetty Park had already given birth to a baby who died five hours after
birth from a polycystic kidney disease. Id. Based on their history, Hetty Park and
her husband were questioning whether the kidney disease was a geneticallycaused disease. Id.
58
Id.
59
Id.
60
Id.
61
Id.
54
BETH MACROED CORRECTED FINAL 060605.DOC
6/6/2005 1:37 PM
FETAL RIGHTS UNDER NEW YORK LAW
817
infant prior to his death, and for the psychological and physical
injuries suffered by Hetty Park as the result of her child’s having
been born with a genetic kidney disease.62
The Becker court held that, while the parents might have a
valid claim for the pecuniary expenses they endured or would
continue to endure in providing care and treatment for their infants,
there could be no recovery by the plaintiff-mothers for the
psychological injuries they endured from having given birth to
impaired infants, based on the court’s decision in Howard.63
Furthermore, the court held that permitting the plaintiffs to recover
for pecuniary loss but precluding recovery for the emotional
injuries was consistent with the court’s decision in Johnson v.
State, in which the court sustained a cause of action for emotional
harm based on the plaintiff’s having been falsely informed by a
hospital that her mother had died.64 In Johnson, the court
recognized the existence of a duty by the hospital not to issue death
notices in a negligent manner; the breach of this duty entitled the
plaintiff to recover for harmful consequences proximately caused
by the breach, including pecuniary loss and emotional harm caused
by the tortious act.65 In distinguishing Johnson from Becker, the
court restated the Johnson rule, which limited the plaintiff’s
recovery to damages for the “proven harmful consequences
proximately caused by the breach.”66 The court explained that, in
Johnson, the causal nexus between the daughter’s emotional
injuries and the hospital’s breach was clear, but that the “same
cannot be confidently said with respect to the birth of a child, the
conception of which was planned and fully desired by the
parents.”67 While parents may suffer from psychological injuries
due to the birth of their child in an impaired state, the parents may
also “experience a love that even an abnormality cannot fully
dampen.”68 Thus, to assess an amount for emotional damages
62
63
64
65
66
67
68
Id.
Id. at 813.
Id. (citing Johnson v. New York, 334 N.E.2d 590 (N.Y. 1975)).
Id. at 814 (citing Johnson, 334 N.E.2d at 593).
Id.
Id.
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
818
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
would require consideration of the “love” factor in mitigation of
the parents’ emotional injuries.69 The court noted that, unlike
Johnson, Becker required consideration of mitigating factors that
would complicate the calculation of damages for the plaintiffs’
emotional injuries—injuries that would ultimately prove too
speculative and uncertain to be deemed a “proven harmful
consequence proximately caused by the breach” of the defendantdoctors’ duties to the plaintiffs.70
Following Becker, in 1980, the Court of Appeals decided
Vaccaro v. Squibb.71 In that case, the plaintiff-mother, Inez
Vaccaro, was prescribed a hormone by her physician to prevent the
miscarriage of her baby, given that she had previously suffered a
stillbirth and a miscarriage.72 The drug caused Vaccaro’s infant to
be born with neither arms nor legs and with other serious injuries.73
Relevant to damages for emotional distress, the plaintiffs brought a
cause of action against the defendants, the physician and drug
manufacturer, for “damages for the injuries to their nervous
systems and emotional damage, personality changes and extreme
mental anguish occasioned by the birth of their daughter without
limbs and with other serious and permanent injuries and congenital
defects” due to the plaintiff-mother’s having ingested the
dangerous hormone during pregnancy.74 The trial court held that
the facts of Vaccaro were more like the facts in Johnson than
Howard because the plaintiffs in Vaccaro alleged that the infant’s
deformities were the direct result of exposure to a drug
administered to the mother by the mother’s physician during
pregnancy.75 The court noted that this was a direct harm to the
69
Id.
Id. Interestingly, the court here said that the legislature would be a better
body than the judiciary to determine whether emotional damages should be
permitted in cases in which the plaintiffs’ emotional injuries stemming from the
prenatal medical malpractice that led to the birth of their infants in an impaired
state might be mitigated by their love for the child. Id.
71
Vaccaro v. Squibb, 418 N.E.2d 386 (N.Y. 1980).
72
Vaccaro v. Squibb, 412 N.Y.S.2d 722, 723 (N.Y. Sup. Ct. 1978).
73
Id.
74
Id. at 724.
75
Id. at 730.
70
BETH MACROED CORRECTED FINAL 060605.DOC
6/6/2005 1:37 PM
FETAL RIGHTS UNDER NEW YORK LAW
819
mother and father caused by the breach of a duty owed by the
defendants to the parents.76 Furthermore, the plaintiff-mother
actually ingested the hormone, and thus, there was a triable issue
of fact as to whether the psychological damages were the natural
consequences of the wrongful act.77 Thus, the trial court denied the
defendants’ motions to dismiss the claims for emotional
damages.78
The appellate court affirmed the trial court’s holding in
Vaccaro,79 in part holding that Vaccaro was more like Johnson
than Tobin or Howard.80 The court explained that, in Vaccaro, a
duty of care was owed by the doctor-defendant to the mother who
ingested the hormone, as it was owed to the daughter falsely
informed of her mother’s death in Johnson.81 This duty derived
from the fact that the defendant-doctor was the mother’s physician,
knew of her prior stillbirth and miscarriage, and selected and
administered the hormone said to be responsible for the infant’s
deformities.82 Unlike Tobin, the plaintiff-mother in Vaccaro was
not a bystander; rather, the mother herself ingested the drug.83
Additionally, in contrast to Howard, in Vaccaro there was
“something more” than the failure to discuss with the parents the
risk of bearing a child with Tay-Sachs syndrome.84 While the
doctor in Howard had committed no affirmative acts or errors, the
doctor in Vaccaro had affirmatively administered to the plaintiffmother a drug that subsequently caused her infant to be born
impaired.85 Thus, the appellate level court held that the mother
could maintain a cause of action for emotional distress, premised
on the theory that she suffered from emotional harm directly
76
77
78
79
80
81
82
83
84
85
Id.
Id.
Id.
Vaccaro v. Squibb, 422 N.Y.S.2d 679, 684 (N.Y. App. Div. 1979).
Id. at 681.
Id. at 681-82.
Id. at 682.
Id.
Id.
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
820
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
caused by the breach of defendant-doctor’s duty to her.86 The
father’s cause of action for emotional distress was dismissed, as he
was not a patient of the doctor, did not ingest the drug, and thus,
was owed no duty, the breach of which would give rise to a
recovery.87
Despite the holdings of both the trial and appellate level courts,
the Court of Appeals held that the plaintiff-mother’s cause of
action for emotional distress could not stand in Vaccaro.88 Citing
Howard and Becker, the majority dismissed the cause of action for
the plaintiff-mother’s emotional injuries because she did not set
forth evidence of any independent injuries.89 This brief but
steadfast application of Howard and Becker would set the stage for
the court’s decision in Tebbutt v. Virostek, the precedent case that
would not be overturned until Broadnax v. Gonzalez, almost
twenty years later.
In 1985, the Court of Appeals decided Tebbutt v. Virostek,90 in
which the alleged negligence of medical care providers directly
resulted in a fetus’s death in utero, although the mother suffered no
physical injuries distinct from the injuries to the fetus.91 In Tebbutt,
the plaintiff’s obstetrician attempted to perform an amniocentesis
three times with no success.92 Prior to the first attempted
amniocentesis, the fetal heart monitor showed the fetus to be viable
86
Id. at 683-84. The court held that “[t]here is a vital interest to be
protected, there is proximate cause, there is demonstrable injury and there is
foreseeability. ‘Thus, the rationale underlying the Tobin case, namely, the real
dangers of extending recovery for harm to others than those directly involved, is
inapplicable to the instant case.’” Id. (citing Johnson v. New York, 334 N.E.2d
590, 593 (N.Y. 1975)).
87
Id. at 684.
88
Vaccaro, 418 N.E.2d at 386.
89
Id.
90
Tebbutt v. Virostek, 483 N.E.2d 1142 (N.Y. 1985).
91
Id. at 1143. See also Broadnax v. Gonzalez, 809 N.E.2d 645, 647-48
(N.Y. 2004).
92
Tebbutt v. Virostek, 477 N.Y.S.2d 776, 777 (N.Y. App. Div. 1984);
Tebbutt, 483 N.E.2d at 1144 (describing the amniocentesis that Dr. Virostek
administered to Marta Tebbutt) (Jasen, J., dissenting). An amniocentesis is a
procedure in which a syringe punctures the womb in order to draw fluid for
testing. Id.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
821
and of normal size for a sixteen-week gestation.93 More than a
month later, despite reassurances that the fetus was normal, the
plaintiff-mother delivered a stillborn baby, bearing three
hemorrhagic blisters, whose size was consistent with sixteen-week
gestation.94 Doctors concluded that it was possible that the failed
amniocentesis attempts caused the fetal death.95 In her claim to
recover for “pain, severe disappointment, anxiety, despondency,
bitterness, and suffering,”96 the plaintiff alleged no physical
injuries apart from those suffered by the fetus.97
The majority in Tebbutt rejected the plaintiff’s claim for
damages for emotional distress.98 The court succinctly explained
that the plaintiff’s claims for emotional damages must be denied
based on Vaccaro v. Squibb, which held that damages for
emotional distress may not be recovered by the parents of children
who are injured in utero but born alive.99 In Vaccaro, the court
“rejected the contention that the defendants owed any duty to the
mother” where the harm done to the child in utero was not
discovered until the birth of the child, some time after the damage
was done.100 Based on the logic of Vaccaro, the Tebbutt majority
rejected the mother’s claim for emotional distress damages.101
While the majority declared that the plaintiff-mother was not
owed a duty by her doctors, the dissent in Tebbutt expressed
considerable concern about the consequences of precluding
emotional distress claims by mothers of fetuses negligently killed
93
Tebbutt, 483 N.E.2d at 1144.
Id. at 1145.
95
Id.
96
Id. at 1143, 1145.
97
Id. at 1143, 1145.
98
Id. at 1143.
99
Id. at 1144 (citing Vaccaro v. Squibb, 418 N.E.2d 386 (N.Y. 1980)).
100
Id. at 1143 (citing Vaccaro v. Squibb, 418 N.E.2d 386 (N.Y. 1980)).
101
Id. at 1143-44. Interestingly, in Vaccaro v. Squibb, 418 N.E.2d 386, 387
n.* (N.Y. 1980), the dissenting judge reflected on the “stultifying effect of what
may be too indiscriminating an application of stare decisis.” In his dissent, the
judge stated that the defendants owed a duty directly to the mother as the patient
of the doctor and the consumer of the implicated drug (the patient had ingested a
prescription drug that caused deformities in her baby). Vaccaro, 418 N.E.2d at
387.
94
BETH MACROED CORRECTED FINAL 060605.DOC
822
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
in utero.102 The dissent explained that the majority had created a
“juridical limbo,” in which a physician’s negligent acts resulting in
the death of an unborn child would be “neither compensated nor
deterred.”103 The dissent concluded that a child killed in utero has
no rights under the law for two reasons: (1) for purposes of the
wrongful death statute, the stillborn child is not considered a
“person” who is owed a duty of care;104 and (2) under the Tebbutt
majority’s rationale, for the purposes of a personal injury action,
the stillborn child is not owed a duty of care.105 Under the
majority’s analysis, if the child in the case were born alive, a
remedy would exist;106 however, if the child were more seriously
injured, resulting in the child’s death, the loss would go
unredressed.107 The dissent thus concluded that “[w]here the law
declares that the stillborn child is not a person who can bring suit,
then it must follow in the eyes of the law that any injury here was
done to the mother.”108 According to the dissent’s logic, the mother
should have been able to bring a claim of emotional distress
resulting from the stillbirth of her child.
C. Broadnax v. Gonzalez
Tebbutt provided the New York courts with a precedent that
was strictly adhered to for nearly twenty years until two cases—
Broadnax v. Gonzalez109 and Fahey v. Canino110—percolated up
through the courts. At the trial level, the plaintiff-mothers sought
damages for emotional distress from their prenatal medical
102
Tebbutt, 483 N.E.2d at 1145 (Jasen, J., dissenting).
Id. at 1144.
104
Id. at 1148 (Jasen, J., dissenting).
105
Id.
106
Id. at 1149 (citing Woods v. Lancet, 102 N.E.2d 691 (N.Y. 1951) (Kaye,
J., dissenting)).
107
Id.
108
Id.
109
Broadnax v. Gonzalez, 759 N.Y.S.2d 499 (N.Y. App. Div. 2003), rev’d,
809 N.E.2d 645 (N.Y. 2004).
110
Fahey v. Canino, 758 N.Y.S.2d 708 (N.Y. App. Div. 2003) rev’d sub
nom. Broadnax v. Gonzalez, 809 N.E.2d 645 (N.Y. 2004).
103
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
823
caregivers based on malpractice that resulted in the deaths of the
fetuses carried by the mothers.111 In both cases, the courts granted
the defendants’ motions for summary judgment, holding that the
plaintiff-mothers could not recover for emotional or psychological
injuries stemming from malpractice resulting in the death of an
unborn child.112 Subsequently, the plaintiffs in both cases
appealed; however, in both cases, the appellate court affirmed the
judgments of the trial court.113
In the early hours of September 24, 1994, Karen Broadnax,
pregnant and almost due to give birth, called her midwife to say
that her water had broken and that she was expelling blood.114 The
midwife told Karen and her husband to come to the birthing center,
but when Karen arrived just over an hour later, she was still
experiencing vaginal bleeding.115 The midwife consulted Karen’s
obstetrician, Dr. Gonzalez, who requested that Karen be
transferred to a hospital.116 Approximately forty-five minutes later,
Karen, her husband, and the midwife arrived at the hospital.117
Although Karen’s obstetrician still had not arrived, the midwife
failed to call or consult the on-call doctor at the hospital.118 When
Dr. Gonzalez arrived two hours later, the fetal heart rate had
already decelerated.119 However, instead of performing an
emergency cesarean section, Dr. Gonzalez conducted a number of
tests, including a vaginal and pelvic examination and a
sonogram.120 Half an hour later, Karen delivered a full-term
stillborn baby by cesarean section who, according to the autopsy,
had died from a placental abruption.121
111
Broadnax, 759 N.Y.S.2d at 500; Fahey v. Canino, No. 40038(U), slip
op. at 2 (N.Y. Sup. Ct. Mar. 5, 2002).
112
Broadnax, 759 N.Y.S.2d at 500; Fahey, slip op. at 5.
113
Broadnax, 759 N.Y.S.2d at 500; Fahey, 758 N.Y.S.2d at 710.
114
Broadnax v. Gonzalez, 809 N.E.2d 645, 646 (N.Y. 2004).
115
Id.
116
Id.
117
Id.
118
Id.
119
Id.
120
Id.
121
Id. at 647.
BETH MACROED CORRECTED FINAL 060605.DOC
824
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
Karen and her husband brought an action against Karen’s
obstetrician, the midwife, and the hospital to recover damages for
emotional distress resulting from the stillbirth of the baby.122 The
appellate level court held that Tebbutt v. Virostek precluded
mothers from “recovering damages for emotional or psychological
harm stemming from the stillbirth . . . [unless they had] suffered a
legally cognizable physical injury distinct from the fetus’s.”123
Given that Karen Broadnax failed to produce evidence of an
independent injury apart from those normally incident to
childbirth, the trial court granted the defendants’ motion for
summary judgment.124 The appellate court affirmed the
judgment.125 Applying Vaccaro and Tebbutt, the court rejected the
possibility of recovering emotional damages for a negligently
caused stillbirth, noting:
There is an absence of evidence that the plaintiff mother
suffered a physical injury distinct from the injury to her
unborn child and separate and apart from that which occurs
in any normal childbirth. Thus, she may not recover
damages for the psychological and emotional harm she
allegedly suffered as a result of the stillbirth of her child.126
Debra Ann Fahey and her husband experienced a loss similar
to that of the Broadnaxes. In August 1999, Debra Ann was told by
her obstetrician, Dr. Canino, that she was carrying twins.127 Two
months later at a regular checkup, Debra Ann informed Dr.
Canino’s partner, Dr. Ruggiero, that she was experiencing lower
back pain and cramping.128 Dr. Ruggiero performed an ultrasound
and concluded that one of the twins was pressed against Debra
Ann’s sciatic nerve, and that this was the source of her pain.129
Two days later, Debra Ann experienced increasingly intense pain
122
123
124
125
126
127
128
129
Id.
Id.
Broadnax, 759 N.Y.S.2d at 500.
Id.
Id.
Broadnax, 809 N.E.2d at 647.
Id.
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
825
and nausea.130 She called Dr. Canino who, relying on Dr.
Ruggiero’s examination, suggested to Debra Ann the pain was
related to the sciatic nerve and that the nausea was probably related
to something she ate for lunch.131 Dr. Canino simply told Debra
Ann to lie down.132 While sitting on the toilet two hours later,
Debra Ann tragically gave birth to one of the twins.133 With the
umbilical cord from the first fetus still attached to her body, Debra
Ann was transported by ambulance to the hospital, where she
delivered the second twin.134 Neither twin lived.135 Debra Ann was
later diagnosed with an “incompetent cervix,”136 a problem that is
detectable by ultrasound and can be remedied with a surgical
procedure.137
The plaintiffs, Debra Ann and her husband, commenced a
medical malpractice action against the defendant-doctors for the
emotional distress caused by Debra Ann’s loss of the twins,
130
Id.
Id.
132
Id.
133
Broadnax, 809 N.E.2d at 647.
134
Id.
135
Id.
136
Id. An incompetent cervix is a weakened cervix that predisposes a
woman to mid-term miscarriage or early (premature) delivery. Special Care
Pregnancies: Incompetent Cervix, University of Pennsylvania Health System, at
http://www.pennhealth.com/health_info/pregnancy/specialcare/articles/cervix/%
20html (last visited Apr. 20, 2005) [hereinafter Special Care Pregnancies].
137
If a doctor suspects that a woman might have an incompetent cervix, she
can perform an ultrasound early in the pregnancy to examine the thickness of the
cervical tissue. Special Care Pregnancies, supra note 136. A surgical procedure
can successfully treat an incompetent cervix eighty-five percent to ninety
percent of the time. Id. This procedure, called cerclage, is usually performed
when the patient is under spinal or epidural anesthesia and involves closing the
cervix with strong stitches for the full term of the pregnancy. Id. After having
cerclage, a woman is usually prescribed medication to help prevent miscarriage.
Id. The stitches are removed around the ninth month of pregnancy or sooner if
labor commences, to prepare for delivery. Id. In a later pregnancy, Debra Ann
Fahey was able to undergo a cerclage procedure to prevent her from delivering
the fetus prematurely, and she was able to carry her pregnancy until the baby
was healthy enough to survive (although the baby was born six weeks
premature). Broadnax, 809 N.E.2d at 647.
131
BETH MACROED CORRECTED FINAL 060605.DOC
826
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
arguing that the extra operations she had endured as a result of the
negligently monitored labor and delivery constituted a “physical
injury.”138 The court granted the defendants’ motion for summary
judgment,139 finding the plaintiff-mother’s “physical injury”
argument without merit140 and her claim for emotional damages
based on personal injury unsupported by her testimony.141
Moreover, the trial court held that so long as Tebbutt provided the
legal framework for prenatal cases, a mother could not recover for
emotional damages resulting from a negligently caused stillbirth
absent proof of her suffering an independent injury.142 Indeed, the
court found “the more reasonable rule to be that which precludes
recovery, not only for the emotional suffering resulting indirectly
from the loss or impairment of the fetus or baby, but also for ‘the
more immediate emotional harm attendant to the mother’s
enduring a negligently caused stillbirth.’” The appellate court
affirmed the trial court’s decision.143
In April 2004, the New York Court of Appeals reversed the
lower courts’ orders granting the defendants’ motions for summary
138
Fahey, slip op. at 2. Plaintiff’s arguments were in response to
defendants’ motion to dismiss on the grounds that a plaintiff may not recover for
emotional distress resulting from a negligently caused stillbirth in the “absence
of any independent, causative physical injury to her own person.” Id.
139
Id. at 5.
140
Id. at 2.
141
Id. at 3. The court held:
[P]laintiff’s testimony demonstrates that her primary concern was over
the plight of the babies, [sic] and their condition . . . there is simply
nothing in the record to support a finding that plaintiff suffered any . . .
psychological trauma as a result of her own condition or experiences,
separable from the distress she felt because of the condition or death of
the fetuses.”
Id. (citations omitted).
142
Id. at 5.
143
Fahey, 758 N.Y.S.2d at 710 (holding that the plaintiffs failed to present
evidence that the mother was independently injured beyond those injuries
naturally caused during childbirth, and as such, the cause of action for emotional
distress was properly dismissed because recovery for psychological damages
resulting from the stillbirth was precluded “in view of the present status of the
law”).
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
827
judgment in both Broadnax and Fahey.144 The court addressed the
issue of when, if ever, a mother could recover damages for
emotional distress resulting from a miscarriage or stillbirth caused
by medical malpractice absent a showing of independent physical
injury to the mother.145 The court recognized that precedent
strongly disfavored claims for emotional distress in cases in which
the plaintiffs had suffered no independent physical injuries.146
However, the court noted that applying a strict interpretation of
this rule in cases of negligently caused stillbirths creates a “logical
gap in which the fetus is consigned to a state of ‘juridical
limbo.’”147 Essentially, infants who were injured in utero but
survived could maintain a cause of action for medical malpractice
against tortfeasors after they were born.148 Furthermore, a pregnant
mother could bring a cause of action for her independent
injuries.149 However, neither party had a cause of action if medical
malpractice had caused the pregnancy to terminate in miscarriage
or stillbirth and the mother was not physically injured beyond the
pain and suffering naturally attendant to childbirth.150 The gap
created by precedent resulted in an uncomfortable dichotomy:
medical caregivers faced liability for injuries to fetuses that
survived, but faced no liability for injuries to fetuses that died in
utero.151
In Broadnax, the defendants argued against the permissibility
of claims for emotional damages resulting from the wrongful death
of a fetus, grounding their challenge in the fact that the defendants’
144
Broadnax, 809 N.E.2d at 647.
Id. at 646.
146
Id. at 648.
147
Id. (citing Tebbutt v. Virostek, 483 N.E.2d 1142, 1144 (Jasen, J.,
dissenting)).
148
Id.
149
Id.
150
Id. In both Appellate Division decisions, the courts held that the
procedures incident to childbirth, miscarriage, or stillbirth are not considered
independent physical injuries to the mother, and thus, do not allow for a cause of
action for emotional distress. See Broadnax, N.Y.S.2d at 500; Fahey, 758
N.Y.S.2d at 710.
151
Broadnax, 809 N.E.2d at 648.
145
BETH MACROED CORRECTED FINAL 060605.DOC
828
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
actions, negligent or otherwise, did not violate a duty to the
expectant mothers; rather, the alleged conduct injured only the
fetuses.152 The court dismissed this argument as “tortured”
reasoning.153 The court explained that, given that prenatal medical
providers owe a duty of care to the developing fetus, the providers
would naturally owe a corresponding duty of care to the mother,
who is the primary patient during the entire pregnancy.154 The
court determined the health of the mother and the fetus to be linked
in the unique situation of pregnancy, but in the same breath,
clarified that the fetus and the mother are each owed a duty of
care.155 Thus, in overturning nearly twenty years of precedent, the
court held that, “even in the absence of an independent injury,
medical malpractice resulting in miscarriage or stillbirth should be
construed as a violation of a duty of care to the expectant mother,
entitling her to damages for emotional distress.”156
Broadnax was decided 6-1 in favor of the plaintiff-mothers.157
In the only dissenting opinion, Judge Reed posed the possibility
152
Id.
Id. The “[d]efendants [argue that] their alleged conduct injured only the
fetuses, and, accordingly, they did not violate a duty to the expectant mothers.
Defendants’ reasoning is tortured.” Id.
154
Id. (citing Woods v. Lancet, 102 N.E.2d 691 (1951)). In Woods v.
Lancet, the plaintiff-infant sustained serious injuries through the negligent
actions of his mother’s physician, such that he was born permanently impaired
and disabled. Woods, 102 N.E.2d at 691-92. The court held that the infant,
injured in utero and later born alive, had the right to maintain an action for the
alleged negligence causing such injury. Woods, 102 N.E.2d at 695.
155
Broadnax, 809 N.E.2d at 648-49. Cf. Tebbutt v. Virostek, 483 N.E.2d
1142, 1146 (N.Y. 1985) (Jasen, J., dissenting). In his dissent, Judge Jensen
noted:
The interests of the mother and the unborn child are intertwined by
nature during the mother’s pregnancy. Due to these relationships, a
tortious act, which results in the death of an unborn child, represents a
breach of a direct duty to the mother. Defendant’s infringement upon
the mother’s freedom from mental distress was occasioned by the
breach of a distinct and independent duty flowing to the mother.
Id.
156
Broadnax, 809 N.E.2d at 649.
157
Id.
153
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
829
that the decision might expose medical caregivers to additional
liability.158 Of great concern to her was that there was no way, at
the time of the decision, to assess or predict the potential effect of
increased liability on the availability and cost of gynecological and
obstetrical care in New York.159
II. BEYOND BROADNAX
In its indication that a fetus enjoys a legal status independent of
the mother, Broadnax v. Gonzalez stands in stark contrast to New
York’s past jurisprudence regarding actions for the wrongful death
of a fetus. In 1969, the New York Court of Appeals affirmatively
stated in Endresz v. Friedberg that actions for the wrongful death
of a fetus are barred in New York.160 In Endresz, the court
discussed at length the reasons for denying the survivors to fetuses
negligently killed by medical malpractice in utero the right to sue
the physician tortfeasors.161 Broadnax, however—in holding that
the fetus is a separate being, that the fetus need not be born to have
rights, and that the difficulty of calculating damages is not a
justification for barring wrongful death suits—may render the logic
of Endresz invalid. In so doing, Broadnax possesses the potential
to work a significant change in the law regarding fetal rights. The
precise impact of Broadnax is as yet unclear; however, in testing
the boundaries of this new precedent, plaintiffs and the lower
courts may compel the reevaluation of the recovery bar for actions
158
Broadnax, 809 N.E.2d at 650; See supra note 12.
Broadnax, 809 N.E.2d at 650.
160
Endresz v. Friedberg, 248 N.E.2d 901, 902, 907 (N.Y. 1969). The
Endresz court upheld the lower court’s dismissal of the plaintiff’s wrongful
death suits and claims for loss of services of the infants, holding that
the fairest and most practical solution . . . the one most in accord with
the dictates of justice, public policy and common sense . . . [is] to leave
the parents of a stillborn fetus, whose death has been caused by a third
party’s wrongful act, to the damages recoverable by them in their own
right and to deny to the distributees any redress by way of a wrongful
death action.
159
Id. at 907.
161
Endresz, 248 N.E.2d at 903-05.
BETH MACROED CORRECTED FINAL 060605.DOC
830
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
grounded in the wrongful death of a fetus. Indeed, only through
clarification by the New York Court of Appeals or the New York
State legislature will stability and predictability come to the law of
torts related to fetal rights.
A. The Unraveling of Endresz: Broadnax’s Implications for
Wrongful Death Law
An action for wrongful death is a lawsuit brought by the
survivors of a decedent whose death resulted from a defendant’s
negligent or wrongful act.162 A majority of jurisdictions today,
with the exception of New York, recognize that a cause of action
lies for the negligently caused death of an unborn child.163 The old
rule,164 which barred actions for the wrongful death of an unborn
child, held that the fetus was not a person for whom recovery could
be made under wrongful death, as the fetus was part of the mother
at the time of the injury.165 This rule was abandoned by most states
162
See supra note 11 and accompanying text. See also 12 AM. JUR. Trials §
317 (2004).
163
James M. Simpson, Growing Recognition of Wrongful Death for
Unborn Children, ADVOCACY, PRACTICE AND PROCEDURES (Int’l Ass’n of Def.
Counsel, Aug. 2001), reprinted in DEFENSE COUNSEL JOURNAL, Oct. 1, 2001, at
487. The following U.S. jurisdictions either explicitly or implicitly recognize a
wrongful death action for the death of an unborn child by statute, state case law,
or federal case law: Alabama, Arkansas, Arizona, Colorado, Connecticut,
District of Columbia, Delaware, Georgia, Hawaii, Idaho, Illinois, Indiana,
Kansas, Kentucky, Louisiana, Maryland, Massachusetts, Michigan, Minnesota,
Mississippi, Missouri, Montana, North Carolina, North Dakota, Nevada, New
Hampshire, New Mexico, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode
Island, South Carolina, South Dakota, Utah, Vermont, Washington, Wisconsin,
and West Virginia. Id.
164
Simpson, supra note 163 (citing Dietrich v. Northampton, 52 Am. Rep.
242, 138 Mass. 14 (1884)). The rule barring actions for the wrongful death of an
unborn child was promulgated by Oliver Wendell Holmes in Dietrich v.
Inhabitants of Northampton in 1884. Id.
165
Simpson, supra note 163 (citing Dietrich v. Northampton, 52 Am. Rep.
242138 Mass. 14 (1884)). This rule is no longer applicable in jurisdictions that
have concluded that unborn children are “persons” within the meaning of the
wrongful death statutes. Simpson, supra note 163. For example, in 2001, the
Supreme Court of Arkansas was asked to reconsider its position in Chatelain v.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
831
in part because advances in medicine “fundamentally changed the
way the modern mind conceptualizes ‘separateness’ between
mother and child.”166 Nonetheless, New York courts maintain a bar
against actions based on the wrongful death of a fetus dating from
the New York Court of Appeals’s decision in Endresz, which
foreclosed this avenue of relief.167
In Endresz, a pregnant woman was injured in a car accident
negligently caused by another driver.168 The injuries caused to her
and her twin fetuses resulted in the stillbirth of both babies.169 The
plaintiff-parents sued on behalf of the unborn twins for wrongful
death.170 In dismissing the plaintiffs’ claims for wrongful death,
the court held that when an unborn child is injured through the
wrongful act of a defendant, “liability attaches only upon
fulfillment of the condition that the child be born alive.”171 The
court thus concluded that there was no right of recovery under
New York law by the personal representative of a stillborn fetus
that had died as a result of injuries received while in utero.172
Importantly, although Broadnax expressly declared that there
was no right of recovery for the wrongful death of a fetus,173 the
Kelley, 910 S.W.2d 215, 219 (Ark. 1995), that a viable fetus was not a “person”
within the meaning of the wrongful death statute. Id. In Arkansas v. Jefferson
Hospital Ass’n, 42 S.W.3d 508, 518 (Ark. 2001), the Court noted that their
holding that a fetus was not a person was based on the Arkansas legislature’s
former characterization of the word “person,” which had changed in the context
of criminal law since Chatelain to include unborn children. Simpson, supra note
163. The court noted that the holding in Chatelain that a fetus was not a person
was seriously undermined by legislative change. Id. After Jefferson Hospital,
the Arkansas legislature amended its wrongful death statute to include unborn
children. Id.
166
Simpson, supra note 163.
167
Endresz v. Friedberg, 248 N.E.2d 901 (N.Y. 1969).
168
Id. at 902.
169
Id.
170
Id.
171
Id. at 905.
172
Id. at 907 (denying the distributees of the fetus a cause of action for the
fetus’s wrongful death).
173
Broadnax, 809 N.E.2d at 649 n.4. In footnote 4 of the majority opinion,
the court noted that in rejecting Tebbutt, it also recognized that a majority of
jurisdictions permit some form of recovery for negligently caused stillbirths or
BETH MACROED CORRECTED FINAL 060605.DOC
832
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
case nonetheless challenges the logic of the arguments asserted in
Endresz in support of a prohibition on such actions.
The first reason advanced in Endresz for barring a cause of
action for the wrongful death of a fetus is that an unborn child is
not a decedent under the wrongful death statute.174 The majority
opinion observed that the law in New York has declined to
attribute the unborn fetus a “legal personality or identity ‘until it
sees the light of day,’”175 and thus, a fetus killed as a result of
medical malpractice could not be legally termed “deceased.”176
However, the Court of Appeals acknowledged in Broadnax that a
mother has a right to sue for emotional damages caused when
medical malpractice results in a miscarriage or stillbirth precisely
because the infant has no such right.177 Additionally, the court
deemed the fetus to be owed a duty of care independent of the
mother.178 By acknowledging that a fetus is owed an independent
duty of care, the Broadnax decision suggests that an unborn fetus
has a legal personality or identity.179 If this is true, then an unborn
miscarriages. The court then proceeded to specifically limit a mother’s recovery
to damages for the emotional distress attending the stillbirth or miscarriage
caused by medical malpractice, and affirmed the holding in Endresz v.
Friedberg barring wrongful death actions under the circumstances of medical
malpractice resulting in stillbirth or miscarriage. Id. The court gives no
reasoning for this statement, other than the implied reasoning of stare decisis in
saying, “[w]e do not depart from our holding in Endresz.” Id.
174
Endresz, 248 N.E.2d at 903. However, the majority does admit that the
statute is silent on this matter. Id. The majority interpreted the legislative intent
to not have included unborn children within the meaning of the wrongful death
statute based on the case law at the time the Decedent Estate law was written in
1847. Id. The Decedent Estate Law became, without major changes, Section 54.1 of the EPTL. Id.; see also N.Y. EST. POWERS & TRUSTS LAW § 5-4.1 (2004).
175
Endresz, 485 N.E.2d at 904.
176
Endresz, 485 N.E.2d at 905 (holding that a “conditional prospective
liability” is created when a fetus is injured through the wrongful acts of the
defendant, and as such, liability for those wrongful acts attaches only if the child
is later born alive).
177
Broadnax, 809 N.E.2d at 648.
178
Id.
179
In Endresz, the court noted that the law had never considered an unborn
fetus as having a separate “juridical existence” or a legal personality or identity
unless it was later born, as part of its reasoning that a fetus did not fall within the
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
833
child would fall within New York’s statutory definition of
“deceased.”180 Thus, in light of the court’s holding in Broadnax,
the first reason advanced in Endresz for denying a cause of action
in wrongful death for an unborn fetus is no longer viable.
The second reason articulated for barring causes of action in
wrongful death for infants killed in utero is that a deprivation of
life should not be actionable unless there has first been a birth.181
Yet, the wrongful death statute is designed to compensate the
decedent’s estate for the loss caused by the decedent’s death.182
Given that the Broadnax court found it illogical to permit doctors
to evade liability when their negligence results in the stillbirth or
miscarriage of a fetus,183 it makes little sense to preclude recovery
for wrongful death when a fetus dies in utero. In both cases, the
fetus dies as a result of the negligence of a third party before it is
born, and in both cases, a loss is occasioned by the death of the
decedent.184 If the logic flowing from Broadnax is extended,185 the
parents, as representatives of the decedent (the unborn fetus), must
be permitted a cause of action for wrongful death simply because
the fetus was deprived of life in the first place.
meaning of “person” for the purposes of the wrongful death statute. Endresz,
248 N.E.2d at 904 (citations omitted). However, now the court in Broadnax has
acknowledged that the fetus is owed a duty of care separate from the expectant
mother, in addition to the duty of care owed to the mother. Broadnax, 809
N.E.2d at 648. Thus, through the decision in Broadnax, the law may now
consider an unborn fetus as having a separate “juridical existence” even when
the fetus does not survive through birth.
180
N.Y. EST. POWERS & TRUSTS LAW § 5-4.1 (2004).
181
Endresz, 248 N.E.2d at 903.
182
Weisberg v. Layne-New York Co., 517 N.Y.S.2d 304 (N.Y. App. Div.
1987) “The policy underlying [New York’s wrongful death statute] is . . . to
compensate the decedent’s estate for loss suffered by his death.” Id. at 306.
183
Broadnax, 809 N.E.2d at 648.
184
The commentary to New York’s Pattern Jury Instructions for wrongful
death actions states that “[i]n order to establish a right to a wrongful death
recovery, the plaintiff need only show that he has a reasonable expectation of
support from the decedent and therefore a pecuniary loss.” 1 N.Y. P.J.I.3d §
2:320, at 1565 (2005).
185
The logic flowing from Broadnax is that the mother must be permitted
to bring a cause of action for injury because the fetus itself cannot bring suit.
Broadnax, 809 N.E.2d at 648.
BETH MACROED CORRECTED FINAL 060605.DOC
834
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
The Broadnax court noted that had the fetus been born, as an
infant it could have sued for its injuries.186 The court also held that
since the fetus was not born, it must be the mother who was
injured; thus, the court conferred upon the mother the right to bring
suit simply because the fetus could not.187 The court’s argument
essentially eliminates the birth requirement for the wrongful death
statute. The wrongful death statute states as a condition of suit that
the cause of action must be one that could have been sued upon
had death not ensued.188 Had the fetus been born but injured in the
womb and survived until at least birth, it could have sued for the
negligent injuries it sustained in the womb.189 However, as the
second prong of Broadnax explains, since those injuries killed the
fetus, the survivors of the fetus must have the right to bring suit
simply because the fetus was could not.190 The very fact that the
fetus was not born, but could have been born, mandates the
existence of a cause of action for wrongful death on behalf of the
fetus’s survivors. Thus, to fulfill the policy reasons behind the
wrongful death statute, the representatives of the unborn fetus must
have a cause of action for wrongful death.
The Endresz court supported its second reason for precluding a
cause of action in wrongful death by stating that “considerations of
justice which mandate the recovery of damages by an infant,
injured in his mother’s womb and born deformed through the
wrong of the third party, are absent where the foetus, deprived of
life while yet unborn, is never faced with the prospect of impaired
mental or physical health.”191 However, as noted, Broadnax
expressly acknowledged that the child in utero is owed a duty of
care by the medical professional treating the expectant mother’s
pregnancy.192 The Broadnax court impliedly held that consigning
the unborn fetus to a state in which it has no rights is an injustice in
186
187
188
189
190
191
192
Id.
Id.
Endresz, 248 N.E.2d at 908 (Burke, J., dissenting).
Broadnax, 809 N.E.2d at 648.
Id.
Endresz, 248 N.E.2d at 903.
Broadnax, 809 N.E.2d at 648.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
835
itself193 that should be resolved by allowing the mother a cause of
action for injuries if the fetus cannot bring suit.194 Thus,
considerations of justice necessitate the recovery of damages by
representatives of the fetus injured and killed in utero simply
because that fetus never had the chance to bring suit in the first
place.
In his dissenting opinion in Endresz, Judge Burke dismissed
the majority’s reliance on birth as a prerequisite for an action in
wrongful death as illogical.195 First, citing language from Woods v.
Lancet, Judge Burke declared that the majority’s reasoning was an
“outmoded, timeworn fiction.”196 He proposed that life, not birth,
should be the criteria by which the court reviews causes of action
in wrongful death.197 Second, Judge Burke relied on an analogous
Wisconsin case, from which he quoted:
If no right of action is allowed, there is a wrong inflicted
for which there is no remedy. Denying a right of action for
negligent acts which produce a stillbirth leads to very
incongruous results. For example, a doctor or midwife
whose negligent acts in delivering a baby produced the
baby’s death would be legally immune from a lawsuit.
However, if they badly injured the child they would be
exposed to liability. Such a rule would produce the absurd
result that an unborn child who was badly injured by the
tortious acts of another, but who was born alive, could
recover while an unborn child, who was more severely
injured and died as the result of the tortious act of another,
could recover nothing.198
Judge Burke’s criticism of this inconsistency is similar to that
raised by the dissent in Tebbutt—that the practitioner who caused a
more serious injury resulting in death would face less liability than
193
Akin to the aforementioned “juridical limbo,” mentioned supra note 19
and accompanying text.
194
Broadnax, 809 N.E.2d at 648.
195
Endresz, 248 N.E.2d at 908 (Burke, J., dissenting).
196
Id. (citing Woods v. Lancet, 102 N.E.2d 691 (1951)).
197
Id. at 908.
198
Id. (citing Kwaterski v. State Farm Mut. Auto. Ins. Co., 34 Wis. 2d 14,
20).
BETH MACROED CORRECTED FINAL 060605.DOC
836
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
the practitioner who caused a more minor injury that the infant
survived.199 The persuasiveness of Judge Burke’s dissent is given
even more support by the Broadnax court’s express approval of the
judge’s reasoning in its discussion of the rationale for permitting a
cause of action by the mother for emotional damages on behalf of
the fetus.200 Given this logic, it is likely that the second reason
cited in Endresz against permitting a cause of action for wrongful
death, specifically that there has been no birth, also has been
impliedly overruled by Broadnax.
The third reason advanced for prohibiting recovery for the
wrongful death of fetuses negligently killed in utero is that
damages for such injuries are difficult to calculate.201 In addressing
this concern, Judge Burke, in his dissent in Endresz, argued that
the difficulty of calculating damages should not preclude
substantive recovery.202 Specifically, Judge Burke noted that the
majority’s reliance on the argument that causation and damages are
too difficult to calculate had been effectively dismissed by the
court in Woods v. Lancet.203 Indeed, the majority in Woods
asserted that “it is an inadmissible concept that uncertainty of proof
can ever destroy a legal right.”204 Judge Burke noted that this
portion of the Woods holding was cited approvingly by a Kentucky
state court in its refusal to dismiss a cause of action for the
wrongful death of a stillborn fetus based solely on the difficulty of
estimating damages.205 Judge Burke thus concluded that the
199
Tebbutt, 483 N.E.2d at 1147-49.
Broadnax, 809 N.E.2d at 648 (holding that Tebbutt wrongly “exposed
medical caregivers to malpractice liability for in utero injuries when the fetus
survived, but immunized them against any liability when their malpractice
caused a miscarriage or stillbirth”).
201
Endresz, 248 N.E.2d at 904 (holding that there are “no elements
whatever upon which a jury could base any conclusion that a pecuniary injury
has been suffered by the plaintiff from the loss of the unborn child”).
202
Id. at 909. Compare Broadnax, 809 N.E.2d at 649, in which the court
dismisses the dissenting judge’s concerns about juries being asked to quantify
the emotional distress that a woman feels upon suffering a miscarriage or
stillbirth.
203
Endresz, 248 N.E.2d at 909 (citing Woods, 303 N.Y. at 356).
204
Endresz, 248 N.E.2d at 909.
205
Id. at 909 (citing Mitchell v. Couch, 285 S.W.2d 901, 906 (Ky. 1955)).
200
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
837
supposed difficulty of calculating damages should not be used to
justify the denial of causes of action for the wrongful death of a
fetus.206 Furthermore, the Broadnax majority was unconcerned that
damages might be difficult to quantify.207 In addressing the
dissent’s concern that juries would be asked to quantify the
emotional distress experienced by a woman who has suffered a
miscarriage or stillbirth, the majority responded that “no one from
any quarter [had come] forward [during the appellate process] to
support any such concerns.”208 It can be inferred from this
statement that unless interested parties voiced concerns regarding
any difficulties in affixing damages, the Broadnax majority would
not consider such concerns sua sponte. Given the court’s sentiment
that the difficulty of affixing damages does not justify adherence to
Tebbutt, it is possible that the court may also find that the difficulty
of affixing damages does not justify adherence to the principle in
Endresz that wrongful death damages are barred in part because it
would be difficult to calculate damages and causation in a claim
for the wrongful death of a fetus.
The fourth reason cited to preclude recovery for wrongful
death by the personal representative of a stillborn fetus is that the
parents would receive an undeserved windfall.209 The Endresz
court noted that, in a given case, a mother could sue for any
independent physical injuries she suffered and the father could sue
for the loss of services, making any award for wrongful death an
“unmerited bounty . . . [as the award] would constitute not
compensation to the injured but punishment to the wrongdoer.”210
However, this argument is undercut by the existence of cases in
which a mother does not suffer any physical injuries from the
The Mitchell court, citing to Woods v. Lancet, 102 N.E.2d 691 (1951), held that
uncertainty of proof, by itself, can never destroy a legal right. Mitchell, 285
S.W.2d at 906. “The questions of causation and reasonable certainty which arise
in these cases are no different in kind from the ones which have arisen in
thousands of other negligence cases decided in this state in the past.” Id.
206
Endresz, 248 N.E.2d at 909.
207
Broadnax, 809 N.E.2d at 649.
208
Id.
209
Endresz, 248 N.E.2d at 904.
210
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
838
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
stillbirth apart from those normally incident to childbirth.211 In
such cases, prior to the Broadnax decision, if the plaintiff-mothers
did not allege any independent physical injuries, they were barred
from asserting causes of action for emotional damages.212 Notably,
Broadnax has removed this bar to recovery.213 Thus, the Broadnax
decision could influence the fourth Endresz factor in one of two
ways: (1) either the court may look to Broadnax as representative
of a current trend to permit greater recovery on behalf of plaintiffparents, thereby rendering the “unmerited bounty” argument
outdated, or (2) the court could decide that because a mother can
now recover for emotional distress, she should not be permitted to
also recover as the representative of the fetus in a cause of action
for wrongful death, given that the combination of the two damage
awards would constitute an “unmerited bounty.”214
If the reasoning in Endresz is outdated and a majority of other
states recognize a cause of action for wrongful death of a fetus,
why then do the New York courts consistently bar wrongful death
actions on behalf of survivors of fetuses negligently killed in
utero? The fundamental reasoning for barring wrongful death
actions may parallel the reasoning cited by the New York state
courts in barring actions for wrongful life—that the issue is one
best addressed by the legislature, not the court.215
An action for wrongful life is “[a] lawsuit brought by or on
behalf of a child with birth defects, alleging that but for the
defendant doctor’s negligent advice, the parents would have not
conceived the child, or if they had, they would have aborted the
fetus to avoid the pain and suffering resulting from the child’s
congenital defects.”216 New York currently does not permit causes
of action for wrongful life.217 While a parent may recover damages
211
See Tebbutt, 483 N.E.2d at 1143; see also Broadnax, 809 N.E.2d at 646.
Broadnax, 809 N.E.2d at 648.
213
Id.
214
Endresz, 248 N.E.2d at 904.
215
See supra note 70 and accompanying text.
216
BLACK’S LAW DICTIONARY 1607 (7th ed. 1999).
217
Sheppard-Mobley v. King, 778 N.Y.S.2d 98, 101 (N.Y. App. Div. 2003)
(holding that “[n]o cause of action may be maintained on behalf of an infant
plaintiff for ‘wrongful life,’ i.e., that he or she would never have been born but
212
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
839
for the increased cost of caring for the child until the age of
majority,218 a child is barred from recovering damages for the
extraordinary expenses that the child will incur upon reaching
majority.219 In the landmark case Becker v. Schwartz, the court
cited two reasons for barring claims for wrongful life.220 First, the
court noted that children who bring wrongful life actions have not
suffered any legally cognizable injuries, there being no
“fundamental right . . . to be born as a whole, functional human
being.”221 Second, the court found that damages would be
impossible to compute, there being no way to provide a remedy
that would place the infants in the place they would have occupied
but for the negligence of the defendants because that place would
have been nonexistence.222 These concerns echo those of the
Endresz court regarding claims for wrongful death, specifically
that an unborn child is not a legally cognizable person in the eyes
of the wrongful death statute and that damages for the wrongful
death of a fetus would be too difficult for a jury to calculate.
In Becker, the majority voiced its discomfort with having to
recognize claims for wrongful life, holding that the court was illequipped to calculate damages based on a comparison between life
in an impaired state and non-existence.223 Indeed, the court noted
that “[r]ecognition of so novel a cause of action . . . is best reserved
for the negligence of the defendants”); see also Sample v. Levada, 779 N.Y.S.2d
96, 99 (N.Y. App. Div. 2004); Becker v. Schwartz, 386 N.E.2d 807, 813 (N.Y.
1978).
218
1 N.Y. P.J.I.3d 2:280, at 1464. “[P]arents may recover the cost of care
and treatment of a disabled child born because of a negligent failure to test for or
advise the parents of the potential for the birth of such a child . . . [t]his recovery
is limited to the extraordinary expenses incurred . . . prior to the child’s 21st
birthday.” Id. (citations omitted).
219
Alquijay by Alquijay v. St. Luke’s-Roosevelt Hosp. Center, 473 N.E.2d
244, 245 (N.Y. 1984) (holding that an infant does not have a cause of action in
wrongful life because he cannot allege any cognizable injury, there being no
right not to be born over being born impaired).
220
Becker, 386 N.E.2d at 812.
221
Id.
222
Id.
223
Becker, 46 N.Y.2d at 412.
BETH MACROED CORRECTED FINAL 060605.DOC
840
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
for legislative, rather than judicial, attention.”224 Although
wrongful death is not a novel cause of action and is recognized in a
majority of states,225 New York continues to reject causes of action
for wrongful death-of-fetus, primarily for reasons that echo its
reasons for rejecting causes of action for wrongful life. Thus,
perhaps as in the case of wrongful life, the cause of action for
wrongful death is also best left to the legislature for a formal
decision. If there is confusion regarding Broadnax’s implications
for wrongful life actions, the legislature could affirmatively
address this issue through an amendment to the wrongful death
statute. State legislatures in South Dakota and Arkansas have
drafted their wrongful death statutes to expressly permit actions on
behalf of survivors of unborn children.226 These codes of these
states could serve as a helpful model if the legislature decides to
properly address the issue of wrongful death actions in New York.
However, just as it has not yet addressed the issue of whether a
cause of action lies for wrongful life, the New York legislature has
stalled in enacting a statute providing that fetuses are persons for
the purpose of the wrongful death statute. In 2003, the New York
State legislature put forward bills in both the State Assembly and
the Senate that, if enacted, would amend the Estates, Powers and
Trusts Law to allow recovery for the wrongful death of a fetus
which dies in the womb through a wrongful act or negligence by a
third party.227 However, these bills have not yet been passed and
remain in the committee stage.228
B. The Application of Broadnax by New York Courts
Broadnax is a fairly recent decision, and thus, New York courts
have not been presented with many occasions in which to apply the
principles articulated in the case. However, three significant lower
224
Id.
Simpson, supra note 163.
226
See S.D. CODIFIED LAWS § 21-5-1 (2004); ARK. CODE ANN. § 16-62102 (2004).
227
S.B. 135, 226th Ann. Leg. Sess., Reg. Sess. (N.Y. 2003); S.A. 5753,
226th Ann. Leg. Sess., Reg. Sess. (N.Y. 2003).
228
Id.
225
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
841
court decisions, Sheppard-Mobley v. King, Shaw v. QC-Medi New
York, Inc., and Stuart v. New York City Health & Hospitals Corp.,
and the very recent Court of Appeals decision to overturn
Sheppard-Mobley v. King, address the Broadnax ruling directly.229
The outcomes of these cases suggest that the courts are struggling
to deal with the expansion of fetal and maternal rights that resulted
from Broadnax. If nothing else, the four decisions indicate some
disagreement among the courts with regard to how best to apply
Broadnax.
In June 2004, in Sheppard-Mobley, the Appellate Division,
Second Department, extended the principles of Broadnax to the
plaintiff-mother’s claim of emotional distress resulting from the
successful birth of a child negligently injured in utero.230 Finding
no reason to limit the Broadnax holding to cases of stillbirth and
miscarriage, the Appellate Division relied on an analysis of
previous Court of Appeals decisions, including Broadnax, Tebbutt,
and Vaccaro, to demonstrate that the court had “repealed the
independent physical injury requirement for all three categories of
birth trauma.”231 First, the court in Sheppard-Mobley noted the
holding in Broadnax that if there is a duty of care owed to the
infant in utero, then surely there is a duty of care owed to the
expectant mother.232 Second, the court held that, in prohibiting a
mother’s recovery for emotional distress damages in the absence of
an independent injury, it had consistently treated the miscarriage,
stillbirth, or live birth of a fetus in an impaired state alike.233 Thus,
miscarriage, stillbirth, and live birth of a fetus in an impaired state
should be treated alike in allowing a mother’s recovery for
emotional distress damages in the absence of an independent
229
Sheppard-Mobley v. King, 778 N.Y.S.2d 98 (N.Y. App. Div. 2004);
Shaw v. QC-Medi New York, 778 N.Y.S.2d 791 (N.Y. App. Div. 2004); Stuart
v. New York City Health & Hospitals Corp., No. 9767/03, slip. op. at 1 (N.Y.
Sup. Ct. Jan. 10, 2005); Sheppard-Mobley v. King, No. 49, 2005 N.Y. LEXIS
1135, (N.Y. App. Div. May 10, 2005).
230
Sheppard-Mobley, 778 N.Y.S.2d at 103-04.
231
Id. at 103.
232
Id.
233
Id.
BETH MACROED CORRECTED FINAL 060605.DOC
842
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
injury.234 Finally, the court noted that if Broadnax overruled
Tebbutt, it should also overrule Vaccaro, which denied damages
for emotional distress to the parents of children injured in utero but
born alive.235 Consequently, the court held that recovery for
emotional damages should be permitted when the defendantdoctor’s negligence results in the live birth of a severely impaired
child.236
In addition to the Second Department’s holding that a mother’s
right to recover for emotional distress under Broadnax extends to
cases involving the live birth of a child in an impaired state,237 in
early 2005, the Queens County Supreme Court held that the
retroactive application of both Broadnax and Sheppard-Mobley
was appropriate.238 The decision in Stuart v. New York City Health
and Hospitals Corp. reflected a turning point in New York
jurisprudence marked by Broadnax and Sheppard-Mobley. The
judge noted that Broadnax and Sheppard-Mobley created a “new
rule” that recognized actions for emotional distress absent physical
injury.239 This rule fulfilled the “commendable purpose” of
expanding the duty of care owed to expectant mothers.240 The
court held that this “commendable purpose” was achieved by
retroactive application, since there was no lawful justification for
the old policy, which did not address a mother’s emotional
wellbeing as dependent on the health of her child.241 Thus, the
Stuart decision reinforced Broadnax and Sheppard-Mobley,
comporting with the “spirit and direction” of the Court of
234
Id.
Id.; see also supra notes 68-87 and accompanying text for a summary of
Vaccaro v. Squibb, 418 N.E.2d 386 (N.Y. 1980)).
236
Sheppard-Mobley, 778 N.Y.S.2d at 103.
237
1 N.Y. P.J.I.3d 2:280, at 1463 (2005); Sheppard-Mobley, 778 N.Y.S.2d
at 103.
238
Stuart, No. 9767/03, slip op. at 3. Retroactive application means that a
change in law will be applied to injured parties that filed lawsuits prior to the
change in law, in that they will be allowed to amend their complaint to include a
cause of action for recovery that the new law permits them. Id.
239
Id.
240
Id.
241
Id.
235
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
843
Appeals’s decisional law in this area.242
From holdings such as the Second Department’s in SheppardMobley and the Queens County Supreme Court’s in Stuart, it
appeared that the New York courts would progressively expand
upon the holding in Broadnax and go to great lengths to permit
recovery for emotional damages when a defendant-doctor’s
negligence resulted in the live birth of a severely impaired child.
However, in May 2005, the Court of Appeals reexamined
Sheppard-Mobley and overturned the Second Department’s
decision, holding that an expectant mother may not recover
damages for emotional injuries when a defendant-doctor’s
negligence causes injury to a fetus that later survives.243 The court
explained that the holding in Broadnax was intended to “fill a gap”
in tort jurisprudence that had exposed doctors to liability for their
negligence when a fetus was born alive, but immunized them when
the fetus died in the womb.244 Further, the court held that the
Broadnax holding had been crafted to eliminate the injustice
created by ignoring a small, but undoubtedly aggrieved, class of
plaintiffs, and that it was this unique injustice that the court sought
to rectify by permitting mothers, even absent an independent
injury, to recover for emotional distress when medical malpractice
resulted in the stillbirth or miscarriage of the fetuses they were
carrying.245 Thus, the Court of Appeals held that the Second
Department had wrongly applied the principles of Broadnax to the
facts in Sheppard-Mobley because, as the court had held many
years earlier in Woods v. Lancet, a child born alive has a cause of
action for the physical injuries it sustained as a fetus through
medical malpractice.246
242
Broadnax v. Gonzalez, 809 N.E.2d 645, 648 (N.Y. 2004). “In
categorically denying recovery to a narrow, but indisputably aggrieved, class of
plaintiffs, Tebbutt is at odds with the spirit and direction of our decisional law in
this area.” Id.
243
Sheppard-Mobley v. King, No. 49, 2005 N.Y. LEXIS 1135, at *1-2 (N.Y.
App. Div. May 10, 2005).
244
Id. at *7.
245
Id. at *7.
246
Id. at *7-8 (citing Woods v. Lancet, 102 N.E.2d 691). It is unclear
whether the recent Court of Appeals’ decision in Sheppard-Mobley will impact
the Queens County Supreme Court’s decision in Stuart. Andrew Harris,
BETH MACROED CORRECTED FINAL 060605.DOC
844
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
The decision by the Court of Appeals to narrow the scope of
Broadnax provides support for an earlier decision by the Appellate
Division, Fourth Department, to narrowly interpret the Broadnax
precedent. Only ten days after Sheppard-Mobley was decided by
the Second Department in June 2004, the Appellate Division,
Fourth Department, in Shaw v. QC-Medi New York, refused to
extend Broadnax to the parents’ claims of emotional distress
absent physical injury to the nonpatient plaintiff mother.247 In that
case, the plaintiffs’ baby was born with severe defects requiring
her to be on a ventilator and to receive twenty-four-hour nursing
care.248 The infant’s mother was diabetic and her condition was
aggravated by stress.249 The nursing staff hired by the plaintiffs
was apprised of the mother’s poor health.250 When one of the
nurses failed to adequately respond to an alarm on the infant’s
ventilator, the plaintiff returned home to find her two-year-old
daughter “sweating profusely, very blue, and barely conscious.”251
The child later recovered, but the mother sued for negligent
infliction of emotional distress, arguing that an independent duty
was owed to her by the defendant nurses because they were “on
notice of her condition and the effect that stress had upon it.”252
The court held that, despite the decision in Broadnax to permit
Expanding ‘Broadnax’; Court of Appeals Soon to Rule On Case Used By Suffolk
Judge to Add Emotional Distress Claim to Neo-Natal Malpractice Suit, N.Y.
L.J., May 10, 2005, at 16 (noting, prior to the publication of the Court of
Appeals decision on Sheppard-Mobley, that if the Court of Appeals overturned
the Second Department’s decision in Sheppard-Mobley, the new holding could
“sweep away” the decision in Stuart as well).
247
Shaw v. QC-Medi New York, 778 N.Y.S.2d 791 (N.Y. App. Div. 2004).
248
Id. The child’s severe defects were not caused by medical malpractice
and are important to the case only in that the defects caused the child to need
twenty-four-hour nursing care and attention.
249
Id.
250
Id. Before the incident in question, the parents in Shaw were frustrated
when their nurses sometimes failed to show up for work. The father wrote a
letter informing the nursing service that his wife’s severe diabetes was being
exacerbated by the stress of the nursing staff’s “lack of professional
commitment,” and that the stress his wife was under was “literally killing her.”
Id.
251
Id. at 792-93.
252
Id. at 793.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
845
recovery for a mother’s emotional distress resulting from
negligence to her fetus, the duty of care owed to a patient-mother
in pregnancy was unique, and thus, the principles of Broadnax
were not applicable to cases in which the mother was a
nonpatient.253 Thus, it appears for now that the principles of
Broadnax may not extend to instances involving a nonpatient
parent.254
It is clear that the New York courts are in a state of transition
regarding their willingness, or unwillingness, to extend the
principles announced in Broadnax to other situations. The Court of
Appeals’s decision in Sheppard-Mobley strongly suggests that the
holding of Broadnax will be applied very narrowly in the future.
However, Sheppard-Mobley only addresses the potential for a
mother to recover emotional damages for prenatal negligence that
resulted in the birth of an impaired baby. Importantly, the holding
of Broadnax remains valid as applied to other cases affecting the
rights of a fetus and the rights of the mother of a fetus negligently
killed. Thus, the precise repercussions of Broadnax for wrongful
death-of-fetus cases still remain to be seen. As additional cases
percolate up through the New York courts on the issue of maternal
and fetal rights as related to medical malpractice suits, the true
scope of Broadnax hopefully will become clearer.
III. EXPANDING MALPRACTICE LIABILITY IN NEW YORK
The recent New York Court of Appeals decision in Broadnax
v. Gonzalez has the potential to either change or altogether
eliminate the current bars to suits for wrongful death in New
York.255 Regardless of whether the Broadnax decision opens the
door to wrongful death litigation, it almost certainly will impact
the field of obstetrical and gynecological care in New York by
expanding medical malpractice liability.256 This concern was
253
Id. at 795.
In other words, only by reason of pregnancy does a mother, absent
independent injury, have a cause of action for emotional damages for negligence
resulting in harm to her child. Shaw, 778 N.Y.S.2d at 795.
255
See supra Part II.
256
Broadnax, 809 N.E.2d at 650. See supra note 12 and accompanying
254
BETH MACROED CORRECTED FINAL 060605.DOC
846
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
highlighted in Judge Reed’s dissent in Broadnax.257 While she
stated that “there is no way . . . to predict . . . the potential effect of
this expansion of liability . . . on the cost and availability of
gynecological and obstetrical services in New York State,”258 it is
a general concern among medical practitioners in New York that
“stifling liability insurance rates could come even closer to
suffocating them” after the Broadnax decision.259 Doctors and
insurance carriers are especially concerned that Broadnax will
result in the filing of an increasing number of lawsuits and, with
“escalating jury awards” and the high costs of defending a lawsuit,
additional lawsuits mean higher liability insurance premiums.260
According to the American Medical Association (AMA), New
York faces a “medical liability insurance crisis that has physicians
retiring early, moving to states where insurance rates are lower and
cutting back on high-risk procedures in an effort to lower insurance
premiums.”261 The AMA reports that New York physicians pay
some of the highest rates of liability insurance in the country, in
the range of up to $200,000 annually.262 Doctors are struggling to
obtain $1 million in malpractice coverage, but jury awards greater
than $1 million are frequent in New York, and the average award
increased from $1.7 million in 1994 to $6 million in 1999.263
Indeed, fear of staggering liability compels many young doctors
not to specialize in obstetrics.264 Further, forty-five percent of the
obstetrical residents who graduated in New York in 2002 have
text.
257
Broadnax, 809 N.E.2d at 650.
Id.
259
Tanya Albert, New York Court Expands Liability in Miscarriage and
Stillborn Cases AMEDNEWS.COM, para. 1 (2004), available at http://www.amaassn.org/amednews/2004/04/26/prsb0426.htm.
260
Adam D. Glassman, The Imposition of Federal Caps in Medical
Malpractice Liability Actions: Will They Cure the Current Crisis in Health
Care, 37 AKRON L. REV. 417, 417-18 (2004).
261
Albert, supra note 259, at para. 7.
262
Hearings, supra note 13.
263
Id.
264
Id. (citing New York Daily News, Feb. 12, 2004).
258
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
847
since left the state to practice elsewhere.265 In response to these and
other statistics, however, it must be noted that the legislative
director of the New York Public Interest Research Group has
asserted that nowhere is there any independent data proving that
doctors and obstetricians are leaving New York.266
It is possible that the fear of liability has led to a decrease in the
number of obstetricians and an increase in the cost of medical
services available in New York State.267 When there are physician
shortages, fewer obstetricians and gynecologists are available for
routine screenings and checkups.268 Consequently, “women lose
care that helps protect fertility, end pelvic pain, or treat cancer
early . . . [women have to] travel longer distances to find a doctor,
have longer waiting periods for appointments, and have shorter
visits once they get there.”269 Increasing medical liability
disproportionately harms pregnant women because they are unable
to get the prenatal and delivery care they need.270 Furthermore,
obstetric shortages disproportionately impact poor and
disadvantaged women.271 These women frequently rely on
community care clinics, which often have to limit the number of
patients they accept because they cannot shift the costs of their
rising insurance premiums to their uninsured patients.272 Medical
care expenses may also increase when doctors, out of fear of
getting sued, practice what is termed “defensive medicine,” where
they order too many, and sometimes needless, medical tests to
265
Id. (citing Long Island Business News, Mar. 28, 2003).
Kathleen Kerr, Docs Don’t See a Future in Babies, NEWSDAY, Oct. 19,
2004, at A06.
267
Hearings, supra note 13.
268
News Release, American College of Obstetricians and Gynecologists,
Medical Liability Survey Reaffirms More Ob-Gyns Are Quitting Obstetrics
(2004), available at http://www.acog.org/from_home/publications/press_
releases/nr07-16-04.cfm [hereinafter Medical Liability Survey].
269
Id.
270
Id.
271
Sarah Domin, Where Have All the Baby-Doctors Gone? Women’s
Access to Healthcare in Jeopardy: Obstetrics and the Medical Malpractice
Insurance Crisis, 53 CATH. U. L. REV. 499, 537 (2004).
272
Id.
266
BETH MACROED CORRECTED FINAL 060605.DOC
848
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
insulate themselves from future lawsuits.273 It is not difficult to
conclude that with fewer physicians, limited community clinic
services, and defensive medical practices, women in a lower socioeconomic bracket would face significant difficulties in accessing
vital gynecological or obstetrical services.
Some doctors fear that the Broadnax decision could expand
liability, such that that they will end up in court for cases that
involved no medical negligence and face jurors who will award
damages for psychological suffering based not on the degree of
harm or fault, but on the emotionally-charged nature of fetal
malpractice cases.274 The vice-chair of the American College of
Obstetricians and Gynecologists expressed concern about the
potential expansion of liability following Broadnax, noting that “if
it were a fair fight, it would not be a problem. But the problem is
that science doesn’t protect us [obstetricians and gynecologists] in
court” when dealing with such emotional issues.275 Similarly,
insurers voice concerns that echo doctors’ concerns. Edward
Amsler, vice president of Medical Liability Mutual Insurance
Company, which insures most of New York’s physicians, noted to
Newsday that children who have been injured through negligence
in utero are very sympathetic plaintiffs and “hence they get huge
jury verdicts.”276
In contrast to doctors’ fears of increased medical malpractice
liability resulting from Broadnax and similar decisions, trial
lawyers doubt whether Broadnax will have any impact on the crisis
of medical malpractice liability facing New York and the rest of
the country; others debate whether there is even a “crisis” at all.277
In one published reaction to Broadnax, Lenore Kramer, past
president of the New York State Trial Lawyers Association,
refuted contentions that Broadnax would increase malpractice
273
Kerr, supra note 266.
Albert, supra note 259, at para. 4.
275
Id. at para. 5.
276
Kerr, supra note 266.
277
John Caher, Liability Widens for Fetal Death Caused by Doctors;
Distress Damages Do Not Require Bodily Harm to Women, N.Y.L.J., Apr. 2,
2004, at 1.
274
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
849
litigation so as to affect liability rates.278 She declared that the
ruling in Broadnax
recognizes a reality of these terrible situations [of medical
malpractice to fetuses] and brings the law into conformity
with what people’s understanding of what justice is . . .
[trial lawyers] sincerely believe that there is no medical
malpractice crisis and that it is a trumped up issue
perpetrated by the insurers.279
In another published reaction, Margaret C. Jasper, one of the
attorneys representing the appellants in Broadnax, stated that if
doctors are concerned about unjustified lawsuits, they will need to
do a better job of “policing their own.”280 Jasper further explained
that even in clear cases of medical malpractice, it is difficult to
bring a malpractice case in New York due to statutes of limitations
and expert testimony requirements.281 Regardless of its impact on
the medical liability insurance crisis, it is clear that the Broadnax
decision was “heralded by plaintiff’s attorneys as having brought
New York out of the dark ages by expanding the amount of
damages potentially recoverable,”282 with some opining that
Broadnax merely comports with a growing national sentiment that
the unborn child is worthy in the eyes of the law.283 Whether an
increase in the amount of available damages will actually have an
impact on the cost of malpractice liability insurance in New York
is yet to be seen.
There are two specific ways in which the New York State
278
Id.
Id.
280
Albert, supra note 259, at para. 15.
281
Id. at para. 16.
282
Marian E. Silber and Maria Elyse Rabar, Medical Malpractice
Litigation; Damages for Stillbirth: Will The Floodgates Be Opened?, N.Y. L. J.,
April 30, 2004, at 3.
283
Albert, supra note 259. In addition to civil remedies, the nation is also
seeking criminal remedies for those who injure a child in utero. For example,
the decision in Broadnax was coincidentally handed down on the same day that
President George W. Bush signed into law the Unborn Victims of Violence Act,
making it a crime to harm the fetus of a pregnant woman during an assault.
Caher, supra note 277.
279
BETH MACROED CORRECTED FINAL 060605.DOC
850
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
legislature could address concerns related to the Broadnax
decision. First, the legislature could directly address the cause of
action for wrongful death by statute, either by affirmatively
denying recovery for these causes of action or by expressly
allowing for these causes of action.284 This method goes directly to
the heart of the matter and in fact would determine the impact that
Broadnax will have on fetal rights litigation related to prenatal
negligence.
Alternatively, the legislature could also address the concerns
resulting from Broadnax by regulating medical malpractice
liability itself. Damages caps and insurance reform are often
suggested as two means of stabilizing premium rates for doctors.
Physicians and the insurance industry generally favor the
imposition of caps on non-economic damages in medical
malpractice liability cases as a solution to rising insurance
premiums.285 This method has been supported primarily by
Republican legislators at both the state and federal level.286
California’s Medical Injury Compensation Reform Act (MICRA)
is one such model of damage cap legislation.287 MICRA places a
$250,000 cap on the amount of compensation awarded to
malpractice victims for their non-economic injuries.288 The New
284
S.D. CODIFIED LAWS § 21-5-1 (2004); ARK. CODE ANN. §16-62-102
(2004). South Dakota and Arkansas have statutes that expressly allow for causes
of wrongful death for the fetus negligently killed in utero. Id.
285
Glassman, supra note 260, at 419 (noting that physicians and the
insurance industry place the blame for escalating malpractice liability insurance
rates on an excess of litigation and high jury awards). Non-economic damages
are defined generally as damages awarded for a litigant’s past and/or future pain
and suffering. Id. at 423 n.27.
286
Id. at 419. The GOP’s objective is to impose federal caps on noneconomic damages in medical malpractice cases and to take the decision out of
the hands of the states. Id.
287
CAL. CIVIL CODE § 3333.2 (Deering 2005); see also Hearings, supra
note 13 (statement to Congress in which the American Medical Association
advocates federal legislation based on California’s medical liability reform act,
known as MICRA).
288
CAL. CIVIL CODE § 3333.2 (Deering 2005). Non-economic damages, as
defined in the California statute, include pain, suffering, inconvenience, physical
impairment, disfigurement and other non-pecuniary injury. Id.
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
851
York legislature could enact a similar cap on non-economic
damages.289 A cap might propel more obstetricians and
gynecologists back into high-risk practices, thus helping to
alleviate the women’s healthcare crisis that might result from the
Broadnax decision. However, opponents to a similar proposal in
the U.S. House of Representatives have stated that
[a cap on non-economic damages in healthcare lawsuits]
offers a “solution” prior to having discovered the root of
the problem. Instead of reducing the occurrence of
frivolous lawsuits, providing direct assistance to health care
providers and communities, and examining every aspect of
this problem [i.e., doctors facing soaring medical
malpractice insurance premiums], this legislation restricts
the legal rights of those who have been truly wronged.290
Insurance reform has been suggested as an alternative means of
reducing or stabilizing doctors’ insurance premium costs.291
Insurance reform is supported primarily by Democratic state and
federal legislators, the Association of Trial Lawyers of America,
state trial lawyer associations, and consumer watchdog groups.292
These groups are critical of federal caps and maintain that caps
have not proven successful in either lowering or stabilizing
premiums.293 Indeed, as noted in a 2003 study released by the
Foundation for Taxpayer and Consumer Rights on the relative
success of California’s MICRA statute, malpractice caps, and other
restrictions on patients’ rights did not actually lower or stabilize
premiums, as insurers and doctors claimed; rather, it was the
implementation of California’s insurance reform initiative in 1988
that reduced California doctors’ premiums by twenty percent over
three years.294 This law resulted in a rate freeze, a rate rollback,
289
Medical Liability Survey, supra note 268.
Glassman, supra note 260, at 424 n.39.
291
FOUNDATION FOR TAXPAYER AND CONSUMER RIGHTS, HOW
INSURANCE REFORM LOWERED DOCTORS’ MEDICAL MALPRACTICE RATES IN
CALIFORNIA (Mar. 7, 2003), available at http://www.consumerwatchdog.org/
healthcare/rp/rp003103.pdf [hereinafter INSURANCE REFORM].
292
Glassman, supra note 260, at 420.
293
Id.
294
INSURANCE REFORM, supra note 291.
290
BETH MACROED CORRECTED FINAL 060605.DOC
852
6/6/2005 1:37 PM
JOURNAL OF LAW AND POLICY
and stringent regulation that reduced premiums in all lines of
insurance, including medical malpractice.295 Thus, the New York
State legislature might be well advised to adopt insurance reforms
similar to those implemented in California and to require that
insurance companies roll back premium rates to offset any
concerns about the rising costs of medical malpractice insurance in
the aftermath of the Broadnax decision.
Clearly, divergent views exist as to whether Broadnax will
affect the medical liability crisis faced by obstetricians and
gynecologists in New York. Some even question whether a crisis
exists at all. However, even if the crisis is “trumped up,” as some
opine, it is almost certain that the legislature will address the crisis,
or potential crisis, through initiatives that either eliminate possible
causes of action for the wrongful death of fetuses, impose caps on
malpractice verdicts, or enact insurance premium reforms. Thus,
while Broadnax’s impact may be a drop in the bucket in terms of
affecting the availability or cost of obstetrical or gynecological
care in New York, it has almost assuredly contributed to fear that
the availability or cost of obstetrical or gynecological care could be
compromised by expanding liability in the area of wrongful death.
Indeed, this fear may be what spurs the legislators to take action.
CONCLUSION
The New York Court of Appeals’s recent decision in Broadnax
v. Gonzalez overturned nearly twenty years of precedent in which
New York courts refused to permit mothers to recover emotional
damages for negligently caused stillbirths or miscarriages absent
independent injuries of their own. In declaring that both the fetus
and the mother are owed a duty of care, and by expanding the
rights of the fetus by assigning a cause of action to the mother,
Broadnax may have far-reaching implications for other causes of
action involving fetal rights, namely, suits for wrongful death.
While the victims of negligence clearly deserve to have their
295
Foundation for Taxpayer and Consumer Rights, Five Dangerous Myths
About California’s Medical Malpractice Restrictions, at http://www.consumer
watchdog.org/healthcare/fs/fs003009.php3 (last visited Apr. 20, 2005).
BETH MACROED CORRECTED FINAL 060605.DOC
FETAL RIGHTS UNDER NEW YORK LAW
6/6/2005 1:37 PM
853
injuries redressed, doctors are justifiably concerned that expanding
liability will result in higher insurance premiums, forcing
obstetricians to abandon high-risk patients or even the practice of
obstetrics and gynecology in New York altogether, thereby
lowering the quality of obstetrical and gynecological care provided
in this state. This result would undermine one of the principal
purposes of medical malpractice liability—to encourage
accountability in medicine and to ensure the availability of highquality healthcare. If an increase in the number of malpractice
lawsuits results in an exodus of obstetrical and gynecological
physicians from the medical field, it must be asked whether this
expansion of liability is beneficial for New York in the long run. In
addressing this question, the courts have faced difficult decisions
and have demonstrated a desire to leave the expansion of tort
liability to the legislature. Given the court’s reluctance to address
this area of the law, the legislature must seriously examine the
trend of expanding fetal rights in New York and the United States
generally and take affirmative steps to either expressly accept or
reject the extension of these rights to wrongful death causes of
action. Only with definitive and clear statutes will this murky area
of fetal rights ever be resolved in New York.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
VEILED MUSLIM WOMEN AND DRIVER’S
LICENSE PHOTOS:
A CONSTITUTIONAL ANALYSIS
Peninna Oren*
INTRODUCTION
On June 4, 2004, the Florida ACLU filed an appeal in the case
of Freeman v. State that remains pending almost one year later.1
The ACLU was appealing the June 6, 2003 decision of a Florida
state intermediary level court upholding the Florida Department of
Highway Safety and Motor Vehicles’ (DHSMV) revocation of a
Muslim woman’s driver’s license on account of the woman’s
refusal to take a photograph for her license without her veil, or
“niqab.”2 The Muslim woman, Sultaana Lakiana Myke Freeman,
believes from her study of the Quran and the Sunnah that
legislation from Allah mandates that she, as a Muslim woman, veil
*
Brooklyn Law School Class of 2005; B.A., Boston University, 2002. I
would like to thank my editor Doug Brooks and the entire Journal of Law and
Policy Executive Board, especially Skye Phillips and Cory Shindel for their
patience and for all of the time and effort they poured into assisting me with my
note. I would like to thank my father, Steve Oren, for teaching me not to accept
the conclusions of others, but rather, to think for myself. I would also like to
thank my mother, Roz Oren, for teaching me to respect those whose beliefs
differ from my own.
1
Telephone Interview with ACLU of Florida (April 18, 2005); Brief for
Appellant at 44, Freeman v. State, No. 2002-CA-2828, 2003 WL 21338619 (Fla.
Cir. Ct. June 6, 2003), available at http://www.aclufl.org/pdfs/Legal%20PDfs/
Freeman%20appeal%20brief.pdf.
2
Freeman v. State, No. 2002-CA-2828, 2003 WL 21338619 (Fla. Cir. Ct.
June 6, 2003). The case was decided by the Florida Circuit Court, Ninth Judicial
Circuit. Id. at *1. The plaintiff referred to her veil, which covers her entire face
with the exception of her eyes, as a niqab. Id.
855
OREN MACROED CORRECTED 062205.DOC
856
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
her face as part of her religious obligation to dress modestly.3
Ms. Freeman challenged the Florida statute that requires a fullface photograph for driver’s licenses4 on the grounds that it
violated Florida’s Religious Freedom Restoration Act of 19985
(FRFRA) and Florida’s state constitution.6 The FRFRA prohibits
the State from substantially burdening an individual’s exercise of
religion absent a compelling state interest and proof by the State
that the law that burdens the individual’s religious freedom is the
least restrictive means of achieving the state’s interest.7 The free
exercise of religion is similarly protected under Article I, Section 3
of the Florida Constitution, which grants the right to religious
freedom.8
Notably, Freeman did not challenge the State’s driver’s license
photo requirement on federal constitutional free exercise grounds
because the Supreme Court has held that there is no federal remedy
for individuals who claim only that their religious practices are
interfered with by a neutral law of general applicability.9 The
Supreme Court has left open the possibility of a “hybrid claim,”
however, when a neutral law of general applicability interferes
3
See Statement by Sultaana Lakiana Myke Freeman, (May 27, 2003),
available at http://www.aclufl.org/issues/religious_liberty/freemanpersonal_
statement.cfm (last visited Apr. 5, 2005).
4
FLA. STAT. ANN § 322.142(1) (West 2004). The Florida statute describing
the requirements for driver’s licenses uses the term “fullface photograph.” Id. §
322.142(1). For the sake of consistency, this note will use the term “full-face
photograph” to describe the driver’s license photographs required by the Florida
law. In Freeman, there was an initial argument about whether a veiled Muslim
woman fulfilled the requirement of a full-face photograph because, although
veiled, the woman was facing the camera when her photograph was taken.
Freeman, 2003 WL 21338619, at *2 n.2. This note, like the Florida court,
assumes that the full-face requirement dictates that a veiled Muslim woman
must unveil for her driver’s license picture.
5
FLA. STAT. ANN § 761.03 (West 2004).
6
FLA. CONST. art. I, § 3; Freeman, 2003 WL 21338619, at *1.
7
FLA. STAT. ANN. § 761.03 (West 2004).
8
FLA. CONST. art. I, § 3.
9
Employment Div., Dep’t. of Human Res. of Or. v. Smith, 494 U.S. 872,
890 (1990). A neutral law of general applicability is now subject only to rational
basis review. Id.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 857
with the free exercise of religion as well as a conjoining
constitutional right.10
This note examines whether laws that require veiled Muslim
women to unveil for their driver’s license photographs violate
these women’s Fourth Amendment right to protection against
unreasonable searches in addition to their right to free exercise of
religion and, therefore, give rise to a “hybrid” claim with an
available federal remedy.11 Part I of this note summarizes the
Freeman case. Part II describes the present status of free exercise
jurisprudence, including hybrid claims.12 Part III.A discusses the
components of a Fourth Amendment claim and analyzes how a
hybrid claim might be asserted.13 Part III.B applies the hybrid
10
Id. at 881. “The only decisions in which we have held that the First
Amendment bars application of a neutral, generally applicable law to religiously
motivated action have involved not the Free Exercise Clause alone, but the Free
Exercise Clause in conjunction with other constitutional protections, such as
freedom of speech.” Id.
11
Id. (noting that “hybrid” claims are the lone claims available to challenge
a neutral law of general applicability under the Free Exercise Clause of the First
Amendment).
12
This note does not address the question of whether driving is a privilege
or a right because the court in Freeman treated driving as a right, despite the fact
that the language of the driver’s license statute in question referred to driving as
a privilege. Freeman v. State, No. 2002-CA-2828, 2003 WL 21338619, at *6
(Fla. Cir. Ct. June 6, 2003). The Florida court stated:
Although the Florida statutes use the term “driving privileges” this does
not mean that driving is a “privilege” rather than a “right.” The Court
recognizes that in Sherbert v. Verner, the U.S. Supreme Court stated
that the distinction between privilege and right is not meaningful when
the benefit in question, i.e., being able to drive a car and thereby
conduct normal life activities, is the same. So even if driving is a
“privilege,” the government may not deny Plaintiff that benefit without
showing that there is a compelling state interest that overrides her right
to free exercise of religion.
Id. (citations omitted). For a discussion of the Supreme Court’s
“unconstitutional conditions” doctrine, see Jason Mazzone, The Waiver
Paradox, 97 NW. U. L. REV. 801 (2003).
13
Although Freeman’s claim was not successful, had Freeman’s attorneys
attempted to make a federal hybrid claim, this claim might not have even made
it to trial had the state made a motion under FED. R. CIV. P. 12(b)(6) to dismiss
the complaint for failure to state a claim upon which relief could be granted. If
OREN MACROED CORRECTED 062205.DOC
858
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
claim analysis to a case such as that of Freeman by weighing a
veiled Muslim woman’s right to Fourth Amendment protection and
free exercise of religion against the State’s interest in the full-face
driver’s license photo requirement. This note concludes that a
Muslim woman who wishes to be photographed for her driver’s
license may assert a hybrid claim, however, given the fact that
there has yet to be a successful hybrid claim, it is doubtful that her
claim would be successful.
I.
FREEMAN V. STATE
In Freeman v. State, the Florida Circuit Court of the Ninth
Judicial District decided the case of Ms. Freeman, whose
previously-issued driver’s license was revoked after she refused to
take a new picture for her driver’s license without her full-face
veil.14 The Florida court evaluated the Freeman case under both
the Florida Constitution,15 which grants the right to religious
freedom, and the FRFRA,16 which prohibits the State from
the court had decided in a pre-trial motion that Freeman did not have a valid
hybrid claim, her case would not have been heard. Therefore, Freeman’s lawyers
were better off challenging the law under Florida’s RFRA, which provides an
available remedy. Indeed, only twelve states have legislation that protects the
free exercise of religion. See infra note 98. In states without free exercise
legislation, a plaintiff would have little to lose by asserting a hybrid claim
(except of course the money spent on attorney’s fees).
14
Freeman, 2003 WL 21338619.
15
FLA. CONST. art. I, § 3 (West 2004). “There shall be no law respecting
the establishment of religion or prohibiting or penalizing the free exercise
thereof . . . .” Id.
16
FLA. STAT. ANN. § 761.03 (West 2004). In relevant part, the statute
reads:
(1) The government shall not substantially burden a person’s exercise
of religion, even if the burden results from a rule of general
applicability, except that government may substantially burden a
person’s exercise of religion only if it demonstrates that application of
the burden to the person:
(a) Is in furtherance of compelling governmental interest; and
(b) Is the least restrictive means of furthering that compelling
governmental interest.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 859
substantially burdening the free exercise of religion unless the
State can prove both a compelling state interest and that the law is
the least restrictive means of achieving the State’s goal.17 The
Freeman court held that Freeman’s right to free exercise of
religion was not substantially burdened, but nevertheless analyzed
the State’s compelling interest in the driver’s license statute,
holding that strict scrutiny review was required because Freeman
alleged an infringement upon her fundamental constitutional right
to free exercise of religion.18
A. Freeman’s Case
On February 21, 2001, the State of Florida issued Sultaana
Lakiana Myke Freeman a driver’s license that contained a picture
of her wearing a full-face veil, or niqab, so that only her eyes were
visible.19 Freeman’s face was similarly covered in the photograph
on her driver’s license from Illinois, where she lived prior to her
move to Florida.20 On November 28, 2001 and December 18,
2001, Freeman received letters from the State of Florida informing
her that her license would be revoked if she did not report to the
DHSMV to be photographed without her veil for her driver’s
license.21 For religious reasons, Freeman refused to comply, and
her license was revoked.22 Freeman then brought an action
challenging Florida’s revocation of her driver’s license under the
FRFRA and the Florida Constitution.23
The court found that the driver’s license requirement did not
substantially burden Freeman because the DHSMV had a practice
(2) A person whose religious exercise has been burdened in violation of
this section may assert that violation as a claim or defense in a judicial
proceeding and obtain appropriate relief.
Id.
17
18
19
20
21
22
23
Id.
Freeman, 2003 WL 21338619 at *1.
Id. at *4.
Id. at *1.
Id.
Id.
Id.
OREN MACROED CORRECTED 062205.DOC
860
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
of accommodating women who veil by having a female employee
photograph them in a private room so that the women’s face and
hair would only be exposed to a female employee and, in certain
situations (for example, when women were pulled over) to law
enforcement officers.24 Because the court found no substantial
burden, it held that the statute did not violate the FRFRA, given
that the Act only precludes the State from substantially burdening
an individual’s free exercise of religion and does not prohibit the
State from enacting a statute that places a lesser burden on an
individual’s free exercise of religion.25
The court also addressed Freeman’s constitutional claim.26
Article I, Section 3 of Florida’s constitution provides: “There shall
be no law respecting the establishment of religion or prohibiting or
penalizing the free exercise thereof. Religious freedom shall not
justify practices inconsistent with public morals, peace, or
safety . . . .”27
The Florida court held that it was required to apply the strict
scrutiny standard of review to Freeman’s constitutional claim to
determine whether the State had a compelling interest to justify its
restriction of a religious practice.28 The court thus analyzed
whether the State had a compelling interest in the statute requiring
full-face driver’s license photographs.29
B. The Freeman Court’s Compelling Interest Analysis
Freeman argued that the State did not have a compelling
interest in restricting her right to have a driver’s license without a
full-face photograph because a driver’s license is not a state
identification card, but rather, is “merely certification of
competence to drive.”30 Freeman relied on three cases upholding
the right of religious Christians to receive driver’s licenses without
24
25
26
27
28
29
30
Freeman, 2003 WL 21338619, at *3.
Id. at *4.
Id. at *1.
FLA. CONST. art. I, § 3.
Freeman, 2003 WL 21338619, at *1.
Id.
Id. at *5.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 861
photographs because of their religious beliefs that photographs
constitute graven images.31 Specifically, Freeman cited Quaring v.
Peterson, in which the Eighth Circuit held that there was no
compelling state interest in the photograph requirement because
individuals who possessed out-of-state licenses that did not contain
photographs were permitted to drive in the state.32 Further,
Freeman cited Bureau of Motor Vehicles v. Pentecostal House of
Prayer33 and Dennis v. Charnes, which held that allowing an
exception for individuals whose religions proscribe the taking of
photographs would not lead to widespread abuse.34
In addition to arguing that her claims should be considered
under the same reasoning as the graven images cases, Freeman
also challenged the utility and accuracy of driver’s license
photographs.35 Specifically, Freeman contended that photographs
are “largely flawed” and can be “easily thwarted” by those who
“change their hair, cover their foreheads and ears, wear large
glasses, shave their heads, grow their beards, or alter their
appearance by other means, including contact lenses and plastic
surgery.”36
Freeman further argued that there are more than 4,000 people
to whom the State of Florida issued photo-less driver’s licenses
and tens of thousands of people from other states with photo-less
driver’s licenses driving in the State of Florida.37 Thus, she argued,
Florida lacked a compelling interest in refusing to grant religious
exceptions to its driver’s license photograph requirement.38
In holding for the State, the Florida court adopted the State’s
arguments as its own analysis.39 The State argued that it had a
31
Id.
Id. (citing Quaring v. Peterson, 728 F.2d 1121 (8th Cir. 1984)).
33
Freeman, 2003 WL 21338619, at *5 (citing Bureau of Motor Vehicles v.
Pentecostal House of Prayer, Inc., 380 N.E.2d 1225 (Ind. 1978)).
34
Freeman, 2003 WL 21338619, at *5 (citing Dennis v. Charnes, 805 F.2d
339 (10th Cir. 1984)).
35
Freeman, 2003 WL 21338619, at *5.
36
Id.
37
Id. at *6.
38
Id.
39
Id. at *3-7.
32
OREN MACROED CORRECTED 062205.DOC
862
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
compelling interest in the full-face photograph requirement
because the requirement promotes safety and security, combats
crime, and protects interstate commerce.40 In response to
Freeman’s argument that photographs are not effective, the State
presented a witness who testified that photographs of faces that
have changed are still more effective than veiled photographs
because some facial features do not change.41 Furthermore, the
State contended that, without a full-face photograph, law
enforcement officers would be at a greater risk when they stopped
individuals, given the extra time necessary to verify the driver’s
identity.42 The State also asserted that, despite the fact that such
intent does not appear in the driver’s license statute, driver’s
licenses are intended for use as identity documents by people in
“society at large to cash checks, rent cars and clear airport
security.”43 Moreover, the State distinguished Freeman’s case from
earlier cases permitting exceptions to the driver’s license
photograph requirements.44 In adopting the State’s analysis, the
court noted that the world is different than it was twenty to twentyfive years ago and that since 1978, when the first of the three cases
cited by Freeman was decided, the increased degree of domestic
terror has amplified the potential for widespread abuse.45
Next, the court rejected the argument that Freeman should be
granted an exception based on the fact that others, including outof-state drivers and those with temporary licenses, are legally
allowed to drive in Florida without a full-face photograph on their
40
Freeman, 2003 WL 21338619, at *4.
Id. at *5.
42
Id.
43
Id.
44
Id. at *7.
45
Id. According to the U.S. Centennial of Flight Commission, the U.S.
Department of Transportation reported that there were 364 hijackings worldwide
from 1968 until 1972. There were no hijackings from February 1991 until
September 11, 2001. Judy Rumerman, U.S. Centennial Flight Commission;
Aviation Security, available at http://www.centennialofflight.gov/essay/
Government_Role/security/POL18.htm (last visited May 18, 2005). There have
been no hijackings since September 11, 2001. Eli Lehrer, The Homeland
Security Bureaucracy, PUBLIC INTEREST, June 22, 2004, at 71.
41
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 863
licenses.46 The court explained that individuals with temporary
licenses that did not contain photographs had already received a
permanent license, and therefore, the State possessed a full-face
photograph of those individuals.47 The court held that the State of
Florida cannot control the laws of other states and must accept
their citizens’ driver’s licenses because full faith and credit is given
to the laws of other states.48 The court held that Florida can,
however, exert control over its own residents and impose
regulations regarding the requirements for obtaining driver’s
licenses within the state.49
Based on its analysis of both parties’ arguments, the court
found that the driver’s license statute promoted public safety and
protected against fraud, and thus, the State had a compelling
interest in the statute that outweighed the seemingly insubstantial
burden the law posed to the free exercise of religion.50 The court
also explained that, given the accommodations put in place by the
DHSMV, the statute was the least restrictive means of furthering
the State’s interest.51 Thus, the court held that the driver’s license
statute did not violate Article I, Section 3 of the Florida
Constitution.52
II. RELIGIOUS FREEDOM CLAIMS
Freeman challenged the revocation of her driver’s license due
to her refusal to unveil under the FRFRA and Florida’s state
46
Freeman, 2003 WL 21338619, at *6.
Id.
48
Id. 28 U.S.C. § 1738 (West 2005) provides:
[a]cts of the legislature of any state, territory or possession of the
United States . . . shall have the same full faith and credit in every court
within the United States and its Territories and Possessions as they
have by law or usage in the courts of such State, Territory or
Possession from which they are taken.
47
Id.
49
50
51
52
Freeman, 2003 WL 21338619, at *6.
Id. at *7-8.
Id.
Id.
OREN MACROED CORRECTED 062205.DOC
864
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
constitution.53 Freeman notably pursued these avenues of relief
based on the lack of a federal remedy. A federal claim under the
First Amendment’s Free Exercise Clause would have been
unsuccessful54 because such a claim, which alleges only a violation
of a person’s freedom of religion, no longer applies to neutral laws
of general applicability.55 This section discusses Supreme Court
case law regarding federal free exercise review and analyzes the
remedies that remain available to individuals whose free exercise
of religion is substantially burdened by neutral laws of general
applicability.
A. Federal Free Exercise Clause Review
The Free Exercise Clause of the First Amendment provides
that “Congress shall make no law respecting an establishment of
religion, or prohibiting the free exercise thereof. . . .”56 In Sherbert
v. Verner, the Supreme Court interpreted the Free Exercise Clause
to mean that “if the purpose or effect of a law is to impede the
observance of one or all religions or is to discriminate invidiously
between religions, that law is constitutionally invalid even though
the burdens may be characterized as only indirect.”57 Under this
interpretation of the Free Exercise Clause, the Supreme Court
53
Freeman, 2003 WL 21338619, at *1.
See The Case of Mrs. Sultaana Freeman, at http://www.aclufl.org/
news_events/archive/2003/freemanrelease052703.cfm. The Florida ACLU
represented Freeman and discusses her case on the organization’s website.
55
Daniel A. Crane, Beyond RFRA: Free Exercise of Religion Comes of Age
in the State Courts, 10 ST. THOMAS L. REV. 235, 238 (1998) (“As a result [of
Flores], the states are no longer bound by any federal standard, whether
statutory or constitutional, to exempt the religiously devout from neutral laws of
general applicability.”). Driver’s license requirements are neutral laws of
general applicability because they were written with the neutral intention of
regulating drivers and not to regulate religious activity, and they are generally
applicable in that they apply to the entire public equally and are not applied
exclusively to religious individuals. Employment Div., Dep’t. of Human Res. of
Or. v. Smith, 494 U.S. 872, 890 (1990).
56
U.S. CONST. amend. I.
57
Sherbert v. Verner, 374 U.S. 398, 404 (1963) (citing Braunfeld v. Brown,
366 U.S. 599, 607 (1961)) (emphasis added).
54
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 865
invalidated a number of neutral, generally applicable laws as they
applied to the religiously observant.58
In Sherbert, the petitioner, a Seventh Day Adventist, was fired
from her job because, for religious reasons, she would not work on
Saturdays.59 While the petitioner sought other employment
following her dismissal, she refused to accept positions that
required her to work on Saturdays and could not find a job that did
not require her to do so.60 The petitioner was subsequently denied
unemployment benefits under the South Carolina Unemployment
Compensation Act, which provided that a person is not eligible for
benefits if “he has failed without good cause . . . to accept available
suitable work when offered to him by the employment office or the
employer.”61 In reviewing the petitioner’s claim, the Sherbert
Court applied a balancing test, equivalent to strict scrutiny review,
in which it balanced the state’s compelling interest in the law
against the substantial burden the law imposed on the plaintiff’s
religious practices.62 The Court found that the substantial burden of
the petitioner’s being required to work on her Sabbath or forgo
state benefits outweighed the state’s interest in preventing
fraudulent claims that would dilute unemployment funds and
disrupt work schedules.63
58
See Hobbie v. Unemployment Appeals Comm’n of Fla., 480 U.S. 136,
146 (1987) (holding that Florida’s refusal to award unemployment
compensation benefits to a Seventh Day Adventist who quit her job because she
would not work on her Sabbath was unconstitutional under the Free Exercise
Clause); Thompson v. Review Bd. of Ind. Employment Sec. Div., 450 U.S. 707,
720 (1981) (holding that the state’s termination of petitioner’s unemployment on
the grounds that the petitioner quit his job violated the Free Exercise Clause
because his religion prohibited making armaments); Wisconsin v. Yoder, 406
U.S. 205 (1972) (excepting the Amish from a general state law requiring that
children remain in school until they are sixteen years of age).
59
Sherbert, 374 U.S. at 399. This balancing test is also called “strict
scrutiny.” See id. at 908-09 (Blackmun, J., dissenting) (referring to the Sherbert
test as strict scrutiny review).
60
Id. at 402 n.3.
61
Id. at 400-01 (citing S.C. CODE ANN. § 68-1-404 (Law Co-op. 1962)).
62
Employment Div., Dep’t. of Human Res. of Or. v. Smith, 494 U.S. 872,
875 (1990) (citing Sherbert v. Verner, 374 U.S. 398 (1963)).
63
Sherbert, 374 U.S. at 407.
OREN MACROED CORRECTED 062205.DOC
866
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
In 1990, however, the Court narrowed its definition of free
exercise. In Employment Division v. Smith,64 the Supreme Court
rejected the application of strict scrutiny review to free exercise
claims.65 In Smith, the Court considered the case of two Native
Americans who were dismissed from their jobs for ingesting
peyote during a religious service and who were subsequently
denied unemployment benefits by the State of Oregon because
ingesting peyote was a criminal offense under Oregon state law.66
The Supreme Court of Oregon applied strict scrutiny and held that,
although the respondents had committed a crime by using peyote,
the purpose of the unemployment law, which precluded the receipt
of benefits by individuals who were dismissed from their jobs for
misconduct, was not to punish individuals for crimes, but rather, to
preserve the fund’s integrity.67 The court held that the burden on
the respondent’s religious practice outweighed the purpose of the
law; therefore, the law was unconstitutional.68
The U.S. Supreme Court overruled the Oregon Supreme
Court’s decision, rejecting the application of strict scrutiny to a
free exercise claim that challenged a neutral law of general
applicability.69 The Court held that the Free Exercise Clause would
be violated were a law to specifically target a religious group or
religious observance, for example, if a statute were to specifically
prohibit “bowing down before a golden calf.”70 The Court noted,
however, that the right to free exercise of religion under the First
Amendment is not unlimited.71 The Court explained that to allow
individuals in all circumstances to practice their religions, even
when their doing so would conflict with existing, generally
applicable law, would “contradict[ ] both constitutional tradition
and common sense” because individuals would be excused from
following the law whenever their religions conflicted with the
64
65
66
67
68
69
70
71
Smith, 494 U.S. at 872.
Id. at 884.
Id. at 872.
Id. at 875.
Smith, 494 U.S. at 875.
Id.
Id. at 878.
Id. at 886.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 867
laws.72 The Court cited a sampling of cases in which it had rejected
the extension of free exercise protection to individuals when state
laws interfered with the practice of religion.73 The Court’s
examples included laws prohibiting polygamy and child labor, and
those requiring the payment of Social Security taxes.74 The Court
noted that it had never invalidated a neutral law of general
applicability when the law interfered only with a person’s right to
free exercise of religion.75
Rather than applying strict scrutiny, the Supreme Court in
Smith76 held that it is up to the “political process” and not the
courts to protect the interests of individuals whose religious
practices are interfered with by a neutral, generally applied state
law.77 Although it rejected the use of the strict scrutiny as the
standard of review for free exercise claims, the Smith Court did not
overrule Sherbert.78 Instead, the Court distinguished the case
before it from Sherbert, stating that strict scrutiny applied only to
“employment compensation” cases, not criminal matters, as
examined in Smith.79 The Supreme Court explained that the issue
72
Id.
Id. at 879-80 (citing Reynolds v. United States, 98 U.S. 145 (1879)
(holding that a person who believed, based on his religion, that a law prohibiting
bigamy should not have been enacted is not immune from prosecution for
violating that law); Prince v. Massachusetts, 321 U.S. 158 (1944) (holding that a
woman who used her child to distribute literature on the street could be
prosecuted for violation of child labor laws, despite the fact that the literature
being distributed was religious); United States v. Lee, 455 U.S. 252 (1985)
(holding that an Amish person was not exempt from paying Social Security
taxes, even though his religion prohibited taking part in governmental support
programs)).
74
Smith, 494 U.S. at 886.
75
Id. at 878-79. “We have never held that an individual’s religious beliefs
excuse him from compliance with an otherwise valid law prohibiting conduct
that the State is free to regulate.” Id.
76
Id. at 872.
77
Id. at 872, 890. The phrase “political process” describes the legislature.
Id.
78
Id. at 884 (holding that “[e]ven if we were inclined to breathe into
Sherbert some life beyond the unemployment compensation field, we would not
apply it to require exemptions from a generally applicable criminal law”).
79
Id.
73
OREN MACROED CORRECTED 062205.DOC
868
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
in Smith was not whether Oregon could deny the respondents
benefits, but rather, whether the Oregon criminal statute, which
generally prohibits the use of peyote, could be applied to
individuals whose religion required its use.80 The Court held that
unemployment compensation cases belong to a separate class of
cases in which exceptions for free exercise of religion are
permitted because “a distinctive feature of unemployment
compensation programs is that their eligibility criteria invite
consideration of the particular circumstances behind an applicant’s
unemployment.”81 The Court further held “that where the state has
in place a system of individual exemptions, it may not refuse to
extend that system to cases of ‘religious hardship’ without
compelling reason.”82
The Court also distinguished Smith from prior Supreme Court
cases in which it had invalidated statutes as applied to the
religiously observant, holding that the burden on the religious
individuals in those cases outweighed the states’ compelling
interests in the challenged statutes.83 The Court held that the cases
in which it had invalidated laws on free exercise grounds involved
both a right to free exercise and a conjoining additional
constitutional claim, and that strict scrutiny is only available in
such “hybrid” cases.84 As a result of Smith, under existing federal
law, a court may still apply strict scrutiny in free exercise cases
involving 1) laws that are not neutral and generally applicable 2)
unemployment compensation, or 3) a free exercise claim that is
80
Id. at 876.
Smith, 494 U.S. at 884.
82
Id. (citing Bowen v. Roy, 476 U.S. 693 (1986)).
83
Id. at 881-82.
84
Id. (citing Catwell v. Connecticut, 310 U.S. 296 (1940)) (containing
conjoining free speech and free press claims); Murdock v. Pennsylvania, 319
U.S. 105 (1943) (containing a conjoining free speech claim); Follet v.
McCormick, 321 U.S. 573 (1944) (same); Pierce v. Society Sisters, 268 U.S.
510 (1925) (containing conjoining claim of parents’ right to direct the education
of their children); Wisconsin v. Yoder, 406 U.S. 205 (1972) (same); Wooley v.
Maynard, 430 U.S. 705 (1977) (decided on free speech grounds, but containing
a conjoining free exercise claim); West Virginia Bd. of Ed. v. Barnette, 319 U.S.
624 (1943) (same).
81
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 869
conjoined with another constitutional claim.85 However, when a
neutral law of general applicability violates only the right to free
exercise of religion, the state must survive only rational basis
review, the lowest form of scrutiny. To prevail under rational basis
review, the state need only prove that its law is rationally related to
a legitimate state interest.86
B. The Federal and State Legislative Response to the Smith
Decision
In response to Smith, Congress enacted the Religious Freedom
Restoration Act of 1993 (RFRA).87 The Senate Report that
accompanied the Act criticized the Smith decision and explained
that “the framers of the Constitution, recognizing free exercise of
religion as an unalienable right, secured its protection in the First
Amendment to the Constitution.”88 The report also stated that
“laws ‘neutral’ towards religion may burden religious exercise as
surely as laws intended to interfere with religious exercise.”89 The
85
Smith, 494 U.S. at 884-85 (holding the Sherbert test inapplicable to
challenges against generally applicable laws on free exercise grounds, but
excepting employment cases from those to which the Sherbert test applies). See
also Swanson v. Guthrie Independent School District, 135 F.3d 694, 700 n.5
(10th Cir. 1998) (“The Smith opinion does not make it clear whether it is
constitutionally sufficient for a law or policy to be neutral and of general
applicability, or whether the policy or law will still have to satisfy some lesser
standard than the compelling interest test.”).
86
S. REP. NO. 103-111, at 7-8 (1993) (stating that the review remaining
after Smith when a neutral law of general applicability interferes with a person’s
religious observance is rational basis review).
87
42 U.S.C.A. § 2000bb (1993). The statute lists as its purpose:
(1) to restore the compelling state interest test as set forth in Sherbert v.
Verner, 373 U.S. 398 (1963) and Wisconsin v. Yoder, 406 U.S. 205
(1972) and to guarantee its application in all cases where free exercise
of religion is substantially burdened; and (2) to provide a claim or
defense to persons whose religious exercise is substantially burdened
by the government.
Id. § 2000bb(b) (1993).
88
S. REP. NO. 103-111, at 2-3 (1993).
89
Id.
OREN MACROED CORRECTED 062205.DOC
870
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
report further criticized the Smith decision, stating that “[b]y
lowering the level of constitutional protection for religious
practices, the decision has created a climate in which the free
exercise of religion is jeopardized.”90
In recognition of these concerns, the RFRA reinstated strict
scrutiny as the test for determining whether a federal or state law
violates the Free Exercise Clause of the U.S. Constitution.91 In
relevant part, the RFRA provided that the “[g]overnment shall not
substantially burden a person’s free exercise of religion . . .
[unless] it is in furtherance of a compelling governmental interest;
and is the least restrictive means of furthering that compelling
interest.”92
In 1997, the Supreme Court, in City of Boerne v. Flores, struck
down the Religious Freedom Restoration Act of 1993 as it applied
to the states, holding that Congress had exceeded its power under
the Enforcement Clause of the Fourteenth Amendment.93 City of
Boerne involved an RFRA challenge to city zoning ordinances by
a Catholic archbishop who was denied a permit to enlarge his
church.94 The Court held that, under the Enforcement Clause,
Congress has the power to make laws that protect people from
state infringement upon their constitutional rights.95 However,
90
Id.
Id. See also City of Boerne v. Flores, 521 U.S. 507, 516-17 (1997).
92
42 U.S.C.A. § 2000 (West Supp. 2004). The RFRA defined
“government” as “a branch, department, agency, instrumentality, and official (or
other person acting under color of law) of the United States or a covered entity.”
The RFRA defined “covered entity” as “the District of Columbia, the
Commonwealth of Puerto Rico, and each territory and possession of the United
States.” Id.
93
Flores, 521 U.S. at 536. The Court held:
When the political branches of the Government act against the
background of a judicial interpretation of the Constitution already
issued . . . in later cases the Court will treat its precedent with the
respect due them under settled principals . . . as the provisions of the
federal statute here invoked are beyond congressional authority, it is the
Court’s precedent, not the RFRA, which must control.
Id.
94
Id.
95
Id. at 517.
91
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 871
because the Supreme Court held in Smith that strict scrutiny is not
available when a general law of neutral applicability interferes
with one’s exercise of religion, Congress cannot be said to be
enforcing the constitutional right of free exercise of religion
through the RFRA because the rights that the RFRA grants are not
provided for by the Free Exercise Clause.96 In other words, the
Court determined that the RFRA was unconstitutional because the
Act provided for rights not granted in the Constitution.97
96
Id. In response to the Supreme Court’s striking down the RFRA,
Congress has passed the Protection of Religious Exercise in Land Use and by
Institutionalized Persons Act (RRLUIPA) which reinstated strict scrutiny as the
test for more narrow instances of government interference with individuals’ free
exercise of religion including land use and zoning regulations and over people
residing in or confined to government institutions. 42 U.S.C. § 2000cc (West
2005). In Elsinore Christian Center v. City Lake of Elsinore, 291 F. Supp. 2d
1083 (C.D. Cal. 2003), the district court for the central district of California
found RLUIPA unconstitutional, holding it exceeds Congress’s enforcement
power. But see U.S. v. Maui County, 298 F. Supp. 2d 1010 (D. Haw. 2003)
(rejecting challenge to RLUIPA on the grounds that RLUIPA violates the
Establishment Clause). See also Corporation of the Presiding Bishop of the
Church of Jesus Christ of Latter-Day Saints v. City of West Linn, 86 P.3d 1140
(Or. Ct. App. 2004) (finding no substantial burden to Plaintiffs under RLUIPA).
For a discussion of the constitutionality of RLUIPA in comparison to the RFRA,
see Michael Paisner, Boerne Supremacy: Congressional Responses to City of
Boerne v. Flores and the Scope of Congress’s Article I Power, 105 COLUM. L.
REV. 537 (2005).
97
Id. Congress’s enforcement power under the Fourteenth Amendment
extends only to enforcing the provisions of the Fourteenth Amendment. In this
case, the petitioner asserted that Congress was enforcing the Privileges and
Immunities Clause of the Fourteenth Amendment, which, in pertinent part,
reads:
No State shall make or enforce any law which shall abridge the
privileges or immunities of citizens of the United States; nor shall any
state deprive any person of life, liberty, or property, without due
process under the law, nor deny to any person within its jurisdiction the
equal protection of the laws.
U.S. CONST. amend. XIV, § 1. However, the Supreme Court held in Smith that
the right to free exercise of religion does not require strict scrutiny when neutral
laws of general applicability interfere only with an individual’s practice of
religion. Therefore, strict scrutiny is not one of the privileges of citizens of the
United States and Congress has no right to direct the state legislation in this
OREN MACROED CORRECTED 062205.DOC
872
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
In response to the Supreme Court’s rejection of both the
Sherbert test and the RFRA, a number of states enacted legislation
requiring a balancing test similar to Sherbert for neutral, generally
applicable laws that impede freedom of religion.98 Florida is
among the states that have enacted Religious Freedom Restoration
Acts in the wake of the Supreme Court’s decisions in Smith and
City of Boerne.99 Florida’s RFRA provides for strict scrutiny
review like Sherbert and the federal RFRA.100
manner. Flores, 521 U.S. at 536 (“RFRA was designed to control cases and
controversies, such as the one before us; but as the provisions of the federal
statute here invoked are beyond congressional authority, it is this Court’s
precedent, not RFRA, which must control.”). Although the Supreme Court has
not ruled on whether the RFRA is constitutional as applied to the federal
government, many circuit courts have held that this is so. See Anne Y. Chiu,
When Prisoners Are Weary and Their Religious Exercise Burdened, RLUIPA
Provides Some Rest for Their Souls, 79 WASH. L. REV. 999, 1004 n. 49 (stating
that the RFRA seems to remain valid as applied to the federal government)
(citing O’Bryan v. Bureau of Prisons, 349 F.3d 399, 400-01 (7th Cir. 2003));
Guam v. Guerrero, 290 F.3d 1210, 1221 (9th Cir. 2002); Henderson v. Kennedy,
265 F.3d 1072, 1073 (D.C. Cir. 2001); Kikumura v. Hurley, 242 F.3d 950, 958
(10th Cir. 2001); Christians v. Crystal Evangelical Free Church (In re Young),
141 F.3d 854, 856 (8th Cir. 1998)).
98
According to RJ&L Religious Liberty Archive, a religious liberty
watchdog organization, twelve states have enacted their own statutes protecting
the free exercise of religion. See http://www.churchstatelaw.com/
statestatutes/index.asp (last visited Feb. 10, 2005). These states include
Alabama, Arizona, Connecticut, Florida, Idaho, Illinois, Missouri, New Mexico,
Ohio, Rhode Island, South Carolina and Texas. See ALA. CONST. amend. NO.
622 (1999); ARIZ. REV. STAT. § 41-1493(1999); CONN. GEN. STAT. § 52-571b
(1993); FLA. STAT. ANN. §§ 761.01-761.05 (1998); IDAHO CODE § 73-401-404
(2000); 775 ILL. COMP. STAT. ANN. 35/15 (West 2004); MO. REV. STAT. §1.302
(2003); N.M. STAT. ANN. § 28-22-1(1978); OKLA. STAT. 51 §§ 251-58 (2000);
R.I. GEN. LAW § 42-80-1 (1956); S.C. CODE ANN. § 1-32-10-60 (1999); TEX.
CIV. PRAC. & REM. CODE ANN. tit. 5, § 110 (1999).
99
FLA. STAT. ANN. §§ 761.01-761.05 (1998).
100
Florida’s RFRA provides:
The government shall not substantially burden a person’s exercise of
religion, even if the burden results from a rule of general applicability,
except that government may substantially burden a person’s exercise of
religion only if it demonstrates that application of the burden to the
person: (b) is the least restrictive means of furthering that compelling
governmental interest.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 873
C. Hybrid Claims
Although no federal legal remedy lies when a neutral law of
general applicability interferes with the right to free exercise of
religion standing alone, the U.S. Supreme Court has acknowledged
that a petitioner may challenge a law on Free Exercise Clause
grounds if the petitioner’s free exercise claim is joined with a
claim based on the violation of another constitutional freedom,
such as freedom of speech.101 In such “hybrid” cases, in which a
generally applicable law is challenged on the basis of the Free
Exercise Clause and another constitutional freedom, strict scrutiny
appears to remain available.102
To date, no circuit court has actually applied strict scrutiny to a
hybrid claim.103 However, of the circuits that have decided cases in
which hybrid claims were asserted,104 with the exception of the
Second and Sixth Circuits, all have recognized the existence of
Id. § 761.03(1)(b).
101
Employment Div., Dep’t. of Human Res. of Or. v. Smith, 494 U.S. 872,
881 (1990).
102
See Crane, supra note 55, at 236 (“The strict scrutiny test of earlier
cases would now be reserved for “hybrid” cases—those involving a combination
of free exercise rights and constitutional rights.”).
103
Leebaert v. Harrington, 332 F.3d 134, 143 (2d Cir. 2003). Hybrid claims
were basically irrelevant from 1993 to 1997 because, during that time, plaintiffs
asserting free exercise claims could do so under the federal RFRA. See infra
Part I.A.
104
The First, Second, Third, Sixth, Seventh, Ninth, Tenth, and D.C.
Circuits have decided cases in which hybrid claims were asserted. See Leebaert
v. Harrington, 332 F.3d 134, 143 (2d Cir. 2003); Civil Liberties for Urban
Believers v. City of Chicago, 342 F.3d 752 (7th Cir. 2003); Tenafly Eruv Ass’n,
Inc. v. Borough of Tenafly, 309 F.3d 144 (3d Cir. 2002); Prater v. City of
Burnside, Kentucky 289 F.3d 417, 430 (6th Cir. 2002); American Family
Assoc., Inc. v. City and County of San Francisco, 277 F.3d 1114 (9th Cir. 2000);
Swanson v. Guthrie Independent School District, 135 F.3d 694, 700 (10th Cir.
1998); EEOC v. Catholic Charities of America, 83 F.3d 455 (D.C. Cir. 1996);
Brown v. Hot, Sexy and Safer Productions, Inc., 68 F.3d 525 (1st Cir. 1995). For
further discussion of the split within in the Circuits regarding the status of
hybrid claims, see Ryan M. Akers, Begging the High Court for Clarification;
Hybrid Rights Under Employment Division v. Smith, 17 REGENT. U. L. REV. 77
(2004-2005).
OREN MACROED CORRECTED 062205.DOC
874
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
such a claim.105 The Second and Sixth Circuits have held that the
Court’s discussion of hybrid claims in Smith is not part of the
Court’s holding, but rather, is merely dicta; the circuits thus
maintain that Smith overruled the Sherbert test for all neutral laws
of general applicability, including those brought as hybrid
claims.106 The circuits that recognize hybrid claims differ in
opinion with regard to whether the claim conjoined with the free
exercise claim must be successful on its own or whether the claim
must simply be one with “a ‘fair probability’ or ‘likelihood’ but
not a certitude of success on the merits.”107
In Swanson v. Guthrie Independent School District, for
105
See Civil Liberties for Urban Believers v. City of Chicago, 342 F.3d 752
(7th Cir. 2003) (recognizing a hybrid claim, but holding that a plaintiff fails to
assert a valid hybrid claim by conjoining a free exercise claim with a meritless
claim). See also Tenafly Eruv Ass’n, Inc. v. Borough of Tenafly, 309 F.3d 144,
165 n.26 (3d Cir. 2002) (recognizing hybrid claims, but noting that the plaintiffs
did not assert such a hybrid rights claim); American Family Assoc., Inc. v. City
& County of San Francisco, 277 F.3d 1114 (9th Cir. 2000) (recognizing hybrid
claims, but holding that the conjoined claim must be colorable, which the
plaintiff’s free speech claim was not); Swanson v. Guthrie Independent School
District, 135 F.3d 694, 700 (10th Cir. 1998) (recognizing hybrid claims, but
holding that in order to succeed on such a claim, a plaintiff must be able to
succeed independently on the claim conjoined with the free exercise claim);
EEOC v. Catholic Charities of America, 83 F.3d 455, 467 (D.C. Cir. 1996)
(recognizing the possibility that the respondents had a valid hybrid claim, but
denying petitioner’s claim on other grounds); Brown v. Hot, Sexy and Safer
Productions, Inc., 68 F.3d 525 (1st Cir. 1995) (recognizing the existence of
hybrid claims, but denying petitioner’s claim because there was no violation of a
privacy right).
106
Leebaert v. Harrington, 332 F.3d 134, 143 (2d Cir. 2003); Prater v. City
of Burnside, Kentucky 289 F.3d 417, 430 (6th Cir. 2002) (holding that Smith
overruled the compelling state interest/least restrictive means test for a neutral
law of general applicability, including those cases in which hybrid claims were
asserted).
107
Swanson v. Guthrie Independent School District, 135 F.3d 694, 700
(10th Cir. 1998) (recognizing the availability of a hybrid claim, but holding that
in order to succeed on such a claim, a plaintiff must be able to succeed
independently on the conjoined constitutional claim). See also Miller v. Reed,
176 F.3d 1202, 1207 (9th Cir. 1999) (citing Thompson v. Anchorage Equal
Rights Comm’n, 165 F.3d 692, 703, 707 (9th Cir. 1999)) (holding that a
conjoined claim does not require a certitude of success).
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 875
example, the Tenth Circuit rejected a hybrid claim brought by
Christian parents who desired that their home-schooled daughter
take classes at a local public school.108 The parents challenged the
local school board’s decision requiring that students be enrolled
either full-time or not at all, and alleged a violation of the Free
Exercise Clause as well as the constitutional right of parents to
direct their children’s education.109 The Tenth Circuit held that,
although “parents have a constitutional right to direct [their
child’s] education, up to a point . . . parents simply do not have a
constitutional right to control each and every aspect of their
children’s education and oust the state’s authority over that
subject.”110 Based on this rationale, the Swanson court found that
the petitioners did not have a valid claim based on their
constitutional right to direct their child’s education.111 The court
held that “it is not sufficient simply to invoke the Free Exercise
Clause as well as another general constitutional claim to trigger the
compelling-interest/narrowly-tailored-rule analysis,” but rather,
there must be a “colorable showing of infringement of recognized
and specific constitutional rights.”112 The Tenth Circuit, in essence,
required that the claim conjoined with the free exercise claim be
one that would succeed independently.113
Conversely, in Miller v. Reed, the Ninth Circuit specifically
held that its test was less stringent than that of other circuits, given
that it did not require a “certitude” that the conjoined claim would
succeed on the merits, but only a “fair probability” or “likelihood”
that such a claim would succeed.114 In Miller, a religious individual
108
Swanson, 135 F.3d at 696-97.
Id. at 697, 699.
110
Id. at 699.
111
Id. at 700 (citing a host of cases in which courts rejected the claims of
parents asserting a constitutional right to direct their child’s education).
112
Id.
113
Id.
114
Miller v. Reed, 176 F.3d 1202, 1207 (9th Cir. 1999) (“We recently held,
to assert a hybrid-rights claim, ‘a free exercise plaintiff must make out a
‘colorable claim’ that a companion right has been violated-that is, a ‘fair
probability’ or a ‘likelihood,’ but not a certitude, of success on the merits.”)
(quoting Thompson v. Anchorage Equal Rights Comm’n, 165 F.3d 692, 703,
109
OREN MACROED CORRECTED 062205.DOC
876
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
who was not a member of an organized religion challenged the
state’s requirement that he submit his social security number in
order to renew his driver’s license.115 The plaintiff claimed that
this requirement interfered with his religious belief because being
identified by a number diminished his identity as an individual and
also that the restriction violated his fundamental right to interstate
travel.116 The Ninth Circuit held that the plaintiff did not articulate
a valid hybrid claim because the conjoining constitutional claim
was “utterly meritless,” given that denying the plaintiff a driver’s
license would not prevent his interstate travel in the same way that
gasoline taxes and toll roads do not violate the right to interstate
travel.117 Although the Miller court distinguished its analysis from
that of more stringent circuits, the court utilized virtually the same
analysis as the Swanson court; it evaluated the conjoining claim
independently and then specifically denied the hybrid claim based
on the weakness of the conjoining claim.118
Both the Ninth and Tenth Circuits’ holdings indicate that,
regardless of whether a probability or a certainty of success is
required, the analysis of a hybrid claim centers on whether the
conjoining claim can survive on its own.119 Requiring a strong or
“colorable” conjoining claim, however, diminishes the utility of a
hybrid claim, as plaintiffs may sue on the conjoining claim
alone.120 As Justice Souter noted in his concurring opinion in
Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, “if a
hybrid claim is one in which a litigant would actually obtain an
exemption from a formally neutral, generally applicable law under
707 (9th Cir. 1999), rev’d on other grounds by 220 F.3d 1134 (9th Cir. 2000)).
115
Miller, 176 F.3d at 1204.
116
Id. at 1204-05 (holding that because the petitioner could still travel
interstate as a passenger that the examined law did not affect the petitioner’s
right to travel, but rather, his operation of a motor vehicle).
117
Id. at 1205-06. The court noted, “Other circuits have adopted similar or
more stringent predicates for a hybrid rights claim.” Id.
118
Id.
119
Swanson, 135 F.3d at 699.
120
Church of the Lukumi Babalu Aye v. City of Hileah, 508 U.S. 520, 567
(1993) (Souter J., concurring) (noting that requiring the conjoining claim of a
hybrid claim to be strong on its own alleviates the need for a hybrid claim).
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 877
another constitutional provision, then there would have been no
reason in what Smith calls the hybrid cases to have mentioned the
Free Exercise Clause at all.”121
The U.S. Supreme Court has yet to decide a hybrid claim. The
Court referenced hybrid claims most recently in 2002, in
Watchtower Bible and Tract Society of New York, Inc. v. Village of
Stratton.122 In Watchtower, a group of Jehovah’s Witnesses
challenged on both free exercise and free speech grounds a village
ordinance requiring that door-to-door canvassers or solicitors
obtain a permit.123 The Court held that it was “unnecessary to
[determine the standard of review] because the breadth of speech
affected by the ordinance and the nature of the regulation make it
clear that the Court of Appeals erred in upholding it.”124 The
question of how strong a conjoining claim must be to support a
valid hybrid claim and obtain strict scrutiny review thus remains
unanswered.
III. SEARCH CLAIMS UNDER THE FOURTH AMENDMENT
In addition to a free exercise claim, a veiled Muslim woman
may assert that requiring her to unveil constitutes an unreasonable
search under the Fourth Amendment. This section analyzes the
components of a Fourth Amendment claim.
The Fourth Amendment protects people “against unreasonable
121
Id. Hileah was not a hybrid case. The petitioners challenged city
ordinances banning ritual sacrifice. Id. The Court invalidated the ordinances,
finding that Smith was inapplicable because the ordinances were not neutral. Id.
122
Watchtower Bible and Tract Society of New York, Inc. v. Village of
Straton, 536 U.S. 150 (2002).
123
Id.
124
Id. at 164. The Court held:
The mere fact that the ordinance covers so much speech raises
constitutional concern. It is offensive not only to the values protected
by the First Amendment, but to the very notion of a free society—that
in the context of every day public discourse a citizen must first inform
the government of her desire to speak with her neighbors and then
obtain a permit to do so.
Id. at 165-66.
OREN MACROED CORRECTED 062205.DOC
878
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
searches and seizures.”125 In his concurring opinion in United
States v. Katz, Justice Harlan set forth a two-prong test for
determining whether an action constitutes a search.126 The test
requires both that the person allegedly searched have a subjective
expectation of privacy in the subject of the claimed search and that
society recognize the person’s expectation as a reasonable one.127
Because the Constitution protects only against unreasonable
searches, once an action is determined to be a search, the Supreme
Court must then determine whether that search was reasonable.128
The Court has determined that individualized suspicion is required
for a search to be deemed reasonable, unless authorities can
establish the existence of “special needs beyond the normal need
for law enforcement.”129
A. Unveiling as a Search
The Supreme Court, in Katz v. United States,130 held that what
a person “seeks to preserve as private, even in an area accessible to
the public, may be constitutionally protected.”131 However, “[w]hat
a person knowingly exposes to the public, even in his own home or
office, is not subject to Fourth Amendment protection.”132 The
Katz Court held that the police’s taping of the petitioner’s phone
calls, made within a public phone booth and taped using a device
attached to the outside of the phone booth, constituted a search.133
125
U.S. CONST. amend. IV.
Katz v. United States, 389 U.S. 347, 362 (1967) (Harlan J., concurring)
(“My understanding of the rule that has emerged from prior decisions is that
there is a twofold requirement, first that a person have exhibited an actual
(subjective) expectation of privacy and, second, that the expectation be one that
society is prepared to recognize as ‘reasonable.’”).
127
Id.
128
City of Indianapolis v. Edmond, 531 U.S. 32 (2000).
129
Edmond, 531 U.S. at 37 (citing Vernonia School Dist. 47J v. Acton, 515
U.S. 646 (1995)).
130
Katz v. United States, 389 U.S. 347 (1967).
131
Id. at 351.
132
Id.
133
Id. at 356-57.
126
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 879
Because the petitioner in Katz had a reasonable expectation that his
conversations were private, the Court reasoned, he “may rely on
the Fourth Amendment.”134
In his concurring opinion in Katz, Justice Harlan articulated a
two-prong test that “emerged from prior decisions” and was to be
applied in cases in which a right to privacy was asserted to
determine whether a search had taken place.135 Justice Harlan’s test
examines 1) whether “a person exhibited an actual (subjective)
expectation of privacy,” and 2) whether “the expectation . . . [is]
one that society is prepared to recognize as reasonable.”136 This
two-prong test was subsequently adopted by the Supreme Court as
the test for determining whether an individual enjoys a reasonable
expectation of privacy in a given case, and therefore, receives
Fourth Amendment protection.137 Thus, under present law, in order
for a veiled Muslim woman to successfully assert a right to privacy
in her face, she must demonstrate not only that she has a subjective
expectation of privacy in her face, but also that society is prepared
to recognize that expectation as reasonable.138
1. The Katz Test: Subjective Expectation of Privacy
In explaining what is meant by something a person “seeks to
preserve as private” the Katz Court cited with approval its earlier
decision in Rios v. United States,139 in which it held that the
134
Id. at 352 (holding that “[o]ne who occupies [a phone booth], shuts the
door behind him, and pays the toll that permits him to place a call is surely
entitled to assume the words he utters into the mouthpiece will not be broadcast
to the world”).
135
Id. at 361 (Harlan J., concurring).
136
Id. (internal quotations omitted).
137
See Hudson v. Palmer, 468 U.S. 517, 525-26 (1984) (holding that the
petitioner failed to satisfy the two-prong test because, although he had a
reasonable expectation of privacy in his jail cell, it was not one society was
prepared to accept as reasonable); see also Smith v. Maryland, 442 U.S. 735,
740 (1979) (holding that the petitioner did not establish either a subjective or an
objective expectation of privacy in the phone numbers he dialed from his
telephone).
138
See also Katz, 389 U.S. at 351 (Harlan J., concurring).
139
Rios v. United States, 364 U.S. 253 (1960).
OREN MACROED CORRECTED 062205.DOC
880
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
admissibility as evidence of a package of heroin that was dropped
on the floor of a taxi cab turned on whether the petitioner dropped
the package before or after the police arrested him.140 If the police
arrested the petitioner without probable cause and thereafter saw
the package of heroin, the search and arrest would be unlawful,
and the package of heroin would be inadmissible.141 However, if
the petitioner held the package in the officers’ view and the
officers then arrested him based upon probable cause, that is, upon
seeing the package of heroin, then evidence of the package would
be admissible.142 The Katz Court cited Rios to emphasize that
when a person makes an effort to preserve something as private, as
the petitioner may have done in the Rios case by hiding the
package of heroin on the floor of the cab, that this “something”
merits constitutional protection.143
Courts have reviewed several cases involving an individual’s
right to privacy in certain physical attributes.144 Because those
courts did not find a subjective expectation of privacy in the
examined physical characteristics, such as one’s voice,
handwriting, hands, and eyes, they did not reach the question of
whether society was prepared to recognize the individuals’
expectations of privacy as reasonable.145
For example, the Sixth Circuit, in United States v. Richardson,
analyzed the right to privacy in one’s hands.146 The Richardson
court held that examining the petitioner’s hands under an
ultraviolet light before arrest and without a warrant did not
140
Id. at 261-62.
Id.
142
Id.
143
Katz, 389 U.S. at 351-52 (citing Rios v. United States, 364 U.S. 253
(1960) (holding that “what [a person] seeks to preserve as private, even in an
area accessible to the public, may be constitutionally protected”).
144
See United States v. Dionosio, 410 U.S. 1 (1978) (analyzing the right to
privacy in one’s voice); United States v. Doe, 457 F.2d 895 (2d Cir. 1972)
(analyzing the right to privacy in handwriting samples); United States v.
Richardson, 388 F.2d 842 (6th Cir. 1968) (discussing the right to privacy in
one’s hands); State v. Shearer, 30 P.3d 995 (Idaho Ct. App. 2001) (analyzing the
right to privacy in one’s eyes.)
145
Id.
146
United States v. Richardson, 388 F.2d 842 (6th Cir. 1968).
141
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 881
constitute a search under the Fourth Amendment.147 However, the
court relied heavily on the fact that the petitioner had agreed to the
search, “gambl[ing] on his ability to convince the officers of his
innocence.”148 The Richardson court did not discuss whether the
petitioner would have had a reasonable expectation of privacy in
his hands had he not voluntarily shown them to the officers.149
In United States v. Dionisio, the Supreme Court considered
whether an individual has a reasonable expectation of privacy in
his voice.150 The Court concluded that there is no reasonable
expectation of privacy in a person’s voice because it is “constantly
exposed to the public” and “repeatedly produced for others to
hear.”151 The Court relied on Katz and determined that a person’s
voice is something that one knowingly exposes to the public;
therefore, it is not subject to Fourth Amendment protection.152
Similarly, in U.S. v. Doe, the Second Circuit held that handwriting
samples could be compelled by subpoena in a grand jury
proceeding because there is “no intrusion into an individual’s
privacy . . . [since] nothing is exposed to the grand jury that has not
previously been exposed to the public at large.”153
State courts have examined similar questions. In State v.
Shearer, the Idaho Court of Appeals rejected a petitioner’s claim
that his right to privacy was violated when he was pulled over by a
147
Id. at 845. In Richardson, FBI agents dusted stolen bank bags with
fluorescein powder, which becomes florescent under ultraviolet light. After the
petitioner and his accomplice retrieved the bags, FBI agents and police dropped
in on the petitioner at work and asked him if they could view his hands under a
light without explaining to the petitioner the purpose of this request.
148
Id.
149
Id.
150
United States v. Dionisio, 410 U.S. 1, 14 (1978) (involving a challenge
on Fourth and Fifth Amendment grounds by two grand jury witnesses who were
held in contempt of court for refusing to provide voice samples).
151
Id.
152
Id. at 14 (citing Katz v. United States, 389 U.S. 347, 351 (1967)).
153
United States v. Doe, 457 F.2d 895, 899 (2d Cir. 1972) (challenging on
Fourth Amendment grounds a judgment of contempt by the Southern District of
New York related to the appellant’s refusal to provide the grand jury with a
handwriting sample).
OREN MACROED CORRECTED 062205.DOC
882
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
police officer and asked to remove his sunglasses.154 The court
held that there is no reasonable expectation of privacy in a person’s
eyes and stated that “taking minimal steps to temporarily conceal a
facial characteristic that is ordinarily and frequently exposed to the
public is, in our view, insufficient to create a legitimate expectation
of privacy.”155
Taken together, these rulings indicate that courts will not
uphold a right to privacy in a feature that is normally in plain view
and that the petitioner generally makes no effort to conceal.156 For
the most part, these decisions rely on the Supreme Court’s
assertion in Dionisio, which provides that “[n]o person can have a
reasonable expectation that others will not know the sound of his
voice, any more than he can expect that his face will be a mystery
to the world.”157 Although these words would seem fatal to a case
asserting a right to privacy in one’s face, the Dionisio Court
qualified its statement by asserting that “[e]xcept for the case of
the rare recluse who chooses to live his life in complete solitude, in
our daily lives we constantly speak and write.”158 Indeed, the Court
concluded its opinion by noting that “nothing [was] being exposed
to the grand jury that [was] not previously . . . exposed to the
public at large.”159 Therefore, it is not clear whether the “rare
recluse” possesses a right to privacy in her handwriting or voice
samples.160 Unlike members of the general public, a recluse
presumably has not exposed the characteristic in question to “the
public at large.”161 Like a recluse, a veiled Muslim woman keeps
154
State v. Shearer, 30 P.3d 995 (Idaho Ct. App. 2001).
Id. at 1000. The Freeman case is easily distinguishable from Shearer
because Freeman constantly wore her veil. Therefore, her head and face were
not “ordinarily and frequently exposed to the public,” unlike the petitioner’s
eyes in Shearer.
156
See United States v. Dionisio, 410 U.S. 1, 14 (1972) (citing United
States v. Doe, 457 F.2d 895, 898-99 (2d Cir. 1972)); State v. Shearer, 30 P.3d
995, 1000 (Idaho Ct. App. 2001).
157
Dionisio, 410 U.S. at 14.
158
Id. (citing United States v. Doe, 457 F.2d 895, 898-99 (2d Cir. 1972)).
159
Dionisio, 410 U.S. at 14.
160
Id.
161
Id.
155
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 883
her face regularly concealed from the public.162 Because no court
has ever ruled on whether there is a privacy interest in the case of a
person who regularly keeps private a physical feature freely
exposed by the general population, such a case would be one of
first impression in the United States.
2. The Katz Test: Expectation of Privacy That Society Is Prepared
to Recognize as Reasonable
The Supreme Court has not set forth a bright line test for
establishing how to evaluate the second prong of Katz, that is,
whether society is prepared to recognize a privacy right as
reasonable.163 However, the Supreme Court has decided several
cases upholding certain privacy expectations as ones that society is
prepared to recognize as reasonable.164 For example, in Minnesota
v. Olson, the Court held inadmissible as evidence a confession
made by an individual who was arrested in a home where he was
staying as an overnight guest after the police had entered without a
warrant and with their guns drawn.165 The Olson Court rejected the
government’s argument that the defendant was not entitled to
Fourth Amendment protection because the place he was staying
was not his home.166 The Court relied on Katz to demonstrate that
Fourth Amendment protection extends beyond one’s home and
focused on the fact that the defendant was an overnight guest in the
searched home.167 The Court explained that society recognizes as
reasonable an expectation of privacy by overnight visitors in a
162
Freeman, 2003 WL 21338619, at *1 (“Plaintiff wears the niqab in front
of all strangers and unrelated Muslim men.”).
163
See Hudson v. Palmer, 468 U.S. 517, 525-26 (1984); Smith v.
Maryland, 442 U.S. 735, 740 (1979). See also Katz v. United States, 389 U.S.
347, 351 (1967) (Harlan J., concurring).
164
See Minnesota v. Olson, 495 U.S. 91 (1990) (holding that an arrest
warrant was required to arrest an overnight guest in the home of a third person);
see also Bond v. United States, 529 U.S. 334 (2000) (involving the manipulation
of a bus passenger’s bag by a law enforcement agent).
165
Olson, 495 U.S. at 94.
166
Id. at 96-99.
167
Id.
OREN MACROED CORRECTED 062205.DOC
884
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
host’s home:
To hold that an overnight guest has a legitimate expectation
of privacy in his host’s home merely recognizes the
everyday expectations of privacy that we all share. Staying
overnight in another’s home is a longstanding social
custom that serves functions recognized as valuable by
society. We stay in others’ homes when we travel to a
strange city for business or pleasure, when we visit our
parents, children, or more distant relatives out of town,
when we are in between jobs or homes, or when we housesit for a friend. We will all be hosts and we will all be
guests many times in our lives. From either perspective, we
think that society recognizes that a houseguest has a
legitimate expectation of privacy in a host’s home.168
Similarly, in Bond v. United States, the Supreme Court
considered whether a search resulted when a government agent
checking a bus for illegal immigrants squeezed a passenger’s bag
located in the bus’s overhead storage bin.169 Finding that society
recognizes as reasonable a passenger’s expectation that his bags
will not be physically manipulated, the Court explained:
When a bus passenger places a bag in an overhead bin, he
expects that other passengers or bus employees may move
it for one reason or another. Thus, a bus passenger clearly
expects that his bag may be handled. He does not expect
that other passengers, or bus employees will, as a matter of
course, feel the bag in an exploratory manner. But this is
exactly what the agent did here. We therefore hold that the
agent’s physical manipulation of the petitioner’s bag
violated the Fourth Amendment.170
The Supreme Court has also provided guidance with regard to
when an expectation of privacy is one that society is not prepared
to recognize as reasonable.171 For example, in Hudson v. Palmer,
168
Id. at 98.
Bond, 529 U.S. at 336 (2000).
170
Id. at 338-39.
171
See Hudson v. Palmer, 468 U.S. 517 (1984) (holding that society is not
prepared to recognize the privacy rights of prisoners in their cells).
169
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 885
the Court considered the case of an inmate who challenged a
random search of his prison cell where his contraband property
was intentionally destroyed.172 The Court held that although
prisoners have a reasonable expectation of privacy in their cells, it
is not one that society is prepared to recognize as legitimate
because “[t]he recognition of privacy rights for prisoners in their
individual cells cannot be reconciled with the concept of
incarceration and the needs and objectives of penal institutions.”173
The Court noted that incarceration is the result of committing a
crime and that its premise is to withhold an individual’s personal
freedoms.174 The Court therefore concluded that a prisoner’s
expectation of privacy in his prison cell is not an expectation that
society is prepared to accept as reasonable.175
This body of Supreme Court case law can be used to determine
on a case-by-case basis whether an expectation of privacy is one
that society is prepared to accept as reasonable, and therefore,
whether a search has occurred.176
B. Unveiling as an Unreasonable Search
The Fourth Amendment to the U.S. Constitution protects only
against unreasonable searches.177 Once a court determines that a
search has occurred, the court must determine whether the search
was reasonable.178 As the Supreme Court held in 2000, in City of
172
Id.
Id. at 526.
174
Id.
175
Id. at 525-26.
176
See Bond v. United States, 529 U.S. 334, 338-39 (2000) (engaging in a
fact-based analysis to determine whether the expectation that law enforcement
would not manipulate a passenger’s bag to discover its contents was one that
society is prepared to accept as reasonable); Minnesota v. Olson, 495 U.S. 91,
95-99 (1991) (engaging in a fact-based analysis to determine whether society is
prepared to accept as reasonable overnight guests’ expectation of protection
against warrantless searches); Hudson, 468 U.S. at 526 (engaging in a fact-based
analysis to determine whether society was prepared to recognize as reasonable
the right of prisoners to protection against unreasonable searches).
177
U.S. CONST. amend. IV.
178
City of Indianapolis v. Edmond, 531 U.S. 32, 37 (2000).
173
OREN MACROED CORRECTED 062205.DOC
886
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
Indianapolis v. Edmond, a search is reasonable if there are “special
needs beyond the normal need for law enforcement” or if the
search is an administrative search with a narrowly limited
purpose.179 In the absence of special needs, individualized
suspicion is required for a search to be considered reasonable.180 In
Edmond, the Court examined Indianapolis’ highway checkpoint
system, whereby cars were stopped and inspected by police from
the outside and sniffed by a drug dog for signs of illegal drug
activity.181 The Court held that police activity of this sort
constitutes a seizure and because such a seizure protects only the
city’s general interest in crime control, it is unreasonable absent
individualized suspicion.182
1. Special Needs
In special needs cases individualized suspicion is not required
for a search to be considered reasonable because, by definition, the
cases are such that “the privacy interests implicated by the search
are minimal, and . . . an important governmental interest furthered
by the intrusion would be placed in jeopardy by a requirement of
individualized suspicion.”183 In stating that the case before it was
not a special needs case, the Edmond Court referenced three cases
in which it had identified special needs.184 Each of the cases cited
by the Court in Edmond involved drug testing that was performed
179
Id. The Court also mentions that in certain circumstances “brief
suspicionless seizures of motorists at fixed Border Patrol checkpoint[s] designed
to intercept illegal aliens” or sobriety checkpoints may also be considered
reasonable. Id. (citing United States v. Martinez-Fuerte, 428 U.S. 543 (1976);
Michigan v. Dept. of State Police v. Sitz, 496 U.S. 444 (1990)). However, such
seizures are not relevant to the discussion of the case of a veiled Muslim
woman.
180
Edmond, 531 U.S. at 37.
181
Id. at 48.
182
Id.
183
Skinner, 489 U.S. at 624.
184
Edmond, 531 U.S. at 37 (citing Vernonia School Dist. 473 v. Acton, 515
U.S. 646 (1995)); Nat’l Treasury Employees v. Von Raab, 489 U.S. 656 (1989);
Skinner v. Railway Labor Executives’ Ass’n, 489 U.S. 602 (1989).
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 887
without individualized suspicion or a warrant.185 In Vernonia
School Dist. 473 v. Acton, the Supreme Court upheld the random
drug testing of student athletes.186 In National Treasury Employees
Union v. Von Raab, the Court upheld the drug testing of employees
who work for U.S. Customs Service and who apply for promotions
to positions that are directly involved with drugs or in which
employees are required as part of their jobs to carry a firearm.187
The Von Raab Court declined to determine whether employees
who handle classified material should be subject to random drug
testing as well and remanded the case to the Fifth Circuit for that
determination.188 In Skinner v. Railway Labor Executives’ Assoc.,
the Court upheld a Federal Railroad Administration regulation
requiring blood and urine tests of employees involved in “major
train accidents” to test for drugs and alcohol.189
The Supreme Court found in all three cases that drug testing
implicated the Fourth Amendment.190 The Court then balanced the
character of the intrusion against the governmental interest
furthered by the intrusion.191 In each case, the Court held that the
intrusiveness of a urinalysis is minimal.192 In determining that the
185
Id.
Vernonia, 515 U.S. at 664-65. The Vernonia holding was expanded in
Board of Educ. of Indep. Schools Dist. No. 92 of Pottawtomie v. Earls, 536 U.S.
822 (2002), in which the Court upheld random drug testing of all students
involved in extracurricular activities. Id.
187
Von Raab, 489 U.S. at 677.
188
Id. at 678.
189
Skinner, 489 U.S. at 607.
190
Vernonia, 515 U.S. at 652; Von Raab, 489 U.S. at 665; Skinner, 489
U.S. at 618.
191
Id.
192
Vernonia, 515 U.S. at 658-59; Skinner, 489 U.S. at 624; Von Raab, 489
U.S. at 672. In Vernonia, the Court cited to Skinner and held that the privacy
interests related to the manner in which the urine was obtained for a urinalysis
were negligible because female students urinated within a stall and male
students urinated in a urinal, but were only viewed from behind. Vernonia, 515
U.S. at 658-59 (citing Skinner, 489 U.S. at 626). Further, the Court held that the
privacy interest in the information that the urinalysis disclosed was minimal as
well because the test only looked for drugs and not any health condition. Id.
(citing Skinner, 489 U.S at 617). Further, the Court held that the required
disclosure of any medications that the students were taking to avoid a false
186
OREN MACROED CORRECTED 062205.DOC
888
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
nature of the intrusion of a drug tests is minimal, the Court in
Vernonia and Von Raab specifically relied on the fact that positive
results of such tests are not reported to law enforcement
authorities.193 In Skinner, which involved blood in addition to urine
testing, the Court cited to its decision in Schmerber v. California,
which held that blood tests are not significant privacy intrusions
because such “tests are commonplace in these days of periodic
physical examinations and experience with them teaches that the
quantity of blood extracted is minimal and that for most people the
procedure involves virtually no risk, trauma or pain.”194
Having found that the privacy interests involved were minimal,
the Court in all three cases found that there were special needs that
outweighed the minimal privacy concern.195 In Vernonia, the Court
held that deterring drug use in school children constituted a special
need.196 Additionally, in Von Raab, the Supreme Court held that
because those working in drug departments or those required to
carry a firearm for the U.S. Customs Service “depend uniquely on
their judgment and dexterity, these employees cannot reasonably
expect to keep from the Service personal information that bears
positive result also was minimal. Id. at 658-59.
193
Vernonia, 515 U.S. at 658-59; Von Raab, 489 U.S. at 663. The Skinner
Court specifically stated that the testing was not for prosecutorial purposes, but
rather, as a way to prevent train accidents. Skinner, 489 U.S. at 620.
194
Skinner, 489 U.S. at 607 (quoting Schmerber v. California, 384 U.S.
757, 770-71 (1966)).
195
Vernonia, 515 U.S. at 661; Von Raab, 489 U.S. at 672; Skinner, 489
U.S. at 607.
196
Vernonia, 515 U.S. at 661. The Court held:
Drugs had not been a major problem in Vernonia schools. In the midto-late 1980’s, however, teachers and administrators observed a sharp
increase in drug use. Students began to speak out about their attraction
to the drug culture, and to boast that there was nothing the school could
do about it. Along with more drugs came more disciplinary problems.
Between 1988 and 1989 the number of disciplinary referrals in
Vernonia schools rose to more than twice the number reported in the
early 1980’s, and several students were suspended. . . . Not only were
student athletes included among the drug users, but, as the District
Court found, athletes were the leaders of the drug culture.
Id. at 648-49.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 889
directly on their fitness.”197 Lastly, the Skinner Court held that
special needs existed, given the significant use of drugs and
alcohol by railway employees.198
2. Limited Purpose Administrative Searches
The Edmond Court also cited three cases in which the Supreme
Court had examined administrative searches conducted without a
warrant.199 In only one of those cases did the Court hold that an
absolute exemption from the warrant requirement was appropriate,
based on the limited nature of the administrative search in
question.200 In New York v. Burger, the Supreme Court upheld a
New York statute permitting authorities to systematically search
197
Von Raab, 489 U.S. at 672. The Court held:
The Government’s compelling interest in preventing the promotion of
drug users to positions where they might endanger the integrity of our
Nation’s borders or the life of the citizenry outweigh the privacy
interests of those who seek promotion to these positions, who enjoy a
diminished expectation of privacy by virtue of the special, and obvious
physical and ethical demands of those positions.
Id. at 679.
198
Skinner, 489 U.S. at 607, n.1 (citing 48 Fed. Reg. 30,726 (1983)).
The FRA noted that a 1979 study examining the scope of alcohol abuse
on seven major railroads found that “[a]n estimated one out of eight
railroad workers drank at least once while on duty during the study
year.” In addition, “5% of workers reported to work ‘very drunk’ or got
‘very drunk’ on duty at least once in the study year,” and “13% of
workers reported to work at least ‘a little drunk’ one or more times
during that period.” The study also found that 23% of the operating
personnel were “problem drinkers,” but that only 4% of these
employees “were receiving help through an employee assistance
program, and even fewer were handled through disciplinary
procedures.”
Id. (internal citations omitted).
199
City of Indianapolis v. Edmond, 531 U.S. 32, 37 (2000) (citing New
York v. Burger, 482 U.S. 691, 700 (1987)); Michigan v. Tyler, 436 U.S. 499,
511 (1978); Camara v. Municipal Court of the City and County of San
Francisco, 387 U.S. 523 (1967)).
200
New York v. Burger, 482 U.S. 691, 700 (1987).
OREN MACROED CORRECTED 062205.DOC
890
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
junkyards without a warrant in order to look for stolen property.201
The Court in Burger found that the owners of commercial property
employed in closely regulated industries have a lesser expectation
of privacy with regard to that property.202 The Court held that even
in closely regulated industries three criteria must be met in order
for warrantless searches to be permitted: 1) “there must be a
‘substantial’ governmental interest that informs the regulatory
scheme pursuant to which the inspection is made”;203 2) “the
warrantless inspection must be ‘necessary to further the regulatory
scheme”;204 and 3) “[t]he statute’s inspection program, in terms of
the certainty and regularity of its application, must provide a
constitutionally adequate substitute for a warrant.”205 The Burger
Court found that the New York junkyard statute fulfilled its threeprong test.206 First, the Court cited to a statement by the governor
of New York approving the statute, in which the governor
emphasized the magnitude of the problem of car theft in the state
and explained that New York had a “substantial interest in
regulating the vehicle dismantling and automobile junk
industry.”207 Second, the Court held that the “regulation of the
201
Id.
Id.
203
Id. at 702 (citing Donovan v. Dewey, 452 U.S. 594, 602 (1981); United
States v. Biswell, 406 U.S. 311, 315 (1972); Colonnade Catering Corp. v.
United States, 397 U.S. 72, 75 (1970)).
204
Burger, 482 U.S. at 703 (citing Dewey, 452 U.S. at 600).
205
Id. at 703.
206
Id. at 708.
207
Id. The Governor stated:
Motor vehicle theft in New York State has been rapidly increasing. It
has become a multimillion dollar industry which has resulted in an
intolerable economic burden on the citizens of New York. In 1976,
over 130,000 automobiles were reported stolen in New York, resulting
in losses in excess of $225 million. Because of the high rate of motor
vehicle theft, the premiums for comprehensive motor vehicle insurance
in New York are significantly above the national average. In addition
stolen automobiles are often used in the commission of other crimes
and there is a high incidence of accidents resulting in property damage
and bodily injury involving stolen automobiles.
Id. (citing Governor’s Message approving L.1979, chs. 691 and 692, 1979 N.Y.
202
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 891
vehicle-dismantling industry reasonably serves the State’s
substantial interest in eradicating automobile theft” because “it is
well established that the theft problem can be addressed effectively
by controlling the receiver of, or market in stolen property.”208
Lastly, the Court held that the statute itself provides for a
“constitutionally adequate substitute for a warrant” because it
informs the operator of a junkyard that searches will be made on a
regular basis and that these searches are “not discretionary acts by
a governmental official.”209 Additionally, the court held that the
“time, place, and scope” of the search are limited because such
inspections can only be made during normal business hours.210
In the two remaining limited purpose administrative search
cases cited by the Edmond Court, the Court also examined
warrantless searches of property.211 In Michigan v. Tyler, the Court
held that no warrant is required for firefighters to enter a building
to fight a fire and that “once in the building, officials may remain
there for a reasonable time thereafter to investigate the cause of the
blaze.”212 The Court noted, however, that additional entries to
investigate a fire require a warrant.213 Further, in Camara v.
Municipal Court of the City and County of San Francisco, the
Court held that warrantless inspections of properties by housing
and public health officials pursuant to San Francisco’s Housing
Code are unconstitutional.214 The above cases instruct that a search
will be deemed unreasonable in the absence of individualized
suspicion unless special needs exist or the search falls into the very
narrow category of a limited purpose administrative search.
Laws 1826, 1826-1827 (McKinney)).
208
Burger, 482 U.S. at 709 (citing 2 W. LAFAVE & A. SCOTT,
SUBSTANTIVE CRIMINAL LAW § 8.10(a) (1986); 2 ENCYCLOPEDIA OF CRIME AND
JUSTICE 789 (Kadish ed. 1983)).
209
Id. at 711.
210
Id. (citing United States v. Biswell, 406 U.S. 311, 315 (1972)).
211
City of Indianapolis v. Edmond, 531 U.S. 32, 37 (citing Michigan v.
Tyler, 436 U.S. 499, 507-09, 511-12 (1978); Camara v. Municipal Court of City
and County of San Francisco, 387 U.S. 523, 534-39 (1967)).
212
Tyler, 436 U.S. at 511.
213
Id.
214
Camara, 387 U.S. at 540.
OREN MACROED CORRECTED 062205.DOC
892
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
IV. FREE EXERCISE AND PROTECTION AGAINST UNREASONABLE
SEARCHES AS A CONJOINED CLAIM
Although Freeman’s lawyer did not so argue, Freeman’s claim
is one that falls into the “hybrid” category of free exercise claims
because Freeman may assert a conjoining constitutional claim that
merits federal attention.215 In addition to her free exercise claim,
Freeman could have argued that the Florida driver’s license statute
was a violation of both her right to free exercise and her right to
Fourth Amendment protection against unreasonable searches. This
section analyzes a claim such as Freeman’s as both a standalone
Fourth Amendment claim, which Freeman might have asserted
independent of any other relief, and as a hybrid claim.
In her case before the Florida Circuit Court, Freeman indeed
asserted a violation of her Fourth Amendment rights; however, the
trial court granted summary judgment in favor of the State on that
claim, holding that Freeman did not have an objective expectation
of privacy in her face.216 As previously noted, there is a split in the
circuit courts with regard to the permissibility of hybrid claims.217
Even assuming the existence of these claims, the Supreme Court
has yet to address how strong a conjoined claim must be to proceed
as a hybrid claim.218 Courts requiring that a conjoining claim be
capable of succeeding on its own render the assertion of free
exercise claims unnecessary, as once a court determines that the
plaintiff has prevailed on her conjoining claim it need not continue
on to analyze the plaintiff’s alternative constitutional claims.219
However, in courts that require that the conjoining claim be
215
See Smith, 494 U.S. at 881 (distinguishing hybrid cases from those to
which the Sherbert test does not apply). See also Crane, supra note 55; see also
supra text accompanying note 85.
216
Brief for Appellant at 44, Freeman v. State, No. 2002-CA-2828, 2003
WL 21338619 (Fla. Cir. Ct. June 6, 2003).
217
See supra note 105 (noting the current status of the law among the
circuits).
218
See supra Part I.B (outlining the present status of hybrid cases).
219
Church of the Lukumi Babalu Aye v. City of Hileah, 508 U.S. 520, 567
(1993) (Souter J., concurring) (noting that requiring the conjoining claim of a
hybrid claim to be strong on its own alleviates the need for a hybrid claim).
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 893
colorable, establishing a standalone Fourth Amendment claim is
the key to the assertion of a viable hybrid claim.220 In those courts,
once the plaintiff demonstrates that she has a colorable Fourth
Amendment claim, the court will analyze the statute under strict
scrutiny, requiring that the state’s compelling interest outweigh the
burden to the plaintiff and that the statute be the least restrictive
means of accomplishing the state’s objective.221
A. Fourth Amendment Claim
For a Fourth Amendment claim to succeed, the individual
asserting such a claim must first demonstrate that a search occurred
and then that the search was unreasonable.222 There is no reported
federal or state case analyzing the Fourth Amendment right of a
Muslim woman who wears a face veil.223 However, cases that have
analyzed the extent of an individual’s Fourth Amendment
protection against unreasonable searches may shed some light on
220
See supra Part I.B.
See Smith, 494 U.S. at 899 (O’Connor J., concurring in part and
dissenting in part) (indicating that the least restrictive means test is appropriate
for determining whether a government regulation of criminal law interferes with
an individual’s right to free exercise of religion). Although, the majority’s
holding in Smith indicates that the least restrictive means test does not apply to
neutral laws of general applicability, it remains the test for hybrid claims, as the
Court specifically excluded those claims from its holding. Id. at 882.
222
See supra Part III.
223
A Westlaw search for “all state and federal cases” using the search
terms “Muslim” and “veil” and “privacy” produced four cases. Ctr. for Nat’l
Sec. Studies v. U.S. Dep’t of Justice, 331 F.3d 918 (D.C. Cir. 2003) (involving
suit under FOIA for information on the detention of people following the
September 11th attacks); Freeman v. State, No. 2002-CA-2828, 2003 WL
21338619 (Fla. Cir. Ct. June 6, 2003); Adsani v. Miller, No. 94 Civ. 9131, 1996
WL 194326 (S.D.N.Y. April 22, 1996) (involving copyright infringement
dispute); State v. Sport and Health Clubs Inc., 370 N.W.2d 844 (Minn. 1985)
(involving discrimination claims in hiring, employment, and promotion based on
religion). Three cases were entirely unrelated to this issue. One was Freeman v.
State. A Westlaw search of all state and federal cases with the terms “Muslim,”
“veil,” and “Fourth Amendment” produced no cases. A search of LexisNexis of
its federal and state cases using the same search terms produced the same
results.
221
OREN MACROED CORRECTED 062205.DOC
894
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
these rights.224
1. A Veiled Woman’s Subjective Expectation of Privacy
According to the test articulated by Justice Harlan in his
concurrence in Katz, in order for a veiled Muslim woman to
successfully assert that requiring her to unveil constitutes a search
under the Fourth Amendment, she must first demonstrate that she
has a subjective expectation of privacy in her face.225 Based on the
analysis of the aforementioned Fourth Amendment cases, which
rely heavily on exposure to the general public of the feature
claimed to be private, it is possible that a court may conclude that a
veiled Muslim woman whose face has not been exposed to the
public has an actual expectation of privacy in her face.226 A woman
who chooses to veil her face protects it from exposure to the
general public.227 Accordingly, a veiled woman has a heightened
expectation of privacy in her face because she seeks to preserve it
as private.228 Unlike cases in which individuals have resisted
orders to produce voice and handwriting samples after an offense
has occurred, a veiled woman chooses to preserve the feature
claimed as private before its production was requested.229 Further,
224
See United States v. Dionosio, 410 U.S. 1 (1978); see also United States
v. Doe, 457 F.2d 895 (2d Cir. 1972); United States v. Richardson, 388 F.2d 842
(6th Cir. 1968); State v. Shearer, 30 P.3d 995 (Idaho Ct. App. 2001).
225
Katz v. United States, 389 U.S. 347, 351 (1967) (Harlan, J., concurring).
226
See supra Part III (detailing the requirements of the Katz test and
distinguishing Freeman’s case from previous cases in which an expectation of
privacy was alleged in a body part).
227
Freeman, 2003 WL 21338619, at *1. “Plaintiff wears the niqab in front
of all strangers and unrelated Muslim men.” Id. (indicating that the plaintiff’s
face is not “ordinarily and frequently exposed to the public”).
228
See Katz, 389 U.S. at 351 (holding that “what [a person] seeks to
preserve as private, even in an area accessible to the public, may be
constitutionally protected”).
229
Compare United States v. Dionisio, 410 U.S. 1 (1978) (involving the
refusal of witnesses to furnish voice samples to a grand jury in an investigation
relating to possible federal statutes prohibiting gambling), with Freeman, 2003
WL 21338610 (reviewing the requirement that veiled Muslim women unveil for
their drivers license photographs).
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 895
in contrast to cases in which the petitioners spoke, wrote, and
walked about freely without attempting to hide their handwriting,
voices, or appearance, a veiled Muslim woman regularly covers
her face and does not expect it to be seen.230 Likewise, unlike an
individual’s eyes that have been temporarily hidden by sunglasses,
a veiled Muslim woman’s face has not previously been exposed to
the public.231 A Muslim woman who chooses to veil has
determined that she is commanded by Allah not to reveal her face
as part of the requirement that she dress modestly.232 Because she
believes that she is choosing to follow the will of God, she dresses
in this manner at all times when she is in public.233 Thus, a veiled
Muslim woman would not reasonably expect the public to see her
230
See United States v. Dionosio, 410 U.S. 1; Freeman, 2003 WL
21338610.
231
See Shearer, 30 P.3d at 1000 (Idaho Ct. App. 2001) (finding no
reasonable expectation of privacy in one’s eyes by a sunglass wearer and
holding that “taking minimal steps to temporarily conceal a facial characteristic
that is ordinarily and frequently exposed to the public is, in our view,
insufficient to create a legitimate expectation of privacy”). But see Freeman,
2003 WL 21338619, at *1 (indicating that the plaintiff’s face is not “ordinarily
and frequently exposed to the public”). Note that in Freeman’s case, Freeman
converted to Islam in 1997 and began veiling at that time. Brief for Appellant at
3, Freeman v. State, No. 2002-CA-2828, 2003 WL 21338619 (Fla. Cir. Ct. June
6, 2003). Therefore, the fact that she at one time did freely expose her face to the
public could be used in argument against Ms. Freeman’s subjective expectation
of privacy in her face.
232
See Statement by Sultaana Lakiana Myke Freeman, May 27, 2003,
available at http://www.aclufl.org/issues/religious_liberty/freemanpersonal_
statement.cfm (last visited Apr. 5, 2005). Ms. Freeman stated:
I discovered veiling to be the ultimate in self-respect and feminism, as
this liberating act sent a clear message that I am not an object of sexual
fulfillment but a person of strong religious conviction. Whether you
believe that the niqab is a requirement of Muslim women or not, the
fact is ? [sic] it is how I have chosen to practice my religion. I wear the
niqab because I believe that according to The Qur’an and Sunnah,
Allah has legislated for the believing woman to dress in this modest
way. Embracing the niqab was a very personal choice, and I thank
Allah for the protection it has afforded me in life, as a woman of faith.
Id.
233
Id. See also Freeman, 2003 WL 21338619, at *1.
OREN MACROED CORRECTED 062205.DOC
896
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
face and could arguably establish a subjective expectation of
privacy in her face under Katz.234
2. A Veiled Muslim Woman’s Objective Expectation of Privacy
The second prong of the Katz test requires that the claimed
privacy right be one that society is prepared to accept as
reasonable.235 The same privacy concerns at issue in Olson and
Bond appear in the case of a veiled Muslim woman.236 Just as
society understands that a person staying with a friend or relative is
entitled to privacy and that a bus passenger’s bags should not be
squeezed for contraband items, so too should society value the
choice that religious people make to dress as their religions
mandate.237 Practicing religion in the United States is, like being
234
See Katz, 389 U.S. at 351 (holding that the petitioner has a right to
privacy in the conversation he has in a phone booth because he is “entitled to
assume that the words he enters into the mouthpiece will not be broadcast to the
world”).
235
Id. at 361 (Harlan, J., concurring).
236
Bond v. United States, 529 U.S. 334 (2000); Minnesota v. Olson, 495
U.S. 91 (1990).
237
See Olson, 495 U.S. at 98 (holding that society is prepared to recognize
a right to privacy of a houseguest in his host’s home). See Bond, 529 U.S. at 339
(recognizing bus passenger’s right not to have his or her bags manipulated as
part of a search for contraband items). See also President George W. Bush,
Remarks by the President at the Islamic Center of Washington D.C. (Sept. 17,
2001) (discussing the fact that Muslim women who wear head coverings should
be treated with respect), available at http://www.whitehouse.gov/news/releases/
2001/09/20010917-11.html (last visited Apr. 16, 2005).
America counts millions of Muslim amongst our citizens and Muslims
make an incredibly valuable contribution to our country. Muslims are
doctors, lawyers, law professors, members of the military,
entrepreneurs, shopkeepers, moms and dads. And they need to be
treated with respect. In our anger and emotion, our fellow Americans
must treat each other with respect. Women who cover their heads in
this country must feel comfortable going outside their homes. Moms
who wear cover must not be intimidated in America. That’s not the
America I know. That’s not the America I value. I’ve been told that
some fear to leave; some don’t want to go shopping for their families;
some don’t want to go about their daily routines because, by wearing
cover, they’re afraid they’ll be intimidated. That should not and will not
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 897
and having an overnight guest, a “longstanding social custom that
serves functions recognized as valuable by society.”238 Indeed, the
Fourth Amendment is implicated when a state’s policy requires
that a Muslim woman unveil because society respects the choice of
Muslim women to follow her religion and wear a veil. Requiring
that a Muslim woman remove her veil for a driver’s license
photograph would be disrespectful to that choice and an intrusion
upon her privacy just as law enforcement’s squeezing a
passenger’s bag intrudes on the privacy of a person’s personal
possessions.239
Further, the case of a veiled Muslim woman is notably distinct
from cases in which prisoners’ expectations of privacy were held
to be unreasonable in light of the unique goals of punitive
confinement.240 The objective of a driver’s license, unlike the
objective of prisons, is not to restrict a person’s privacy, but rather,
to ensure safety on the roads and to enable the state to verify that
individuals on the road have fulfilled certain state requirements.241
stand in America.
Id.
238
Olson, 495 U.S. at 98.
Bond, 529 U.S. at 339. The rationale of the Court in holding that the law
enforcement officer’s squeezing of a bus passenger’s bag violates the Fourth
Amendment was that if society expects individuals to value something as
private, then this expectation applies to law enforcement officers as well. Id.
Additionally, removing a veiled Muslim woman’s veil without her consent
would likely constitute criminal battery in most, if not every, state. See FLA.
STAT. ANN. § 784.03(1)(a)(1) (West 2003). “The offense of battery occurs when
a person: Actually and intentionally touches or strikes another person against the
will of the other.” Id.
240
Hudson v. Palmer, 468 U.S. 517 (1984).
241
See, e.g., FLA. STAT. ANN. § 322.263 (West 2004).
It is declared to be the legislative intent to: (1) Provide maximum safety
for all persons who travel or otherwise use the public highways of the
state. (2) Deny the privilege of operating motor vehicles on public
highways to persons who, by their conduct and record, have
demonstrated their indifference for the safety and welfare of others and
their disrespect for the laws of the state and the orders of the state
courts and administrative agencies. (3) Discourage repetition of
criminal action by individuals against the peace and dignity of the state,
its political subdivisions, and its municipalities and impose increased
239
OREN MACROED CORRECTED 062205.DOC
898
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
Whereas there is no societal expectation of privacy in a prison cell,
society indeed recognizes the privacy rights of members of free
society. Thus, individuals lawfully applying for driver’s licenses
are likely entitled to Fourth Amendment rights.242 Based on this
reasoning, a court could find that a veiled Muslim woman’s
expectation of privacy in her face is one that society is expected to
recognize as reasonable.
3. Unveiling as an Unreasonable Search
If requiring a veiled Muslim woman to unveil is considered a
search, then absent individualized suspicion, special needs, or the
classification of the search as a limited purpose administrative
search, it would be deemed unreasonable, and therefore, a violation
of the Fourth Amendment.243 In the case of a driver’s license
photograph requirement, individualized suspicion is lacking
because such a policy is a broad one that applies to all individuals
seeking driver’s licenses and is unrelated to particularized
suspicion.244 Consequently, a search of a woman’s veil could only
be deemed reasonable if it were characterized as a search related to
special needs or as a limited purpose administrative search.245
Special needs cases are cases in which “the privacy interests
implicated by the search are minimal, and where an important
governmental interest furthered by the intrusion would be placed in
and added deprivation of the privilege of operating motor vehicles upon
habitual offenders who have been convicted repeatedly of violations of
traffic laws.
Id.
242
See Hudson, 468 U.S. at 527 (citing Lanza v. New York, 370 U.S. 139,
143-44) (internal quotations omitted). “A prison shares none of the attributes of
privacy of a home, an automobile, an office, or a hotel room.” Id. This quotation
implies that there is a recognized privacy right in the locations the Court lists
that are representative of free society.
243
City of Indianapolis v. Edmond, 531 U.S. 32, 37 (2000) (listing the
categories of searches that are considered reasonable).
244
See FLA. STAT. ANN. § 322.142(1) (West 2004) (requiring a full-face
photograph for all those seeking a driver’s license).
245
Edmond, 531 U.S. at 37.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 899
jeopardy by a requirement of individualized suspicion.”246
Photographing the face of a Muslim woman who chooses to veil
would likely not fall into the special needs category. First, in
special needs cases, the privacy interests are necessarily
minimal.247 However, the intrusion into a Muslim woman’s veil to
view her face may not be considered minimal; indeed, the nature of
this intrusion is decidedly distinct from the intrusions previously
examined by the Supreme Court.248 For example, required
unveiling is unlike urine testing, during which an attendant hears a
person urinating or sees the back of a man while he is urinating. As
noted by the Supreme Court in Vernonia, people often use public
bathrooms, where the sounds of their excreting urine can be heard
by others.249 Therefore, requiring that a person supervise students
or employees by hearing them urinate is not a substantial invasion
of the students’ or employees’ privacy.250 However, in the case of
a veiled Muslim woman, the woman’s face is never unveiled to the
public,251 and therefore, the invasion of her privacy is substantial.
The fact that the driver’s license photograph requirement applies to
the population as a whole does not diminish the fact that the
driver’s license requirement violates a veiled Muslim woman’s
Fourth Amendment right. With respect to society generally, the
vast majority of individuals have no privacy interest in their faces,
and therefore, the driver’s license requirement is valid as applied to
them.252 However, because veiled Muslim women have a uniquely
significant expectation of privacy in their faces, required unveiling
246
Skinner v. Railway Labor Executives Ass’n, 489 U.S. 602, 624 (1989).
Id.
248
Vernonia School Dist. 473 v. Acton, 515 U.S. 646 (1995); Nat’l
Treasury Employees v. Von Raab, 489 U.S. 656 (1989); Skinner v. Railway
Labor Executives Ass’n, 489 U.S. 602 (1989).
249
See Vernonia, 515 U.S. at 658. “[The conditions of urine sample
collection] are nearly identical to those typically encountered in public
restrooms, which men, women, and especially schoolchildren see daily.” Id.
250
Id. “Under such conditions, the privacy interests comprised by the
process of obtaining the urine sample are in our view negligible.” Id.
251
See supra note 162 and accompanying text.
252
See supra Part III.A.1. (discussing a person’s expectation of privacy in a
physical attribute).
247
OREN MACROED CORRECTED 062205.DOC
900
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
is a great invasion that is unjustified by the state’s asserted interest
in safety. Moreover, while urinalysis tests can be designed so that
they do not detect any health information about the person being
tested other than whether he or she has used illicit drugs,253 the
photograph of a veiled Muslim woman is a revelation of her face—
the very characteristic she seeks to protect against exposure.254 The
case of veiled Muslim women is similarly distinct from specials
needs cases with regard to the degree to which the searched
individual’s private information is revealed publicly. In special
needs drug testing cases, individuals subjected to testing may be
required to reveal to testers certain illnesses for which they are
being medicated to avoid a false positive results; in such cases, the
exposure of this information can be limited to the testers.255
Further, this information may be completely anonymous, as the
testers who perform urinalysis tests and see the students’ or
employees’ forms may have never seen the subjects of the tests
themselves and may have no additional information about the
tested individuals.256 In the case of a veiled Muslim woman,
however, because a driver’s license will certainly contain both the
woman’s photograph and name, anonymity cannot serve to protect
the woman from an association with the private characteristic she
has revealed.257
In determining the existence of a special need, the Supreme
Court, in Vernonia, Von Raab, and Skinner, identified an existing
problem among the class of people upon which the states whose
policies were in question sought to impose a search.258 Therefore,
253
See Vernonia, 525 U.S. at 658.
See Freeman, 2003 WL 21338619, at *1.
255
See id. (stating that respondent student could have requested that the
medical information that he disclosed only be viewed by the laboratory
performing the test and not by his coaches or teachers).
256
Vernonia, 515 U.S. at 658-59.
257
See Freeman, 2003 WL 21338619, at *1.
258
Vernonia, 515 U.S. at 661 (noting a problem among high school athletes
taking drugs); Von Raab, 489 U.S. at 672 (noting the problem that the people it
sought to test for drugs had access to drugs and/or weapons); Skinner, 489 U.S.
at 607 (noting the problem of railroad accidents resulting from employees
operating under the influence of drugs and alcohol).
254
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 901
in order to establish that a special need exists in the case of a veiled
Muslim woman, a state would have to identify a problem among
this class that compels such a search.259 There was no evidence
mentioned in the Freeman case that among such a class of women
there is a particularly high incidence of driver’s license fraud or
even that identifying women who have been pulled over has been a
problem.260 Therefore, it is unclear that a special need exists.
Additionally, in order for required unveiling to constitute a
special needs case there would have to be a special need beyond
the normal need for law enforcement.261 It is unclear how the
state’s interest would fit into the category of special needs, as it
appears that the state’s goals are directly tied to routine law
enforcement. As in Edmond, where the state’s interest in
controlling illegal drug activity was viewed as part of the normal
need for law enforcement,262 in this case, protecting against fraud
and identifying drivers is similarly part of law enforcement
activity. Further, in contrast to Vernonia and Von Raab, where the
results of positive drug tests were not given to law enforcement
authorities, in this case, the state license database in Freeman was
specifically maintained for a law enforcement purpose in order to
assist police officers in doing their jobs.263
Lastly, the case of a veiled Muslim woman likely does not fall
into the very limited category of administrative searches that are
permissible absent individualized suspicion. First, the only case the
Edmond court cited that blanketly allowed administrative searches
applied only to commercial property, in which people have a lower
expectation of privacy.264 In Burger, the object of the search was a
259
Special needs necessitates that “an important governmental interest
furthered by the intrusion would be placed in jeopardy by a requirement of
individualized suspicion.” Skinner, 489 U.S. at 624.
260
See Freeman, 2003 WL 21338619.
261
Skinner, 489 U.S. at 624.
262
Edmond, 531 U.S. at 48.
263
Freeman, 2003 WL 21338619, at *4 (finding Florida’s compelling state
interest in the driver’s license requirement is promoting safety and security,
combating crime, and protecting interstate commerce).
264
Burger, 482 U.S. at 699 (citing Donovan v. Dewey, 452 U.S. 594, 59899 (1981)) (holding that “[a]n expectation of privacy in commercial premises,
OREN MACROED CORRECTED 062205.DOC
902
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
junkyard that likely had little personal value to its owner and
offered a reduced expectation of privacy, given that junkyards are
frequently used to conduct illegal activity in the dismantling and
selling of parts from stolen vehicles.265 Moreover, in Tyler, the
Court held that fire investigators may only enter a building without
a warrant when the building is burning and the primary purpose of
the entry is to put out the fire.266
In both Tyler and Camara, the Court demonstrated that there is
a very high expectation of privacy in one’s home and, absent an
emergency situation, the Court will not allow a warrantless
search.267 A veiled Muslim woman has a significant expectation of
privacy in her face similar to the expectation of privacy that one
has in his or her home.268 Therefore, she cannot be compelled to
unveil in a non-emergency situation absent individualized
suspicion.269 The state may not circumvent this requirement by
analogizing the woman’s expectation of privacy to that enjoyed by
the individuals in the administrative search cases reviewed by the
Court because, in the case of a veiled Muslim woman, the thing
she expects to keep private is a physical feature that by nature is
highly personal, rather than commercial property, which is by
definition impersonal. For this reason, it appears that the search of
a Muslim woman’s veil may be characterized neither as a special
needs case nor a limited purpose administrative search.
B. Application of the Hybrid Strict Scrutiny Test
Were a veiled Muslim woman to prevail in her claim that the
state’s driver’s license photograph requirement violated her Fourth
Amendment rights, she might be able to assert a hybrid claim,
which would elevate the level of review of her claim to strict
however, is different from, and indeed less than, a similar expectation in an
individual’s home”).
265
Id.
266
Michigan v. Tyler, 436 U.S. 499, 511 (1978).
267
See id. at 511; Camara v. Municipal Court of the City and County of
San Francisco, 387 U.S. 523 (1967).
268
See infra Part IV.A.1.
269
Id.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 903
scrutiny.270 This section analyzes a hybrid claim of this nature
through a balancing of the burden that required unveiling imposes
upon a Muslim woman who veils and the state interests asserted in
Freeman.271
1. Substantial Burden
As the Supreme Court held in Sherbert, the test of substantial
burden looks to whether “the purpose or effect of a law is to
impede the observance of one or all religions or is to discriminate
invidiously between religions that law is constitutionally invalid
even though the burden may be characterized as only indirect.”272
In describing the burden upon a Sabbatarian imposed by the state’s
disqualification of those who did not accept employment offers
from the receipt of unemployment benefits, the Sherbert Court
noted:
The [lower court’s ruling upholding the unemployment
compensation policy] forces her to choose between
270
See, e.g., Swanson v. Guthrie Indep. School Dist., 135 F.3d 694, 700
(10th Cir. 1998) (recognizing the compelling interest test as the appropriate test
for a hybrid claim).
271
Recently there have been a number of bills in the House, including the
Real ID Act of 2005 which passed in the House and was referred to Senate
Committee on the Judiciary on February 17, 2005, seeking to set federal
standards for state driver’s licenses. Real I.D. Act of 2005, H.R. 418, 109th
Cong. (2005). See, e.g., An Act Making Emergency Supplemental
Appropriations for Defense, the Global War on Terror, and Tsunami Relief, for
the Fiscal Year Ending September 30, 2005, and for Other Purposes, H.R. 1268,
109th Cong. (2005) (version including driver’s license provisions referred to
Senate Subcommittee). Driver’s License Security and Modernization Act, H.R.
368, 109th Cong. (referred to House Subcommittee on Immigration, Border
Security, and Claims March 2, 2005), available at thomas.loc.gov. These bills
have yet to pass in the Senate and become law. However, were Congress to pass
a national standard for state driver’s licenses this analysis would not change
because this note discusses whether an exception should be made to driver’s
license laws requiring fullface photographs for veiled Muslim women. The
question of whether driver’s laws requiring a full face photographs should be put
in place for the general citizenry or whether there should be a federal scheme for
state licenses is outside the scope of this Note.
272
Id. at 404 (citing Braunfeld v. Brown, 366 U.S. 599, 607 (1961)).
OREN MACROED CORRECTED 062205.DOC
904
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
following the precepts of her religion and forfeiting
benefits on the one hand, and abandoning one of the
precepts of her religion in order to accept work, on the
other hand. Governmental imposition of such a choice puts
the same kind of burden upon the free exercise of religion
as it would a fine imposed against appellant for her
Saturday worship.273
The burden imposed by requiring that a veiled Muslim woman
unveil for her driver’s license photograph is tantamount to the
burden imposed in Sherbert because, in both cases, the religious
individuals must choose between following their religions and
receiving a state benefit.274 As in Sherbert, a policy requiring
exposure by a woman who believes that showing her face is
prohibited by her religion, as the court found that Freeman does,275
forces the woman to either forgo a driver’s license, and therefore
lose the privilege of driving, or to decide not to follow her religion
and be awarded a driver’s license.276 A Muslim woman who
chooses to veil does so because she believes that this is what Allah
requires.277 She thus is in the same position as the petitioner in
273
Sherbert, 374 U.S. at 404.
Id.
275
Freeman, 2003 WL 21338619, at *2.
276
Sherbert, 374 U.S. at 404. The Court noted:
The ruling [of the lower court in Sherbert, denying the appellant
unemployment benefits because she refused to accept work that
required that she work on Saturday] forces her to choose between
following the precepts of her religion and forfeiting benefits, on the one
hand, and abandoning one of the precepts of her religion in order to
accept work, on the other hand. Governmental imposition of such a
choice puts the same kind of burden upon the free exercise of religion
as would a fine imposed against appellant for her Saturday worship.
Id. Driving is necessary to Freeman’s lifestyle. See State: Terrorists May Benefit
if Veiled Muslim Woman Gets License, Fox News, available at
http://www.foxnews.com/story/0,293388410,00.html. “After the hearing
Freeman complained that without a license, she can’t even drive to the store to
buy diapers for her six-month old son.” Id.
277
See Statement by Sultaana Lakiana Myke Freeman (May 27, 2003),
available at http://www.aclufl.org/issues/religious_liberty/freemanpersonal_
statement.cfm.
274
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 905
Sherbert of either following a tenet of her religion or receiving a
state benefit.278 Therefore, under the Sherbert test, required
unveiling would likely constitute a substantial burden to veiled
Muslim women who are required to take full-face photographs in
order to obtain driver’s licenses.
2. Compelling State Interest
The Freeman court upheld the driver’s license photograph
requirement as applied to veiled Muslim women on the basis that
the state has a compelling interest in the statute because it
promotes public safety and protects against fraud.279 On closer
examination, however, there are flaws in many of the arguments
favoring safety and security upon which the court relied.
a. Speedily Identifying Pulled-Over Drivers
Among the interests asserted by the state and accepted by the
Freeman court as compelling was the state’s interest in speedily
identifying pulled-over drivers.280 It is unquestionable that the state
has an interest in identifying pulled-over drivers. However, it is not
clear that requiring that Muslim women unveil for their driver’s
license photographs will, in actuality, help to achieve that goal. As
discussed, required unveiling may constitute a search for Fourth
Amendment purposes, and therefore, absent individualized
suspicion, police officers will be unable to compel a veiled Muslim
woman to remove her veil once they have pulled her over so that
they may match her face with the photograph on her driver’s
license.281 Moreover, because driver’s license pictures are often
unflattering and many people may change in appearance
278
Sherbert, 374 U.S. at 404.
Freeman, 2003 WL 21338619, at *7.
280
Id.
281
See supra Part III.B. (discussing how required unveiling may constitute
an unreasonable search under the Fourth Amendment). See also Knowles v.
Iowa, 525 U.S. 113 (1998) (holding that the full search of a vehicle after the
driver received a citation for speeding absent probable cause violated the Fourth
Amendment).
279
OREN MACROED CORRECTED 062205.DOC
906
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
subsequent to taking such a picture, identifying an individual by
her driver’s license picture is not a certainty. Strict scrutiny also
requires that the state’s method constitute the least restrictive
means of accomplishing its ends.282 In this instance, a less
restrictive means of furthering the state’s interest would be to grant
these woman an exception to the driver’s license photograph
requirement while adding an additional requirement that those
women carry with them when they drive certain documents, such
as a birth certificate or a social security card, verifying their
identity as the person granted the driver’s license. Because the
state has not used the least restrictive means of furthering its goal,
the state’s interest in speedily identifying drivers does not appear
to outweigh the burden the requirement imposes upon a veiled
Muslim woman.
b. To Protect against Driver’s License Fraud
The Florida court in Freeman also found a compelling state
interest in the use of driver’s license photographs to protect against
driver’s license fraud.283 Although a full-face photograph may
assist in the prevention of fraud in the case of an unlicensed driver
who borrows the driver’s license of a licensed driver,284 the
likelihood of such an instance of fraud is extraordinarily rare. Most
people who are driving have valid licenses and have no need to use
another’s.285 Moreover, were an individual to drive illegally
282
See Thomas v. Review Bd. of Indiana Employment Sec. Div., 450 U.S.
707, 718 (1981) (holding that “the state may justify an inroad on religious
liberty by showing that it is the least restrictive means of achieving some
compelling interest”).
283
Freeman, 2003 WL 21338619, at *7.
284
Id. at *4 (outlining the state’s argument that the purpose of a driver’s
license photographs is for speedy identification and to combat fraud).
285
Because driver’s license fraud and driving without a license can only be
documented if the perpetrators are caught, it is difficult to find any statistics on
the frequency with which driving without a license occurs. However, the
statistics below were somewhat indicative of the frequency with which people
drive without a valid driver’s license. Roughly eighty-seven percent of those
driving in fatal crashes have a valid driver’s license. See AAA Foundation Study
on Unlicensed Drivers, Table A.1, License Status of Drivers Involved in Fatal
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 907
without a license, it is unlikely that such a person would bother to
locate another person’s valid license and, if pulled over, attempt to
pass it off as her own.286 Further, if an illegal driver decided to use
another individual’s license as her own when pulled over, she
would likely borrow the license of someone with similar features,
given the photograph requirement. It is no more likely that women
who choose to veil will “share” their driver’s licenses than friends
or family members who look alike will “share” theirs.287
Moreover, full-face license photographs will only prevent fraud
in relation to driving if there is a reasonable suspicion that the
driver has committed another violation. The Supreme Court held in
Delaware v. Prouse that it is a violation of the Fourth Amendment
for police officers to randomly pull over drivers on the highway
without “an articulable and reasonable suspicion that a motorist is
unlicensed or that an automobile is unregistered or that either the
vehicle or an occupant is otherwise subject to seizure for a
violation of the law.”288 The Prouse Court held that the state’s
interests in public safety were not sufficiently furthered by the
Crashes in the United States 1993-1999 (June 2000), available at
http://www.aafoundation.org/pdf/UnlicensedToKill2.pdf. Of the 13.5 percent of
driver’s without valid license, only 3.6 percent have never been issued a driver’s
license. Id. The rest have driver’s licenses that are either suspended, revoked,
canceled, or expired. Id. Because this study is composed of drivers involved in
fatal crashes, it is likely that the percentage of drivers on the road with valid
licenses is even higher since it is more likely that those without a valid license
will be involved in a fatal accident since such drivers either never fulfilled the
license criteria or had their licenses taken away because of a tendency to commit
driving infractions. Regardless, the study shows that the vast majority of drivers,
86.5 percent, have a valid license. See Delaware v. Prouse, 440 U.S. 648, 660
(1979) (holding that the state’s interests in public safety were not sufficiently
furthered by the chance that the individual whom law enforcement officers
chose to pull over would in fact be in violation of the law).
286
See Prouse, 440 U.S. at 660 (1979) (holding the state’s interests in
public safety were not sufficiently furthered by the chance that the individual
whom law enforcement officers chose to pull over would in fact be in violation
of the law).
287
The AAA Foundation Study on Unlicensed Drivers, supra note 285,
makes no mention of a danger of people with valid licenses “lending” their
licenses to unlicensed drivers.
288
Prouse, 440 U.S. at 663.
OREN MACROED CORRECTED 062205.DOC
908
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
chance that the individual whom law enforcement officers
randomly chose to pull over would in fact be in violation of the
law.289 The court explained:
It seems common sense that the percentage of all drivers on
the road who are driving without a license is very small and
that the number of licensed drivers who will be stopped in
order to find one unlicensed operator will be large indeed.
The contribution to highway safety made by discretionary
stops selected from among drivers generally will therefore
be marginal at best. . . . Much of the same can be said about
the safety aspects of automobiles as distinguished from
drivers . . . .
....
. . . The marginal contribution to roadway safety possibly
resulting from a system of spot checks cannot justify
subjecting every occupant of every vehicle on the roads to a
seizure-limited in magnitude compared to other intrusions
but nonetheless constitutionally cognizable—at unbridled
discretion of law enforcement officials.290
Given that drivers can only be pulled over based on an
articulable suspicion and not at random, a veiled Muslim woman
cannot be pulled over unless she has committed a driving
infraction or is driving a vehicle that is unlicensed or subject to
seizure.291 In such a case, the need for a positive full-face
identification would be significantly lessened because the person
who committed the violation would be in law enforcement
officer’s presence, and therefore, the need to identify the individual
in order to ascertain whether the driver’s identity matches that of
the individual for whom the police are searching is moot because
the offender has been caught.
Further, as the Prouse Court observed, the percentage of
drivers that are unlicensed is small, making the probability that an
unlicensed driver would be discovered based on a random check
289
290
291
Id. at 660.
Id.
Id. at 663.
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 909
minimal.292 Likewise, because of the small number of Muslim
women who choose to wear a full-face veil,293 it is improbable that
this minute percentage of veiled Muslim women will take
advantage of the fact that they are not required to take full-face
pictures and craft ways to commit criminal acts in which they
benefit from the fact that they do not have driver’s licenses with
full-face photographs. As articulated in Prouse, the minimal risk of
unlicensed individuals driving on state roads does not justify the
institution of a police policy of pulling over drivers at random.294
For the same reason, the minimal risk that a veiled Muslim woman
will allow another veiled Muslim woman to use her driver’s
license as her own is insufficient to justify invading the privacy
rights of all veiled Muslim women who apply for driver’s licenses.
Therefore, the state’s interest in protecting against driver’s license
fraud is not sufficiently compelling to outweigh the burden that an
unveiling requirement places upon a veiled Muslim woman.
c. To Protect against Identity Theft
The Freeman court noted an additional compelling state
interest in the state’s use of driver’s license photos in that licenses
are commonly used as form of identification.295 However, as
292
Id. at 660.
See Freeman, 2003 WL 21338619, at *2. “[M]ost Muslims do not veil
to the extent the plaintiff does, and that she is in a small minority of Salfeeha
Muslim women who refuse to remove their veils when they have their pictures
taken for identification.” Id. There are an estimated 6 to 7 million Muslims in
the United States. See http://www.cair-net.org/asp/populationstats.asp (last
visited Nov. 9, 2004). Presumably, half of the Muslim population, around 3.5
million, are women. Only a small minority of these women wear the full face
veil. Even if ten percent wear the full face veil, when in reality it is most
probably a much smaller percentage, this would be 350,000 women. This
number is roughly .01 percent of the population of the United States.
294
Prouse, 440 U.S. at 660.
295
Freeman, 2003 WL 21338619, at *6 (finding that protecting interstate
commerce from widespread identity theft and fraud is a compelling state
interest.). But see Freeman, 2003 WL 21338619, at *2, regarding the small
percentage of women who wear the full-face veil. The fact that such a small
number of individuals choose to wear the veil and would require an exception
293
OREN MACROED CORRECTED 062205.DOC
910
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
Freeman noted, Florida’s Motor Vehicles Statute describing the
legislative intent for driver’s licenses makes no mention that one of
the purposes of a driver’s license is to serve as identification.296
Although as a general state interest it seems sound that private
industry should be able to have a uniform policy regarding what it
considers valid forms of identification, the state may maintain this
policy while still allowing for an exception for a small minority of
its residents.297 As the Eighth Circuit stated in Quaring v.
Peterson, “the state may still achieve its interest . . . because
people may freely refuse to do business [with the respondent] if
she is unable to present adequate identification.”298 Because
from the full-face photograph requirement negates the Freeman court’s
argument that such an exception would lead to “widespread abuse.” Id. at *6
(emphasis added).
296
FLA. STAT. ANN. § 322.263 (West 2003) states:
It is declared to be the legislative intent to: (1) Provide maximum
safety for all persons who travel or otherwise use the public highways
of the state. (2) Deny the privilege of operating motor vehicles on
public highways to persons who, by their conduct and record, have
demonstrated their indifference for the safety and welfare of others and
their disrespect for the laws of the state and the orders of the state court
and administrative agencies.
Id.
297
In fact, it does not seem as though private industry is entirely reliant on
driver’s licenses as forms of identification. See Identification (ID) Requirements
for GRE Tests, at http://www.gre.org/idreq.html (listing a driver’s license as a
valid form of identification but stating that if the license lacked a photograph it
is not valid). A search of a number of bank websites using the term
“identification” uncovered no statement by any bank indicating what form of
identification it will consider valid. See Citizens Bank, Important Information
About Online Security, at http://www.citizensbank.com/misc/online_security.
asp (indicating “[w]hen you call us, come to a branch or visit us online, we will
ask you for some information to verify your identity”). See also
http://citibank.com (search for term “identification” came up with no relevant
results). See also http://fleet.com/home.asp (search for term “identification”
came up with no relevant results). See also http://www.wau. com/servlet/
wamu/index.html (search for term “identification” came up with no relevant
results).
298
Quaring, 728 F.2d at 1127 (upholding the right of religious Christians
who believe that photographs are forbidden graven images to be exempted from
the state’s driver’s license photo requirement). See also supra note 293
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 911
individuals do not have a Fourth Amendment right to protection
against unreasonable searches by private entities, a business can
require any reasonable type of identification it desires in restricting
access to its services.299 Additionally, the fact that driver’s licenses
are commonly used as valid forms of identification by airlines and
other entities does not limit the ability of states to formulate
exceptions to their driver’s license requirements. The Federal
Aviation Administration (FAA) and others may determine for
themselves the types of identification that are required for
individuals to board planes or make use of other private services;
at the same time, a state may choose not to burden its religiously
observant citizens by allowing them an exemption from the
driver’s license photo requirement.300
Moreover, if the purpose of the driver’s license photograph
requirement is to protect individuals who wear a full-face veil from
identity fraud, this legitimate state interest would be outweighed by
the burden the requirement imposes on the very group it aims to
protect. Therefore, a court may find that a state cannot demonstrate
that its legitimate interest in speedy identification and the
prevention of fraud outweighs the burden the photograph
requirement imposes upon veiled Muslim women. It thus is
possible that a court may find that a state is required to grant an
exception from the driver’s license photograph requirement to
veiled Muslim women.
(discussing the small number of Muslim women who veil and, therefore, would
require an exemption demonstrating the minute effect such an exemption would
have on businesses).
299
Lebron v. National R.R. Passenger Corp., 513 U.S. 374, 411-12 (1995).
“The conduct of a private entity is not subject to constitutional scrutiny if the
challenged action results from the exercise of private choice and not from state
influence or coercion.” Id.
300
Note that although the FAA’s website contains a great deal of
“Passenger Information,” including “Airline Contact Information,” “Baggage
Size Requirements” “Check Airport Status,” “Passenger Health and Safety
Information,” “Travel Tips,” “Using Child Safety Seats,” and “Wait Times at
Airport Security Checkpoints,” it does not state what it considers valid forms of
identification. See http://www.faa.gov/passengers/index.cfm (last visited Apr.
16, 2005).
OREN MACROED CORRECTED 062205.DOC
912
6/22/2005 3:49 PM
JOURNAL OF LAW AND POLICY
CONCLUSION
Although there is no federal protection apart from rational
basis review for an individual claiming that a neutral law of
general applicability interferes with her free exercise of religion,
veiled Muslim women may still argue violations of their federal
constitutional rights to free exercise of religion and Fourth
Amendment protection through a hybrid claim.301 However,
because an argument that exposing a veiled Muslim woman’s face
is a violation of the Fourth Amendment is revolutionary302 and
because no circuit court has yet applied strict scrutiny based on a
hybrid claim,303 it seems unlikely that a person in Freeman’s
position will find recourse in a federal court. Even if the courts are
unwilling to provide veiled Muslim women with a remedy,
however, state legislatures retain the authority and the
responsibility to provide accommodations for the religiously
observant within their jurisdictions.304 The seemingly negative
reception of Freeman’s case by the national media and the
American public suggests that legislatures may choose not to carve
out an exception for such a marginalized group.305 However, in
301
See supra Part I (outlining the present state of free exercise claims).
See supra Part II.A (analyzing a Muslim woman’s Fourth Amendment
right to privacy in her face).
303
Part I.B (outlining the status of hybrid claims).
304
See Smith, 494 U.S. at 890 (holding it is up to the “political process” to
protect the interests of individuals whose religious practice is interfered with by
a neutral law of general applicability).
305
Debbi Gardiner, Fla. Muslims See Veil Case as Distraction, BOSTON
GLOBE, June 15, 2003, at A12 (citing to Muslims who criticize Ms. Freeman for
putting this issue into the spotlight and feel that it reflects poorly on Muslims);
Susan Taylor Martin, A Fight for Religion or Something More, ST. PETERSBURG
TIMES, June 15, 2003, at A2 (“Still you have to wonder. Why would someone
who is fighting so hard to protect one basic right-freedom of religion-adopt the
dress code of an Islamic sect that has denied right to so many women in
Afghanistan and Saudi Arabia.”); Gloria Kaplan Sulkin, Driver’s Photos, CHI.
TRIB., June 13, 2003, at 22 (arguing in a letter that “sanity has prevailed in the
case of Sultaana Freeman”); License Is Unveiled; Allowing Woman to Hide
Face in Identification Photo Would Have Been Foolish, THE COLUMBUS
DISPATCH, June 10, 2003, at A10; License Must Do Its Job, SUN-SENTINEL
(Fla.), June 3, 2003, at A18.
302
OREN MACROED CORRECTED 062205.DOC
6/22/2005 3:49 PM
MUSLIM WOMEN AND DRIVER’S LICENSE PHOTOS 913
denying veiled Muslim women an exception to driver’s license
photo requirements, legislatures make broader statements about the
value of religious freedom and their attitudes toward minority
religious groups. In order to promote religious freedom and
accommodate the religiously observant, legislatures should
examine the actual motivations behind what seem to be general
policy requirements and determine whether state goals truly
necessitate requiring religious minorities to forgo sacred practices
or lose state benefits.
RIES MACROED CORRECTED 053105.DOC
6/6/2005 1:43 PM
DUTY-TO-PROTECT CLAIMS BY INMATES
AFTER THE PRISON RAPE ELIMINATION
ACT
David K. Ries∗
Roderick Johnson was on probation in Texas for a nonviolent
burglary offense.1 In January 2000, Johnson’s probation was
revoked and, within nine months, he was transferred to a
maximum-security prison.2 The prison officials responsible for
Johnson’s cell assignment there “knew that Johnson was
homosexual and possessed an effeminate manner,”3 but placed him
in the prison’s general population after telling him “‘we don’t
protect punks on this farm.’”4 Soon afterward a prison gang
asserted ownership over Johnson and forced him into daily sex
acts.5 Throughout his eighteen-month stay in prison, Johnson
∗
Brooklyn Law School Class of 2006; B.A. Vassar College, 1998. I thank
the members of the Journal of Law & Policy as well as the various practitioners
who took the time to answer my questions. Special thanks go to all those who
read drafts of this Note: Ursula Bentele, Liz Budnitz, Eve Cary, Matt Keller,
Claire Kelly, Skye Phillips, Kathryn Razin, and Cory Shindel. Lastly, thanks to
Jean Kaminsky, for her love and support.
1
Roderick Keith Johnson v. Gary Johnson, 385 F.3d 503, 512 (5th Cir.
2004).
2
Id.
3
Id.
4
Id.
5
Adam Liptak, Ex-Inmate’s Suit Offers View into Sexual Slavery in
Prisons, N.Y. TIMES, Oct. 16, 2004, at A1.
“The Crips already had a homosexual that was with them,” Mr.
Johnson explained. “The Gangster Disciples, from what I understand,
hadn’t had a homosexual under them in a while. So that’s why I was
automatically, like, given to them.” According to court papers and
[Johnson’s] own detailed account, the Gangster Disciples and then
other gangs treated Mr. Johnson as a sex slave. They bought and sold
915
RIES MACROED CORRECTED 053105.DOC
916
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
“sought help from guards, filed numerous ‘life-endangerment’
forms, and wrote letters to prison administrators,” but he remained
in the facility’s general population.6 During those eighteen months,
Johnson was passed among various prison gangs, and rape became
a routine part of his prison life.7
Prison rape, disturbingly a running joke in popular culture,8 is
conservatively estimated to occur 12,000 times a year and affect
nearly thirteen percent of the nation’s prisoners.9 In 1994, in
Farmer v. Brennan,10 the Supreme Court held that no legitimate
penological purpose is served by allowing rape to occur within
him, and they rented him out. Some sex acts cost $5, others $10.
Id.
6
Johnson, 385 F.3d at 513.
Liptak, supra note 5, at A1. “‘I was forced into oral sex and anal sex on a
daily basis,’ said Mr. Johnson, who has been living in a boarding house [in
Austin, Texas] since his release in December [2003]. ‘Not for a month or two.
For, like, 18 months.’” Id.
8
Sabrina Qutb & Lara Stemple, Selling a Soft Drink, Surviving Hard Time
Just What Part of Prison Rape Do You Find Amusing?, S.F. CHRON., June 9,
2002, at D2, available at http://www.sfgate.com/cgi-bin/article.cgi?file=/
chronicle/archive/2002/06/09/IN181350.DTL.
9
The Prison Rape Elimination Act, 42 U.S.C. § 15601(2) (2005)
(“[E]xperts have conservatively estimated that at least 13% of the inmates in the
United States have been sexually assaulted in prison”). The estimate of 12,000
rapes comes from the corrections industry. Eli Lehrer, A Blind Eye, Still Turned:
Getting Serious About Prison Rape, NAT’L REVIEW, June 2, 2003, at 10. “Even
if this is the actual number, it still represents more rapes than are reported
annually against women in New York City, Los Angeles, Philadelphia, Boston,
San Diego, and Phoenix combined.” Id. Higher estimates have been made.
Extrapolating from the findings of a study of Nebraska’s prison system by
Professor Cindy Struckman-Johnson of the University of South Dakota to the
national level, Human Rights Watch cites a total of more than 140,000 inmates
who have been anally raped while in prison. HUMAN RIGHTS WATCH, NO
ESCAPE: MALE RAPE IN U.S. PRISONS 130 (2001) [hereinafter NO ESCAPE],
available at http://www.hrw.org/ reports/2001/prison. These rapes occur within
a national prison system of federal and state facilities that, in 2001, held at least
24,147 prisoners known to be HIV-positive. LAURA MARUSCHAK, U.S. DEP’T
OF JUSTICE, HIV IN PRISONS, 2001 2 (2004) [hereinafter HIV IN PRISONS 2001],
available at http://www.ojp.usdoj.gov/bjs/pub/pdf/hivp01.pdf.
10
511 U.S. 825 (1994).
7
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
917
prisons.11 Further, the Court held that inmates who are raped in
prison due to the “deliberate indifference” of prison officials suffer
cruel and unusual punishment within the meaning of the Eighth
Amendment to the U.S. Constitution.12 This decision was handed
down fourteen years after Justice Blackmun noted in his dissent in
United States v. Bailey that rape was a fact of prison life.13
Johnson’s case suggests that this continues to be true in 2005.14
Once a prisoner has been raped, or “turned out” in prison
parlance,15 that prisoner (or “punk”)16 can expect to be
continuously raped by other sexual predators and shared among
prison gang members throughout his sentence.17 Alternatively, the
prisoner’s body may become the property of a single dominating
prisoner.18 This occurs despite the Supreme Court’s statement that
sexual assault is “simply not part of the penalty that criminal
11
Farmer, 511 U.S. at 833 (holding that “allowing the beating or rape of
one prisoner by another serves no ‘legitimate penological objective’”).
12
Id. at 832-33.
13
United States v. Bailey, 444 U.S. 394, 421 (1980) (Blackmun, J.,
dissenting) “A youthful inmate can expect to be subjected to homosexual gang
rape his first night in jail or, it has been said, even in the van on the way to jail.
Weaker inmates become the property of stronger prisoners or gangs, who sell
the sexual services of the victim.” Id. (footnote omitted).
14
Johnson v. Johnson, 385 F.3d 503 (5th Cir. 2004).
15
NO ESCAPE, supra note 9, at 90-91.
16
Id. at 93.
17
Ruiz v. Johnson, 37 F. Supp. 2d 855, 929 (S.D. Tex. 1999).
The evidence before this court revealed a prison underworld in which
rapes, beatings, and servitude are the currency of power. Inmates who
refuse to join race-based gangs may be physically or sexually assaulted.
To preserve their physical safety, some vulnerable inmates simply
subject to being bought and sold among groups of prison predators,
providing their oppressors with commissary goods, domestic services,
or sexual favors.
Id.
18
Johnson v. Johnson, 385 F.3d 503 (5th Cir. 2004). In presenting the facts
of the case, the Court of Appeals described Roderick Johnson’s experience,
writing that “[i]n October 2000, not long after his arrival in the general
population, a prison gang member named Hernandez asserted ‘ownership’ over
Johnson, forcing Johnson to become his sexual servant.” Id. at 512.
RIES MACROED CORRECTED 053105.DOC
918
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
offenders pay for their offenses against society.”19
The 108th Congress endeavored to rid U.S. prisons of sexual
assault by passing the Prison Rape Elimination Act of 2003
(PREA).20 Congress incorporated into the statute a cause of action
under Farmer for inmates who have been subjected to rape.21
Additionally, Congress authorized funding for the Bureau of
Justice Statistics to study this issue and created the National Prison
Rape Elimination Commission to recommend national standards
for eradicating prison rape.22 The PREA, by mandating the
collection of records and the creation of standards for prison
management, may aid future plaintiffs who, like Johnson, bring
legal claims against prison officials who fail in their duty to protect
prisoners from sexual assault by other prisoners.
Both Congress and the Supreme Court have now expressed the
need for prison administrators to address inmate-on-inmate rape in
19
Farmer v. Brennan, 511 U.S. 825, 834 (1994) (quoting Rhodes v.
Chapman, 452 U.S. 337, 347 (1981)). The Court held that “[prison] conditions
that cannot be said to be cruel and unusual under contemporary standards are not
unconstitutional. To the extent that such conditions are restrictive and even
harsh, they are part of the penalty that criminal offenders pay for their offenses
against society.” Farmer, 511 U.S. at 834.
20
Prison Rape Elimination Act of 2003, Pub. L. No. 108-79, 117 Stat. 972
(2003) (codified at 42 U.S.C. §§ 15601-15609). For discussion of the lobbying
effort behind the passage of the Prison Rape Elimination Act, see James E.
Robertson, Compassionate Conservatism and Prison Rape: The Prison Rape
Elimination Act, 30 NEW ENG. J. ON CRIM. & CIV. CONFINEMENT 1, 3-8 (2004).
21
42 U.S.C. § 15601(13) (2005). Presuming that Congress knew the case
law on prison rape, the PREA was passed in part to further inform the meaning
of liability for deliberate indifference to rape.
The high incidence of sexual assault within prisons involves actual and
potential violations of the United States Constitution. In Farmer v.
Brennan, the Supreme Court ruled that deliberate indifference to the
substantial risk of sexual assault violates prisoners’ rights under the
Cruel and Unusual Punishments Clause of the Eighth Amendment.
Id. (citation omitted). “Fighting prison rape is also affirmatively mandated by
the Constitution.” 149 CONG. REC. H7765 (daily ed. July 25, 2003) (statement of
Rep. Robert C. Scott).
22
42 U.S.C. §§ 1560304, 15606 (2005). See infra Parts III.A., III.B for
discussion of the Prison Rape Elimination Act.
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
919
correctional facilities.23 However, Congress’s effort to eliminate
prison rape may be undermined by the discretion the statute affords
to federal and state executive bodies in implementing preventive
programs.24 Officials responsible for state prison systems may
continue to deny the extent of prison rape within their facilities,
while facility administrators may continue to tolerate its
existence.25 Therefore, to have their constitutional rights
23
MATHEW BENDER & CO. CIVIL RIGHTS ACTIONS TREATISE § 2.09(C)
(2004). “Prison officials have a clearly established duty not to be deliberately
indifferent to physical or sexual assaults or the possibility of such assaults on
inmates by other inmates.” Id. (citing Farmer v. Brennan, 525 U.S. 825 (1994));
42 U.S.C. § 15601(13) (2005) (incorporating the Supreme Court’s holding into
the PREA by stating that “[i]n Farmer v. Brennan, the Supreme Court ruled that
deliberate indifference to the substantial risk of sexual assault violates prisoners’
rights under the Cruel and Unusual Punishments Clause of the Eighth
Amendment”) (citation omitted).
24
See Olga Giller, Note and Comment, Patriarchy on Lockdown:
Deliberate Indifference and Male Prison Rape, 10 CARDOZO WOMEN’S L.J. 659,
688-89 (2004) (writing of the PREA that “eradicating prison rape takes more
than passing legislation and setting legal precedent. . . . While physical changes
in prison administration will assist in ameliorating the scourge of prison rape,
only structural change will prove lasting”). See also Carla I. Barrett, Note, Does
the Prison Rape Elimination Act Adequately Address the Problems Posed by
Prison Overcrowding? If Not, What Will?, 39 N. ENG. L. REV. 391 (2005).
Opponents to the PREA feel that the Act is simply an empty gesture
that does show high-level governmental recognition of the problem of
prison rape, but does not provide for any real remedy. Robert Weisberg
and David Mills, writing for MSN Slate, claim that “the main thing the
law aims to do is collect data,” a goal that will prove difficult because
of “unreliable observations and underreporting inherent in prison
assault” and redundant because many reports produced by various
organizations across the country provide the same information that the
government seeks to obtain.
Id. at 427 (citing Robert Weisberg & David Mills, Violence Silence: Why No
One Really Cares About Prison Rape, MSN SLATE, Oct. 1, 2003, at
http://slate.msn.com/id/2089095).
25
Within state systems, prison officials deny that the problem of prison
rape is as substantial as prisoners, or even prison staff, report. NO ESCAPE, supra
note 9, at 133-35. Whereas staff at three Nebraska state prisons estimated that
sixteen percent of inmates “were being pressured or forced into sexual contact”
in a 1996 study, id. at 135, over half of all state corrections departments officials
surveyed by Human Rights Watch reported that sexual assault occurred too
RIES MACROED CORRECTED 053105.DOC
920
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
vindicated, inmates will be forced to bring suits that rely on
Farmer, but will prevail only when they can show that “deliberate
indifference” by wardens or other prison officials contributed to
the assaults they endured.26 Federal courts may grant
administrators discretion in prisoners’ rights cases,27 but this
discretion should not include the power to ignore the threat of
sexual assault to prisoners under their control.
This note explores the potential uses of the PREA in litigation
brought by inmates against prison officials for “deliberate
indifference” to the threat of rape. Part I describes the procedures
used and the obstacles faced by prisoners who bring lawsuits
challenging the conditions of their imprisonment. Part II discusses
the Supreme Court’s “deliberate indifference” standard for holding
prison supervisors liable for Eighth Amendment violations and
examines the application of this standard by the federal courts to
claims brought by prisoners in response to assaults within prisons.
Part III presents a review of the PREA and outlines some of the
prison programs that could be implemented as a result of the
legislation. Lastly, Part IV of this note discusses ways in which
future plaintiffs who sue prison officials for deliberate indifference
to a risk of sexual assault will be able to use the PREA in their
lawsuits.
infrequently to maintain data. Id. at 133.
Penal security staffs will also, if not encourage, then definitely tolerate
a homosexual relationship by a potentially troublesome prisoner,
theorizing that a prisoner who is getting some degree of emotional and
sexual gratification from his prison “wife” is less likely to cause trouble
than a prisoner who is not because he’s comfortable and, once
emotionally attached, he will not want to lose his “wife”.
WALTER RIDEAU & RON WILKBERG, LIFE SENTENCES: RAGE AND SURVIVAL
BEHIND BARS 88-89 (1992).
26
Farmer v. Brennan, 511 U.S. 825, 834 (1994).
27
Bell v. Wolfish, 441 U.S. 520, 548 (1979) (“[T]he operation of our
correctional facilities is peculiarly the province of the Legislative and Executive
Branches of our Government, not the Judicial.”).
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
921
I. PRISONER LITIGATION
Under 42 U.S.C. Section 1983 (Section 1983), inmates can sue
prison staff for violations of their Eighth Amendment right to be
free from cruel and unusual punishment.28 Lawsuits alleging harms
against a prisoner perpetrated by another prisoner are brought as
“condition-of-confinement” claims29 on the theory that when an
individual is held in custody, there is a corresponding duty
assumed by the government to ensure that the individual will
remain safe.30 Only in the last quarter of the twentieth century have
prisoner challenges to conditions of confinement been recognized
by the Supreme Court as valid claims under the Eighth
Amendment.31 This expansion of constitutional protection has
28
42 U.S.C. § 1983 (2005). In relevant part, the statute states:
Every person who, under color of any statute, ordinance, regulation,
custom, or usage, of any State or Territory or the District of Columbia,
subjects, or causes to be subjected, any citizen of the United States or
other person within the jurisdiction thereof to the deprivation of any
rights, privileges, or immunities secured by the Constitution and laws,
shall be liable to the party injured in an action at law, suit in equity, or
other proper proceeding for redress, except that in any action brought
against a judicial officer for an act or omission taken in such officer’s
judicial capacity, injunctive relief shall not be granted unless a
declaratory decree was violated or declaratory relief was unavailable.
Id. See infra Part I.A. Federal officials are sued for violations of prisoners’ rights
in actions known as “Bivens claims” pursuant to the Supreme Court’s decision
in Bivens v. Six Unknown Named Agents of the Federal Bureau of Narcotics,
403 U.S. 388 (1971). For city officials who are responsible for the conditions of
city jails, liability under Section 1983 is controlled by the Supreme Court’s
decision in Monell v. Dep’t of Social Serv., 436 U.S. 658 (1978).
29
See infra part I.B. Flint v. Ky. Dep’t of Corr., 270 F.3d 340, 352 (6th Cir.
2001). “Where the harm is perpetrated by another prisoner, rather than by a
government official, the claim is characterized as one of ‘conditions of
confinement,’ rather than of ‘excessive use of government force.’” Id. (quoting
Thaddeus-X v. Blatter, 175 F.3d 378, 400-01 (6th Cir. 1999)).
30
DeShaney v. Winnebago County Dep’t of Social Serv., 489 U.S. 189,
199-200 (1989). “[W]hen the State takes a person into custody and holds him
there against his will, the Constitution imposes upon it a corresponding duty to
assume some responsibility for his safety and general well-being.” Id.
31
Rhodes v. Chapman, 452 U.S. 337, 345-46 (1981) (describing expansion
RIES MACROED CORRECTED 053105.DOC
922
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
since been limited by further decisions of the Supreme Court and
by Congress’s passage of the Prison Litigation Reform Act.32
A. Prisoner Claims under Section 1983
Section 1983 provides a right of action for a person who
suffers “a deprivation of any rights, privileges, or immunities
secured by the Constitution” under color of law.33 State prisoners
may bring claims under this statute34 for violations of their First
Amendment rights to expression,35 association,36 and religion,37 as
well as their due process rights,38 privacy rights under the Fourth
Amendment,39 and Eighth Amendment rights.40 Federal prisoners
are able to bring similar claims against federal prison officials,
known as “Bivens claims,” following the Supreme Court’s
of Supreme Court’s interpretation of the Eighth Amendment “beyond the
barbarous physical punishments at issue in the Court’s earliest cases” to reach
conditions “that are ‘totally without penological justification”) (quoting Gregg v.
Georgia, 428 U.S. 153, 183 (1976)).
32
See infra Part I.C.
33
42 U.S.C. § 1983 (2005).
34
Monroe v. Pape, 365 U.S. 167 (1961) (discussing reasons for applying
Section 1983 to the states and stating of the Act that one “aim was to provide a
federal remedy where the state remedy, though adequate in theory, was not
available in practice”).
35
Procunier v. Martinez, 416 U.S. 396 (1974) (hearing a claim brought by
California state inmates under § 1983 to challenge censorship of prisoner mail).
36
Turner v. Safley, 482 U.S. 78 (1987) (hearing a claim brought by
Missouri state inmates to challenge prison restrictions on inmate-to-inmate
communications as well as inmate marriages).
37
O’Lone v. Estate of Shabazz, 482 U.S. 342, 347 (1987) (hearing a claim
brought by New Jersey state inmates under § 1983 challenging prison regulation
that prevented them from attending a Muslim congregational service).
38
Sandin v. Conner, 515 U.S. 472, 476 (1995) (hearing a claim brought by
a Hawaii state inmate under § 1983 for a deprivation of procedural due process
in connection with a prison disciplinary hearing).
39
Hudson v. Palmer, 468 U.S. 517, 520 (1984) (hearing a claim brought by
a Virginia state inmate under § 1983 challenging a search of his cell for
contraband as unreasonable under the Fourth Amendment).
40
Hudson v. McMillian, 503 U.S. 1 (1992) (hearing a claim brought by a
Louisiana state inmate under § 1983 for guards’ use of excessive force).
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
923
decision in Bivens v. Six Unknown Named Agents of Federal
Bureau of Narcotics.41 The Supreme Court has held that
supervisors of state employees may be held liable for their
subordinates’ conduct under Section 1983 only when there is an
“affirmative link” between a violation of a plaintiff’s constitutional
rights and the supervisor’s own official conduct.42 In the context of
an inmate’s lawsuit, the Court held that Section 1983 “contains no
state-of-mind requirement independent of that necessary to state a
violation of the underlying constitutional right.”43 Thus,
supervisors can only be held liable for their own culpable conduct,
not for that of their subordinates through respondeat superior.44
Supervisors may be liable for “failure to supervise”45 as well as
“failure to train” employees under their control.46 A supervisor’s
41
Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics,
403 U.S. 388 (1971) (holding that federal courts have jurisdiction over claims of
constitutional rights violations under 28 U.S.C. § 1331); see also Michael Irvine,
Excerpts from a Jailhouse Lawyer’s Manual, Fifth Edition: Chapter 17: Using
42 U.S.C. § 1983 and 42 U.S.C. § 1331 to Obtain Relief from Violations of
Federal Law, 31 COLUM. HUM. RTS. L. REV. 305, 349 (2000) [hereinafter
Jailhouse Lawyer’s Manual § 1983].
42
Rizzo v. Goode, 423 U.S. 362, 371 (1976) (holding that supervisors of
the Philadelphia police department were not liable for unconstitutional
mistreatment by police officers because “there was no affirmative link between
the occurrence of the various incidents of police misconduct and the adoption of
any plan or policy by [the defendants]—express or otherwise—showing their
authorization or approval of such misconduct”). See Kit Kinports, The Buck
Does Not Stop Here: Supervisory Liability in Section 1983 Cases, 1997 U. ILL.
L. REV. 147, 151-52 (1997) [hereinafter Kinports].
43
Daniels v. Williams, 474 U.S. 327, 329-30 (1986) (affirming summary
judgment for defendant where inmate claimed a negligent deprivation of liberty
without due process under the Fourteenth Amendment because he suffered
injuries caused by defendant’s misplacement of a pillowcase in a jailhouse
stairway).
44
Kinports, supra note 42, at 153 (describing that Monell v. Dep’t of
Social Serv., 436 U.S. 658 (1978), “signaled the Court’s unwillingness to
impose respondeat superior liability on supervisors”).
45
Monell, 436 U.S. at 694 n.58 (“By our decision in Rizzo v. Goode, 423
U.S. 362 (1976), we would appear to have decided that the mere right to control
without any control or direction having been exercised and without any failure
to supervise is not enough to support § 1983 liability.”).
46
City of Canton v. Harris, 489 U.S. 378, 388 (1989) (holding that
RIES MACROED CORRECTED 053105.DOC
924
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
direct participation in a constitutional rights violation also will
make that supervisor liable under Section 1983, as will the creation
of policies or customs that lead to a foreseeable violation.47 Noting
that the Supreme Court has not established a standard for
supervisory liability under Section 1983 separate from the
constitutional standard applied to violations of the specific right
alleged, Professor Kit Kinports has identified various factors that
courts of appeals use in assessing supervisors’ culpability:
[T]he courts of appeals tend to agree that five interrelated
factors ought to be considered in applying that
[constitutional] standard and determining whether a
particular supervisor is liable on the facts of a given case:
(1) the extent to which prior similar incidents have
occurred; (2) the supervisor’s response to those prior
incidents; (3) the supervisor’s response to the specific
incident on which the suit is based; (4) the extent to which
the supervisor can be considered a cause of the violation;
and (5) the nature of the supervisor’s awareness of the
constitutional misconduct.48
Since state governments are immune from legal claims by way
of the Eleventh Amendment,49 courts must find individual officials
“inadequacy of [government employee] training may serve as the basis for §
1983 liability only where the failure to train amounts to deliberate indifference
to the rights of persons with whom the [employees] come into contact”). For
cases applying “failure to train” liability to supervisors, see Kinports, supra note
42, at 165-68.
47
Jailhouse Lawyer’s Manual § 1983, supra note 41, at 308.
48
Kinports, supra note 42, at 169.
Although the courts agree that these are the relevant considerations,
they have not been consistent in applying them. As a result, the courts
have reached contrary outcomes in similar cases, seemingly without
any regard to the particular standard of supervisory liability they
purport to be applying. And all too often, they have been unduly
generous in ruling in favor of supervisory officials.
Id.
49
Papasan v. Allain, 478 U.S. 265, 276 (1986).
The [Eleventh] Amendment provides: “The Judicial power of the
United States shall not be construed to extend to any suit in law or
equity, commenced or prosecuted against one of the United States by
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
925
liable for their injuries to provide prisoners with redress. Individual
officials whose conduct is within the scope of Section 1983 may
nevertheless be immune from liability. The Supreme Court held in
Imbler v. Pachtman, for example, that legislators, judges, and
prosecutors receive absolute immunity when sued under Section
1983, unless they commit “willful deprivations of Constitutional
rights.”50 Other government officials avoid liability under Section
1983 when they act in their “official capacities.”51 However,
pursuant to the Supreme Court’s decision in Monell v. Department
of Social Services, officials can be sued under Section 1983 for
their execution of a government policy or custom when that policy
or custom is the “moving force of the constitutional violation”
alleged.52 This has provided the means for holding government
supervisors liable when those supervisors represent government
policy.53
Government officials may be found liable for damages when
their conduct is beyond the scope of their official capacities; thus,
they are said to have “qualified immunity.”54 The doctrine of
Citizens of another State, or by Citizens or Subjects of any Foreign
State.” This language expressly encompasses only suits brought against
a State by citizens of another State, but this Court long ago held that the
Amendment bars suits against a State by citizens of that same State as
well.
Id. (citing Hans v. Louisiana, 134 U.S. 1 (1890)).
50
Imbler v. Pachtman, 424 U.S. 409, 429 (1976).
51
Will v. Mich. Dep’t of State Police, 491 U.S. 58, 71 (1989) (holding that
“neither a State nor its officials acting in their official capacities are ‘persons’
under § 1983”).
52
Monell v. Dep’t of Social Serv., 436 U.S. 658, 694 (1978).
[I]t is when execution of a government’s policy or custom, whether
made by its lawmakers or by those whose edicts or acts may fairly be
said to represent official policy, inflicts the injury that the government
as an entity is responsible under § 1983. . . . [T]his case unquestionably
involves official policy as the moving force of the constitutional
violation.
Id.
53
Id. at 694.
54
Id. at 707 (Powell, J., concurring). “It has been clear that a public official
may be held liable in damages when his actions are found to violate a
constitutional right and there is no qualified immunity.” Id. (citing Wood v.
RIES MACROED CORRECTED 053105.DOC
926
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
qualified immunity deems government officials liable only when
“it would be clear to a reasonable official that his conduct was
unlawful in the situation he confronted.”55 This applies to all 1983
claims and severely limits the ability of plaintiffs to win damages
under Section 1983.56 To defeat a government official’s motion for
qualified immunity, a plaintiff must first demonstrate that a
violation of federal or constitutional law indeed occurred and then
proceed to show that the violated law was clearly established and
that a reasonable official would have understood his conduct to be
a violation.57 When defendants prevail on grounds of qualified
Strickland, 420 U.S. 308 (1975); Procunier v. Navarette, 434 U.S. 555 (1978)).
55
Saucier v. Katz, 533 U.S. 194, 202 (2001).
56
Sheldon Nahomad, From the Courtroom to the Street: Court Orders and
Section 1983, 29 HASTINGS CONST. L. Q. 613, 637-38 (2002).
The damages remedy functions not only to deter unconstitutional
conduct but also to compensate innocent people as a matter of
corrective justice. Regrettably, however, the Supreme Court has all too
often emphasized the possible over-deterrence of government officials
and employees at the expense of providing corrective justice to those
harmed by unconstitutional conduct. It is fair to say that this move has
been based on the Court’s intuition about the non-meritorious nature of
many 1983 claims, to say nothing of its concern for federalism, and its
apparent distaste for many 1983 plaintiffs, especially prisoners. It was
on such grounds, for example, that the Court transformed qualified
immunity, originally a defense to liability, into an immunity from suit,
effectively converting it, primarily for instrumental reasons, into a kind
of absolute immunity.
Id. (footnotes omitted).
57
Saucier, 533 U.S. at 201-02.
A court required to rule upon the qualified immunity issue must
consider, then, this threshold question: Taken in the light most
favorable to the party asserting the injury, do the facts alleged show the
officer’s conduct violated a constitutional right? . . . [T]he next,
sequential step is to ask whether the right was clearly established. . . .
The relevant, dispositive inquiry in determining whether a right is
clearly established is whether it would be clear to a reasonable officer
that his conduct was unlawful in the situation he confronted.
Id. Relying on Supreme Court decisions is the most persuasive means for a
plaintiff to claim that the right allegedly violated was clearly-established at the
time of their injury. Daugherty v. Campbell, 935 F.2d 780, 784 (6th Cir. 1991)
(holding that the Sixth Circuit looks “first to decisions of the Supreme Court” in
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
927
immunity, plaintiffs’ constitutional harms may be left
uncompensated, and only plaintiffs who seek injunctive relief or
take solace in further establishing the law for future litigants will
be satisfied.58
For plaintiffs using Section 1983 to remedy a constitutional
harm, injunctive relief is generally available as are both
compensatory and punitive damages.59 Only when state and
municipal officials are sued in their “individual capacities,”
however, can they be held liable for monetary damages.60 Plaintiffs
suing officials in their “official capacities” can expect to receive at
most injunctive relief.61 For prisoners, both the injunctive relief
and monetary relief available under Section 1983 and through
Bivens claims have been severely limited by the Prison Litigation
determining whether rights are clearly established for purposes of qualified
immunity). Courts of Appeals decisions clearly establish the law in their own
circuits. The Supreme Court’s decision in Hope v. Pelzer, held that a
Department of Justice report warning prison officials that their conduct might
violate the Eighth Amendment provided notice sufficient to deny them qualified
immunity. 536 U.S. 730 (2002). This raises the question of what role
administrative regulations might generally play in clearly establishing law for
qualified immunity purposes. See Amanda K. Eaton, Optical Illusions: The
Hazy Contours of the Clearly Established Law and the Effects of Hope v. Pelzer
on the Qualified Immunity Doctrine, 38 GA. L. REV. 661, 709-10 (2004).
58
John C. Jeffries, The Right-Remedy Gap in Constitutional Law, 109
YALE L.J. 87, 113 (1999).
Qualified immunity disfavors the backward-looking remedy of cash
payments to victims of past harms and, in so doing, opens the door to
forward-looking remedies requiring investments in the future.
Structural reform injunctions walk through that door. They direct
resources toward preventing future harms rather than compensating
past injuries, thereby implementing the bias in favor of the future that
qualified immunity invites and allows.
Id.
59
See Daniel D. Williams, Twenty-Eighth Annual Review of Criminal
Procedure: VI. Prisoners’ Rights: Procedural Means of Enforcement Under 42
U.S.C. § 1983, 87 GEO. L.J. 1940, 1944-45 (1999). A prevailing plaintiff may
also recover attorney’s fees. 42 U.S.C. § 1988 (2005).
60
Jailhouse Lawyer’s Manual § 1983, supra note 41, at 309.
61
Id.
RIES MACROED CORRECTED 053105.DOC
928
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
Reform Act.62 The Supreme Court also has reined in the discretion
of federal courts to fashion injunctions to prevent violations of
prisoners’ constitutional rights.63 Therefore, even successful
prisoner-plaintiffs face difficulties in obtaining meaningful relief.
B. Challenges to Conditions of Confinement
Until late in the twentieth century, the Eighth Amendment’s
prohibition against “cruel and unusual punishments” was applied
only to criminal sentences.64 However, through its decisions in
Estelle v. Gamble and Rhodes v. Chapman, the Supreme Court
62
See infra Part I.C.
Lewis v. Casey, 518 U.S. 343 (1996) (reversing a decision of the Ninth
Circuit that upheld an injunction ordering improvements to a prison system’s
law libraries).
The actual-injury requirement would hardly serve the purpose we have
described above— of preventing courts from undertaking tasks
assigned to the political branches —if once a plaintiff demonstrated
harm from one particular inadequacy in government administration, the
court were authorized to remedy all inadequacies in that administration.
The remedy must of course be limited to the inadequacy that produced
the injury in fact that the plaintiff has established.
Id. at 357 (emphasis in original); Bell v. Wolfish, 441 U.S. 520, 544 (1979)
(holding that judges ought to be discouraged from managing prisons).
The court might disagree with the choice of means to effectuate
[security] interests, but it should not “second-guess the expert
administrators on matters on which they are better informed. . . .
Concern with minutiae of prison administration can only distract the
court from detached consideration of the one overriding question
presented to it: does the practice or condition violate the Constitution?”
Id. at 544 (quoting Wolfish v. Levi, 573 F.2d 118 (2d Cir. 1978)).
64
Hudson v. McMillian, 503 U.S. 1, 18 (1992) (Thomas, J., dissenting)
Until recent years, the Cruel and Unusual Punishments Clause was not
deemed to apply at all to deprivations that were not inflicted as part of
the sentence for a crime. For generations, judges and commentators
regarded the Eighth Amendment as applying only to torturous
punishments meted out by statutes or sentencing judges, and not
generally to any hardship that might befall a prisoner during
incarceration.
Id.
63
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
929
interpreted the Cruel and Unusual Punishments Clause of the
Eighth Amendment to include prison conditions.65 These decisions
came during a period in which the federal courts had employed a
“totality-of-circumstances” analysis to issue structural injunctions
intended to prevent prison conditions from becoming overly
harsh.66
In Estelle v. Gamble, a case brought as a challenge to the level
of medical care offered in a Texas state prison, a prisoner claimed
that he had received inadequate treatment for an injury he
sustained while performing a prison work assignment.67 The
Supreme Court used its established Eighth Amendment doctrine of
measuring punishments against “evolving standards of decency” to
hold that the constitutionality of conditions of imprisonment could
65
Estelle v. Gamble, 429 U.S. 97 (1976); Rhodes v. Chapman, 452 U.S.
337 (1981). Arguments against this extension of the Eighth Amendment persist
in Supreme Court dissents by Justice Thomas and in commentary. See, e.g., Sara
Rose, Comment, “Cruel and Unusual Punishment” Need Not Be Cruel,
Unusual, or Punishment, 24 CAP. U.L. REV. 827 (1995).
66
Russel W. Gray, Note, Wilson v. Seiter: Defining the Components of and
Proposing a Direction for Eighth Amendment Prison Condition Law, 41 AM.
U.L. REV. 1339, 1352 (1992) (citing Courts of Appeals decisions from 1970
through 1985 that applied the totality of circumstances test); John Jeffries, supra
note 58, at 111-12. Professor Jeffries argues that remedies exceeded rights when
courts ordered more structural injunctions, because
courts increasingly focused on prophylactic precautions against the risk
of constitutional violations. Over time, specific remedial strategies that
recurred in one case after another assumed a life of their own. They
underwent a subtle transformation from ad hoc remedies for
independently demonstrated unconstitutionality of confinement to
normative criteria for assessing the acceptability of prison operation. In
effect, remedies became quasi-rights. . . . Whether this phenomenon is
described as remedy exceeding right or as remedy implicitly redefining
right or as remedy merely becoming a “criter[ion] by which . . .
lawfulness is judged” is for present purposes immaterial. The important
point is that in structural reform litigation, courts prospectively and
selectively impose requirements that in other remedial contexts would
not be constitutionally compelled.
Id. at 111-12 (quoting in part Note, Complex Enforcement: Unconstitutional
Prison Conditions, 94 HARV. L. REV. 626, 638 (1981)).
67
Estelle, 429 U.S. at 98.
RIES MACROED CORRECTED 053105.DOC
930
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
be assessed under the Eighth Amendment.68 The Court held,
however, that whereas the “evolving standards of decency”
assessment provided an objective test for determining whether a
punishment was cruel and unusual, claims regarding conditions of
confinement required an inquiry into the subjective intent of the
prison’s officers.69 The Court adopted a “deliberate indifference”
standard for evaluating the actions of prison officials.70 The Court
required a finding of “wanton infliction of pain” in order to hold
prison officials liable for unconstitutional prison conditions.71 This
standard was intended to limit the liability of prison officials to the
creation or support of conditions that could genuinely be deemed
“cruel” under the Eighth Amendment.72 Although the Court
deemed the medical care offered to the prisoner a condition of his
confinement, it denied that any inadequacies in the prisoner’s care
constituted cruel and unusual punishment.73 The Court maintained
that, in general, defendant prison officials should be found at fault
under this standard only when the conditions are “wanton,” as only
68
Id. at 102. “[W]e have held repugnant to the Eighth Amendment
punishments which are incompatible with ‘the evolving standards of decency
that mark the progress of a maturing society.’” Id. (quoting Trop v. Dulles, 356
U.S. 86, 101 (1958)).
69
Id. at 105-06.
[A]n inadvertent failure to provide adequate medical care cannot be
said to constitute “an unnecessary and wanton infliction of pain” or to
be “repugnant to the conscience of mankind”. . . . In order to state a
cognizable claim, a prisoner must allege acts or omissions sufficiently
harmful to evidence deliberate indifference to serious medical needs. It
is only such indifference that can offend “evolving standards of
decency” in violation of the Eighth Amendment.
Id.; Gray, supra note 66, at 1357-58 (explaining that the Supreme Court’s stateof-mind requirement for conditions of confinement derives from the Eighth
Amendment’s explicit proscription only of punishment).
70
Estelle, 429 U.S. at 104 (“We therefore conclude that deliberate
indifference to serious medical needs of prisoners constitutes the ‘unnecessary
and wanton infliction of pain’ proscribed by the Eighth Amendment.”) (citation
omitted).
71
Id. at 104.
72
Wilson v. Seiter, 501 U.S. 294, 297-98 (1991) (discussing the holding of
Estelle v. Gamble).
73
Estelle, 429 U.S. at 107-08.
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
931
then can the prison conditions at issue be considered
“punishment.”74
Having accepted that conditions of confinement could impose
constitutional harms on prisoners, the Supreme Court soon was
faced with the task of determining more precisely which conditions
merited judicial scrutiny. In Hutto v. Finney,75 a majority of the
Court held that the “interdependence of the conditions producing
the violation”76 justified “a comprehensive order to insure against
the risk of inadequate compliance.”77 Shortly thereafter, the Court
examined a challenge under the Eighth Amendment to
overcrowded prison conditions.78 In Rhodes v. Chapman, the Court
considered “whether the housing of two inmates in a single cell . . .
is cruel and unusual punishment”79 as a condition of confinement
or as the root cause of other harms suffered by inmates.80 The
majority decision held that the so-called “double-celling” of
inmates was not unconstitutional per se;81 rather, “restrictive and
74
Seiter, 501 U.S. at 300. For a discussion of the subjective element of the
Eighth Amendment’s application to conditions of confinement, as defined in
Farmer v. Brennan, see infra Part II.A.
75
Hutto v. Finney, 437 U.S. 678, 684 (1978) (affirming an injunction
against the Arkansas prison system “that placed limits on the number of men
that could be confined in one cell, required that each have a bunk, discontinued
the ‘grue’ diet, and set 30 days as the maximum isolation sentence”).
76
Id. at 688.
77
Id. at 687.
78
Rhodes v. Chapman, 452 U.S. 337 (1981).
79
Id. at 339.
80
Id. at 340.
Asserting a cause of action under 42 U. S. C. § 1983, [plaintiffs]
contended that “double celling” at [their facility] violated the
Constitution. The gravamen of their complaint was that double celling
confined cellmates too closely. It also was blamed for overcrowding at
[the prison], said to have overwhelmed the prison’s facilities and staff.
As relief, respondents sought an injunction barring petitioners, who are
Ohio officials responsible for the administration of SOCF, from
housing more than one inmate in a cell, except as a temporary measure.
Id.
81
Id. at 350. “The question before us is . . . whether the actual conditions
of confinement . . . are cruel and unusual.” Id. (emphasis added).
RIES MACROED CORRECTED 053105.DOC
932
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
even harsh” conditions were a constitutional “part of the penalty
that criminal offenders pay for their offenses against society.”82
The Court emphasized, however, that conditions of confinement
may not be “grossly disproportionate to the severity of the crime
warranting imprisonment.”83
Further, the Rhodes decision refocused the inquiry conducted
by the court in Estelle and Hutto by examining whether the
allegedly unconstitutional condition caused “unquestioned and
serious deprivations of basic human needs” comprising “the
minimal civilized measure of life’s necessities.”84 The Court’s
subsequent decision in Wilson v. Seiter85 accepted this inquiry as
the objective test for evaluating prison conditions under the Eighth
Amendment, holding that, whether alone or in isolation, conditions
of confinement are cruel and unusual when they deprive prisoners
of “a single, identifiable human need.”86 In subsequent condition
of confinement cases, the Supreme Court recognized that “human
needs” include food, clothing, shelter, medical care, and reasonable
82
Id. at 347. “[C]onditions that cannot be said to be cruel and unusual
under contemporary standards are not unconstitutional. To the extent that such
conditions are restrictive and even harsh, they are part of the penalty that
criminal offenders pay for their offenses against society.” Id.
83
Id. “Conditions must not involve the wanton and unnecessary infliction
of pain, nor may they be grossly disproportionate to the severity of the crime
warranting imprisonment.” Id.
84
Id.
In Estelle v. Gamble, we held that the denial of medical care is cruel
and unusual because, in the worst case, it can result in physical torture,
and, even in less serious cases, it can result in pain without any
penological purpose. In Hutto v. Finney, the conditions of confinement
in two Arkansas prisons constituted cruel and unusual punishment
because they resulted in unquestioned and serious deprivations of basic
human needs. Conditions other than those in Gamble and Hutto, alone
or in combination, may deprive inmates of the minimal civilized
measure of life’s necessities.
Id.
85
501 U.S. 294 (1991).
86
Id. at 304. See also Gray, supra note 66, at 1384-85 (discussing Wilson v.
Seiter’s rejection of the “totality of circumstances” test in favor of the “core
conditions” or “single identifiable human need” test).
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
933
safety.87 The Supreme Court’s condition of confinement decisions
also clarified that state officials may be held responsible for the
constitutional harms to which inmates are exposed during
imprisonment.88
In Helling v. McKinney, the Court extended the protection of
the Eighth Amendment to prospective violations, noting that it
would be “odd” for courts to ignore the threat of future
constitutional harms.89 In that case, a prisoner challenged the
conditions of his confinement based on the imminent danger posed
by the secondhand smoke to which he was exposed by his
cellmate.90 Recognizing the potential validity of the prisoner’s
claim, the Court remanded the case to the trial court for a
determination regarding whether the conditions complained of
were sufficiently serious to satisfy both the objective and
87
DeShaney v. Winnebago County Dep’t of Social Serv., 489 U.S. 189,
199-200 (1989), and citing cases, infra note 88.
88
Helling v. McKinney, 509 U.S. 25, 33 (1994). “It is cruel and unusual
punishment to hold convicted criminals in unsafe conditions.” Id. (citing
Deshaney, 489 U.S. at 200; Youngberg v. Romeo, 457 U.S. 307, 315-316
(1982)); Wilson v. Seiter, 501 U.S. 294, 311 (1991) (White, J., concurring)
(“[H]aving chosen to use imprisonment as a form of punishment, a State must
ensure that the conditions in its prisons comport with the ‘contemporary
standard of decency’ required by the Eighth Amendment.”) (citing DeShaney,
489 U.S. at 198-200); id. at 199-200 (1989).
[W]hen the State by the affirmative exercise of its power so restrains an
individual’s liberty that it renders him unable to care for himself, and at
the same time fails to provide for his basic human needs—e. g., food,
clothing, shelter, medical care, and reasonable safety—it transgresses
the substantive limits on state action set by the Eighth Amendment and
the Due Process Clause.
Id.; see also Will A. Smith, Comment, Civil Liability for Sexual Assault in
Prison: A Challenge to the “Deliberate Indifference” Standard, 34 CUMB. L.
REV. 289, 309-11 (2003) (describing applications of this dicta from DeShaney as
a “state-created-danger theory” of liability).
89
509 U.S. 25 (1994).
90
Id. at 28. “The complaint . . . alleged that respondent was assigned to a
cell with another inmate who smoked five packs of cigarettes a day. . . .
Respondent sought injunctive relief and damages for [prison officials]
subjecting him to cruel and unusual punishment by jeopardizing his health.” Id.
RIES MACROED CORRECTED 053105.DOC
934
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
subjective tests of deliberate indifference.91 The Court held that a
challenge to a future harm requires a likelihood and seriousness
beyond what “a scientific and statistical inquiry” can provide.92 A
prisoner must show both that the future harm threatens to deprive
him of an identifiable human need and “that the risk of which he
complains is not one that today’s society chooses to tolerate.”93 In
Farmer v. Brennan, the Court resolved that protection against
sexual assault is a human need warranting protection under the
Eighth Amendment.94
C. The Prison Litigation Reform Act: An Obstacle to Prisoner
Rape Suits
In 1996, the course of prison litigation was altered still further
through Congress’s enactment of the Prison Litigation Reform Act
(PLRA).95 The PLRA was passed to address a perceived deluge of
91
Id. at 35.
Id. at 36. “[W]ith respect to the objective factor, determining whether
McKinney’s conditions of confinement violate the Eighth Amendment requires
more than a scientific and statistical inquiry into the seriousness of the potential
harm and the likelihood that such injury to health will actually be caused by
exposure to [secondhand smoke].” Id.
93
Id. For a more complete discussion of the Court’s treatment of imminent
dangers in Helling v. McKinney, see Katherine L. Frazier, Comment,
Constitutional Law – Helling v. McKinney: Future Risks of Harm Actionable
Under the Eighth Amendment, 25 U. MEM. L. REV. 1479 (1995).
94
Farmer v. Brennan, 511 U.S. 825, 832-33 (1994).
[T]he treatment a prisoner receives in prison and the conditions under
which he is confined are subject to scrutiny under the Eighth
Amendment. . . . [P]rison officials must ensure that inmates receive
adequate food, clothing, shelter, and medical care, and must take
reasonable measures to guarantee the safety of the inmates. In
particular . . . prison officials have a duty to protect prisoners from
violence at the hands of other prisoners. . . . [G]ratuitously allowing the
beating or rape of one prisoner by another serves no legitimate
penological objective [nor] squares with evolving standards of decency.
Id. (quotations and citations omitted).
95
Prison Litigation Reform Act, Pub. L. No. 104-134, 110 Stat. 1321
(1996) (codified as amended in 18 U.S.C. § 3636, 28 U.S.C. § 1915, 28 U.S.C. §
1346, 42 U.S.C. § 1997, and other scattered sections).
92
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
935
frivolous and meritless prisoner claims that overwhelmed the
federal courts.96 The legislation significantly limits the types of
claims and remedies that are available to inmates.97 Indeed, in
1997, the number of prisoner civil rights suits filed decreased by
thirty-one percent;98 by 2000, that number decreased further to
forty percent less than before passage of the PLRA.99
Prior to the enactment of the PLRA, the Attorney General was
responsible for certifying that each state prison system’s grievance
procedure was in compliance with standards issued pursuant to the
Civil Rights of Institutionalized Persons Act.100 A court would
scrutinize a prison’s administrative grievance procedure before
96
Jennifer Winslow, Comment, The Prison Litigation Reform Act’s
Physical Injury Requirement Bars Meritorious Lawsuits: Was It Meant To?, 49
UCLA L. REV. 1655, 1658 (2002).
PLRA proponents declared their intention to curtail the number of
frivolous and meritless inmate suits clogging the federal judiciary.
They then used exaggerated examples of inmate complaints to suggest
subtly that all inmate suits are frivolous and meritless. While
proponents provided assurances that meritorious inmate suits would not
be affected by the PLRA, they made little effort to acknowledge that
meritorious inmate suits do exist.
Id.
97
42 U.S.C. § 1997e(c) (2005). “The court shall . . . dismiss any action
brought with respect to prison conditions . . . if the court is satisfied that the
action is frivolous, malicious, fails to state a claim upon which relief can be
granted, or seeks monetary relief from a defendant who is immune from such
relief.” Id.
98
Randall S. Jeffrey, Restricting Prisoners’ Equal Access to the Federal
Courts: The Three Strikes Provision of the Prison Litigation Reform Act and
Substantive Equal Protection, 49 BUFFALO L. REV. 1099, 1108 n.29 (2001)
(citing Crawford-El v. Britton, 523 U.S. 574, 597 n.18 (1998) (citing ADMIN.
OFFICE OF THE U.S. COURTS, JUDICIAL BUSINESS OF THE UNITED STATES
COURTS, STATISTICAL TABLES FOR THE FEDERAL JUDICIARY 28 tbl. C-2 (2000)).
99
Brian J. Ostrom et al., Congress, Courts and Corrections: An Empirical
Perspective on the Prison Litigation Reform Act, 78 NOTRE DAME L. REV. 1525,
1525 (2003).
100
Civil Rights of Institutionalized Persons Act, Pub L. No. 96-247, 94
Stat. 349 (1980) (codified at 42 U.S.C. §§ 1997-1997j (1994)). For discussion of
how this certification program was administered ineffectively by the Justice
Department, see Note, Resolving Prisoners’ Grievances Out of Court: 42 U.S.C.
§ 1997e, 104 HARV. L. REV. 1309, 1320 (1991).
RIES MACROED CORRECTED 053105.DOC
936
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
deciding whether to dismiss the inmate’s claim for failure to
exhaust his administrative remedies.101 The PLRA, however,
limited judicial scrutiny of grievance procedures to the simple
question of whether a prison made any remedies available through
a grievance procedure.102 Thus, under the PLRA, federal courts
will dismiss a Section 1983 suit brought by a prisoner if
administrative remedies are available and the prisoner has failed to
exhaust them.103 The PLRA also raised the stakes for dismissals of
claims by enacting a “three strikes” provision for prisoners’ court
fee waivers.104 Prisoners whose lawsuits were “dismissed on the
grounds that [they were] frivolous, malicious, or fail[ed] to state a
claim” on three prior occasions would become ineligible for fee
waivers in all future actions or appeals.105
Under the PLRA, in order to prevail on a claim for
compensatory damages, a prisoner must demonstrate a physical
101
42 U.S.C. § 1997 (1994). See Lynn S. Branham, The Prison Litigation
Reform Act’s Enigmatic Exhaustion Requirement: What It Means and What
Congress, Courts and Correctional Officials Can Learn From It, 86 CORNELL L.
REV. 483 (2001).
102
Branham, supra note 101, at 498. “The only substantive requirement
remaining on the face of [42 U.S.C. § 1997(e)] that administrative remedies
must meet in order for the exhaustion requirement to apply is that the remedies
be ‘available.’” Id.
103
42 U.S.C. § 1997(e) was amended by the Prison Litigation Reform Act
to make exhaustion of administrative remedies a requirement of litigation. “No
action shall be brought with respect to prison conditions under section 1983 of
this title, or any other Federal law, by a prisoner confined in any jail, prison, or
other correctional facility until such administrative remedies as are available are
exhausted.” 42 U.S.C. § 1997e(a) (2005).
104
28 U.S.C. § 1915(g) (2005).
In no event shall a prisoner bring a civil action or appeal a judgment in
a civil action or proceeding [in forma pauperis] if the prisoner has, on 3
or more prior occasions, while incarcerated or detained in any facility,
brought an action or appeal in a court of the United States that was
dismissed on the grounds that it is frivolous, malicious, or fails to state
a claim upon which relief may be granted, unless the prisoner is under
imminent danger of serious physical injury.
Id.
105
Id. Jeffrey, supra note 98, at 1133.
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
937
injury.106 Congress categorically denied recovery for mental and
emotional injuries to prisoners without evidence of “actual injury,”
a requirement that the federal courts already maintained.107 The
PLRA also placed limitations on the injunctive relief available to
inmates.108 The statute mandates that federal courts may order only
narrowly-drawn injunctions that address the likelihood of a
specific injury’s reoccurring.109 Despite these restrictions,
prisoners continue to bring suits asking courts to enjoin prison
supervisors and improve unsafe conditions.110
106
42 U.S.C. § 1997e (2005). “Limitation on recovery. No Federal civil
action may be brought by a prisoner confined in a jail, prison, or other
correctional facility, for mental or emotional injury suffered while in custody
without a prior showing of physical injury.” Id.
107
See Carey v. Piphus, 435 U.S. 247, 264 (1978) (“[A]lthough mental and
emotional distress caused by the denial of procedural due process itself is
compensable under § 1983, we hold that neither the likelihood of such injury nor
the difficulty of proving it is so great as to justify awarding compensatory
damages without proof that such injury actually was caused.”); Slicker v.
Jackson, 215 F.3d 1225, 1229 (11th Cir. 2000) (holding that “compensatory
damages under § 1983 may be awarded only based on actual injuries caused by
the defendant and cannot be presumed or based on the abstract value of the
constitutional rights that the defendant violated”) (citing Carey, 435 U.S. at 264;
Memphis Community School District v. Stachura, 477 U.S. 299, 309 (1986));
But see Durrell v. Cook, 71 Fed. Appx. 718, 719 (9th Cir. 2003) (holding that
“mental injury suffices for Eighth Amendment cruel and unusual punishment
cases”). For a discussion of how the physical injury requirement fails to further
the goals of the PLRA and may be unconstitutional, see Jennifer Winslow,
Comment, The Prison Litigation Reform Act’s Physical Injury Requirement
Bars Meritorious Lawsuits: Was It Meant To?, 49 UCLA L. REV. 1655 (2002).
108
18 U.S.C. § 3626(a) (2005). “Prospective relief . . . shall extend no
further than necessary to correct the violation of the Federal right of a particular
plaintiff.” Id. § 3626(a)(1)(A). “Preliminary injunctive relief must be narrowly
drawn, extend no further than necessary to correct the harm the court finds
requires preliminary relief, and be the least intrusive means necessary to correct
that harm.” Id. § 3626(a)(2).
109
Id. § 3626(a). “The court shall not grant or approve any prospective
relief unless the court finds that such relief is narrowly drawn, extends no further
than necessary to correct the violation of the Federal right, and is the least
intrusive means necessary to correct the violation of the Federal right.” Id.
110
See Nami v. Fauver, 82 F.3d 63 (3d Cir. 1996) (inmates housed in
protective custody sought injunctive relief challenging conditions of their
RIES MACROED CORRECTED 053105.DOC
938
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
In the context of prison sexual assault cases, the PLRA creates
numerous procedural hurdles. Researchers have noted that the
Act’s exhaustion requirement is particularly problematic because
inmates fear retaliation from their assailants if they make use of
administrative grievance procedures by reporting threats or even
actual assaults.111 Further, the physical injury requirement calls for
evidence that prison medical or psychiatric facilities may not
reliably collect or maintain.112 Moreover, in seeking injunctive
relief, prisoner-plaintiffs basing their claims on continuing threats
of assault have a difficult burden in proving the likelihood that
they will be attacked again in the future.113 Finally, the PLRA
could limit relief to an order concerning a specific cellmate’s
protective custody status, excluding broader suits that would
protect other inmates as well.114
confinement in relation to general population inmates); Skinner v. Uphoff, 234
F. Supp. 2d 1208 (D. Wyo. 2002) (inmates who suffered a risk of assault from
their conditions of confinement at a Wyoming state prison prevailed in having
the court instruct prison management to consent to enforcement of specific
administrative regulations already in place at the facility).
111
See Flint v. Ky. Dep’t of Corr., 270 F.3d 340 (6th Cir. 2001) (deciding a
suit brought by a father after his son was murdered in prison by another inmate
who mistakenly “accused Flint of ‘ratting’ on him”); Cindy Struckman-Johnson
& David Struckman-Johnson, Sexual Coercion Rates in Seven Midwestern
Prison Facilities for Men, 80 THE PRISON J. 379, 380 (2000). “[M]any
researchers have noted that sexual assault is likely to be underreported by male
inmates because of fears of reprisals, unwillingness to be a ‘snitch,’ and fear of
being labeled a homosexual or weak.” Id. (citations omitted).
112
Butler v. Dowd, 979 F.2d 661, 669 (8th Cir. 1992) (affirming plaintiffs’
award of nominal damages for multiple rapes because no “objective medical
evidence supporting their physical injuries or detailing the extent of their
emotional injuries” was presented at trial).
113
Id. at 674 (“Although [plaintiff] claims that he is still subject to threats
from fellow inmates, he does not claim that he is still subject to sexual
assault.”).
114
Cf. Miller v. French, 530 U.S. 327, 347 (2000) (holding that “[t]he
PLRA has restricted courts’ authority to issue and enforce prospective relief
concerning prison conditions, requiring that such relief be supported by findings
and precisely tailored to what is needed to remedy the violation of a federal
right”).
RIES MACROED CORRECTED 053105.DOC
6/6/2005 1:43 PM
THE PRISON RAPE ELIMINATION ACT
939
II. SUPERVISORY LIABILITY FOR INMATE-ON-INMATE ASSAULTS
In Rhodes v. Chapman, the Supreme Court addressed the
problem of prison overcrowding and held that officials have no
general duty to make prisons “free of discomfort.”115 Officials in
that case were not liable for inmate-on-inmate assaults under the
Eighth Amendment where those assaults were claimed to be a
result of overcrowded conditions and the double-celling of
inmates.116 In Farmer v. Brennan, the Supreme Court heard the
case of a prisoner who claimed an Eighth Amendment violation
grounded in the fact that prison officials had imposed
unconstitutional conditions on the prisoner specifically and had
allowed the prisoner to be assaulted by other inmates.117 The
“deliberate indifference” standard defined in that decision has
115
Rhodes v. Chapman, 452 U.S. 336, 349 (1981). “[T]he Constitution
does not mandate comfortable prisons, and prisons . . . which house persons
convicted of serious crimes, cannot be free of discomfort.” Id.
116
Id. at 342-43, 348-49. “As to violence, the [trial] court found that the
number of acts of violence at [the prison] had increased with the prison
population, but only in proportion to the increase in population. Respondents
failed to produce evidence establishing that double celling itself caused greater
violence . . . .” Id. at 342-43. This led the Court to hold that “there is no
evidence that double celling under these circumstances either inflicts
unnecessary or wanton pain or is grossly disproportionate to the severity of
crimes warranting imprisonment.” Id. at 348. But see id. at 375. “There is no
dispute that the prison was violent even before it become overcrowded, and that
it has become more so. Contrary to the contention by the majority, I do not
assert that violence has increased due to double celling. I accept the finding of
the District Court that violence has increased due to overcrowding.” Id.
(Marshall, J., dissenting) (emphases and citations excluded).
117
Farmer v. Brennan, 511 U.S. 825 (1994).
[Plaintiff’s] complaint alleged that respondents either transferred
petitioner to [a particular facility] or placed petitioner in [that prison’s]
general population despite knowledge that the penitentiary had a
violent environment and a history of inmate assaults, and despite
knowledge that petitioner, as a transsexual who “projects feminine
characteristics,” would be particularly vulnerable to sexual attack by
some [of the prison’s] inmates. This allegedly amounted to a
deliberately indifferent failure to protect petitioner’s safety, and thus to
a violation of petitioner’s Eighth Amendment rights.
Id. at 830-31.
RIES MACROED CORRECTED 053105.DOC
940
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
since been applied to claims in which there is a connection
between a prison official’s conduct and the assault of an inmate by
another inmate.118 Prisoners have sought to hold supervisors liable
in this way for conditions such as those unsuccessfully challenged
in Rhodes v. Chapman.119 Some circuits have held that when a
prisoner’s particular double-cell assignment leads to a violent
assault, a prison supervisor may be found liable for deliberate
indifference to the prisoner’s safety.120
A. THE FARMER DELIBERATE INDIFFERENCE STANDARD
Following Estelle v. Gamble, federal courts differed with
respect to the mental state required to subject prison officials to
liability for prison conditions.121 Ten years after it considered the
objective seriousness of injuries suffered by prisoners to find
Eighth Amendment violations in Rhodes v. Chapman, the Court
refocused the Eighth Amendment test on an inquiry into the
mindset of prison officials in Wilson v. Seiter.122 The Court
cautioned that a condition’s effect on an inmate was not
determinative of a violation of the Eighth Amendment; rather, a
prison official’s actions in subjecting an inmate to a specific
condition of confinement, taken with deliberate indifference to the
violation of an inmate’s rights, would determine the existence of
an Eighth Amendment violation.123
Almost twenty years after defining the deliberate indifference
standard for conditions of confinement claims in Estelle v.
118
See infra Part II.B.
See infra Part II.C. (discussing cases arising out of assaults that
allegedly occurred because of prisons’ use of double-celling).
120
See infra Part II.C.
121
Compare Wilson v. Seiter, 501 U.S. 294, 297 (1991) (“In Estelle v.
Gamble . . . we said . . . only the unnecessary and wanton infliction of pain
implicates the Eighth Amendment”) (emphasis in original), with Stokes v.
Delacambre, 710 F.2d 1120, 1126 (5th Cir. 1983) (describing that the decision
in Smith v. Wade, 461 U.S. 30 (1983), “specifically upheld a standard of
recklessness”).
122
Wilson v. Seiter, 501 U.S. 294 (1991).
123
Id. at 303.
119
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
941
Gamble, the Court decided Farmer v. Brennan and more clearly
defined for the federal courts the requirements of the deliberate
indifference standard.124 This clarification came in the context of a
prison rape claim.125 Dee Farmer, an eighteen-year-old transsexual
convicted of credit card fraud, was placed in the general population
of a federal prison and was raped numerous times by fellow
inmates.126 Farmer brought a pro se civil suit against the prison’s
warden and guards for failing to protect her from an ongoing threat
of foreseeable sexual assault.127
The Farmer decision reaffirmed that conditions of confinement
could be violative of the Eighth Amendment.128 Further, the Court
set forth a two-prong test for determining when prison officials
could be held liable for constitutional rights violations related to
prison conditions.129 Under the first prong, courts must consider
whether the prison conditions were “objectively, sufficiently
serious.”130 This prong makes use of the standard set forth in
124
Farmer v. Brennan, 511 U.S. 825, 828 (1994).
Id. at 830 (stating that “according to petitioner’s allegations, petitioner
was beaten and raped by another inmate in petitioner’s cell. . . . [P]etitioner then
filed a Bivens complaint, alleging a violation of the Eighth Amendment”).
126
Id. at 829 (describing that the plaintiff, Dee Farmer, “underwent
estrogen therapy, received silicone breast implants, and submitted to
unsuccessful ‘black-market’ testicle-removal surgery”); Farmer v. Brennan, 81
F.3d 1444, 1445 (7th Cir. 1996). “Farmer is serving a twenty year federal
sentence for credit card fraud, which was imposed in 1986 when she was 18
years old.” Id.
127
Farmer v. Brennan, 511 U.S. 825, 829 (1994). This note uses the female
pronoun to refer to Dee Farmer because, as stated by the Seventh Circuit in a
decision following the Supreme Court’s remand of the case, “Farmer uses the
female pronoun to refer to herself, despite the fact that she is still biologically
male. We will respect that preference . . . .” Farmer v. Brennan, 81 F.3d 1444,
1445 n.1 (7th Cir. 1996).
128
Farmer, 511 U.S. at 832. “[T]reatment a prisoner receives in prison and
the conditions under which he is confined are subject to scrutiny under the
Eighth Amendment.” Id. (quoting Helling v. McKinney, 509 U.S. 25, 31
(1993)).
129
Farmer, 511 U.S. at 834. “[A] prison official violates the Eighth
Amendment only when two requirements are met.” Id.
130
Id. at 834 (citing Wilson v. Seiter, 501 U.S. 294, 298 (1991); Hudson v.
McMillian, 503 U.S. 1, 5 (1992)).
125
RIES MACROED CORRECTED 053105.DOC
942
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
Rhodes v. Chapman, which requires that the conditions challenged
must deprive a plaintiff of a human need.131 To claim that prison
officials failed to prevent a violation of an inmate’s need for safety
under the objective prong, a plaintiff must demonstrate “conditions
posing a substantial risk of serious harm.”132
Under the second prong, the Court must determine whether
prison officials had a “sufficiently culpable state of mind.”133 In
clarifying the subjective prong of the standard, the Court
considered both civil standards for liability as well as different
criminal mental states.134 Although prisoners’ rights litigation
consists of civil lawsuits, the Court adopted something akin to the
criminal standard of “subjective recklessness.”135 The result was a
131
Id. at 834 (holding that “a prison official’s act or omission must result in
the denial of the minimal civilized measure of life’s necessities”) (citing Rhodes
v. Chapman, 425 U.S. 337, 347 (1981)).
132
Id. at 834. “For a claim (like the one here) based on a failure to prevent
harm, the inmate must show that he is incarcerated under conditions posing a
substantial risk of serious harm.” Id. (citing McKinney, 509 U.S. at 35).
133
Farmer, 511 U.S. at 837. As described by Judge Posner in a Seventh
Circuit prisoner rape case:
Prison employees who act with deliberate indifference to the inmates’
safety violate the Eighth Amendment. But to be guilty of “deliberate
indifference” they must know they are creating a substantial risk of
bodily harm. If they place a prisoner in a cell that has a cobra, but they
do not know that there is a cobra there (or even that there is a high
probability that there is a cobra there), they are not guilty of deliberate
indifference even if they should have known about the risk, that is,
even if they were negligent—even grossly negligent or even reckless in
the tort sense—in failing to know. But if they know that there is a cobra
there or at least that there is a high probability of a cobra there, and do
nothing, that is deliberate indifference.
Billman v. Indiana Dep’t of Corr., 56 F.3d 785, 788 (7th Cir. 1995) (citations
omitted).
134
Farmer, 511 U.S. at 836-37. “With deliberate indifference lying
somewhere between the poles of negligence at one end and purpose or
knowledge at the other,” the Court considered the civil and criminal standards of
“recklessness,” as well as the civil liability standard of “gross negligence.” Id. at
836-37; see also id. at 836 n.4 (dismissing gross negligence as “typically
meaning little different from recklessness as generally understood in the civil
law”).
135
Id. at 839-40 (stating “subjective recklessness as used in the criminal
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
943
deliberate indifference standard “somewhere between the poles of
negligence at one end and purpose or knowledge at the other.”136
The threshold for this standard is that an official must have been
aware or should have been aware of the unconstitutional conditions
established under the objective prong.137 The Court noted that, in
Dee Farmer’s case, the record contained statements by some
defendants admitting knowledge of Farmer’s vulnerability to
sexual assault.138 The Supreme Court generally limited the liability
of prison supervisors, however, by suggesting that “it remains open
to the [defendant] officials to prove that they were unaware even of
an obvious risk to inmate health and safety.”139
Despite this difficult standard, the language of the Farmer
decision suggests that demonstrating prison conditions that present
a general threat of rape may be sufficient to satisfy both the
objective and subjective prongs of the deliberate indifference
test.140 Courts thus might presume that a defendant had notice of a
law is a familiar and workable standard that is consistent with the Cruel and
Unusual Punishments Clause”).
136
Id. at 836.
137
Id. at 837 (holding “a prison official cannot be found liable under the
Eighth Amendment for denying an inmate humane conditions of confinement
unless the official knows of and disregards an excessive risk to inmate safety”).
138
Id. at 848-49.
For example, in papers filed in opposition to respondents’ summaryjudgment motion, petitioner pointed to respondents’ admission that
petitioner is a “non-violent” transsexual who, because of petitioner’s
“youth and feminine appearance” is “likely to experience a great deal
of sexual pressure” in prison. And petitioner recounted a statement by
one of the respondents, then warden of the penitentiary in Lewisburg,
Pennsylvania, who told petitioner that there was “a high probability that
[petitioner] could not safely function at USP-Lewisburg” an incident
confirmed in a published District Court opinion.
Id. (citations omitted).
139
Farmer, 511 U.S. at 844.
140
Id. at 842-43.
For example, if an Eighth Amendment plaintiff presents evidence
showing that a substantial risk of inmate attacks was longstanding,
pervasive, well-documented, or expressly noted by prison officials in
the past, and the circumstances suggest that the defendant-official being
sued had been exposed to information concerning the risk and thus
RIES MACROED CORRECTED 053105.DOC
944
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
threat to the plaintiff because the prison’s conditions posed such an
obvious risk to all inmates.141 Plaintiffs may show that a risk was
“long-standing, pervasive, well-documented or expressly noted by
prison officials in the past” and then link this information to their
supervisors’ awareness of the threat.142
Under Farmer, “[a] failure to give advance notice is not
dispositive”;143 rather, “a factfinder may conclude that a prison
official knew of a substantial risk from the very fact that the risk
was obvious.”144 Prior to Farmer, a prison guard or supervisor
could be held deliberately indifferent only when an inmate
reported a threat from a particular, identified aggressor.145 In
Farmer, however, the Court held that prison officials may be held
liable when they are aware of a risk to an inmate despite their not
knowing by whom that inmate is threatened.146 The Farmer
decision focused the inquiry of courts on the conduct of prison
officials rather than on prisoners’ own steps to inform prison
“must have known” about it, then such evidence could be sufficient to
permit a trier of fact to find that the defendant-official had actual
knowledge of the risk.
Id.
141
Thus, when a defendant is aware of prison conditions generally, “a fact
finder may conclude that a prison official knew of a substantial risk from the
very fact that the risk was obvious.” Id. at 842.
142
Id.
143
Id. at 848.
144
Id. at 842.
145
McGill v. Duckworth, 944 F.2d 344, 349-50 (7th Cir. 1991) (deciding
inmate-on-inmate rape claim prior to Farmer and holding that “[o]ther circuits
have held that failure to tell prison officials about threats is fatal and have
dismissed such claims at the pleading stage”) (citing Ruefly v. Landon, 825 F.2d
792 (4th Cir. 1987); Blankenship v. Meachum, 840 F.2d 741 (10th Cir. 1988)).
146
Farmer v. Brennan, 511 U.S. 825, 843 (1994). “Nor may a prison
official escape liability for deliberate indifference by showing that, while he was
aware of an obvious risk to inmate safety, he did not know that the complainant
was especially likely to be assaulted by the specific prisoner who eventually
committed the assault.” Id. For examples of courts using this language in
determining deliberate indifference, see Greene v. Bowles, 361 F.3d 290, 294
(6th Cir. 2004); Krein v. Norris, 309 F.3d 487, 491-92 (8th Cir. 2002); Lopez v.
LeMaster, 172 F.3d 756, 762 n.5 (10th Cir. 1999).
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
945
officials of aggressive conduct by other inmates.147
The Court also held that defendants may prevail “if they
responded reasonably to the risk even if the harm ultimately was
not averted.”148 Indeed, the official’s response could preclude a
prisoner’s claim for an injunction, even if that response postdates
litigation.149 A prisoner seeking injunctive relief must demonstrate
a threat of future harm “in light of the prison authorities’ current
attitudes and conduct”;150 thus, the defendants “could prevent
issuance of an injunction by proving, during the litigation, that they
were no longer unreasonably disregarding an objectively
intolerable risk of harm and that they would not revert to their
obduracy upon cessation of the litigation.”151
B. Supervisors’ Liability for Reported or Obvious Threats of
Harm
Following from the Supreme Court’s precedents stating that
personal safety is a human need that cannot be deprived under the
Eighth Amendment,152 prisoners who are the victims of assault
may try to hold prison supervisors directly liable for not preventing
147
Farmer, 511 U.S. at 843-44. “If, for example, prison officials were
aware that inmate rape was so common and uncontrolled . . . it would be
obviously irrelevant to liability that the officials could not guess beforehand
precisely who would attack [victims of assault].” Id.
148
Id. at 844.
149
Id. at 847 n.9 (“[E]ven prison officials who had a subjectively culpable
state of mind when the lawsuit was filed could prevent issuance of an injunction
by proving, during the litigation, that they were no longer unreasonably
disregarding an objectively intolerable risk of harm and that they would not
revert to their obduracy upon cessation of the litigation.”).
150
Id. at 845 (emphasis added).
151
Id. at 847 n.9.
152
Helling v. McKinney, 509 U.S. 25, 33 (1994). “The [Eighth]
Amendment, as we have said, requires that inmates be furnished with the basic
human needs, one of which is ‘reasonable safety.’ It is ‘cruel and unusual
punishment to hold convicted criminals in unsafe conditions.’” Id. (citing
Deshaney v. Winnebago County Dep’t of Social Serv., 489 U.S. 189, 200
(1989); Youngberg v. Romeo, 457 U.S. 307, 315-16 (1982)).
RIES MACROED CORRECTED 053105.DOC
946
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
their harm.153 Without evidence establishing a direct connection
between an official’s own conduct and an inmate’s harm, however,
a warden or corrections official will be liable only for inmate-oninmate assaults the circumstances of which were within the
official’s control.154 As with other types of Section 1983 claims,
wardens and other supervisors will not be held liable for claims
against prison guards through respondeat superior.155 Rather, the
series of factors listed by Professor Kinports and quoted above will
be applied to determine supervisory liability in prisoner assault
cases.156
153
See, e.g., Johnson v. Johnson, 385 F.3d 503, 512 (5th Cir. 2004) (“This
is a § 1983 suit brought by a former Texas prisoner against . . . defendant prison
officials [who] failed to protect him from prison gangs who repeatedly raped
him.”); Greene v. Bowles, 361 F.3d 290, 292 (6th Cir. 2004) (reviewing a
“[Section] 1983 suit against Warden Brigano and other prison officials resulting
out of an attack on Greene by another inmate”); Weiss v. Cooley, 230 F.3d
1027, 1029 (7th Cir. 2000) (addressing a claim by the plaintiff “that the jail
officials put him with other inmates who the officials knew would attack him”).
154
LaMarca v. Turner, 662 F. Supp. 647, 663 (S.D. Fla. 1987). “Where the
defendants hold supervisory positions, vicarious liability will not suffice.” Id.
(citing Monell v. Dep’t of Social Serv., 436 U.S. 658 (1978) for the proposition
that a municipality will not be vicariously liable for the acts of its employees
absent evidence that the injury inflicted was the result of official policy).
155
“Respondeat superior or vicarious liability will not attach under §
1983.” City of Canton v. Harris, 489 U.S. 378, 385 (1989) (citing Monell, 436
U.S. at 694-95).
156
Kinports, supra note 42, at 169.
[T]he courts of appeals tend to agree that five interrelated factors ought
to be considered in . . . determining whether a particular supervisor is
liable on the facts of a given case: (1) the extent to which prior similar
incidents have occurred; (2) the supervisor’s response to those prior
incidents; (3) the supervisor’s response to the specific incident on
which the suit is based; (4) the extent to which the supervisor can be
considered a cause of the violation; and (5) the nature of the
supervisor’s awareness of the constitutional misconduct.
Id. These same factors have been used by courts considering supervisory
liability in inmate-on-inmate assault cases. For consideration of the occurrence
of prior similar incidents, see Skinner v. Uphoff, 234 F. Supp. 2d 1208, 1214 (D.
Wyo. 2002). The court granted plaintiffs’ motion for summary judgment on a
claim of unconstitutional conditions at a prison, in part because the plaintiffs
demonstrated that “between one hundred and three hundred inmate assaults”
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
947
In a case brought under Section 1983, one of the factors to be
considered in assessing supervisory liability is the supervisor’s
awareness of the constitutional misconduct for which the plaintiff
brings suit.157 Under the Farmer standard, a plaintiff-inmate has
occurred at the prison during a six year period without sufficient response from
prison supervisors.” Id. at 1214. The supervisor’s response to prior incidents was
considered in Ware v. Jackson County, 150 F.3d 873 (8th Cir. 1998). The court
in that case affirmed a denial of the defendant county’s motion for judgment as a
matter of law in a suit over the plaintiff’s rape by a subordinate corrections
officer. The court stated that the corrections supervisors “were informed of the
first set of allegations against [the subordinate]. However, there is no evidence
that they were instructed to increase their supervision of [the subordinate].” Id.
at 884. The defendant-supervisor’s response to the incident before the court was
a factor in Giroux v. Somerset County, 178 F.3d 28, 34 (1st Cir. 1999). By that
decision, the court reversed summary judgment of inmate’s claim against a
supervisor, where “[a] juror could find that [defendant]’s abdication of his
responsibility [to alert subordinates on his shift of plaintiff’s protective status],
in the face of such a known danger to [plaintiff]’s safety, was a reckless
dereliction of duty.” Id. at 34. The extent to which the supervisor himself caused
the alleged incident was an important consideration in Hale v. Tallapposa
County, 50 F.3d 1579 (11th Cir. 1995). Summary judgment for a jail supervisor
was reversed where the plaintiff “presented sufficient proof of causation to
survive summary judgment,” by presenting evidence
sufficient to support a reasonable jury determination that the excessive
risk of violence [at the facility] flowed from an atmosphere of
deliberate indifference reflected in [the supervisor]’s failure to classify
or segregate violent from non-violent inmates, assign inmates to cells
or beds, adequately train the jailers, and adequately supervise and
monitor the inmates.
Id. at 1584-85. Finally, the court looked to the supervisor’s awareness of
constitutional misconduct in Skinner v. Uphoff, 234 F. Supp. 2d 1208 (D. Wyo.
2002). Prison supervisors were found liable on a failure-to-discipline claim
because they “admitted their knowledge of a substantial risk of serious harm to
inmates, as well as their failure to discipline subordinates or take any other
corrective action. In light of the undisputed facts, their failure to disciple
amount[ed] to deliberate indifference.” Id. at 1216. Also, in Daskalea v. District
of Columbia, the court affirmed the district’s liability for negligent supervision
where the plaintiff inmate’s Eighth Amendment rights were violated by
“persistent, open and notorious conduct,” which supervisors “failed to notice, let
alone stop.” 227 F.3d 433, 445 (D.C. Cir. 2000).
157
Kinports, supra note 42, at 180-81.
The final factor that the courts of appeals tend to consider in
RIES MACROED CORRECTED 053105.DOC
948
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
the best chance of proving the subjective prong of deliberate
indifference when the inmate can show that he reported a threat of
assault to the supervisor-defendant.158 For example, a plaintiff in
Arkansas who had been sexually assaulted in prison claimed that
the warden of his facility was liable because the warden had
reviewed some of the prisoner’s requests for a change in cell
assignment.159 The trial court granted the warden judgment as a
matter of law, holding that he lacked subjective knowledge of the
risk to the plaintiff.160 The Eighth Circuit, in Spruce v. Sargent,
reversed this finding on the basis of evidence that the warden had
also received reports that were filed by the prisoner after his
cellmate forced him to perform oral sex.161 The court held that this
evidence was sufficient for a jury to find that the warden had the
notice required for deliberate indifference.162
In determining whether to hold supervisors liable under Section
1983 for inmate-on-inmate assaults, courts also consider “the
extent to which the supervisor can be considered a cause of the
violation.”163 Officials can be linked to a deprivation of
determining a supervisor’s liability for her subordinate’s
unconstitutional behavior is the nature of the supervisor’s awareness of
the risk of constitutional injury. The greater the supervisor’s awareness
of the problem, the more culpable she seems and the more likely the
courts are to conclude that their particular standard of supervisory
liability is met.
Id.
158
McGill v. Duckworth, 944 F.2d 344, 349-50 (7th Cir. 1991). “Other
circuits have held that failure to tell prison officials about threats is fatal and
have dismissed such claims at the pleading stage.” Id. (citing Fourth Circuit and
Tenth Circuit cases).
159
Spruce v. Sargent, 149 F.3d 783, 786 (8th Cir. 1998).
160
Id. at 785.
161
Id. at 786.
162
Id. The Court of Appeals held “that there was sufficient evidence from
which a jury could conclude that Warden Sargent knew Spruce was subject to an
excessive risk of harm from sexual assault.” Id.
163
Kinports, supra note 42, at 178.
Given Section 1983’s requirement that the defendant “subject[ ]” the
plaintiff to a violation of her constitutional rights or “cause[ ] [her] to
be [so] subjected,” and the Supreme Court’s requirement in Rizzo v.
Goode of an “affirmative link” between the supervisory official and the
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
949
constitutional rights when their failure to train subordinates caused
those subordinates to act with deliberate indifference.164 The
connection between the resulting harm and the supervisor must be
established through proof that the failure to train employees
actually led to the violation of the individual’s constitutional
rights.165
In Lopez v. LeMaster, the Tenth Circuit found that plaintiff
Genaro Lopez failed to establish such a connection between a
government supervisor and the inmate-on-inmate assault he
suffered.166 Lopez was an arrestee in the Jackson County jail in
Oklahoma and was placed in a general population cell, where he
was threatened by another inmate.167 Although he reported the
threat to his jailer, the jailer returned Lopez to the same cell; Lopez
was subsequently beaten so severely as to leave jail the next day
with a concussion and a strained spine.168 Lopez brought suit
against the sheriff who supervised the jail, claiming that the jailer
plaintiff’s constitutional injury, causation issues often arise in cases
involving supervisory liability.
Id.
164
See City of Canton v. Harris, 489 U.S. 378 (1989) (finding municipality
liability in § 1983 suit for inadequate training of police that led to constitutional
rights violations); Walker v. City of New York, 974 F.2d 293, 297 (2d Cir.
1992) “[M]any § 1983 claims against municipalities [are molded] into ‘failure to
train’ or ‘failure to supervise’ claims. It is only by casting claims in this way that
plaintiffs can link an actual decision by a high level municipal official to a
challenged incident.” Id. The court went on to discuss the application of City of
Canton v. Harris to a claim that New York City was liable for its failure to train
police not to cover up exculpatory evidence or to commit perjury.
165
See Wilson v. Town of Mendon, 294 F.3d 1, 6 (1st Cir. 2002). “Liability
will attach to the municipal employer . . . where a specific deficiency in training
is the ‘moving force’ behind a constitutional injury.” Id. (citing City of Canton
v. Harris, 489 U.S. 378, 388-89, 391); Currier v. Doran, 242 F.3d 905, 923 (10th
Cir. 2001). “[W]hen a superior’s failure to train amounts to deliberate
indifference to the rights of persons with whom his subordinates come into
contact, the inadequacy of training may serve as the basis for § 1983 liability.”
Id. (quotations and citations omitted).
166
Lopez v. LeMaster, 172 F.3d 756, 760 (10th Cir. 1999).
167
Id. at 758.
168
Id. at 758-59.
RIES MACROED CORRECTED 053105.DOC
950
6/6/2005 1:43 PM
JOURNAL OF LAW AND POLICY
to whom he complained was not properly trained.169 This failureto-train claim could not be sustained, however, because following
his assault Lopez could not identify the jailer to whom he had
reported the incident.170 In addressing Lopez’s appeal of a
summary judgment order against him, the Tenth Circuit held that
the identity of the jailer was necessary in order to establish the
sheriff’s liability.171 Lopez was unable to demonstrate that the
particular jailer to whom he had complained had in fact not been
trained or that his training had been insufficient.172 Lopez’s suit
survived, however, on his separate claim that the sheriff was
generally responsible for the conditions at the jail.173
In Section 1983 cases against government supervisors, courts
also consider the supervisors’ responses to the alleged
constitutional violations as well as similar prior incidents in
determining liability.174 Putting in place measures that are known
169
Id. at 760 (“Appellant alleges that his jailer’s acts and omissions were
the result of Sheriff LeMaster’s failure to provide adequate training and
supervision of jail personnel.”)
170
Id.
171
Id.
172
Lopez v. LeMaster, 172 F.3d 756, 760 (10th Cir. 1999).
Appellant has presented no evidence concerning deficiencies in training
of the particular jailer involved in his case. Nor has he shown that the
county had a policy of providing its jailers with insufficient training in
the areas closely related to his ultimate injury from which we might
infer that his particular jailer’s training also was insufficient.
Id.
173
Id. at 760-62. See infra text accompanying notes 179-82.
174
Kinports, supra note 42, at 174-78. “As a general rule, the courts are
more likely to find a supervisor liable the less adequate the remedial steps she
has taken in response to prior violations.” Id. at 174.
The third factor that some courts of appeals consider in determining a
supervisor’s liability for her subordinate’s constitutional wrong is the
nature of the supervisor’s response to the particular incident that led to
the suit. Although some courts refuse to take this factor into account on
the grounds that the supervisor’s conduct subsequent to the
constitutional violation cannot in any way have contributed to it, other
courts view the supervisor’s failure to respond appropriately to the
violation as evidence that she was deliberately indifferent to it or
acquiesced in it and therefore met whatever standard of culpability the
RIES MACROED CORRECTED 053105.DOC
THE PRISON RAPE ELIMINATION ACT
6/6/2005 1:43 PM
951
to be ineffective or enacting policies that are then not enforced will
not satisfy the duty of officials to prevent harm.175 Once an
awareness of a threat to inmates is established, supervising
officials have a duty to discipline subordinates who are complacent
about potential harm to inmates.176 Officials may be found
deliberately indifferent if, after a general risk to inmate safety
becomes known, they fail in their duty to prevent future assaults.177
Courts have found that plaintiffs’ claims are sufficient when they
assert that unsafe conditions at corrections facilities were caused
by supervisors’ deliberate indifference and the claims are
supported by evidence that prison officials knew of a general risk
of rape to inmates.178
court has chosen.
Id. at 177.
175
See Daskalea v. District of Columbia, 227 F.3d 433, 442 (D.C. Cir.
2000) (affirming liability of the