THE CONSTRAINT PRINCIPLE:
ORIGINAL MEANING AND CONSTITUTIONAL PRACTICE
Lawrence Solum
Carmack Waterhouse Professor of Law
Georgetown Law Center
Legal Theory Workshop
UCLA School of Law
Thursday, March 2, 2017, 5:30p.m.-7:00p.m.
Room 1314 (Law Building)
Light refreshments will be available at 5:15 pm and you are welcome to come a little
early to mingle.
Draft, 02/15/2017. For UCLA Workshop. Please
Don’t Cite Or Quote Without Permission.
MEMORANDUM
February 15, 2017
To:
Readers at UCLA
From:
Lawrence Solum
Re:
Suggestions for Reading “The Constraint Principle”
Thank you for reading “The Constraint Principle: Original Meaning and Constitutional
Practice.” This article forms part of a larger project, in which I defend each of the major
claims made by constitutional originalism. The prior articles in the series include:
Lawrence B. Solum, The Fixation Thesis: The Original Meaning of the
Constitutional Text, 91 Notre Dame L. Rev. 1 (2015).
Lawrence B. Solum, Originalism and Constitutional Construction, 82 Fordham L.
Rev. 453 (2013).
Lawrence B. Solum, Originalist Methodology, U. Chi. L. Rev. (forthcoming
2016-17).
You have received two versions of the paper—a short version and the full paper.
Of course, I would be thrilled and grateful if you read the entire draft, but the short
version includes these sections from the longer paper:
(1) Part I.A, "Originalism as a Family of Constitutional Theories (pages 5-9),
(2) Part II.C.2, “A Minimalist Version of the Constraint Principle: Consistency of
Constitutional Doctrine,” (pages 23-25),
(3) Part IV, "Justifying the Constraint Principle,” (pages 54-79), and
(4) Part VII, “Pairwise Comparisons: Rivals, the Status Quo, and Compromise”
(104-132).
Those sections are a total of 60 pages—about the length of a typical law review article.
The most important arguments appear in Part IV, pages 54-79.
For those who read the full paper, I am especially interested in reactions to the following
questions:
(1) Have I correctly identified “Constraint as Consistency” as the formulation of
the constraint principle that provides the most plausible dividing line between
originalism and nonoriginalism? This issue is discussed in Part II.C, pages 22-26.
(2) Does the claim that there are pro tanto justifications for the Constraint
Principle seem reasonable? Does it follow that the real action in evaluating the
Constraint Principle should the countervailing reasons (i.e., the reasons for allow
courts to decide cases in ways that are inconsistent with the text)? This issue is
clarified by the discussion in Part III.B.2.d, on page 39.
(3) Does the argument that the proper method for evaluating constitutional
theories involves wide (and not narrow) reflective equilibrium seem reasonable?
This issue is discussed in Part III.B.1 from pages 32-36.
Thank you again!
D R A F T February 15, 2017
THE CONSTRAINT PRINCIPLE:
ORIGINAL MEANING AND CONSTITUTIONAL PRACTICE*
LAWRENCE B. SOLUM**
TABLE OF CONTENTS
Introduction .................................................................................................................... 2
I. The Role of the Constraint Principle in Contemporary Constitutional Theory .......... 5
A. Originalism as a Family of Constitutional Theories ............................................. 5
B. Constraint and the Debate Over Originalism and Living Constitutionalism ........ 9
C. What Hangs on Acceptance or Rejection of the Constraint Principle? ............... 17
II. Contribution, Constraint, and Restraint ................................................................... 18
A. The Distinction Between Contribution and Constraint ....................................... 19
B. The Distinction Between Constraint and Restraint ............................................. 20
C. Three Versions of the Constraint Principle .......................................................... 22
D. Defeasibility Conditions and Mending Constructions ........................................ 26
III. Framing the Debate Over the Constraint Principle ................................................ 28
A. The Types of Justification ................................................................................... 28
B. The Methods of Justification ............................................................................... 32
C. The Structure of Argumentation .......................................................................... 41
D. The Role of Pairwise Comparison ...................................................................... 53
IV. Justifying the Constraint Principle ......................................................................... 54
A. Constraint and the Rule of Law .......................................................................... 55
B. Constraint and Legitimacy .................................................................................. 74
C. The Relationship Between the Rule of Law and Legitimacy .............................. 79
V. Alternative Justifications for the Constraint Principle? ........................................... 79
A. Is the Constraint Principle the Law? ................................................................... 80
B. Is the Constraint Principle Implied by Writtenness? ........................................... 81
C. Does the Constraint Principle Lead to Good Consequences? ............................. 81
D. Is the Constraint Principle the Only Alternative to Judicial Lawlessness? ......... 82
VI. Objections and Answers......................................................................................... 82
*
© 2017 by the author. Permission is hereby granted to reproduce this work in whole or in part
(including distribution via the Internet) for educational or scholarly purposes. The author requests that any
reproduction provide a citation to the work that includes the author’s name, the title, and the journal or
website from which the article was obtained.
**
Carmack Waterhouse Professor of Law, Georgetown University Law Center. I owe thanks to
participants in a Faculty Workshop at Georgetown Law and at the Originalism Works in Progress
Conference for helpful comments and suggestions. I am especially grateful to Richard Fallon Greg Klass,
Marty Lederman, and Michael Seidman.
Lawrence B. Solum
A. The Canonical Cases Objection .......................................................................... 83
B. The Disruptive Force of an Originalist Big Bang ............................................... 86
C. Three Versions of the Injustice Objection ........................................................... 87
D. The Systemic Bias Objection .............................................................................. 95
E. The Objection from Majoritarianism ................................................................... 96
F. The Objection from Exclusion and Illegitimacy .................................................. 97
G. The Dead Hand Problem ................................................................................... 100
H. Objections that Do Not Clash with the Constraint Principle ............................ 101
VII. Pairwise Comparisons: Rivals, the Status Quo, and Compromise ..................... 104
A. Pairwise Comparison of Originalism with Nine Rivals .................................... 105
B. The Constitutional Status Quo........................................................................... 127
C. Constitutional Compromise ............................................................................... 130
D. Completing the Project of Pairwise Comparison .............................................. 132
VIII. A Restatement of the Case for the Constraint Principle .................................... 132
A. Summing Up: The Core Case for Constraint .................................................... 133
B. Reflective Equilibrium: Considered Judgments and Outcomes ........................ 134
C. Pairwise Comparisons Revisited ....................................................................... 134
D. Does It Matter Whether We Call It Originalism? .............................................. 136
Conclusion ................................................................................................................. 136
INTRODUCTION
Originalism is a family of constitutional theories that agree in principle on the
proposition that the original meaning of the constitutional text should constrain
constitutional practice. Call this idea the “Constraint Principle.”1 Provisionally,2 the
Constraint Principle is the claim that (at a minimum) the content of constitutional
doctrine and the decision of constitutional cases should be consistent with the original
meaning of the constitutional text. The aim of this article is to explicate and justify the
Constraint Principle. Originalists also agree on a second idea, that the linguistic meaning
(or more precisely, “communicative content”) of the constitutional text is fixed when
each provision is framed and ratified. We can call this second idea the “Fixation Thesis;”
that thesis is defended elsewhere.3 Together, constraint and fixation form the core of
contemporary originalist theory.
Originalists agree on fixation and constraint, but they disagree about other things. The
dominant strain of contemporary originalism emphasizes the public meaning of the
constitutional text, but other versions of originalism focus on the original intentions of the
framers or the original methods of constitutional interpretation and construction. The
1
The Constraint Principle is capitalized to indicate that it is being used as a proper noun phrase that
names the principle of constraint formulated in this Article. A similar convention will be used other
elements of the theory on offer here and in related work.
2
For a more complete statement of the Constraint Principle that is explicated in this Article, see infra
Part II.C.2, p. 23 (articulating a minimalist version of the Constraint Principle).
3
Lawrence B. Solum, The Fixation Thesis: The Original Meaning of the Constitutional Text, 91 NOTRE
DAME L. REV. 1 (2015).
2
The Constraint Principle
I. THE ROLE OF THE CONSTRAINT PRINCIPLE IN CONTEMPORARY CONSTITUTIONAL
THEORY
Perhaps the most important debates in contemporary constitutional theory cluster
around the disputes between originalists and living constitutionalists. Sometimes the
disagreement sometimes seems to focus on the idea of fixation, but the most important
disagreements are about other issues. One important issue concerns determinacy, with
some nonoriginalists taking the position that meaning of the constitutional text cannot
constrain because it is indeterminate; other nonoriginalists admit that the text is only
moderately underdeterminate, but suggest that the most important issues of constitutional
law are within the zones of underdeterminacy created by open texture or vagueness.
Ultimately, questions about the degree to which the meaning underdetermines
constitutional practice are empirical or theoretical—and not normative.
The extent of constitutional underdeterminacy is an important question, but there is an
even more fundamental normative disagreement among contemporary constitutional
theorists.15 To the extent that the constitutional text is clear, originalists believe that it is
binding: we owe a duty of fidelity to the original meaning of the Constitution.
Nonoriginalists think otherwise. Although they may believe that the constitutional text is
worthy of respect and consideration, they reject the claim that the meaning of the text
provides hard limits on constitutional practice. Somewhat contentiously, we can say that
nonoriginalists believe that judges have the power to override the linguistic meaning of
the text. That is, nonoriginalists reject the Constraint Principle.
The function of this Part of the Article is to investigate the role of the Constraint
Principle in contemporary constitutional theory. That investigation can begin by
elucidating the nature of “originalism,” starting with the observation that originalism is a
family of constitutional theories.
A. Originalism as a Family of Constitutional Theories
What is originalism? In prior work, I have argued that originalism is a family of
constitutional theories organized around two core ideas, fixation and constraint.16 Almost
all originalists agree with both the idea that the meaning of constitutional text is fixed at
the time each provision is framed and ratified and the idea that this fixed original
15
For an informative discussion of the importance of indeterminacy in debates about originalism, see
Heidi Kitrosser, Interpretive Modesty, 104 GEO. L. J. (forthcoming 2016).
16
Lawrence B. Solum, What Is Originalism? in THE CHALLENGE OF ORIGINALISM: ESSAYS UN
CONSTITUTIONAL THEORY (Grant Huscroft and Bradley W. Miller eds., Cambridge University Press,
2011). The idea that originalism is a family of theories organized around the core ideas of fixation and
constraint is widely accepted. See, e.g., William Baude, Is Originalism Our Law?, 115 COLUM. L. REV.
2349, 2362 (2015) (stating “originalism is a family of theories united by principles of fixation and
constraint); New Originalism combines the “fixation thesis” with the “constraint principle”; Leslie C.
Griffin, Hobby Lobby: The Crafty Case That Threatens Women's Rights and Religious Freedom, 42
HASTINGS CONST. L.Q. 641, 655-56 (2015); Ian Bartrum, Two Dogmas of Originalism, 7 WASH. U.
JURISPRUDENCE REV. 157 (2015) (describing fixation and constraint as the two dogmas of originalism); see
also Solum, supra note 3, at 6 n. 38 (collecting sources that accept the notion that originalism is a family of
theories organized around the Fixation Thesis and the Constraint Principle or advancing similar ideas about
the nature of originalism).
5
Lawrence B. Solum
meaning should constrain constitutional practice. We can express these two ideas in a
preliminary way as follows:
The Fixation Thesis: The meaning (or more precisely communicative content) of the
constitutional text is fixed at the time each provision is framed and/or ratified.17
The Constraint Principle: Constitutional practice, including the elaboration of
constitutional doctrine and the decision of constitutional cases, should be constrained
by the original meaning of the constitutional text. At a minimum, constraint requires
that constitutional practice be consistent with original meaning (as specified below).
Originalists agree on fixation and constraint, but they disagree on other questions. The
most important area of disagreement concerns the nature of original meaning. Among the
important variations are the following:
Public Meaning: The original meaning is the public meaning of the constitutional
text.
Framers’ Intentions: The original meaning is provided by the framer’s intentions.
Intentionalism has further variants, including forms that focus on purposive
intentions and communicative intentions.
Ratifiers’ Understandings: The original meaning is given by the understandings of
the ratifiers of each provision.
Original Methods: The original meaning is the meaning that would be given to the
text by the original methods of constitutional interpretation and construction.
Any particular originalist theory will combine fixation and constraint with an
understanding of the nature of original meaning. For example, Public Meaning
Originalism combines the Fixation Thesis and Constraint Principle with the Public
Meaning Thesis18—the claim that the original meaning of the constitutional text is a
function of the conventional semantic meaning of the words and phrases in the public
available context of constitutional communication. Theories of the nature of original
meaning might be thought of as modules that can plug into the theoretical structure of
originalism as a whole.
Originalists also disagree about the extent of constitutional underdeterminacy and the
role of constitutional construction. Some originalists may believe that the original
meaning is “thick” (rich in communicative content) and hence that constitutional
underdeterminacy is rare or nonexistent. Other originalists may believe that the answers
to some constitutional questions are underdetermined by the constitutional text:
underdetermination can result from language that is vague, open-textured, or irreducibly
17
For a more precise explication of the Fixation Thesis, see Solum, supra note 3, at 15-178. For
qualifications and clarifications, see id. at 30-42. The version explicated in that article is: “The object of
constitutional interpretation is the communicative content of the constitutional text, and that content was
fixed when each provision was framed and/or ratified.” See id. at 15. This formulation is substantially the
same as that in text, which is formulated to be more accessible to readers unfamiliar with theoretical
framework articulated in the prior article.
18
See Lawrence B. Solum, The Public Meaning Thesis (unpublished work-in-progress, August 20,
2105) (articulating and defending the thesis that the original meaning of the constitutional text is its public
meaning).
6
The Constraint Principle
ambiguous, or from gaps or contradictions within the text. And yet another group of
originalists may believe that specific provisions of the text are unclear but that the text as
a whole is determinate because it implicitly contains default rules—for example, a default
rule of deference to elected officials where the text does not otherwise determine the
outcome.
Ecumenical originalism acknowledges the range of reasonable disagreement among
originalists about the extent of underdeterminacy. But from a sectarian perspective, each
originalist takes a stand on this question. For the purpose of this Article, my position on
this question is an undefended assumption: the constitutional text is neither fully
determinate nor radically indeterminate: many constitutional questions are answered by
the communicative content but others are not. Let us call the thesis that significant
constitutional underdeterminacy exists the “Fact of Constitutional Underdeterminacy.”19
Finally, some originalists do (and all originalists should) embrace a distinction
between “interpretation” and “construction.” The notion of a distinction between
interpretation and construction goes back at least as far as 1839 when it was articulated
(in a different form) by Franz Lieber in his Legal and Political Hermeneutics.20 The
modern version of the distinction appears (perhaps for the first time, perhaps not) in
twentieth-century treatises on contract law by Williston and especially Corbin;21 and has
been deployed in many judicial decisions.22 It is important to note that some originalists
reject the interpretation-construction distinction, while others may accept the distinction
as a matter of abstract theory but reject its significance on the ground that the original
meaning of the constitutional text is sufficiently thick to settle all or almost all
constitutional questions.23
For the purposes of this Article, I will use the words “interpretation” and
“construction” in stipulated technical senses, as follows:
Constitutional Interpretation: The phrase “constitutional interpretation” is stipulated
to refer to the activity of that discerns the communicative content (linguistic
meaning) of the constitutional text.
19
See Solum, , supra note 3, at 10-13.
FRANCIS LIEBER, LEGAL AND POLITICAL HERMENEUTICS 43–44, 111 n.2 (Roy M. Mersky & J. Myron
Jacobstein eds., Wm. S. Hein & Co. 1970) (1839). Lieber’s version of the distinction does not explicitly
differentiate communicative content and legal content. See Greg Klass, Interpretation and Construction 1:
Francis Lieber, New Private Law: Project on the Foundations of Private Law,
http://blogs.harvard.edu/nplblog/2015/11/19/interpretation-and-construction-1-francis-lieber-greg-klass/;
Greg Klass, Interpretation and Construction 2: Samuel Williston, New Private Law: Project on the
Foundations of Private Law, https://blogs.harvard.edu/nplblog/2015/11/23/interpretation-and-construction2-samuel-williston-greg-klass/; Greg Klass, Interpretation and construction 3: Arthur Linton Corbin,
http://blogs.harvard.edu/nplblog/2015/11/25/interpretation-and-construction-3-arthur-linton-corbin-gregklass/; see also Ralf Poscher, The Hermeneutical Character of Legal Construction, (November 28, 2015).
Simone Glanert and Fabien Girard (eds), Law’s Hermeneutics: Other Investigations (London: Routledge,
2016). Available at SSRN: http://ssrn.com/abstract=2696486.
21
4 Williston, Contracts §§ 600-02 (3d ed.1961); 3 Corbin, Contracts §§ 532-35 (1960 & Supp.1980).
22
See Fausel v. JRJ Enters., Inc., 603 N.W.2d 612, 618 (Iowa 1999); In re XTI Xonix Technologies
Inc., 156 B.R. 821, 829 n. 6 (D.Ore.1993); Berg v. Hudesman, 115 Wash.2d 657, 663, 801 P.2d 222, 226
(1990). More examples are collected in Solum, supra note 11, at 486-87.
23
See Solum, Originalism and Constitutional Construction, supra note 11 (discussing originalist
positions on the interpretation-construction distinction).
20
7
Lawrence B. Solum
Constitutional Construction: The phrase “constitutional construction” is stipulated
to refer to the activity that determines the content of constitutional doctrine and the
legal effect of the constitutional text (including the decision of constitutional cases
by the courts).24
Interpretation-Construction Distinction: The phrase “Interpretation-Construction
Distinction” will be used to designate the distinction as articulated via the stipulated
definitions immediately above.
The phrase “new originalism”25 has no definite content, but it will be used in this
Article to name a version of originalism that embraces the Fixation Thesis and the
Constraint Principle. Some new originalists emphasize public meaning; others are
focused on original methods or the original law. Given the Fact of Constitutional
Underdeterminacy and the Interpretation-Construction Distinction, many (but not all)
new originalists believe in the existence of what we can call “construction zones.” Such
zones encompass the cases and issues with respect to which the communicative content
of the constitutional text does not fully constrain constitutional practice.
Different versions of originalism might adopt different approaches to the construction
zone. One possibility is that decisions in the construction zone should be guided by a
principle of deference: for example, the courts could defer to Congress (or other political
institutions) where the meaning of the constitutional text was unclear.26 Or construction
might be guided by a “presumption of liberty.”27 There are many other possibilities,
including the use of moral readings, original purposes, or common law methods in the
construction zone. This issue is important, but will be set aside for the remainder of this
Article.
Finally, one recent strand of originalist theory emphasizes the idea that originalism can
be viewed as a theory of constitutional change. This has been implicit in the originalist
claim that the constitutional actors (especially judges) lack the power to adopt
constitutional constructions that are the function equivalent of constitutional
amendments.28 But there are both theoretical and empirical questions implicated by this
originalist observation. Stephen Sachs articulates the idea that originalism is a theory of
constitutional change as follows:
What originalism requires of legal change is that it be, well, legal; that it be
lawful, that it be done according to law .... The originalist claim is that each change
in our law since the Founding needs a justification framed in legal terms, and not
24
These definitions were presented in Solum, supra note 11, at 457 (2013).
See Evan S. Nadel, The Amended Federal Rule of Civil Procedure 11 on Appeal: Reconsidering
Cooter & Gell v. Hartmarx Corporation, 1996 ANN. SURV. AM. L. 665, 691 n. 191 (“An example of the
‘textualism’ to which I refer is the ‘new originalism’ theory often associated with Justice Scalia.”); Randy
E. Barnett, An Originalism for Nonoriginalists, 5 Loy. L. Rev. 611, 620 (1999); Keith Whittington, The
New Originalism, 2 GEO. J.L. & PUB. POL’Y 599 (2004).
26
See Solum, The Fixation Thesis, supra note 3, at 5-6; Solum, Originalism and Constitutional
Construction, supra note 11, 511–23 (discussing Paulsen and Lawson’s default rules approach).
27
See RANDY E. BARNETT, RESTORING THE LOST CONSTITUTION (2d ed. 2013).
28
See Randy E. Barnett, An Originalism for Nonoriginalists, 45 LOY. L. REV. 611, 635 (1999) (stating,
“a proper respect for the writtenness of the text means that those committed to this Constitution have no
choice but to respect the original meaning of its text until it is formally amended in writing.”).
25
8
The Constraint Principle
just social or political ones. To put it another way, originalists believe that the
American legal system hasn't yet departed (even a little bit) from the Founders' law
in the way that the colonies threw off the British yoke or the states got rid of the
Articles of Confederation.29
This position could entail a modified understanding of the constraint principle: if the
original methods of constitutional interpretation and construction can be used to authorize
still other methods and those methods authorize amending constitutional constructions,
then originalism could, in theory, be consistent with some violations of the constraint
principle. Whether it does so or not involves complex questions about originalist
theory—are original methods of construction part of the Constitution?—and history—did
the original methods authorize amending constructions as a matter of historical fact?
These questions will be set to the side for the remainder of this article.
B. Constraint and the Debate Over Originalism and Living Constitutionalism
What role does the Constraint Principle play in debates about originalism and living
constitutionalism? Before we answer that question, we need to recognize that in this area
of constitutional theory, the terminology itself is contested. Because terms like
“originalism” are contested, debates about originalism can give rise to confusion and
misunderstanding. Why is it the case that some of the most basic terminology in
constitutional theory is the subject ongoing contestation? And how should we proceed
given the unstable nature of the terminology? Answering these questions requires us to
step back from the terminological disputes and examine the reasons why the meaning of
the word “originalism” is contested.
1. Metalinguistic Negotiation: What Should Count as “Originalism,”
“Nonoriginalism,” and “Living Constitutionalism”?
Like the term “originalism,” the word “nonoriginalism” and the phrase “living
constitutionalism” do not have precise and universally accepted definitions. Both the
words and the concepts for which they stand are disputed. Sometimes these disputes
about the categories of constitutional theory are just as sharp and rancorous as the
substantive disagreements with which they are associated. For example, in a recent
essay, Martin Redish and Matthew Arnould characterize the use of the term “originalism”
to describe new originalist constitutional theorizing as “Orwellian.”30 This
characterization is misleading given the transparency of recent originalist theorizing,31
but Redish and Arnould’s use of a powerful epithet is strong evidence that at least some
constitutional theorists are heavily invested in particular characterizations of
“originalism.”
29
Stephen E. Sachs, Originalism as a Theory of Legal Change, 38 HARV. J.L. & PUB. POL'Y 817, 82021 (2015); see also William Baude, Is Originalism Our Law?, 115 COLUM. L. REV. 2349, 2367 (2015)
(paraphrasing Sachs as arguing, “Originalism . . . can be recast as an account of lawful change.”).
30
Martin H. Redish & Matthew B. Arnould, Judicial Review, Constitutional Interpretation, and the
Democratic Dilemma: Proposing a “Controlled Activism” Alternative, 64 FLA. L. REV. 1485, 1509 (2012)
31
See Solum, Originalism and Constitutional Construction, supra note 11, at 527-529.
9
The Constraint Principle
(1) Identity: the legal content of the doctrine is identical to the communicative
content of some provisions of the constitutional text;
(2) Deduction: the legal content of the doctrine can be deduced from legal
content that satisfies condition (1).
(3) Application: the legal content of the doctrine results from application of
legal content that satisfies condition (1) or condition (2).
Constraint as Derivability has the following corollary, which states the same idea in
negative form:
Corollary to Constraint as Derivability: Proposition P is (or should not be) not a
legally correct proposition of constitutional doctrine if its content is not derivable
from the communicative content of the constitutional text by identity, deduction,
or application.
And this corollary implies that there can be no constitutional doctrine that is created by
supplementation of the constitutional text to resolve irreducible ambiguities, to precisify
vague or open textured provisions, to fill in the content of constitutional gaps, or to
resolve constitutional contradictions. All of the content of constitutional doctrine must be
derived from the text or from the application of the text to the facts.
Recall that living constitutionalism is stipulated to be the view that the content of
constitutional doctrine should change in response to changing values and circumstances.
The third provision allows the derivation of new legal content from the application of
valid legal content to new facts. Because new facts can come into being, it follows that a
very limited form of living constitutionalism is consistent with Constraint as Derivability.
2. A Minimalist Version of the Constraint Principle: Consistency of Constitutional
Doctrine
Constraint as Derivability, the Maximalist Version of the Constraint Principle, is the
most restrictive form of originalism that is plausible given the normative commitments of
originalism. Given the variations among originalists, the Maximalist Version cannot
serve as a least common denominator among originalists.52
To identify the least common denominator, we need to identify the version of the
Constraint Principle upon which all or almost all originalists could agree. When we say
“agree,” I do not mean that all originalists could agree that a minimalist version of the
52
Some originalists might object to the notion that Constraint as Consistency can serve as a least
common denominator. In this regard it is important to remember that the claim is that consistency is the
minimal form of constraint that can serve as the dividing line between originalists and nonoriginalists. The
claim is not that Constraint as Consistency is somehow the better formulation because it can serve as the
least common denominator. Constraint as Consistency is a normative principle and it would need to be
defended (against more robust forms of consistency and against less robust forms of contribution) on
normative grounds.
Of course, originalists who affirm a more rigorous form of constraint are free to define “originalism” in
a way that excludes self-identified originalists who do not share their understanding of the Constraint
Principle. And they can also engage in metalinguistic negotiation over the the meaning of “originalism”—
arguing that Constraint and Consistency should not serve as the dividing line. In this Article, I operate on
the basis of stipulations and set these questions to the side.
23
Lawrence B. Solum
Constraint Principle is the best version of constraint or even that it is sufficient for
legitimacy (or for some other normative goal). Rather, I mean that all or almost all
originalists would agree that the satisfaction of the Minimalist Version provides a level of
constraint that is normatively justified (better than less constrained alternatives)—even if
some originalists believe that more constraint is justified as well. In other words, the
Minimalist Version acts as a floor but not a ceiling on the level of constraint from an
ecumenical originalist perspective.
The form of constraint that can best play the role of the Minimalist Version of the
Constraint Principle is “Constraint as Consistency,” which can be define as follows:
Constraint as Consistency. Constraint as Consistency is the conjunction of three
requirements and three qualifications as follows:
Requirement One: Constitutional doctrines and the decisions of
constitutional cases must be consistent with the “translation set.” The
translation set consists of the set of doctrines that themselves directly translate
the communicative content of the text into doctrine and the set of doctrines that
are the logical implications of that set.
Requirement Two: All of the communicative content of the constitutional
text and its logical implications must be reflected in the legal content of
constitutional doctrine.
Requirement Three: All of the content of constitutional doctrine must be
fairly traceable to the direct translation set, with traceable content including
precisifications, implementation rules, and default rules presupposed (or
otherwise supported) by the text.
Qualification One: Requirements One, Two, and Three operate only to the
extent that the communicative content of the constitutional text is epistemically
accessible; they are not violated by departures from unknown communicative
content.
Qualification Two: If Requirements One, Two, and Three are not satisfied,
then constitutional practice should be brought into compliance with constraint
over time, giving due regard to the effects of constitutional change on the rule
of law.
Qualification Three: Requirements One, Two, and Three are defeasible in
limited and extraordinary circumstances, as specified by the best theory of
defeasibility.
For the sake of simplicity in exposition, restatements of Constraint as Consistency will
include the two requirements but not the qualifications—except as the qualifications are
relevant to a particular issue or concern. The third qualification, defeasibility, is explored
in depth at the end of this Part of the Article.53
Constraint as Consistency allows for the creation of constitutional doctrines that
cannot be derived from the text by logical implication or application—so long as these
additional doctrines are consistent with and fairly traceable to the communicative content
of the constitutional text. Constraint as Consistency thus provides a mechanism for
dealing with the Fact of Constitutional Underdeterminacy and hence for creating rules of
constitutional law in the case of irreducible ambiguity, vagueness or open texture, gaps,
53
See infra Part II.D, p. 26.
24
The Constraint Principle
or contradictions. Any violation of Constraint as Consistency will also be a violation of
Constraint as Derivation, but the reverse relationship does not hold. Relatedly, any
violation of the weaker Contribution Principle will also violate Constraint as
Consistency—and again, the reverse does not obtain.
3. A Mixed Version of the Constraint Principle: Constraint: Constraint by Core
Structural Provisions and Deference Otherwise
There are many possible versions of the Constraint Principle. For ease of explication,
most of the discussion that follows will be limited to Constraint as Derivability (the
Maximalist Version) and Constraint as Consistency (the Minimalist Version), but a
complete account of the Constraint Principle would need to consider all of the possible
variations.
Because a full account of the possible forms of constraint is impracticable, we can use
one example for the purposes of illustration. Consider a version of constraint that
exemplifies what I call a “mixed” theory. I will use the term “mixed” to refer to theories
that adopt at least the Minimalist Version of the Constraint Principle for some subdomain
of constitutional questions. Consider a mixed theory that adopts the Minimalist Version
for what are sometimes called the “hardwired” provisions of the Constitution (the clauses
that deal with the structure of the national government), but which adopts a form of
nonoriginalism for other constitutional questions. Let us formulate this mixed theory (the
“Hardwired Only Constraint Principle”) as follows:
The Hardwired Only Constraint Principle. The content of constitutional doctrine
should reflect the following requirements and permissions:
(1) in cases in which a constitutional provision governing the structure of the
national government provides a clear rule (understood as a rule that is not
significantly vague or open textured), constitutional practice must be consistent
with the clear rule,
(2) in cases in which a constitutional provision governing the structure of the
national government is vague or open textured, constitutional practice may be
inconsistent with the original meaning of the constitutional text, and
(3) in cases which a constitutional provision governs individual rights, the
content of constitutional doctrine must incorporate the clear and undisputed core
(if there is one) of the enumerated individual rights provisions of the constitution,
but it may go beyond the core.54
We can call this theory “mixed” because it does contain an originalist element. The first
element respects the Constraint Principle with respect to clear rules of constitutional
structure. Many nonoriginalists seem to admit that originalism is a plausible view in
cases governed by clear structural provisions, such as the requirement that the President
be at least 35 years of age, the division of Congress into two houses, the provision
54
This formulation does not deal with the question of unenumerated individual rights. Because the
Hardwired Only Constraint Principle is introduced solely for the purpose of illustrating the idea of a mixed
version of the constraint principle, it is not fully formulated. A complete version of the principle would
need to deal with the unenumerated rights question and many other questions as well.
25
Lawrence B. Solum
constitution itself is illegitimate because the ratification process excluded a variety of
groups is consistent with theories that deny that contribution principle, but difficult to
square with theories that give great weight to original meaning but allow overrides in
limited circumstances.
In the discussion that follows, the focus will be on comparison of Public Meaning
Originalism with Constraint as Consistency to each of the nine most important forms of
nonoriginalism. The comparison begins in next Part, which examines two clusters of
arguments that favor originalism over many of its rivals. The first set of arguments
clusters around the idea of the rule of law; the second set clusters around the notion of
legitimacy. Generic objections to originalism are considered in Part VI. The generic
arguments for and against originalism are then reframed in the context of particular
pairwise comparisons in Part VII.103
***
Sometimes critics of originalism complain about the lack of normative justifications.
And, truth be told, there was a time when some originalists seemed to argue in a way that
failed to recognize the obvious: originalism is a theory that seeks to guide constitutional
practice. Like all theories about how individuals and institutions should act, originalism
stands in need of some kind of normative foundation.
But we should not confuse the failure of early originalists to address the topic with the
quite different notion that it is difficult to provide the normative foundations for
originalism. The idea that there are good reasons to act consistently with the text of the
United States Constitution is shared by most constitutional actors. It is not that pro tanto
justifications for the constraint principle are hard to find. There are so many of them and
they are so obvious that they may be seen by many nonoriginalists as trivially true but
obviously not decisive.
This suggests that the real action in the normative debate about originalism is
elsewhere. Nonoriginalists sometimes hide behind a smokescreen of theoretical
objections to the general idea of constraint by any legal text, but as we shall see, many of
these, if taken seriously, would lead to absurd consequences and radical alterations in
legal practice generally. The real action arises from the fact that constraint by the text
will lead to some results that nonoriginalists (and many originalists as well) find
normatively unattractive. The question then becomes whether this results-oriented
approach to constitutional theory provides appropriate reasons to reject originalism.
But before we deal with the objections, we need to understand the case for the
constraint principle.
***
IV. JUSTIFYING THE CONSTRAINT PRINCIPLE
This Part identifies and defends two clusters of pro tanto justifications for adherence
to the Constraint Principle. The first cluster focuses on the idea of the rule of law:
constraint supports the rule of law, whereas lack of constraint undermines the rule of law.
103
See infra Part VII, p. 104.
54
The Constraint Principle
The second cluster focuses on legitimacy; constraint enhances constitutional legitimacy.
Each of the two clusters consists of a set of overlapping but partially independent
reasons. The rule of law and legitimacy provide pro tanto justifications for constraint,
but they are not, by themselves, decisive reasons. To determine whether the balance of
reasons supports the Constraint Principle, we will examine the objections to originalism
in Part VI104 and engage in pairwise comparison of originalism with its rivals in Part
VII.105 We can begin the examination of the case for constraint with the idea of the rule
of law.
A. Constraint and the Rule of Law
The first cluster of arguments for constraint focuses on the idea of the rule of law. The
ideal of the rule of law is part of political morality. Sometimes the ideal is articulated as
a contrast, made famous by John Adams, and summarized in the notion of “the rule of
law and not of men.”106 Another formulation emphasizes the rule of law values, such as
predictability, certainty, stability, consistency, generality, and publicity of the laws.107
And yet another approach articulates the rule of law as a set of principles.108 Each of
these ideas about the rule of law plays some role in the cluster of rule-of-law
justifications for originalism; rather than beginning with a general investigation, let us
proceed to the particular arguments.
Sometimes the rule of law is discussed as if adherence to the rule of law is all or
nothing, but this is not the case. The rule of law is best understood as a scalar and not a
binary.109 A system of governance can be more or less compliant with the rule of law. At
one extreme is a system in which a tyrant governs by whim; at the other extreme, would
be a system in which an eternal, rigid, and highly particularized code of law provides a
preordained decision in every case. In the more fortunate parts of the actual world,
governance will lie in between these extremes. The arguments in the rule of law cluster
aim to show that observance of the Constraint Principle is a reasonable way to realize the
ideal of the rule of law to a greater degree than nonoriginalist practice that violates
constraint.
The first argument in the cluster plays off the idea that the constrained rule of law of
can be distinguished from the unconstrained rule of individual persons. I shall call this
the “Judicial Tyranny Argument” for reasons that will become apparent.
104
See infra Part VI, p. 82.
See infra Part VII, p. 104.
106
See John Adams, Novanglus Papers, BOSTON GAZETTE, no. 7 (1774); see also Steven G. Calabresi,
A Critical Introduction to the Originalism Debate, 31 Harv. J.L. & Pub. Pol'y 875, 882 (2008) (“At the end
of the day, Justice Brennan's primary concern was that the text of the Constitution be construed to produce
what he deemed to be good consequences. Doing this makes “the rule of law and not of men” impossible,
which leads to very bad long-term consequences.”).
107
Lawrence B. Solum, The Supreme Court in Bondage: Constitutional Stare Decisis, Legal
Formalism, and the Future of Unenumerated Rights, 9 U. PA. J. CONST. L. 155, 165 (2006).
108
Lawrence B. Solum, Equity and the Rule of Law, in NOMOS XXXVI: THE RULE OF LAW 120 (1994).
109
See Lawrence B. Solum, Legal Theory Lexicon: Scalars and Binaries,
http://lsolum.typepad.com/legal_theory_lexicon/2014/09/legal-theory-lexicon-072-scalars-andbinaries.html (Last revised on November 15, 2015).
105
55
Lawrence B. Solum
1. The Judicial Tyranny Argument
The first argument in the rule-of-law cluster is that the Constraint Principle provides
the best feasible alternative to judicial tyranny. This argument is based on a philosophical
conception of tyranny that distinguishes tyranny from both injustice and violations of
democratic legitimacy. The philosophical conception of tyranny is focused on the
distinction between rule by decree and the rule of law. Violations of the Constraint
Principle by the Justices of the Supreme Court constitute rule by decree and hence are a
form of judicial tyranny.
a) Tyranny as Rule by Decree
In the Nicomachean Ethics, Aristotle develops an account of a “law” (nomos in the
ancient Greek) and its relationship to a “decree” (psēphismata). Richard Kraut, the
distinguished Aristotle scholar, explained the difference as follows. Kraut’s exposition
begins with the idea of person who is “lawful” (nominos):
[W]hen [Aristotle] says that a just person, speaking in the broadest sense is
nominos, he is attributing to such a person a certain relationship to the laws, norms,
and customs generally accepted by some existing community. Justice has to do not
merely with the written enactments of a community’s lawmakers, but with the
wider set of norms that govern the members of that community. Similarly, the
unjust person’s character is expressed not only in his violations of the written code
of laws, but more broadly in his transgression of the rules accepted by the society
in which he lives.
There is another important way in which Aristotle’s use of the term nomos
differs from our word ‘law’: he makes a distinction between nomoi and what the
Greeks of his time called psēphismata—conventionally translated as ‘decrees’. A
decree is a legal enactment addressed solely to present circumstances, and sets no
precedent that applies to similar cases in the future. By contrast a nomos is meant
to have general scope: it applies not only to cases at hand but to a general category
of cases that can be expected to occur in the future.110
Aristotle’s conception of tyranny is rule by decree (psēphismata); the ideal of the rule of
law is governance by laws (nomoi).
b) Two Distinctions About Judgments: First versus Second Order and Public
versus Private
We can restate this last point by introducing a set of distinctions between types of
normative judgments. Begin with the distinction between what we can call “first order
judgments” and “second order judgments”:
•
110
First Order Judgments: A first order normative judgment is a judgment that
attributes a normative characteristic to some state of affairs, person, or action.
Thus, the proposition expressed by “lying is wrong” is a first order moral
RICHARD KRAUT, ARISTOTLE 105-06 (2002).
56
The Constraint Principle
•
judgment. The proposition expressed by “flag burning is protected by the First
Amendment” is a first order legal judgment.
Second Order Judgments: A second order normative judgment is a judgment
about how first order questions should be settled. Typically, second order
judgments have two parts: (1) a judgment about which institution, person, or
other entity should have authority with respect to normative first order
judgments in a given domain, and (2) a judgment about how the favored actor
should make its decision. The proposition expressed by “the constitutionality
of state statutes that forbid same sex marriage should be determined by the
Supreme Court and not by state elected officials” is a second order legal
judgment. Likewise, proposition, “the Supreme Court should adopt only those
constitutional doctrines that are constitutional text,” expresses as second order
judgment.
To this distinction we can add the distinction between public and private normative
judgments:
•
•
Public Judgments: A normative judgment is public if the judgment is based on
a norm (such as a standard or rule) that is publicly accessible. For example, a
normative judgment that an action is unlawful on the basis of a statute that is
public (accessible to citizens) is a public legal judgment.
Private Judgments: A normative judgment is private if the judgment is based
on a belief or opinion that is not publicly accessible. For example, if a judge
bases a decision on her personal belief that one outcome is preferable, the
judge has made a private normative judgment.
If judges rely on their own private, first-order judgments of fairness as the basis for the
resolution of disputes, then it follows inexorably that their judgments will be decrees
(psēphismata) and not decisions on the basis of a second order, public judgment—in
other words, not on the basis of a nomos. In other words, a judge who decides on the
basis of her own private judgments about which outcome is fair in particular cases is
making decisions that are tyrannical in Aristotle’s sense.
c) The Argument from Tyranny Distinguished from Arguments from Injustice and
Democracy
It is not clear that the word “tyranny” has a clearly defined meaning in ordinary
discourse.111 In public discourse, the charge of judicial tyranny could be shorthand for
arguments that judges are acting contrary to justice or that they are violating a principle
of democratic legitimacy.112 There is nothing wrong with using the word “tyranny” in
these ways, but we are using the word “tyranny” in a precise and technical sense:
111
The OED gives several definitions, including “oppressive or unjustly severe government” with
similar definitions for tyrant. (http://0www.oed.com.gull.georgetown.edu/search?searchType=dictionary&q=tyranny&_searchBtn=Search).
112
See, e.g., James E. Wright, III, Bush v. Orleans Parish School Board: The Second Battle of New
Orleans, Chronicles of the Case and the Judge, 61 Loy. L. Rev. 135, 161 (2015) (describing “judicial
tyranny” argument that focused on decision by unelected judges).
57
Lawrence B. Solum
“tyranny” is rule by decree and not something else. For this reason, tyrants may be just
and democracies can be tyrannical.
(1) Tyranny May Be Consistent with Substantial Justice (Except Insofar as Justice
Includes the Rule of Law)
“How can this be?” you may ask. “Aren’t decisions that are motivated by fairness the
very opposite of tyranny?” This objection misunderstands the nature of tyranny in the
philosophical sense used here. Whether a decision is tyrannical or lawful is a different
question than the question whether the decision is just or unjust. There can be unjust
laws and just tyrants. In a perfect world, we could have the rule of law and the content of
the law would be perfectly just. And it is frequently the case that tyrannical rule will also
be unjust. But in the actual world, it is unlikely that the rule of law will be perfectly just
and it almost impossible to achieve a state of affairs where the rule of law is perfectly just
(according to the best or true theory of justice) and in which there is universal agreement
among citizens that this is the case.
The impossibility of achieving near universal agreement or even wide consensus that
any set of laws is perfectly just is a function of the pluralism that has characterized the
United States from the beginning—a topic we have already discussed in connection with
the distinction between deep and shallow reasons.113
Our private, first-order judgments about the all-things-considered requirements of
fairness do not agree. So, in any given case, a decision that the judge believes is required
by fairness will be seen by others quite differently. At best, the decision will be viewed
as a good faith error of private judgment about fairness. More likely, those who disagree
will describe the decision as a product of ideology, personal preference, or bias. At
worst, the decision will be perceived as the product of arbitrary will or self-interest. In no
event, will a decision based on a controversial first order private judgment of fairness be
viewed as outcome of a nomos—a publicly available legal norm.
(2) Rule by Decree May Be Consistent with Majoritarian Democratic Legitimacy
(Except Insofar as a Thick Conception of Democracy Incorporates the Rule of
Law)
The Argument from Judicial Tyranny is not based on democratic legitimacy for two
reasons, one relating to the internal structure of the argument and the other concerning the
possibility of democratic tyranny.
First, the internal structure of the argument does rely on any premise about democratic
legitimacy. The key premises are (1) rule by decree is tyrannical, (2) violations of the
Constraint Principle by judicial decision constitute rule by decree, and (3) tyranny
undermines the rule of law and is for that reason normatively undesirable. None of these
premises refers to democratic legitimacy.
Second, decisions that satisfy the requirement of democratic legitimacy can be
tyrannical. This is obviously true in the case of executive tyranny. Suppose, for example,
that a democratically elected President were to rule by decree, disregarding both judicial
decisions and statutes, and instead governing through ad hoc executive orders. The
113
See supra Part III.A.2, p. 29.
58
The Constraint Principle
decisions of the President could have democratic legitimacy; for example, suppose that
the President was elected by a clear majority on a platform of rule by decree. Likewise,
legislative tyranny is also possible in theory. If Congress were to enact decrees (ex post
facto laws governing particular situations) rather than laws, then it would engage in
legislative tyranny. Again, assuming that a majority of the members of Congress were
elected on a platform of rule by decree, their actions would have been simultaneously
democratic and tyrannical.
It might be argued that the conception of democratic legitimacy is thick enough to rule
out the argument made in the immediately prior paragraph. This would certainly be true
if our concept of democratic legitimacy were formulated so as to exclude rule by decree.
But this point does not deny the point that has just been made—which could be
reformulated so as to make it clear that the argument only applies to thin conceptions of
democratic legitimacy.
d) Constraint Results in the Rule of Law and not Rule by Decree
Decisions that accord with the Constraint Principle are lawful because they are made
on the basis of a legal norm (e.g., a rule or standard) that governs a general category of
cases for the future (until the constitutional text is amended). To the extent that the
Constraint Principle governs constitutional decisionmaking, constitutional practice cannot
be characterized as a form of tyranny.
It is hard to see how there could be a serious objection to the argument that the
Constraint Principle obviates judicial tyranny in constitutional practice. Of course, there
are rhetorical moves that could be made. One could argue that the Constitution itself is
tyrannical because it lacks democratic legitimacy, but this argument is using the word
“tyranny” as a synonym for “nondemocratic” and hence is misplaced as an objection with
respect to the narrow claim made in the prior paragraph. Likewise, one might argue that
purportedly originalist judges actually decide on the basis of ideology and hence that
their decisions are tyrannical (in the relevant sense). If this were true, it would be a
problem for originalism114, but it is misplaced as an objection to the Judicial Tyranny
argument as a justification for the Constraint Principle.
e) Violations of Constraint by the Supreme Court Constitute Rule by Decree and
Hence Judicial Tyranny
But from the fact that the Constraint Principle requires lawful decisionmaking, it does
not follow that the alternatives are tyrannical. That requires a separate argument. We can
begin by noting that there are possible arrangements that avoid tyranny through means
other than the Constraint Principle. Consider, for example, a parliamentary system—
where the parliament itself governs only through laws (and not decrees) and where the
courts decide cases on the basis of laws and do not override the laws by decree. This
system avoids judicial tyranny, but lacks a written constitution and hence the Constraint
Principle.
114
See infra Part VI.H.3.
59
Lawrence B. Solum
This paper is not about the question whether a regime in which there is a written
constitution combined with the Constraint Principle is superior to a lawful parliamentary
system properly authorized by a constitution.115 Instead, the topic at hand is whether the
Constraint Principle should be affirmed given the written constitution that we actually
have and the institution of judicial review (which might be better described as the
combination of constitutional supremacy and a judicial duty of lawfulness). In order to
answer that question, we need to examine each of the major alternatives to originalism
and see how they fare with respect to judicial tyranny.
The comparison of originalism (with the Constraint Principle) and the alternatives
(without the Constraint Principle) will proceed in two stages. In this Subsection, we will
take a preliminary look at some of the alternatives and develop a generic argument that
combining constitutionalism with judicial supremacy in the absence of constraint will
produce judicial tyranny. In a subsequent Part, we will take a look at each of the major
alternatives in greater depth.116
Nonoriginalists who embrace judicial supremacy may reply to the objection in the
following way:
The rule of law can be achieved without the Constraint Principle, because there
are alternative methods of achieving lawfulness and avoiding rule by decree.
Indeed, constitutional practice currently does not comply with the Constraint
Principle but it does realize the rule of law. And to the extent that the status quo
does not fully comply with the rule of law, it could be brought into compliance by
means other than adoption of the Constraint Principle.
But is this reply correct? Originalists disagree with the contention that current
constitutional practice instantiates the rule of law and they will argue that the alternatives
to originalism cannot avoid the problem of tyranny.
Of course, our legal system as a whole does realize the rule of law to a substantial
degree. At any given point in time, the system of constitutional doctrine is likely to be
relatively stable. The decrees of the Supreme Court are elaborated by the lower federal
Courts of Appeal and the highest courts of the several states and these elaborations enable
lawyers and trial court judges to operate on the basis of a relatively stable set of legal
rules—so long as lower courts adhere to the doctrine of vertical stare decisis and public
officials and citizens engage in a practice of acquiescence the decisions reached by the
Supreme Court in particular cases. The Argument from Judicial Tyranny can concede
this point: judicial tyranny by the Supreme Court can coexist with substantial realization
of the rule of law at other levels of the system.
Nonoriginalists might concede this point but press for the normative significance of
Supreme Court tyranny. The argument might be articulated as follows:
Even if we concede that the Supreme Court is tyrannical, this level of tyranny is
tolerable. The label “tyranny” has negative connotations, but in practice the
Supreme Court is a benevolent tyrant. So long as the decisions of the Court are
tolerably just and the values of the rule of law (predictability, certainty, uniformity,
115
Parliamentary democracy could be authorized by a written constitution or by an unwritten
constitution (as in the case of the United Kingdom). The important point is that the current United States
Constitution clearly does not authorize government by parliamentary democracy.
116
See infra Part VII, p. 104.
60
The Constraint Principle
even-handedness, and publicity) are realized to a substantial degree in the system as
a whole, judicial tyranny is better than originalism, which would lead to other
problems.
This line of reply shifts the main focus on debate to the disadvantages of originalism.
Fair enough: it is possible that judicial tyranny can be justified. But this line of reply
accepts that the Argument from Judicial Tyranny provides a pro tanto reason for the
Constraint Principle. The affirmative objections to the Constraint Principle are
considered separate below.117 Perhaps, some defenders of nonoriginalism would make
the more radical argument that there is nothing wrong, even pro tanto, with tyranny, but
this position seems implausible.
A more promising line of reply would aim to show that the alternatives to constraint
do not involve tyranny. Consider, for example, the Multiple Modalities theory. One
might argue that the complex practice of constitutional argument itself constrains judicial
decisionmaking. But is this really the case? Of course, there will be cases in which all of
the modalities point in the same direction, and in these cases, judges will be constrained.
But by hypothesis, these are cases in which the modalities reach the same result as that
required by originalism (and the Constraint Principle). The test cases for the Multiple
Modalities view are those in which the modalities conflict. Recall that in such cases, the
Multiple Modalities account rejects the idea that here is a second-order public standard
for the resolution of conflicts among the modalities: there is no hierarchy or lexical
ordering of the various modes of constitutional argument. This suggests that what is
doing the work in such cases is the private first-order judgments of the Justices—but that
is rule by decree and hence tyranny. Another line of reply might appeal to the standards
of argumentation that are shared by competent legal practitioners—these standards might
be argued to discipline the court and hence to avoid judicial tyranny. But this argument
seems quite implausible, because the standards of the legal community are very
latitudinarian. Indeed, the community of judges, lawyers, and scholars does not even
agree that compliance with the Multiple Modalities theory is required.
Each of the major alternatives to originalism must be evaluated in this way, but we
will reserve that task for the penultimate Part of this Article.118 At this stage, however,
we can preview the more detailed arguments to come. Almost all of the major rivals of
originalism relies, in the end, on private first-order judgments by the Justices. And if this
is correct, then only a few of the alternatives to originalism avoids the judicial tyranny
objection.
The most important rival to originalism that avoids judicial tyranny is Thayerianism—
in both its unconstrained and representation-reinforcement forms. If one accepts a very
strong principle of Thayerian deference, combined with an ordering rule for resolving
conflicts between branches of the national government and between the national
government and the states, then judicial tyranny can be avoided. Of course, if the
principle of deference privileged the President, the result would be executive tyranny—
which is arguably worse than the judicial variety. But this problem could be avoided by
limiting deference to duly enacted statues. A system of legislative supremacy limited by
117
118
See infra Part VI, p. 82.
See Part VIII, p. 104.
61
Lawrence B. Solum
the requirement to legislate by law and not decree avoids the problem of tyranny, but it
does this by rejecting the idea of a judicially enforceable written constitution.
f) The Normative Significance of Tyranny
What is the normative significance of judicial tyranny? Why should we care that the
fundamental structures of government and individual rights are determined on the basis
of ad hoc decrees issued by a committee of nine unelected officials who are appointed for
life turns? To some, the answers to these questions may seem obvious.
***
Consider the following thought experiment:
You are a member of the constitutional convention of a newly formed nation. You
are asked to vote on two proposals regarding the fundamental institutional
arrangements, including the structure of government and individual rights. The
proposals are:
Rule by Decree of an Unelected Committee: The structure of government
and the rights of persons will be decided on a case-by-case basis by a
committee of nine persons, selected by the executive and confirmed by the
legislature. Members of the committee shall serve for life. They shall
have the ultimate power to decide all fundamental questions, including
what questions are fundamental.
Rule by a Written Constitution: The structure of government and the rights
of persons shall be decided in advance in a written constitution. The
constitution shall be interpreted and applied by a court that shall be
constrained by the communicative content of the text. The judges shall
serve life terms but shall be removed from office by the legislature if their
interpretations are clearly inconsistent with the communicative content of
the text. The constitution may be amended by supermajority vote of the
national legislature and the regional legislatures.
The constitutional convention is at the final stage and you must choose between
these two proposals—the other alternatives have been voted down and may not be
reconsidered at this time. One of the considerations in your deliberations is
which proposal best realizes the rule of law. How would you rank the proposals
on that dimension? Would the rule of law ranking be a pro tanto reason to favor
one proposal over the other?
Before you fight the hypothetical, pause for a moment to answer the questions. It is
obvious that rule by a written constitution does a better job of instantiating the rule of
law. And it is obvious that this would be at a minimum a pro tanto reason to favor rule by
a written constitution over rule by decree of an unelected committee.
“Okay,” you say, “But I do want to fight the thought experiment. I don’t favor rule by
decree. There are many alternatives to constraint by a written constitution. What about
Thayerian deference? What about moral readings? What about constraint by common
law methods?” Fair enough. The thought experiment is designed for a limited purpose,
to highlight the relationship of the rule of law to the choice between rule by decree and
62
The Constraint Principle
rule by a written constitution. Each of the alternatives to constraint must be considered
on its own merits. Those comparisons are coming.
***
Tyranny is rule by decree, but what is wrong with that? One answer to that question
lies in the idea of the rule of law and the values with which it is associated, including
predictability, certainty, stability, and consistency. Rule by decree undermines
predictability and certainty for obvious reasons: ad hoc decision of particular cases makes
the law less predictable and certain as compared to the alternative: in the case at hand, the
alternative is the Constraint Principle, which creates predictability and certainty by
conforming constitutional practice to the communicative content of the constitutional
text. Likewise, rule by decree undermines the stability of law, because it enables
constitutional change based on the preferences of judges and especially by the
preferences of a majority of the Supreme Court. Finally, rule by decree undermines
consistency (treating like cases alike), because it enables differential treatment based on
the unconstrained decisions made by judges and officials. The relationship between
constraint and the rule of law values will be considered in greater depth below.119
But the judicial tyranny argument directs are attention to another defect of rule by
decree. Rule by decree is rule by a tyrant—someone with unconstrained power who rules
by deciding particular cases or issuing particularized directives. As bad as a
constitutional monarchy might be, the rule of laws issued by a monarch would be
preferable to rule by decree—so long as the monarch was bound by the constitution and
governed through the promulgation of general laws that applied to all persons, including
the monarch.
Rule by decree entails the fundamental rights and powers of individuals are subject to
the whim of the tyrant. Living under these conditions necessary undermines freedom, as
that concept is understood in its republican sense. Individuals are dependent on the
tyrant.120
The republican conception of political liberty . . . defines freedom as a sort of
structural independence—as the condition of not being subject to the arbitrary or
uncontrolled power of a master. Pettit, who has done more than anyone else to
develop this republican conception of freedom philosophically, puts it thus: a
person or group enjoys freedom to the extent that no other person or group has “the
capacity to interfere in their affairs on an arbitrary basis”.121
The connection between the rule of law and the republican conception of freedom from
domination and dependence illuminates a basic reason for valuing the rule of law. The
rule of law and not of individual persons does more than serve the rule-of-law values
(publicity, stability, certainty, and the rest). The rule of law is an essential safeguard of
119
See infra Part IV.A.2, p. 67 (presenting the rule-of-law values argument).
See PHILIP PETTIT, REPUBLICANISM: A THEORY OF FREEDOM AND GOVERNMENT (1999).
121
See Frank Lovett, Republicanism, Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/entries/republicanism/#LibNonDom (April 15, 2014) (quoting Pettit, supra note
120, at 165).
120
63
Lawrence B. Solum
one of the most important aspects of human freedom—the independence of the individual
from domination by others.
The psychological impact of the loss of freedom may depend in part on the substance
of the decrees. If you applaud the outcome in a particular case (whether it be Obergefell
v. Hodges122 or Citizens United v. FEC123) you might be less bothered by the fact that it is
the product of rule by decree than those who are opposed to the outcome. But you might
still be bothered. After all, the case that you like does not bind the Supreme Court, and a
change in the court’s personnel would result in a reversal of the outcome. If the next
election goes against you, a swing justice retires, and the votes align the other way, your
most cherished substantive right could vanish—at the whim of the Supreme Court. If the
right in question is one you view as essential to your life plan or your standing as an
equal citizen, then you are reduced to dependence on the Court and its decrees.
One of the most insidious aspects of an unconstrained Supreme Court is that it
disguises rule by decree as the rule of law. It takes for itself the power to make the
fundamental decisions regarding the constitution of society, including the scope of
individual rights and liberties, but then wraps its rule by decree in the false flag of
constitutional constraint. The decisions of the Court are presented as derived from the
Constitution and hence as settled, pending constitutional amendment. But absent
constraint, the presentation is illusion and not reality. This state of affairs threatens
republican freedom in an especially pernicious way—and brings to mind the dystopias of
speculative fiction.
g) Objections to the Argument from Judicial Tyranny
How might a nonoriginalist answer the argument from judicial tyranny? Here are two
possible objections.
(1) Objection One: The Supreme Court’s Decisions Are Laws Not Decrees
The first objection goes to the question whether violations of the Constraint Principle
constitute “decrees” on the one hand or whether they are more properly described as
“laws” on the other hand. It is important to remember that this question is framed in a
technical vocabulary: “decrees” are ad hoc decisions of particular matters that do not
create general rules, whereas “laws” do create rules (or rule-like legal norms).
One answer to the judicial tyranny objection might focus on the doctrine of stare
decisis. It might be argued that the Supreme Court does, in fact, legislate (make “laws”),
because its holdings create binding rules. This argument has a superficial appeal. Some
Supreme Court decisions do alter the legal landscape in ways that create law-like
consequences. If a statute is struck down by the Supreme Court and the political
branches acquiesce in this decision, then the consequence of the Supreme Court’s action
is very much like a repealing statute. To the extent that the rule of law prevailed before
the statute was enacted, repeal by judicial decree would not undermine the rule of law.
So, there are cases where rule by decree will not undermine the rule of law values of
predictability, stability, certainty, and consistency. But can we generalize from these
122
123
576 U.S. ___ (2015).
558 U.S. 310 (2010).
64
The Constraint Principle
cases to the conclusion that Supreme Court never or rarely governs by decree. Of course,
there are examples where the Supreme Court’s decisions clearly do undermine the rule of
law. The current disorder with respect to state laws restricting abortion is a fairly clear
example. The pattern of Supreme Court decisions has created substantial uncertainty
regarding the score of the power of states to regulate abortion and that uncertainty invites
restrictive legislation and subsequent litigation, with inconsistent outcomes as different
judges with different ideological priors reach inconsistent outcomes.
One of the underlying causes of this disorder is a function of the chaotic state of the
doctrine of stare decisis. There are at least three different operative conceptions of the
doctrine of stare decisis in contemporary American jurisprudence:
Fact Bound Holdings. One view is that the holding of a case is limited to the legally
salient facts. This view entails that judicial violations of the Constraint Principle
constitute decrees—as it denies that holdings can generate rules.124
Ratio Decidendi. Another view is that the holding of case is the narrowest rule
implied the reasoning necessary to the decision of the case given the legally salient
facts and the arguments made by the parties.125 This view results in holdings that are
more like laws, but allows for considerable variations in future cases based on
factual differences and new arguments raised by future parties.
Legislative Holdings. Yet another view is that courts have the power to create
legislative holdings by clearly stating the rule that they intend to create. This view
results in holdings that are laws, not decrees—assuming that such holdings are truly
binding.126
It is not clear that American courts, including the Supreme Court, follow any consistent
practice with respect to the doctrine of stare decisis. Courts seem to use all three
theories—perhaps choosing the theory of stare decisis to justify outcomes reached on
other grounds. If the practice of stare decisis is disordered in this way, the result is that a
superficial appearance of rule-like holdings in particular cases is undermined by a metalevel practice that results in rule by decree.
There is another problem with the idea that the doctrine of stare decisis results
obviates judicial tyranny. The Supreme Court does not consider itself bound by the
doctrine of stare decisis: it is free to overrule its prior decisions and to decide new cases
in ways that fail to adhere to its prior holdings. The Court does give some weight to its
prior decisions, but that weight is light and variable. The most influential modern
statement of the Court’s practices is found in Planned Parenthood of Southeastern
Pennsylvania v. Casey, quoted in full in the accompanying footnote.127 Of course, this is
a very large topic, but I think it is fair to say that it would be difficult to argue that Casey
124
See Arthur L. Goodhart, Determining the Ratio Decidendi of a Case, 40 YALE L. J. 161 (1930).
See Lawrence B. Solum, The Supreme Court in Bondage, supra note 107, at 155-208.
126
Id.
127
505 U.S. 833 (1992). The court stated, “Even when the decision to overrule a prior case is not, as in
the rare, latter instance, virtually foreordained, it is common wisdom that the rule of stare decisis is not an
“inexorable command,” and certainly it is not such in every constitutional case,” and “when this Court
reexamines a prior holding, its judgment is customarily informed by a series of prudential and pragmatic
considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of
law, and to gauge the respective costs of reaffirming and overruling a prior case.” Id. at 854–55.
125
65
Lawrence B. Solum
creates a rule-like approach to horizontal stare decisis. And because the Supreme Court
does not bind itself, lower courts and officials are permitted to guess whether the current
Justices will (or will not) follow past precedents. Open flouting of prior precedents might
result in summary reversal, but given the fluidity in vertical stare decisis, it seems likely
that lower court judges (and lawyers for extrajudicial constitutional actors) will be able to
create plausible arguments that the actual holding of the Supreme Court is
distinguishable.
In sum, it is far from clear that the doctrine of stare decisis is sufficient to transform
the Supreme Court’s decisions from decrees to laws. Indeed, the disordered state of the
current doctrine of stare decisis suggests that it may have the opposite effect—creating
even more uncertainty, instability, unpredictability, and inconsistency.
(2) Objection Two: The Rule by Decree Objection Applies to All Common-Law
Decisionmaking
The second objection is that the judicial tyranny argument proves too much—because
it suggests that all common-law decisionmaking involves judicial tyranny. Articulating
this objection will be a large task—because it will require an account of the common-law
process. Nonetheless, the gist of the objection is clear enough. Common-law
decisionmaking is like the Supreme Court’s constitutional jurisprudence. If the latter is
rule by decree, then so is the former. So, if the rule by decree objection justifies the
Constraint Principle, it will also invalidate all of the common law.
Of course, there is something to this objection. Critics of the common law have long
complained that it undermines the rule of law. If they are right, then the proper solution
would seem to be codification and not the extension of the evil to the constitutional
realm. Two wrongs do not make a right—especially when they are just two different
wrongs!
There are, however, ways in which common law decisionmaking could avoid the rule
by decree objection. For example, the declaratory theory of the common law conceives
of the common-law process as declaratory and not legislative. Common-law judges
recognize existing norms and customs. So long as the common law is limited to judicial
decision on the basis of widely shared and deeply held social norms and customs, the
common law will not constitute rule by decree. It is only if judges view themselves as
having the power to make common law (through legislative decrees) that the problem
arises.
Common-law judging could avoid the judicial tyranny objection in a second way, by
adhering to a strong version of the doctrine of stare decisis or precedent. So long as
common law courts view themselves as bound by prior decisions, common law making is
constrained. But if common law courts view themselves as permitted to overrule prior
decisions on the basis of private first-order judgments about what the law ought to be,
then common-law judging will be inconsistent with the rule of law and will constitute a
form of judicial tyranny.
Finally, notice that the second objection to the Judicial Tyranny Argument does not
actually defend theories that authorize violations of the constraint principle from the
charge they are tyrannical; rather, it argues that the status quo includes other practices that
many be tyrannical. If the common law can be made consistent with the rule of law and
66
The Constraint Principle
would thereby be improved, the argument that the common law now has tyrannical
elements does not seem very persuasive as an objection to constraint.
***
Experience suggests that the judicial tyranny argument is “difficult to swallow.” I
image a reader asking the following question: “How could it possibly be the case that I
live in a judicial tyranny? Surely I would know it already if that were the case. I
certainly didn’t learn that the Court was a tyrant in law school; my colleagues don’t say
that violations of constraint are tyrannical.” This hypothetical reader seems to lack
access to the intuitions that motivate the objection.
But perhaps you already have intuitive access to the notion that rule by decree is
tyrannical. Call to mind a Supreme Court decision to which you object on normative
grounds—a decision you find both wrong and unjustified. Perhaps for you that is
Citizens United or maybe it is Obergefell v. Hodges. Does this decision conjure the idea
of judicial tyranny? Now imagine that the composition of the Supreme Court changes
and the Court reverses this decision. You are very pleased! The rule of law has been
restored! And now imagine that unexpectedly, the composition of the Court changes
again. A younger justice resigns due to ill health, and unexpectedly a President of your
party is not reelected. The new Justice votes to reinstate the despised decision. Would
the label “tyrannical” be out of place?
Here is another way to access the force of the judicial tyranny argument. Imagine that
the President rather than the Supreme Court assumes the power to adopt amending
constructions of the Constitution. A very popular President, a demagogue, abrogates to
himself the power to overrule the Supreme Court and the constitutional text using the
newly created device of “Executive Constitutional Decrees.” The academy howls,
Senators protest, but political forces are such that the President prevails. With the
precedent established, the next President continues to exercise this power. Would you be
inclined to call the exercise of this power “tyrannical”? If you would, then does it really
make a difference that the power of rule by decree is currently exercised by the Supreme
Court (a committee of nine unelected judges appointed for life terms) and not the
President (hopefully elected by a majority of voters)?
***
2. The Rule-of-Law Values Argument
The second argument in the rule of law cluster is that violations of the Constraint
Principle undermine the rule of law values. There is no definitive list of these values, but
publicity, stability, and consistency are frequently counted among them. The rule of law
values provide a set of overlapping and partially redundant pro tanto reasons to adopt the
Constraint Principle. Consider first the value of publicity.
67
Lawrence B. Solum
a) Constraint Enhances the Publicity of Law
Theories of the rule of law frequently articulate “publicity” as a requirement of the
rule of law.128 Putting it another way, publicity is one of the rule-of-law values. The
thesis of this subsection is that observing the constraint principle enhances the publicity
of constitutional law. This argument is closely related to another argument, which
focuses on transparency and its relationship to the legitimacy of constitutional law. The
publicity argument focuses on our ability to know what the law is in advance and to act
on the basis of that knowledge.
In what sense does the failure to observe the Constraint Principle undermine the ruleof-law value of publicity? There are two aspects of an unconstrained judiciary that
undermine the publicity of law: first, the true basis of unconstrained decisions is not
reliable provided by judicial opinions, and second, the legal norms created by judicial
decisions are frequently ambiguous or obscure.
The first aspect of the publicity argument stems from the reluctance of judges to admit
that their decisions violate the Constraint Principle. This reluctance may lead to
circumlocution and avoidance. Most obviously, when the actual basis for a judicial
decision lies in the first-order private judgments of the judge, there will be a temptation to
attribute the decision instead to the “meaning” of the constitutional provision or to
precedent. Given the importance of stare decisis to the system, the failure to disclose true
reasons makes it difficult to determine the content of legal doctrine. By way of contrast,
judges who observe the Constraint Principle can clearly state the basis of their decisions;
for them, legal content is based on the communicative content of the text.
The second aspect of the publicity argument is closely related to the first. In an
unconstrained system, the precise legal content of the law may be unnecessarily
ambiguous, obscure, or open textured. This lack of clarity can arise in many ways. For
example, judges may conceal the role of their own first order private judgements by use
of a flexible balancing test or an all-things-considered standard expressed in terms of thin
legal concepts like “reasonableness” or “just.” Because there may be intersubjective
disagreement about how such standards apply to particular cases, they obscure the real
basis for the decision, substituting the appearance of publicity for the real thing.
These general considerations are illustrated by the Supreme Court’s decision in
International Shoe Co. v. Washington.129 Famously, Shoe initiated a process of
transformation, replacing the highly formalist scheme of categorical rules from Pennoyer
v. Neff130 with a highly abstract and indeterminate formula: “due process requires only
that in order to subject a defendant to a judgment in personam, if he be not present within
the territory of the forum, he have certain minimum contacts with it such that the
maintenance of the suit does not offend 'traditional notions of fair play and substantial
justice.”131 Subsequent cases provided a baroque elaboration of this formula, including
distinctions between general and specific jurisdiction, an obscure purposeful availment
128
LON L. FULLER, THE MORALITY OF LAW 42, 44 (rev. ed. 1964); Jeremy Waldron, The Concept and
the Rule of Law, 43 GA. L. REV. 1, 24 (2008).
129
326 U.S. 310 (1945).
130
95 U.S. 714 (1878).
131
326 U.S. at 316.
68
The Constraint Principle
test, and a five factor interest-balancing test.132 The alternative originalist approach,
suggested but not adopted by Justice Scalia, would involve a return to relatively clear and
determinant standards that prevailed at the time the Fourteenth Amendment was
adopted.133 This is a complex area of law, and a full treatment certainly requires an
extended treatment, but for present purposes, the Supreme Court’s personal jurisdiction
jurisprudence illustrates the mechanisms by which violations of constraint can create
publicity problems that are solved by adherence to original meaning.
At this point, defenders of originalism might argue that the doctrine of stare decisis
provides a sufficient degree of publicity to the law. Of course, the decisions of the
Supreme Court are public and binding on the lower federal courts via the doctrine of
vertical stare decisis. They are not binding on the Court itself. 134 This results in both
formal overrulings and what is sometimes called “stealth overrulings”135—decisions that
vitiate the gravitational force of a precedent that formally remains “on the books.” The
Supreme Court says that it gives its prior opinions substantial weight and has articulated
the factors that it purports to apply when deciding whether to overrule,136 but there are
reasons to doubt that the factors stand in the way of Justices who wishes to overrule a
decision that they oppose on the basis of their first-order private judgments of political
morality. On the great issues of the day, it surely not unreasonable to believe that the
outcome of the next election and the timing of new appointments to the Supreme Court
mean that the content of the law is “up for grabs.”
This problem is compounded by substantial uncertainty about the true scope of the
doctrine of vertical stare decisis. That doctrine requires a distinction between holdings
and dicta—a murky topic at best. There are at least three competing theories of vertical
stare decisis. The first theory, associated with Arthur Goodhart, is that that the holding of
a case is limited by legally salient facts.137 This theory results in extraordinarily narrow
holdings—as the ability of lawyers to find legally relevant factual differences between
cases is very substantial indeed.138
The second theory, the classic theory of the ratio decidendi, limits the holding of a
case to the rule implied by the actually articulated reasoning necessary to the decision on
132
There is a vast literature on the Supreme Court’s personal jurisdiction jurisprudence. See, e.g.,
Howard B. Stravitz, Sayonara to Fair Play and Substantial Justice?, 63 S.C. L. REV. 745 (2012); Ryne H.
Ballou, Civil Procedure-Be More Specific: Vague Precedents and the Differing Standards by Which to
Apply "Arises Out of or Relates to" in the Test for Specific Personal Jurisdiction, 35 U. ARK. LITTLE ROCK
L. REV. 663 (2013); Bernadette Bollas Genetin, The Supreme Court's New Approach to Personal
Jurisdiction, 68 SMU L. REV. 107 (2015).
133
Lawrence Rosenthal, Does Due Process Have an Original Meaning? On Originalism, Due Process,
Procedural Innovation . . . and Parking Tickets, 60 OKLA. L. REV. 1, 24-25 (2007).
134
The question whether nonoriginalist Supreme Court decisions should have binding vertical stare
decisis effect on state court judges is explored by Lee Strange. See Lee J. Strang, State Court Judges are
Not Bound by Nonoriginalist Supreme Court Interpretations, 11 FIU LAW REVIEW 327 (2016).
135
See Barry Friedman, The Wages of Stealth Overruling (With Particular Attention to Miranda v.
Arizona), 99 Geo. L.J. 1 (2010).
136
Planned Parenthood v. Casey, 505 U.S. 833 (1992).
137
See Arthur L. Goodhart, Determining the Ratio Decidendi of a Case, 40 YALE L.J. 161 (1930)
(elaborating theory of the ratio decidendi).
138
Over time, a long succession of cases might result in the emergence of a broad rule, but the opposite
is frequently the case, with inconsistencies among cases resulting in rules becoming more complex and
uncertain as time goes on.
69
Lawrence B. Solum
the basis of the legally salient facts and the arguments presented by the parties.139 This
theory results in both narrow holdings (because the necessary reason is limited by salient
facts and the parties’ arguments) and substantial uncertainty (because both the content of
the reasoning and the rule that is logically implied are subject to contestation and unlikely
to be the subject of intersubjective agreement in ideologically charged disputes).
The third theory affords great weight to we-hold-that statements, essentially giving
such statements what Ronald Dworkin called “enactment force”140 because such
statements provide the best prediction likely future behavior of the Court.141 But even if
the prediction theory of precedent is correct, it does not follow that propositions
introduced by “we hold that” are sufficient to provide the kind of publicity required by
the rule of law. Notice first that the predictive theory implicitly accepts the idea that the
Supreme Court engages in rule by decree: the job of lawyers and lower court is to predict
the decrees of the tyrant. Moreover, the predictive value of we-hold-that statements
degrades over time as the views of the Justices change and old Justices retire and new
ones are appointed.
No one theory of precedent dominates the other. Indeed, one might suspect that
different judges hold different theories and that the same judge is likely to pick and
choose among the theories depending on how judges wants the case before them to come
out. Perhaps someone can make the case that rule by Supreme Court decree provides
substantial publicity, but given the absence of binding horizontal stare decisis and the
complexity and uncertainty of vertical stare decisis, making the case is difficult.
Moreover, it should be noted that the claim here is not that rule by decree provides no
publicity at all. Much of the content of constitutional law is clear and settled. The
question is whether adoption of the Constraint Principle would enhance the publicity of
law. The case that it would is compelling—so long as the communicative content of the
constitutional text is not radically indeterminate. Of course, the extent to which the
content is determinate is itself a complex empirical question, which is bracketed for the
purpose of this article. But even at this preliminary stage, there are good reasons to doubt
claims of radically indeterminacy.142
Finally, the it should be noted that the publicity argument favors constraint over some
but not all of the alternatives. Consider, for example, Unconstrained Thayerianism (the
view that courts should always defer to Congress and that Congress should govern
through general legislation but should not consider itself constrained by the Constitution).
Because Unconstrained Thayerianism substitutes Congress for the Constitution but
requires Congress to govern through legislation that itself is public, it does not share the
publicity problems associated with a Supreme Court that promulgates the constitution
through an unconstrained process of case-by-case decisionmaking. Once again, the
139
See Lawrence B. Solum, How NFIB v. Sebelius Affects the Constitutional Gestalt, 91 WASH. U.L.
REV. 1, 22 (2013) (There is, to be sure, a formalist version of the doctrine that is rooted in the idea of the
ratio decidendi: the holding of a case is the rule that is logically implied by the stated reasons necessary to
the resolution of the case on the facts before the appellate court and the legal arguments presented by the
parties. But there is another tradition of thinking about stare decisis that views the holding of a case as the
rule that best predicts the future behavior of a court from the opinions expressed by the judges.)
140
See RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 111 (1977).
141
See Solum, supra note 139, at 22.
142
See Lawrence B. Solum, On the Indeterminacy Crisis: Critiquing Critical Dogma, 54 U. CHI. L.
REV. 462 (1987); see also Heidi Kitrosser, Interpretive Modesty, 104 GEO. L. J. (forthcoming 2016).
70
The Constraint Principle
question whether the Constraint Principle is best depends on pairwise comparisons with
particular alternatives—the subject of Part VII of this Article.
b) Constraint Supports the Stability of Law
The stability of law is usually counted as a virtue, but too much stability counts as a
vice. Nonetheless, it should be clear that ceteris paribus the stability of law is a good
thing, whereas instability of law has no distinctive virtue that is independent of the
content of the law. Why is this so? One of the reasons that we value the stability of law
is that it enables individuals and institutions to formulate long-run plans. When the law is
unstable, planning may be impossible. And even if long-range planning is possible under
conditions of legal instability, individuals and institutions may need to invest resources in
order to prepare for multiple contingencies: stability of law reduces the costs of long run
planning.
These abstract points can be made more concrete by considering the various ways in
which the content of constitutional law affects planning. Consider, for example, the
ability of Congress to legislate concerning complex social and economic problems. The
Constraint Principle entails that the basic rules of the road are fixed, subject only to
constitutional amendments which are likely to be infrequent given the amendment
process. There has only been one significant example of constitutional instability
associated with the amendment process: the ratification of the Eighteenth Amendment
authorizing prohibition in 1919 followed by its repeal by the Twenty-First Amendment in
1933. Moreover, because change by constitutional amendment requires a supermajority,
the costs of instability are likely to taken into consideration during deliberation over
constitutional change.
The Constraint Principle promotes constitutional stability, which provides a pro tanto
reason for constraint as opposed to many (but not all) of its rivals.
c) Constraint Reinforces Legal Certainty
Consider finally the value of legal certainty. Certainty of the law is, ceteris paribus, a
good thing, but uncertainty about the content of the law is only very rarely desirable.
Whereas publicity refers to accessibility of the public to the content of constitutional
doctrine and stability refers to the persistence of doctrine over time, the value of certainty
is focused on the question whether the content of constitutional doctrine at any given time
is certain or uncertain. In this regard, it is helpful to distinguish two different kinds of
uncertainty, which we shall call “risk” and “ignorance” (in lieu of the alternative
vocabulary which distinguishes uncertainty as “risk” from “Knightian uncertainty”).143
Uncertainty as risk is a scalar: the content of the law can be more or less uncertain.
The Constraint Principle increases certainty about the content of constitutional law as
compared to most of the important alternatives. At any given time, many questions of
constitutional law will not have been addressed by the Supreme Court. Some of these
questions give rise to circuit splits or splits among the state appellate courts. Other
questions are only addressed by the federal trials courts, the decisions of which lack stare
decisis effect. And yet other questions have never been addressed by any court. The
143
See FRANK H. KNIGHT, RISK, UNCERTAINTY, AND PROFIT (1921).
71
Lawrence B. Solum
Constraint Principle generates certainty with respect to many of these open questions.
Many of the rivals of constraint do not, for reasons that have already been explored. This
kind of uncertainty is particularly acute during periods when the future composition of
the Supreme Court is difficult to ascertain and with respect to issues where the
preferences of the Justices are unclear.
3. The Politicization Argument
The rule of law is not the rule of politics: this idea needs to be clarified and qualified,
but it captures an important insight about political morality. Of course, law and politics
are closely related. Politics shapes the law, and widely held ideas about democratic
legitimacy suggest that the shaping of law by democratic politics is a good thing.
Moreover, the word “politics” is sometimes understood as referring to statecraft, the
activity that aims to shape communities and nation-states for the common good: in this
sense, the “politicization” of constitutional law would seem to be a virtue and not a vice.
But there is another sense of the word “politics” that associates that term with partisan
and ideological conflict. Understood in this sense, the politicization of constitutional law
is a vice and not a virtue. Hardly anyone is willing to publicly state that the Supreme
Court should vote along partisan lines on constitutional issues—and there is a widely
shared sense that constitutional deliberation should be something other than ideological
strife. Originalism makes this explicit: originalists believe that constitutional practice
should be guided by the communicative content of the constitutional text and not by
partisan politics or ideology. But originalism is not alone in this regard. The multiple
modalities view does not admit of a partisan or ideological modality. Common law
constitutionalism posits a distinctive common-law method that contrasts with partisan
and ideological approaches to governance. The moral readings theory argues that moral
philosophy, not politics, should guide the process of constitutional interpretation and
construction.
Nonetheless, the rejection of the Constraint Principle may, in fact, lead to the
politicization of constitutional law. Take the moral readings theory as an example. Firstorder private judgments about political morality are likely to be divergent given the fact
of pluralism. Even if morality is objective as a matter of metaphysics, human experience
suggests that there is no method for the discovery of moral truths that leads to reliable
intersubjective agreement: metaphysical objectivity (assuming the objective view is
correct) and epistemology subjectivity seem to coexist in the realm of political morality.
Moreover, many moral disagreements are political and ideological disagreements as well.
Contemporary constitutional law is replete with examples, from Roe v. Wade144 to
Citizens United.145
The Supreme Court is subject to political pressures and can become thoroughly
politicized. It is true that the Supreme Court consists of Justices appointed for life terms,
subject to removal through a supermajoritarian process, which requires a majority vote of
impeachment by the House146 and a two-thirds vote for conviction by the Senate, but this
is not sufficient to isolate the Court from politics, because the appointments process
144
410 U.S. 113 (1973).
558 U.S. 310 (2010).
146
U.S. Const. Art I, sec. 2, cl. 5.
145
72
The Constraint Principle
allows political actors, especially the President to influence the composition of the Court.
Under these circumstances, a Court that decides on the basis of the first-order private
judgments of the Justices on matters of political morality can become politicized—even if
the individual members of the Court view their own decisions are based on true moral
readings.
Similar points can be made about many of the other rivals of originalism. The
common-law process may provide some discipline, but given the power of the Supreme
Court to overrule its own prior decisions, it seems likely that a common-law court can
become thoroughly politicized while couching its decisions in the language of the
common-law process. Moreover, contemporary views of the common law frequently
equate “common law” with “judge made law”—leading to a collapse of common law
constitutionalism into the Supreme Court as a superlegislature. Likewise, the multiplemodalities view makes it difficult, if not impossible, to provide intersubjective agreement
on the question whether a given contested Supreme Court decision is wrong, again
opening the door to the politicization of the Court.
The politicization of constitutional law is clearly inconsistent with the rule of law.
And the more serious the politicization, the more the rule of law will be eroded. In times
of intense partisan conflict there is the possibility of a downward spiral of politicization,
with each party attempting to staff the Court with politically reliable judicial operatives.
The bottom of such a downward spiral could involve very serious erosion of the rule of
law. The high politics of principled disagreement about matters of political morality
could be replaced by the low politics of rigging elections and deciding cases on the basis
of an explicit calculus of political costs and benefits.
At this point, it might be objected that the Constraint Principle cannot insulate the
Court from politicization. Original meaning can be disputed, and politics can operate in
the construction zones, where the original meaning does not fully determine doctrine and
outcomes. This is a fair criticism, but we need to recall that the realization of the rule of
law is a scalar and not binary. The relevant question is not whether an originalist court
would be completely apolitical or a common-law court would be thoroughly politicized.
The relevant question is whether the Constraint Principle can reduce the level of
politicization by providing an objective standard for legally correct decisions that can
serve as the focus for intersubjective agreement in a wide variety of cases. That depends,
of course, on epistemological questions about the discoverability of original meaning and
on further questions about constitutional construction. A rule of law approach to the
construction zone may be a necessary complement to the Constraint Principle. There are
a variety of such approaches ranging from default rules (like a principle of Thayerian
deference when constitutional meaning is unclear) to procedural options, like a strong
doctrine of constitutional stare decisis. Those questions are outside the scope of this
Article, but my judgment is that the comparative superiority of originalism with respect
to politicization of constitutional law is likely to be established once those questions are
addressed in due course.
Let us step back and take stock of the cluster of rule of law arguments. Three
arguments have been advanced: (1) the argument from judicial tyranny, (2) the rule-oflaw values argument, and (3) the politicization argument. These arguments all make the
same basic point, that the Constraint Principle advances the rule of law, but each of the
three arguments approaches the question from a different angle. The arguments are not
73
Lawrence B. Solum
fully independent—they interlock in a variety of ways by sharing assumptions and value
premises. But each argument illuminates the connection between the Constraint Principle
and the rule of law in a distinct way.
B. Constraint and Legitimacy
The second cluster of arguments focuses on constitutional legitimacy. Legitimacy is a
complex concept.147 Legitimacy should be distinguished from justice or rightness. It is
possible for a just law to lack legitimacy (because it was imposed by an unelected
dictator), but nonetheless be substantively just. Likewise, a legitimate law (made by an
elected legislature employing the proper procedures) might be unjust. We can think of
legitimacy as a process value: that a law is legitimate is a reason to consider it
authoritative, providing a pro tanto reason for action that stems from characteristics of
the law other than the moral rightness of its substantive content.
There are at least three reasons to affirm the Constraint Principle that connect with the
idea of legitimacy. The first is the familiar argument that the communicative content of
the constitutional text possesses democratic legitimacy, but that violations of constraint
by unelected judges have this characteristic to a much lesser degree. The second
legitimacy argument focuses on the idea of transparency: judicial decisions complying
with constraint can be made transparently, but violations of the Constraint Principle are
frequently disguised through dissimulation, obfuscation, or deception. The third
argument focuses on judicial role; it argues that the role of judges (and most other
officials almost all of the time) does not include an authorized power of constitutional
revision.
1. The Democratic Legitimacy Argument
The democratic legitimacy argument is one of the most familiar in debates over
originalism. The gist of the argument is based on the idea that the Constitution (and
especially the public meaning of the constitutional text) was enacted by “We the People”
and therefore the original public meaning is legitimate, because it has a democratic
pedigree. Moreover, the Constitution is subject to amendment through democratic
processes—and hence to democratic control. These facts provide a pro tanto reason for
complying with the Constraint Principle. If (1) the polity has through democratic
processes ratified a text and retains the power to modify the text, and (2) the text is a
constitution the communicative content of which creates constitutional law, then the
combination of these two facts constitutes a reason for constitutional actors (judges and
other officials) to act in compliance with the text.
The argument itself is simple, but things become more complex once we consider
objections. One familiar objection is that most of the provisions of the Constitution are
old and were enacted by persons now dead. This argument has some superficial appeal,
but if it were accepted it would deprive almost all law of democratic legitimacy. Huge
numbers of statutes were enacted by legislators who were not themselves elected by
147
For an introduction to the concept of legitimacy, see Fabienne Peter, Political Legitimacy, Stanford
Encyclopedia of Philosophy (2016).
74
The Constraint Principle
contemporary majorities. We believe that legislation retains democratic legitimacy until
it is properly repealed. This objection is discussed further below.148
Another objection focuses on the fact almost every provision of the Constitution now
in force was enacted before the franchise was fully extended by five amendments: (1) the
Reconstruction Amendments (race and former condition of servitude)), (2) the Nineteenth
Amendment (women’s suffrage), (3) the Twenty-Third Amendment (limited voting rights
for residents of the District of Columbia), (4) the Twenty-Fourth Amendment (prohibiting
poll taxes), and (5) the Twenty-Sixth Amendment (extending the right to vote to 18 year
olds). The only amendment to complete ratification after the Twenty-Sixth Amendment
is the Twenty Seventh and a majority of the ratification process for that amendment took
place before the Twenty Sixth. Arguably, all of the Constitution lacks full democratic
legitimacy.
Moreover, massive number of Americans are disenfranchised because they have been
convicted of a felony and there is no immediate prospect that this will be remedied.
Hence, it might be argued that even future constitutional amendments would
democratically legitimate, unless and until this problem of disenfranchisement is
addressed. And this objection would appear to apply to many statutes as well.
Progressive legislation from the turn of the century was enacted before women had the
franchise. New Deal legislation was enacted at a point during a period when most
African Americans were disenfranchised. And the same is true of contemporary
legislation, the regulations promulgated various agencies and executive departments, and
the decisions of the Supreme Court.
At this point, most readers will have recognized the reductio, the pattern of argument
that denies the Constitution of 1789 democratic legitimacy leads to the conclusion that
almost none of the law now in effect is democratically legitimate. The problem that leads
to the reductio is the false assumption that democratic legitimacy is a binary, when in fact
it is a scalar. Moreover, in a society in which the franchise has expanded over time,
democratic legitimacy is relative to the standards of the time at which a given law (hence
constitutional provision) is enacted. The original constitution possesses a substantial
degree of democratic legitimacy; constitutional provisions enacted after the extension of
suffrage to women are more legitimate. There is further discussion of this problem
below.149
What about the rivals of constraint? Here it is important to recognize that particular
alternatives differ with respect to democratic legitimacy. One group of theories (e.g.,
multiple modalities, moral readings, common law constitutionalism) are committed to
juristocracy—rule by unelected judges appointed for life terms. The Constraint Principle
is clearly superior to them with respect to democratic legitimacy. Another group of
theories (unconstrained Thayerianism and representation-reinforcement Thayerianism,
for example) are actually superior to the Constraint Principle: this group is organized by a
principle of democratic legitimacy. The democratic legitimacy argument provides a pro
tanto reason to favor the originalism incorporating constraint over some but not all of the
rivals of originalism that reject the constraint principle.
148
149
See infra Part VI.G, p. 100 (discussing dead hand problem).
See infra Part VI.F, p. 97 (discussing exclusion objection).
75
Lawrence B. Solum
2. The Transparency Argument
The second argument in the legitimacy cluster focuses on the relationship between
transparency and legitimacy. The core of the argument is that abidance with the
Constraint Principle is the best alternative to serious violations of the requirement of
transparency that is required for the legitimacy of law. This argument is closely related to
the publicity argument within the rule-of-law cluster. The difference is that transparency
ties to legitimacy, whereas publicity is bound to the rule of law. The two arguments rely
on common premises, but differ with respect to the value they invoke.
The transparency argument is based in part on the observation that nonoriginalists are
reluctant to frankly acknowledge that they have rejected the Constraint Principle; we
have already taken not of this phenomenon in our discussion of the view that the
Supreme Court is a superlegislature.150 The opinions of the Supreme Court have not
explicitly claimed a power to violate the Constraint Principle, and when the issue arises
the Court almost always claims that its decision is allowed or compelled by the original
meaning of the constitutional text. One suspects that individuals who wish to be
nominated for the Supreme Court are well advised to be avow their allegiance to the
Constraint Principle (or something close to it) and to disavow the Supreme Court’s power
to adopt amending constructions.
The Constraint Principle can be affirmed publicly and transparently by constitutional
actors, but denial of the Constraint Principle by constitutional actors is almost always
done privately or expressed publicly in ways that obfuscate the fact that the Constraint
Principle is being denied. It is widely assumed that political transparency is required for
the legitimacy of law and that a secret decisionmaking procedure is illegitimate—except
in special contexts where the need for secrecy is great (such as certain national security
matters). Action on the basis of principles that cannot be made public lack an important
form of legitimacy.
Nonoriginalists might reply to the argument from legitimacy by arguing that false
allegiance to the Constraint Principle is a noble lie. Consider the following version of the
argument:
The Constraint Principle is superficially attractive to citizens because they are
naïve. This naïveté is illustrated by law students, who enter law school as
formalists (assuming that the communicative content of legal texts does and should
constrain judges) and leave as legal realists (who realize that judges do and should
engage in policymaking). It takes intensive training to grasp the truth of legal
realism, and it is simply not practical for ordinary citizens to acquire such training.
For this reason, judges are warranted when they affirm the Constraint Principle as
an abstract truth but act contrary to the constitutional text in practice. This is a
noble lie, and it is justified because it is necessary for the common good.
Notice once again, that this argument actually accepts that that the Argument from
Transparency offers a valid pro tanto reason for acceptance of the Constraint Principle.
The noble lie argument does not defeat the pro tanto force of the transparency argument:
Lies are only “noble” if they are supported by very good reasons.
150
See supra Part I.B.3.f), p. 17.
76
The Constraint Principle
Once again, different rivals of constraint fare differently with respect to transparency.
For example, the superlegislature theory is completely transparent, but for that reason is
likely outside the feasible choice set.151 Common law constitutionalism fares poorly on
the transparency scale—if one believes that the common-law method disguises judging
on the basis of private first-order judgments of political morality (or even worse, politics
and ideology) by cloaking the real basis for judicial decisions in legalistic language of the
common-law method. The moral readings theory is transparent in theory, but in its
transparent form, it seems unlikely that it will be openly avowed in any case in which it
clashes with the constraint principle.
3. The Judicial Role Argument
The gist of the argument is that the judicial role does not include the power to amend
the constitution, and therefore amending judicial constructions have a legitimacy
problem. By contrast, no one would think that a judge who complies with the
constitution is acting illegitimately for that reason.
Although the primary target of the judicial role argument is the judiciary, a similar
argument applies to other officials when they act outside the Article V amendment
process. As applied to the President, the idea is that the President has executive power,
but lacks the power to amend the Constitution. The role of President does not include the
power to amend, circumvent, or nullify the constitution. As presented on this occasion,
the focus of the official roles argument will be on judges—but we will briefly consider
other officials, including the President.
What evidence do we have for the notion that the role of officials such as the President
and Justices of the Supreme Court does not include the power to amend the Constitution?
One powerful piece of evidence is the oath that these officials are required to swear or
affirm. For example, the President of the United States must swear the following oath:
I do solemnly swear (or affirm) that I will faithfully execute the Office of
President of the United States, and will to the best of my Ability, preserve, protect
and defend the Constitution of the United States.152
And the remaining officials of the government, including the Justices of the Supreme
Court are bound by oath as well.
The Senators and Representatives before mentioned, and the Members of the
several State Legislatures, and all executive and judicial Officers, both of the
United States and of the several States, shall be bound by Oath or Affirmation, to
support this Constitution; but no religious Test shall ever be required as a
Qualification to any Office or public Trust under the United States.153
The rules that constitute the government of the United States are provided in the
Constitution. Those rules specify roles. And the plan that established these roles
151
See Lawrence B. Solum, Constitutional Possibilities, 83 IND. L. J. 307 (2008).
U.S. Const. Art. II, Sec. 1.
153
U.S. Const. Art. VI.
152
77
Lawrence B. Solum
provides that individuals empowered to serve in official roles must are bound to support
the rules—that is “this Constitution.”
The phrase “this Constitution” is significant. The first word in the phrase is “this” and
that word is an indexical.154 It specifies that the relevant “Constitution” is “this”
Constitution; in other words, the constitutional text refers to itself. Officials are not
bound to obey constitutional decisions of the Supreme Court or the constitutional
conventions created by Congress or the President. They are bound by the constitutional
text. And this does not mean that they are somehow bound to mere marks on a piece of
parchment. In context, they are required by oath or affirmation to bind themselves to
communicative content (or linguistic meaning) of the constitutional text. If the Fixation
Thesis is true, then means they are bound by the original meaning of the constitutional
text.
Justices of the Supreme Court are required to bind themselves to “this Constitution”
by oath or affirmation. They swear an oath that commits them to what amounts in
substance to the Constraint Principle. This oath is a prerequisite to their assumption of
the office, and the oath defines the limits of the role of judge. The same is true of other
officials, including all other state and federal judges. This means that the officials of the
United States, including Supreme Court justices, have at least a pro tanto reason to
perform their official actions in a way that is constrained by the original meaning of the
constitutional text.
Of course, it is possible for officials to disregard their oath or affirmation and to act in
ways that are inconsistent with the rules that constitute the United States government.
Thus, the President might assume the powers of a dictator and dissolve Congress and
suspend the operation of the Supreme Court. But if the President did this, he would be
acting contrary to his commitment to the scheme of agency established by the
Constitution. If the President succeeded (through military force or demagogic
persuasion), he would have effectively dissolved the Government of the United States
and established a new constitutional regime. Such an action is ultra vires—it is not an
action that is within the scope of his role as a constituent component of the group agent.
And hence he action would be illegitimate.
Similarly, Justices of the Supreme Court can act in ways that are outside the scope of
their roles in the government of the United States. Given the system established by the
Constitution, the United States Supreme Court is given the power to resolve cases and
controversies by offering authoritative interpretations and constructions of the
constitutional text. Given the power of the lower courts to enforce their judgments
through binding coercive orders (including injunctions and writs) and given the rules of
vertical stare decisis, this entails that the Justices can attempt to evade the restrictions on
their official role that bind them to the original meaning of the constitutional text in a
particularly insidious way. They can attempt to convert ultra vires actions into authorized
actions by judicial fiat. But the Supreme Court does not have the power to convert falsity
into truth or faithlessness into fidelity. The Justices can claim their actions are authorized
and they may be able to coerce cooperation by other officials—but when they do this, the
act contrary to the reasons that apply to them in their official capacity.
154
See Christopher R. Green, 'This Constitution': Constitutional Indexicals as a Basis for Textualist
Semi-Originalism, 84 NOTRE DAME L. REV. 1607 (2009).
78
The Constraint Principle
***
At this point some readers may object that the oath does not require that judges
respect the constitutional text, but instead is a more abstract and flexible commitment.
Perhaps the oath is to the constitutional doctrine enunciated by the Supreme Court
through a process of common law decisionmaking? Maybe the oath is to the best moral
reading of our practices as a whole and thus authorizes constructions that override the
constitutional text? Or perhaps the oath requires judges to engage in a complex
argumentative practice that allows them to nullify the constitutional text on the basis of
historical practice or constitutional values? These possibilities might be advanced by
sophisticated constitutional theorists, but to me they seem silly. The oath says none of
these things: parsing “this Constitution” as “my favorite constitutional theory that
allows overrides of the constitutional text” might be charitably construed as the result of
an unconscious process of motivated reasoning. Less charitably, this move might be
characterized as deliberate obfuscation or deception.
***
C. The Relationship Between the Rule of Law and Legitimacy
There are ways that the Supreme Court could ameliorate the effect of violations of the
Constraint Principle on the rule of law. For example, the Court might adopt a rigid
doctrine of vertical and horizontal stare decisis that approximates legislation. Rather than
deciding “one case at a time,” the Court could include a statute-like legislative holding in
each opinion. The Court could then police the lower federal courts for violations of its
holdings. The Court could adopt a notice-and-comment procedure, inviting amicus briefs
in cases in which it announces that it plans to “hold hearings” with an aim to revising or
repealing a legislative holding. This way of proceeding would implement the view that
the Court is a superlegislature.155
Turning the Supreme Court into a superlegislature that holds hearings and issues
legislative holdings would minimize the rule-of-law problems that attend its current
practices, but it would maximize the legitimacy problems. No one should think that an
unelected superlegislature that explicitly governs as such has democratic legitimacy. And
the superlegislature version of the Court abandons any pretense of proper judicial role.
The Constraint Principle fares better than the status quo with respect to both
legitimacy and the rule of law. It is difficult for nonoriginalist theories to achieve both
simultaneously. Some forms of nonoriginalism overcome rule of law problems at the
expense of legitimacy; others achieve legitimacy, but undermine the rule of law. These
tradeoffs are discussed in more detail below when we turn to pairwise comparisons.156
V. ALTERNATIVE JUSTIFICATIONS FOR THE CONSTRAINT PRINCIPLE?
Originalism is a family of constitutional theories unified by the Fixation Thesis and
the Constraint Principle. Originalists differ on a variety of matters, including the
justifications for originalism. In this Part of the Article, we will briefly investigate some
155
156
See infra Part VII.A.6, p. 115.
See infra Part VII, p. 104.
79
Lawrence B. Solum
especially prevalent among legal academics of a certain age—those who went to law
school and started their academic careers before the turn of the century and especially
those whose attitudes were shaped by constitutional debates in the early 1980s. There
are many possible explanations for this phenomenon. For one thing, most legal scholars
were told that originalism is obviously wrong in law school and this message was
reinforced so many times in some many ways that it became unshakeable dogma, a fixed
point in the web of belief.
But there may be another reason. At a very high level of generality, our
understandings of the law are shaped by what might be called the “jurisprudential
gestalt,” a set of ideas about the nature of law that organize our beliefs about the law at
a very high level of generality. The jurisprudential gestalt organizes our beliefs about
which theories are contenders versus those that are nonstarters. It shapes our beliefs
about which narratives about the development of legal thought versus the stories that
implausible. The dominant jurisprudential gestalt of the late twentieth century was
shaped by the way that legal thinkers reacted to, absorbed, and transformed the insights
of the American legal realists in light of the New Deal and the Warren Court. The
dominant jurisprudential gestalt attempted to reconcile the realist critique of legal
formalism with the idea that legal reasoning infused with policy and principle could still
be “law”—an ambition that was shared by the legal process school, Ronald Dworkin,
and many others.
The problem with claims that the central ideas of originalism are true or correct is
that these ideas do not fit the dominant jurisprudential gestalt. Originalism is a variety
of formalism, and if “we” know anything it is that formalism is wrong. Therefore, there
must be something wrong with originalism. If the Fixation Thesis is true, then the
Constraint Principle must be incorrect. If the Constraint Principle is correct, then the
communicative content of the constitutional text must be radically (or at least
substantially) indeterminate. If the radical indeterminacy thesis is false, then maybe we
should take another look at the Fixation Thesis. But there must be a flaw somewhere.
***
VII. PAIRWISE COMPARISONS: RIVALS, THE STATUS QUO, AND COMPROMISE
One of the themes of the Article is that an evaluation of the Constraint Principle
requires pairwise comparisons of originalism with the alternatives. But in this Article, we
cannot hope to accomplish an in-depth pairwise with each many alternatives to the
Constraint Principle. In the long run, each rival view requires its own article or
monograph. On this occasion, all that can be provided is a sketch that draws out the
implications of the preceding analysis for further investigation of the pairwise
comparisons.
Pairwise comparison with proceed in three stages. First, we will compare originalism
to each of the nine rivals identified above. Second, we will compare originalism to what
we can call the “constitutional status quo”—which I argue is a form of eclecticism.
Third, we can compare originalism to compromise theories that partially accept the
Constraint Principle (for a limited domain), but reject constraint for some set of
constitutional issues.
104
The Constraint Principle
A. Pairwise Comparison of Originalism with Nine Rivals
The discussion that follows is crucially important. We have already seen that “it takes
a theory to beat a theory” and hence that pairwise comparison of originalism with rival
theories is required. This is the only way to conduct the debate over originalism in a
rigorous way. To beat originalism, the proponents of a rival view need only show that
their view is superior to originalism. To produce a compelling argument for the
Constraint Principle, originalism with constraint must be compared to each and every
rival. Recall that there are nine rival views (some of which have subtypes):
1.
2.
3.
4.
5.
6.
7.
8.
9.
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
The Supreme Court as Superlegislature
Thayerianism
Constitutional Antitheory
Constitutional Rejectionism
Each of the nine theories will be discussed in turn.
1. Multiple Modalities
The view that constitutional law is a complex argumentative practice with multiple
modalities of constitutional argument is a powerful one because it captures important
features of the actual practice of constitutional law. We begin with a brief explication of
the approach, and then turn to the question whether the multiple-modalities approach
rejects the constraint principle, and end with a discussion of the normative comparison
between the multiple-modalities approach with originalism (and the Constraint Principle).
Various forms of the multiple-modalities approach have been articulated by articulated
by Phillip Bobbitt,210 Stephen Griffin, 211 and by Mitchell Berman and Kevin Toh.212
This view is closely related to Richard Fallon’s influential approach, which he labels as
“constructivist.”213
The gist of Bobbitt and Griffin’s version of this approach is that there are multiple
modalities or a plurality of methods for establishing the truth or validity of a proposition
of constitutional law. Bobbitt’s list of the modalities includes the following:
210
See PHILIP BOBBITT, CONSTITUTIONAL INTERPRETATION 12-13 (1991).
See Stephen M. Griffin, Pluralism in Constitutional Interpretation, 72 Tex. L. Rev. 1753, 1753
(1994) (“Pluralistic theories of constitutional interpretation hold that there are multiple legitimate methods
of interpreting the Constitution.”).
212
Mitchell N. Berman & Kevin Toh, Pluralistic Nonoriginalism and the Combinability Problem, 91
TEX. L. REV. 1739 (2013).
213
Richard H. Fallon, A Constructivist Coherence Theory of Constitutional Interpretation, 100 HARV.
L. REV. 1189 (1987).
211
105
Lawrence B. Solum
historical (relying on the intentions of the framers and ratifiers of the
Constitution);
textual (looking to the meaning of the words of the Constitution alone, as they
would be interpreted by the average contemporary “man on the street”);
structural (inferring rules from the relationships that the Constitution mandates
among the structures it sets up);
doctrinal (applying rules generated by precedent);
ethical (deriving rules from those moral commitments of the American ethos
that are reflected in the Constitution); and
prudential (seeking to balance the costs and benefits of a particular rule).214
The multiple modalities approach qualifies as a form of nonoriginalism to the extent that
it denies that the textualist modality operates as a constraint on the others. This
understanding of the multiple modalities approach can be illustrated graphically, as in
Figure 1: Nonoriginalist Multiple Modalities.
Structural
Doctrinal
Textual
Historical
Ethical
Constitutional
Interpretation
&
Construction
Prudential
Figure 1: Nonoriginalist Multiple Modalities
This understanding of the multiple modalities approach is premised on the idea that the
modalities are “flat”—they lack hierarchical structure. This premise leads to the
conclusion that the textualist modality (which we shall assume is equivalent to “original
meaning”) can be overridden by any of the other modalities on the basis of the process of
constitutional argumentation. This conclusion entails the further consequence that the
multiple modalities approach denies the Constraint Principle, and hence is a form of
nonoriginalism. Thus, the multiple modalities approach is a version of nonoriginalism.
This understanding of the multiple modalities approach is not inevitable. One can
imagine a variation in which the modalities are hierarchically structured with a lexical
ordering that prioritizes the textualist modality (and hence original meaning). This
variation would look something like the following diagram, which incorporates the
interpretation construction distinction:
214
Bobbitt, Constitutional Interpretation, supra note 210, at 12-13 (emphasis added and paragraph
structure altered for clarity).
106
The Constraint Principle
StepOne:Interpretation
• Textualist
StepTwo:Construction
(ConstrainedbyOriginalMeaning)
• Historical
Doctrinal
• Structural
• Ethical
• Prudential
Figure 2: Originalist Variant of Multiple Modalities
On this alternative understanding of the multiple modalities approach, we would have a
form of originalism, because the constitutional doctrines that emerge from Step Two
must be consistent with the original meaning that is identified in Step One. Bobbitt
himself affirms the nonoriginalist version and hence rejects the Constraint Principle.
Compatibilist forms of the multiple modalities approach may be important, but for the
purposes of this investigation of the Constraint Principle, our consideration is limited to
the incompatibilist forms of pluralism, multiple-modalities, and constructivism.
As a normative matter, the multiple-modalities view has serious problems with the
rule of law. Precisely because the modalities are not hierarchical ordered, the multiple
modalities view allows first-order private judgments to play a decisive role in
constitutional law as a practical matter. So long as the Justices are capable of making
arguments from the modalities for the outcome they desire, their decisions will have the
appearance of compliance with the disciplining force of the complex argumentative
practice of constitutional law.
Here is another way of putting it: in every constitutional case that comes before the
Supreme Court, the lawyers make a reasonable case on both sides using the modalities of
constitutional argument. For this reason, the multiple modalities approach does a poor
job providing the grounds for intersubjective agreement that decision for one side or the
other was legally incorrect. As a consequence, the three rule of law arguments for
constraint apply with great force in the context of pairwise comparison with pluralism.
The multiple modalities approach creates the danger of judicial tyranny; it is inconsistent
with the rule of law values of publicity, stability, and certainty; and the multiple
modalities approach creates does not provide a mechanism for preventing the
politicization of constitutional law.
Similarly, the multiple modalities approach does not compare well with the Constraint
Principle with respect to legitimacy. Law as a complex argumentative practice lacks
democratic legitimacy for obvious reasons: complex argumentation by the justices might
take democratic values into account but it does not prioritize consistency with
democratically legitimate legal norms. As a matter of ideal theory, the multiple
modalities approach might be transparent, but this approach lends itself to violations of
107
Lawrence B. Solum
transparency because of the manipulability of the modalities of constitutional argument.
But the multiple modalities argument does fare well with respect to judicial role: one of
the strengths of this approach is that it requires judges to act on the basis of a coherent
theory of what counts as a legal consideration.
2. Moral Readings
Ronald Dworkin calls his general interpretive method (including his view of
constitutional interpretation and construction) “constructive interpretation.”215 A similar
approach is found in the work of James Fleming, who uses Dworkin’s phrase “Moral
Readings” to describe his constitutional theory.216 The discussion that follows will focus
on Dworkin’s view. Because Dworkin’s theory is much discussed, I will assume that
readers are familiar with the core notion of fit and justification. On my reading of
Dworkin’s theory, the content of constitutional law is determined by the moral theory that
best fits and justifies the legal materials as a whole. Thus, the content of constitutional
doctrine is primarily determined by the moral theory that best fits constitutional texts,
institutions, practices, and judicial decisions on constitutional matters.
Dworkin is sometimes hard to pin down, and his theory was elaborated in many texts
over the course of five decades. One understanding of Dworkin is the following: because
constructive interpretation aims to make our practices “the best that they can be” in light
of the institutional history, his theory implies that meanings can change over time. Our
“moral readings” of the constitutional text are not fixed, but instead evolve in response to
changing circumstances and our evolving constitutional values.217 This would make
Dworkin a nonoriginalist, but it might allow him to accept the Constraint Principle. In a
footnote, I suggest that that Dworkin may actually accept fixation (in his own way),218
but for the purposes of this Article, the key question is whether he also accepts the
Constraint Principle.
In his most explicit discussion of the relationship between what we are calling
communicative content and legal content, Dworkin juxtaposes his view with that of
Laurence Tribe. Here is the full passage, which includes an internal (double-indented)
quotation from Tribe:
Tribe endorses a very strong form of textual fidelity. Tribe states:
I nonetheless share with Justice Scalia the belief that the Constitution’s
written text has primacy and must be deemed the ultimate point of
departure, that nothing irreconcilable with the text can properly be
215
RONALD DWORKIN, LAW’S EMPIRE 62–86 (1986).
See Fleming, supra note 182.
217
The phrase “moral reading” is Dworkin’s. See RONALD DWORKIN, FREEDOM’S LAW: THE MORAL
READING OF THE AMERICAN CONSTITUTION 2–3 (1996), but it is now strongly associated with James
Fleming’s Dworkinian theory of constitutional interpretation and construction. See James E. Fleming,
Fidelity, Change, and the Good Constitution, 62 AM. J. COMP. L. 515, 515 (2014).
218
In 1997, Dworkin seems to have embraced some version of the notion that semantic content is fixed
at the time a text is written. See Dworkin, The Arduous Virtue of Fidelity: Originalism, Scalia, Tribe, and
Nerve, 83 Fordham L. Rev. 1249, 1251–52 (1997).
216
108
The Constraint Principle
considered part of the Constitution; and that some parts of the Constitution
cannot plausibly be open to significantly different interpretations.219
That is a stronger statement of textual fidelity than I [Dworkin] would myself
endorse, because, as I said, precedent and practice over time can, in principle,
supersede even so basic a piece of interpretive data as the Constitution’s text when
no way of reconciling them all in an overall constructive interpretation can be
found. I agree with the Tribe of this statement, however, that the text must have a
very important role: We must aim at a set of constitutional principles that we can
defend as consistent with the most plausible interpretation we have of what the text
itself says, and be very reluctant to settle for anything else.220
Dworkin does not use the same conceptual vocabulary as we have been employing here,
but his point can be translated. Dworkin recognizes that the communicative content of a
text is fixed at the time the text is written. But in the case of the constitutional text, the
legal content of constitutional doctrine can change, because the “constructive
interpretation” of the law as a whole can override the communicative content. In other
words, Dworkin accepts fixation as a thesis about “interpretation” (communicative
content), but rejects the Constraint Principle.
Dworkin’s rejection of the constraint principle is an essential and not an accidental
feature of his theory. Dworkin’s theory is holistic: his fictional judge Hercules constructs
the theory that best fits and justifies institutional practice as a hole. And his theory is
dynamic: as institutional practice (including judicial decisions) changes over time, the
content of the theory that best fits and justifies the law as a whole will change. Because
his theory is holistic and dynamic, the content of constitutional law cannot be constrained
(in the sense of constraint as consistency) by the communicative content of any single
text. Using the typology developed above, we would classify Dworkin as a
nonoriginalist.
Pairwise comparison of originalism (with the Constraint Principle) and constructive
interpretation (or moral readings) suggests that both the rule of law and legitimacy
provide pro tanto reasons to favor constraint. It is built into the moral readings approach
that officials (especially judges) will need to rely on their first-order private judgments of
political morality when engaged in constitutional construction. This leads to rule by
decree and hence judicial tyranny, undermines the rule of law values of publicity,
stability, and certainty, and creates the risk of politicization, given interaction between the
demands of the theory and the political nature of the appointments process.
Likewise, constructive interpretation fares poorly with respect to some but not all of
the legitimacy. The notion of judges as philosopher kings is antithetical to democratic
legitimacy. If practiced in good faith, constructive interpretation would be transparent,
but given political realities, judges who practice this method might conceal the role their
values play in their decisions. Similarly, constructive interpretation could, as a matter of
ideal theory, avoid politicization altogether—with the Supreme Court engaged in
apolitical philosophical deliberation, but in nonideal theory, the focus on direct resort to
219
Laurence H. Tribe, Comment, in ANTONIN SCALIA, A MATTER OF INTERPRETATION: FEDERAL
COURTS AND THE LAW 23 (1997).
220
Dworkin, supra note 218, at 1259–60.
109
Lawrence B. Solum
the moral believes of the Justices might facilitate politicization of the appointments
process.
3. Common Law Constitutionalism
In his book entitled, The Living Constitution,221 David Strauss has argued for a view
that we might call, “Common Law Constitutionalism.” Again, we can consider two
versions of the theory. The nonoriginalist version of Common Law Constitutionalism
would affirm that constitutional law is best conceived as the outcome of a common-law
process that takes into account the original meaning of the constitutional text, but it not
constrained by that meaning: this would be a form of nonoriginalism, because it would
explicitly reject that Constraint Principle.
But we can also imagine an originalist version of Common Law Constitutionalism.
The originalist version would affirm that constitutional law is best conceived as the
outcome of a common-law process but includes some version of the Constraint Principle.
Assuming that Common Law Constitutionalism accepts the Fixation Thesis, the resulting
theory would qualify as a member of the originalist family of constitutional theories.
What is Strauss’s actual position? Consider the following passage from The Living
Constitution:
We cannot make an argument for any constitutional principle without purporting
to show, at some point, that the principle is consistent with the text of the
Constitution. That is an essential element of our constitutional culture. And no
provision of the Constitution can be overruled in the way a precedent can, or
disregarded in the way original understandings often are.222
Strauss’s book is lucid and elegant, but very compact—and so there are some ambiguities
here.223 On the one hand, this passage might be read as an affirmation of the minimalist
version of the Constraint Principle, that is Constraint as Consistency. Strauss seems to
affirm both elements of Constraint as Consistency. The first element requires consistent
of the content of constitutional doctrine with the communicative content of the
constitutional text: Strauss explicitly says that constitutional principles must be
“consistent with the text” and “no provision of the Constitution can be overruled.” The
second element requires that every element of the text be reflected in constitutional
doctrine: Strauss explicitly says that no provision can be “disregarded.”
But Strauss also says that “original understandings” can be disregarded, but it is not
clear what he means by “understandings.” To be fair to Strauss, we would need to
undertake a painstaking examination of all his writings and precisely reconstruct his
operative conceptions including “meaning,” “understanding,” and “text.” My impression
is that such a reconstruction would reveal that Strauss employs the term meaning in
multiple senses, that by understandings he means application beliefs, not communicative
content (or linguistic meaning), and that his theoretical apparatus collapses the
221
DAVID A. STRAUSS, THE LIVING CONSTITUTION (2010).
Id.
223
Strauss’s position is clarified in his recent introduction to the Harvard Law Review’s Supreme Court
issue. See David A. Strauss, Foreword: Does the Constitution Mean What It Says?, 129 HARV. L. REV. 1,
57 (2015) (implicitly rejecting the Constraint Principle by stating that “original understandings are binding
for a time but then lose their force”). We might call this position “Short-Run Originalism.”
222
110
The Constraint Principle
interpretation-construction distinction. Hence, Strauss’s theoretical writings
underdetermine the question whether his version of Common Law Constitutionalism is
originalist or nonoriginalist. As of the writing of The Living Constitution, he did not have
a theoretically precise answer to the question whether the legal content of constitutional
doctrine must be consistent with the communicative content of the constitutional text;
subsequent work suggests that he may have come down decisively as a nonoriginalist—at
least with respect to the long-run implementation of the Constitution.224
Whatever position Strauss would ultimately take on the Constraint Principle, we can
formulate a nonoriginalist version of common-law constitutionalism. This version of
common-law constitutionalism would allow judges to make decisions that are
inconsistent with the constitutional text so long as the decisions were made in accord with
common law methods.
Pairwise comparison of originalism with common law constitutionalism leads to
conclusions that are similar to those reached with respect to the modalities and
constructive interpretation. On the surface, common law constitutionalism looks legal,
but it suffers from the rule of law defects that ground criticisms of the common law in
general. Because the common-law process is so open ended (absent a strong doctrine of
stare decisis), it opens the door for judicial tyranny. The same feature of the common law
undermines the rule of law values of publicity, stability, and certainty. As a matter of
ideal theory, common law judges are apolitical, but in the actual world, the looseness of
common-law constraints invites politicization.
Common law constitutionalism does not fare well with respect to democratic
legitimacy for familiar reasons. In theory, it scores well on transparency, but in practice
there are substantial worries that common law reasoning can disguise the real motives for
judicial decisions. The same point can be made with respect to politicization.
4. Popular Constitutionalism
The phrase “popular constitutionalism”225 can be used to describe many different
ideas. In one sense, popular constitutionalism is a positive theory of constitutional
change: the people themselves determine the content of constitutional law. In another
sense, popular constitutionalism is a normative theory: the content of constitutional
doctrine should be determined by “We the People.” One version of normative popular
constitutionalism would be largely consistent with the Constraint Principle: if one argued
that “We the People” are the authors of the constitutional text and therefore it is higher
law that should constraint the people’s agents (including judges and other officials), that
would provide a normative justification for constraint.
But there is a form of normative popular constitutionalism that would be inconsistent
with the constraint principle. If “We the People” authorized the President to make
transformative appointments to the Supreme Court and the Justices then adopted
amending constructions of the constitutional text that were authorized by the people, then
a normative popular constitutionalist might argue that these violations of the Constraint
224
See Strauss, id.
See Larry Alexander & Lawrence B. Solum, Popular? Constitutionalism?, 118 HARV. L. REV. 1594
(2005).
225
111
Lawrence B. Solum
Principle are democratically legitimate and hence justified. Bruce Ackerman’s work
might be read along these lines, but this is not the occasion to offer a textually supported
interpretation of his large corpus of work on the Constitution.226
Let us call the version of normative popular constitutionalism that sanctions
constitutional change outside of the amendment process, “Extratextual Normative
Popular Constitutionalism.” For ease of expression, I shall use the phrase “popular
constitutionalism” to refer to this view—unless otherwise specified. As defined, popular
constitutionalism authorizes violations of the Constraint Principle and should be
classified as a form of nonoriginalism.
The pairwise comparisons of originalism with the first three rivals of originalism
(multiple modalities, constructive interpretation, and common law constitutionalism)
revealed a good deal of similarity among the three approaches. Popular constitutionalism
is distinctly different. Popular constitutionalism can be consistent with the rule of law—
so long as the mechanisms of popular constitutional change are such that new
constitutional regimes have clearly defined starting points and content. And popular
constitutionalism fares well on the criterion of democratic legitimacy—which is baked
into the very idea of constitutional change authorized by “We the People!”
But it is not clear the popular constitutionalism gets off the ground as a normative
account of constitutional change. Whereas the Article V process provides clear, public,
and transparent criteria for constitutional amendments, popular constitutionalism outside
the Article V process leads to questions about when constitutional change has been
authorized by the people. Consider, for example, the claim that the New Deal Settlement
was authorized by the election of Franklyn Delano Roosevelt to an unprecedented fourth
term. It may be true that Roosevelt ran, in part, against the Supreme Court, but it is not
clear that he ran against constitutional constraint; indeed, it seems more likely that he
presented New Deal constitutionalism as restoration of what we are calling the Constraint
Principle. Constitutional change through transformative appointments may result in
substantial uncertainty about the content of constitutional law, especially if the Justices
disagree among themselves about the “original meaning” of the transformative
moment.227
5. Multiple Meanings
A fifth version of nonoriginalism is based on the idea that texts have “multiple
meanings” rather than a single fixed meaning: call this the “Multiple Meanings Theory”
of constitutional interpretation.228 The gist of the argument would go something like this:
226
See Bruce A. Ackerman, Transformative Appointments, 101 HARV. L. REV. 1164, 1166 (1988) (“As
traditionalist Justices retired, however, the Senate gave enthusiastic support to a series of transformative
appointments, which had precisely the consequences that Roosevelt's opponents feared—and Reagan's
opponents fear today. Within a few short years, the New Deal majority had decisively repudiated the rightsoriented jurisprudence so elaborately developed by its Republican predecessors.”).
227
See Kurt T. Lash, The Constitutional Convention of 1937: The Original Meaning of the New
Jurisprudential Deal, 70 FORDHAM L. REV. 101 (2001) (describing disagreements among the Justices about
the meaning of the New Deal).
228
My reconstruction of the argument has been influenced by a work-in-progress by Cass Sunstein, see
Cass R. Sunstein, There Is Nothing that Interpretation Just Is,
112
The Constraint Principle
texts do not have a single meaning (in the linguistic sense); instead, they have multiple
meanings. Because there are multiple meanings, we must select between them, and this
process of selection must be guided by normative considerations. Originalism is false,
because it privileges one of the multiple meanings (such as original intent or original
public meaning) and therefore begs the crucial normative questions as to which meaning
should govern in any particular case.
Some of the premises of the Multiple Meanings Theory are correct, but from them it
does not follow that the Constraint Principle is false. Mark Greenberg makes the point
that there is more than type of linguistic content in the context of a discussion229 of Smith
v. United States230—the Supreme Court case in which the question was whether offering
to trade a gun for cocaine constituted use of a firearm for the purpose of a penalty
enhancement provision.231
As the contemporary study of language and communication has made clear,
there are multiple components and types of linguistic content. In Smith, there are at
least two types of linguistic content plausibly associated with the statutory text that
would yield opposite outcomes in the case. First, there is the semantic content of
the statutory text—roughly, what is conventionally encoded in the words. Second,
there is the communicative content—roughly, what the legislature intended to
communicate (or meant) by enacting the relevant text.232
So far, so good. Semantic content is distinct from communicative content. Moreover,
the Gricean speaker’s meaning of an utterance is not necessarily identical to the meaning
that the audience actually takes from the utterance.
We can translate Greenberg’s point into constitutional terms. For the sake of
simplicity, we can focus on four distinct “meanings”:
•
•
•
•
Framers’ Meaning: The content that the authors of a constitutional provision
intended to convey to the relevant readers (e.g., the public) through the
readers’ recognition of the framers’ communicative intentions.
Original Semantic Meaning: The content that competent readers of a
constitutional provision would have attributed to a constitutional provision,
given the conventional semantic meanings of the words and phrases and their
syntactic structure, but without consideration of context.
Ratifiers’ Meaning: The content that the ratifiers (or the subset of ratifiers who
were competent speakers of English and who actually read the text) actually
attributed to a constitutional provision.
Reasonable Contemporary Meaning: The meaning that a reasonable
contemporary reader would attribute to the text given contemporary semantics
and syntax and the counterfactual assumption that the text was written in the
contemporary period.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2489088 (August 29, 2014), and by exchanges with
Richard Fallon. My version of the argument should not be taken as representing their positions.
229
Mark Greenberg, The Moral Impact Theory of Law, 123 YALE L.J. 1288, 1291–92 (2014)
230
508 U.S. 223 (1993).
231
Id.
232
Greenberg, supra note 229, at 1291–92.
113
Lawrence B. Solum
The first three forms of meaning are all originalist in the sense that they accept the
fixation thesis. Framer’s intentions are fixed at the time each provision is authored—and
original clause meaning is similarly fixed. Ratifiers’ understandings are fixed at the time
of ratification. If the Multiple Meanings theory were limited to these three meaning
types, it would be a form of originalism. It would differ from Public Meaning
Originalism of Original Intentions originalism because it would allow judges to pick and
choose between the three forms of original meaning. And because Original Semantic
Meaning is acontextual, it creates substantial construction zones.233
The Multiple Meanings theory explicitly rejects the Constraint Principle when it adds
Reasonable Contemporary Meanings to the list from which constitutional actors may
choose. This is not to say that the Multiple Meanings views rejects the idea of constraint
altogether. It imposes a form of constraint, but that form of constraint is untethered from
the communicative content of the constitutional text. It would, for example, allow judges
to base their decisions on meanings created by linguistic drift—the phenomena by which
the conventional semantic meanings of words change over time. This opens the door to a
process of constitutional change that begins with constitutional actors arguing for a new
meaning of a constitutional word or phrase: “commerce” should be understood as “social
interaction” and not as “trade in goods.” Once the new usage becomes established,
constitutional actors are then entitled to rely on the new sense of the word. Because
shifts in meaning through linguistic drift are not themselves limited by the Constraint
Principle, this version of the Multiple Meanings theory authorizes the creation of
constitutional doctrines that are inconsistent with original meaning.
So, the question whether the Multiple Meanings theory violates the Constraint
Principle depends on the precise set of meanings that is authorized by a particular version
of the theory. Versions of Multiple Meanings that are limited to “original meanings” do
not violate constraint, but once a nonoriginal meaning is admitted to the list, the version
that allows the nonoriginal meaning should be classified as nonoriginalist (in the sense in
which that term is used in this Article): those forms of the Many Meanings theory violate
the Constraint Principle.
The multiple meanings theory embraces a loose version of the Constraint Principle,
but the ability of the Justices to pick and choose among the possible meanings of the text
creates problems of both the rule-of-law and legitimacy.
The rule of law problems will be especially acute if the Justices are allowed to pick
and choose among the possible meanings of the constitutional text on a case by cases
basis. And these problems will be further exacerbated if one of the allowable meanings is
contemporary acontextual semantic meaning: this is because bare semantic meaning is
sparse and therefore opens the door to a very large zone of underdeterminacy and hence
judicial discretion as a functional matter. Unless the space is filled with a restraining
default rule (such as deference to Congress), all of the rule of law and legitimacy
233
More precisely, acontextual semantic content is sparse. See Victoria F. Nourse, Elementary
Statutory Interpretation: Rethinking Legislative Intent and History, 55 B.C. L. REV. 1613, 1658 (2014)
(Semantic content is exceedingly sparse, as Scott Soames and other philosophers of language have shown.).
In terms more familiar to lawyers, the literal meaning of the constitutional text, if read without any
reference to context, could mean many different things, corresponding to the many different contextually
enriched meanings that the text would have in possible worlds with the same text but different contexts.
114
The Constraint Principle
problems for the other views we have considered so far will apply with equal force to the
multiple meanings view.
6. The Supreme Court as Superlegislature
Consider yet another possible view. A constitutional theorist might affirm the view
that the Supreme Court should frankly operate as a superlegislature and explicitly
embrace the power to override the constitutional text by promulgating amending
constructions (judicial doctrines that are inconsistent with the communicative content of
the constitution). This view is rarely embraced explicitly as a normative theory. The
label “superlegislature” (or “super-legislature”) is frequently used to express a
criticism—as it was when first introduced by Justice Brandeis (joined by Justice Holmes)
in 1931.234 Max Lerner made the criticism explicit in 1933:
Viewed thus the Court through its power to veto legislation has also the power to
channel economic activity. In that sense, it has been often called a super-legislature,
exercising powers tantamount to the legislative power, but more dangerously since
it is not subject to the same popular control.235
This critical usage of “superlegislature” has been echoed by subsequent scholars.236 And
the Supreme Court itself denied that it “sits a super-legislature” in Griswold v.
Connecticut.237
There is logical space for nonoriginalists to explicitly embrace the notion that the
Supreme Court acts a perpetual constitutional convention238 (with the power to adopt
amending constructions by a majority vote). And this view has been articulated by Brian
Leiter in his essay, Constitutional Law, Moral Judgment, and the Supreme Court as
Super-Legislature. 239 Leiter writes,
234
Jay Burns Baking Co. v. Bryan, 264 U.S. 504, 534, 44 S. Ct. 412, 421, 68 L. Ed. 813 (1924)
(Brandeis J., dissenting (To decide, as a fact, that the prohibition of excess weights ‘is not necessary for the
protection of the purchasers against imposition and fraud by short weights,’ that it ‘is not calculated to
effectuate that purpose,’ and that it ‘subjects bakers and sellers of bread’ to heavy burdens, is, in my
opinion, an exercise of the powers of a super-Legislature-not the performance of the constitutional function
of judicial review.).
235
Max Lerner, The Supreme Court and American Capitalism, 42 YALE L.J. 668, 696 (1933).
236
William W. Van Alstyne, The Demise of the Right-Privilege Distinction in Constitutional Law, 81
Harv. L. Rev. 1439, 1447 (1968) (“A court need not “weigh” or “balance”; it need simply apply the literal
mandate of a given constitutional provision flatly to forbid government from conditioning its largess on any
waiver of such a provision regardless of the circumstances. A court may thus avoid any unseemly
appearance of acting as a superlegislature.”).
237
381 U.S. 479, 482 (1965).
238
The idea of the Supreme Court as a perpetual constitutional convention was suggested to me by Kurt
Lash’s important article on the constitutional theory of the New Deal Court. See Kurt T. Lash, The
Constitutional Convention of 1937: The Original Meaning of the New Jurisprudential Deal, 70 FORDHAM
L. REV. 459, 462-63 (2001) (“Acting, in effect, as a constitutional convention, the New Deal Court had the
responsibility to draft the charter for post-New Deal judicial review. Unanimously rejecting the common
law method of Lochner and Swift, the members of this New Deal Convention declared that judicial
interference with the political process henceforth required, at the very least, some clear textual
justification.”).
239
Brian Leiter, Constitutional Law, Moral Judgment, and the Supreme Court as Super-Legislature, 66
HASTINGS LAW JOURNAL 1601 (2015).
115
Lawrence B. Solum
[A]ll political actors know that the U.S. Supreme Court often operates as a
super-legislature, and thus that the moral and political views of the Justices are
decisive criteria for their appointment. This almost banal truth is, however, rarely
discussed in the public confirmation process, but is common knowledge among
political and legal insiders. To be sure there is media speculation about the political
predilections of the nominees, but their actual moral and political views are treated
as off limits in the real confirmation process. This anti-democratic secrecy is, in my
view, deeply wrong and must be replaced with a realistic acknowledgment of the
role of the Supreme Court as a political actor of limited jurisdiction.240
Leiter is not explicitly advocating the proposition that the Supreme Court should act as a
superlegislature; rather his point is that its role as a superlegislature should be made
public and transparent.
But it is at least possible that some nonoriginalists privately hold the view that the
Supreme Court should act as a superlegislature—adopting amending constructions of the
constitutional text on the basis that these amendments are desirable. One problem with
this position is that it is difficult to affirm in public, because it seems likely that any
constitutional actor who affirmed this position transparently would meet substantial
political resistance. Presidents do not openly affirm that they select Justices who will
override the constitutional text. Judicial nominees are likely to identify as originalists,
even if those “in the know” doubt the sincerity of their protestations.241 But even if the
superlegislature view is not made public, it may in fact be the view affirmed in private by
Presidents, Senators, Justices, and constitutional theorists.
It almost goes without saying that the superlegislature view can be formulated so that
it is inconsistent with the Constraint Principle. Indeed, the use of the phrase
“superlegislature” implies that the Court goes beyond interpretation and is playing the
role of a constitutional convention that creates and amends constitutional provisions. For
the purposes of this essay, we will assume that the Superlegislature Theory has this
feature and hence that it is a form of nonoriginalism.
Pairwise comparison of the superlegislature view with the Constraint Principle is
stark. If the Supreme Court has the power to rewrite the Constitution on a case by case
basis, we have rule by judicial decree and hence judicial tyranny. Depending on the
circumstances, this may lead to problems with stability and certainty as well. On the
other hand, the superlegislature theory potentially satisfies the publicity requirement—so
long as the Supreme Court’s opinions themselves make the rules clear. Similarly, the
superlegislature view results in problems with democratic legitimacy and politicization,
but not with transparency.
7. Thayerianism: Constrained, Unconstrained, and Representation-Reinforcement
I will use the term “Thayerianism” to designate a family of theories that include a
strong principle of deference to democratic institutions, usually Congress but variations
of Thayerianism might include deference to the national executive and to elected officials
240
Id. at 1-2.
Confirmation Hearing on the Nomination of Elena Kagan to Be an Associate Justice of the Supreme
Court of the United States: Hearing Before the S. Comm. on the Judiciary, 111th Cong. 62 (2010)
(statement of Elena Kagan).
241
116
The Constraint Principle
at the state level. The Thayerian family includes at least three important views that reject
or partially reject the Constraint Principle. I will call three two theories “Constrained
Thayerianism,” “Unconstrained Thayerianism” and “Representation Reinforcement
Thayerianism.” Let us stipulate as follows:
Constrained Thayerianism: This view includes two principles: (1) courts should
defer to Congress and should refrain from exercising the power of judicial review to
strike down statutes duly enacted by Congress, and (2) Congress should act
consistently with the original public meaning of the constitutional text unless the first
principle authorizes inconsistent action.
Unconstrained Thayerianism: This view includes two features: (1) Courts should
defer to Congress and refrain from exercising the power of judicial review to strike
down statutes enacted by Congress; (2) Congress should not consider itself bound by
the original meaning of the constitutional text and hence should enact statutes that
violate the original meaning (when there are good reasons for so doing).
Representation Reinforcement Thayerianism: This view has two elements: (1) courts
should defer to Congress and should refrain from exercising the power of judicial
review to strike down statutes duly enacted by Congress unless the exceptions
specified in the second element apply, and (2) courts should exercise the power of
judicial reviews when required to protect discreet and insular minorities or to protect
the democratic process (for example, by protecting the right to vote and the right to
engage in political speech).
a) Constrained Thayerianism
Constrained Thayerianism is partially consistent with the Constraint Principle. The
core idea is that courts should defer to Congress, but that Congress should consider itself
constrained by the constitutional text. In essence, this view transfers the final authority
over the meaning and effect of the Constitution from Congress to the Supreme Court.
Constrained Thayerianism is almost a version of originalism. It accepts the Constraint
Principle with one important exception: the role of the Supreme Court. For the purposes
of this discussion, I will simply assume that the original meaning of the Constitution
requires the Supreme Court (and the lower federal courts) to apply the Constitution as the
supreme law of the law. Although the modern description of the Supreme Court’s role
refers to the concept of “judicial review,” the more traditional formulation conceptualizes
this role in terms of a duty to follow the law.242 To the extent that Constrained
Thayerianism authorizes the courts to depart from the original understanding of the
phrases “judicial power” and “Supreme Court,” it violates the Constraint Principle.
But otherwise, Constrained Thayerianism is a form of originalism. It requires that
Congress’s action be consistent with the original communicative content of the
constitutional text. Thus, the normative question in pairwise comparison of originalism
with Constrained Thayerianism is whether the balance of reasons favors a violation of
constraint with respect to the role of the courts in constitutional interpretation and
construction while also favoring constraint by original meaning in all other cases.
242
See PHILIP HAMBURGER, LAW AND JUDICIAL DUTY (2008).
117
Lawrence B. Solum
Many of the usual arguments for and against constraint drop out in this pairwise
comparison. For example, the rule of law arguments (from judicial tyranny, the rule of
law values, and politicization of the courts) do not favor complete constraint over
constrained Thayerianism. Democratic legitimacy actually favors Constrained
Thayerianism, and transparency is neutral in this context. The legitimacy argument from
judicial role does favor the Constraint Principle over Constrained Thayerianism.
Likewise, most of the arguments against constraint apply with equal force to Constrained
Thayerianism, including the arguments from canonical cases, injustice, the dead hand,
and exclusion.
This leaves the majoritarian objection to originalism. On the one hand,
majoritarianism seems to favor Congress as the ultimate authority in constitutional
interpretation and construction: after all, Congress is elected, whereas the Supreme Court
is not. On the other hand, Constrained Thayerianism is antidemocratic in an important
respect—because a constrained Congress is disabled from making decisions on the basis
of democratic politics whenever the will of the majority is contrary to the original
meaning of the constitutional text.
Nonetheless, democratic politics might play an important role in the construction
zone—the set of issues and cases in which the communicative content of the
constitutional text underdetermines legal effect. Democratic Thayerianism could adopt a
majoritarian approach to constitutional construction with respect to vague and opentextured constitutional provisions. Pairwise comparison on this issue is not possible
without specifying the theory of constitutional construction that originalism should
employ in the construction zone, but for the purposes of this article, that theory has not
yet been specified. As noted above,243 one approach to the construction zone would
involve deference to Congress and hence would be identical to Constrained Thayerianism
in the construction zone. But other approaches (such as a presumption of liberty) would
differ substantially.
A full pairwise comparison of the full Constraint Principle with Constrained
Thayerianism would require investigation of the relative institutional competence of the
Supreme Court and Congress as the final authority on constitutional interpretation and
construction. The conventional wisdom is that the judiciary is better suited to this role,
but the conventional wisdom may be wrong. A full investigation of this question is
beyond the scope of this Article, and for that reason, this aspect of pairwise comparison
cannot be completed here.
Constrained Thayerianism occupies an important position in the conceptual space of
constitutional theory, but because it accepts so much of the Constraint Principle, it can be
set aside for the present purposes. If Constrained Thayerianism provides the best theory
of constitutional interpretation and construction, then originalism will largely have won
the day.
b) Unconstrained Thayerianism
“Unconstrained Thayerianism” incorporates the idea of judicial deference to Congress
that is common to all forms of Thayerianism, but it rejects the Constraint Principle.
Thus, Unconstrained Thayerianism authorizes congressional actions that are inconsistent
243
See supra, text accompanying notes 26-27.
118
The Constraint Principle
with the original meaning of the constitutional text, and it gives Congress the power to
authorizes violations by other institutions (e.g., the states, the national executive, and the
courts). Unconstrained Thayerianism does incorporate a principle of constraint, but that
principle only requires that Congress act through generally applicable legislation—and
hence, it forbids Congress to take on the role of a court of last resort with the power hear
petitions for the enactment of statutes that overrule particular judicial decisions. One can
imagine other versions of Unconstrained Thayerianism with different features, but on this
occasion, pairwise comparison will focus on the stipulated version.
Unconstrained Thayerianism avoids the Judicial Tyranny Argument: so long as
Congress acts through prospective statutory enactments of general application, there can
be no rule by decree. Likewise, Unconstrained Thayerianism avoids the politicization of
the judiciary, although it does this by politicizing constitutional law via a more direct
mechanism, transfer of constitutional authority from a judicially-interpreted written text
to an unconstrained Congress.
Although Unconstrained Thayerianism fares better than many of the other rivals of
constraint with respect to the rule of law values, it is still at a disadvantage when
compared to the Constraint Principle. Constraint provides a substantial degree of
publicity, stability, and certainty—for reasons discussed above.244 Precisely because the
Constitution is relatively difficulty to amend, constrained constitutional law provides
stable, certain, and public rules governing the structure of government and fundamental
rights. An unconstrained Congress has the ability to rewrite the Constitution by majority
vote in each session.
Unconstrained Thayerianism fares beats the Constraint Principle with respect to
democratic legitimacy and ties constraint on transparency, but it involves a
transformation of judicial role by requiring courts to uphold statutes even when they
clearly violate the Constitution. Assuming that a duty to act in accordance with law is a
core component of the role of courts and judges, Unconstrained Thayerianism faces a
legitimacy problem on this score.
Unconstrained Thayerianism runs into many of the objections that are made to
originalism and it fares worse than originalism on some. Unconstrained Thayerianism
cannot be squared with any of the canonical cases and it authorizes injustice of any
kind—so long as the injustice is approved by Congress. But Unconstrained
Thayerianism fares better than constraint with respect to majoritarianism, exclusion, and
the dead hand.
Unconstrained Thayerianism requires that we reject the Constitution. And for this
reason, many readers may believe that it is simply outside the feasible choice set. Indeed,
some readers may see discussion of Unconstrained Thayerianism as a “red herring” or
“straw man”—a rival of originalism that is so far outside the mainstream of constitutional
discourse, both in public and in the academy, that its introduction into the debate draws
attention away from the real and important issues.
My own sense is that Unconstrained Thayerianism is, for now, outside the feasible
choice set. There are scenarios in which Unconstrained Thayerianism might become our
political future, but they involve external shocks that change fundamental attitudes
towards the Constitution—a crisis with the normative force of the Great Depression or
one of the World Wars. Nonetheless, Unconstrained Thayerianism occupies an important
244
See supra Part IV.A.2, p. 67
119
Lawrence B. Solum
role in conceptual space. Parliamentary democracy is an important option, and
Unconstrained Thayerianism represents the closest approximation to that option so long
as the Constitution remains a formal part of American law.
The most important questions that inform pairwise comparison of Unconstrained
Thayerianism with the Constraint Principle concern institutional design. Would
protection of individual rights be better served by Unconstrained Thayerianism or by
constraint? If Congress assumed the power to alter the basic structure of government,
how would it use that power? And what consequences would follow? These are complex
questions that can only be answer by the methods of the social sciences. And they
certainly cannot be resolved in this Article.
c) Representation Reinforcement Thayerianism
“Representation Reinforcement Thayerianism” differs from Unconstrained
Thayerianism because it incorporates a representation-reinforcement principle. The most
prominent advocate of this theory is John Hart Ely,245 and his articulation of the theory
has a strong connection with Footnote Four of the Caroline Products decision.246
Representation Reinforcement Thayerianism adopts an extratextual view about the role of
the judiciary in protecting the democratic process and discrete and insular minorities as
the basis for a set of exceptions to a general rule of deference to elected officials. To
some degree these exceptions may be supported by the text: freedom of speech and press
as well as Section One of the Fourteenth Amendment might support some aspects of the
representation-reinforcement principle. But it is not the original meaning of these
provisions that governs, rather the representation reinforcement principle determines the
legal content of constitutional doctrines that are associated with these provisions. For this
reason, Representation Reinforcement Thayerianism rejects the Constraint Principle.
We have already discussed most of the issues raised by Representation Reinforcement
Thayerianism. The chief advantages of this approach are captured by the arguments from
majoritarianism and the illegitimacy of exclusion; those arguments are discussed above.
Representation Reinforcement Thayerianism is subject to rule of law problems, especially
once we recognize the subjectivity of judgments about the preconditions for adequate
representation.247 As with each of the important rivals of originalism, a full discussion of
these issues would require an extended treatment.
d) Pairwise Comparison with Originalism and the Constraint Principle
The strength of Thayerianism is democratic legitimacy. Moreover, this theory scores
well on publicity and transparency as well—so long as the legislature governs through
general laws that are made public and applied in accord with their public meaning.
Thayerianism also avoids the problem of judicial tyranny, and rule by generally
applicable laws is not rule by decree and hence is not tyranny in any form. Thayerianism
just is politicization, but so long as the politics are democratic, this may be a virtue and
not a vice.
245
JOHN HART ELY, DEMOCRACY AND DISTRUST (1980).
United States v. Carolene Products Co., 304 U.S. 144, 152 n. 4 (1938).
247
See supra Part VI.E, p. 96 & Part VI.F, p. 97.
246
120
The Constraint Principle
This suggests that the pairwise comparison of Thayerianism with originalism raises
issues that are quite different than those discussed thus far. There is a reason for that.
The case for originalism advanced in this Article focuses on the issues that are relevant to
a comparison of the jurisprudential status quo (in some sense) with originalism. The
status quo is judicial supremacy and Congressional supremacy. For this reason, this
particular pairwise comparison requires consideration of a set of issues that are mostly
outside the scope of this Article—a separate and extended treatment is required.
8. Antitheories: Particularism, Pragmatism, Eclecticism, and Opportunism
The first seven forms of nonoriginalism are all theories of constitutional interpretation
and construction. As such, they share the common assumption that constitutional theory
can offer a method that should guide constitutional practice. There is another family of
views about constitutional theory that rejects this assumption. Let us call the family of
views that rejects this assumption “Constitutional Antitheory.” Constitutional Antitheory
is related to antitheoretical views in other branches of moral and political philosophy.
Notably, there are a variety of antitheoretical views in normative ethics.248 One way of
thinking about antitheoretical positions focuses on the notion that there cannot be a
decision procedure for ethics. Hence, antitheory stands in opposition to utilitarianism
(act so that your action maximizes utility) or certain interpretations of Kant (act so that
the maxim of your action could be willed as a universal law of nature). “Moral
particularism” is one form of antitheoretical thinking in ethics; as the name implies, this
view emphasizes the priority of the particular,249 but there are other forms of
constitutional antitheory.
a) Forms of Constitutional Antitheory
Let us use the phrases “constitutional particularism,” “constitutional pragmatism,”
“constitutional eclecticism,” and “constitutional opportunism” to name four
antitheoretical approaches to constitutional practice. These four antitheories have
something in common: they reject the notion that constitutional theory can provide a
systematic method (or decision procedure) for constitutional practice. But each of the
four antitheories has distinctive characteristics.
(1) Constitutional Particularism
Constitutional particularism is the view that constitutional practice should be guided
by the fact-sensitive assessment by constitutional actors of particular cases (or decisions).
Paradigmatically, particularism lauds the assessment of constitutional by trial judges who
decide the constitutional cases with the benefit of a rich factual context involving live
248
See ANTI-THEORY IN ETHICS AND MORAL CONSERVATISM (Stanley G. Clarke & Evan Simpson eds.
1989); SOPHIE GRACE CHAPPELL, INTUITION, THEORY, AND ANTI-THEORY IN ETHICS (2015); NICK FOTION,
THEORY VS. ANTI-THEORY IN ETHICS: A MISCONCEIVED CONFLICT (2014).
249
For an introduction, see Jonathan Dancy, Moral Particularism, Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/entries/moral-particularism/ (2013); see also JONATHAN DANCY, ETHICS WITHOUT
PRINCIPLEs (2004); Brad Hooker, Moral Particularism—Wrong and Bad in MORAL PARTICULARISM
(Hooker and Little eds. 2000).
121
Lawrence B. Solum
testimony experienced in real time. Appellate judges are at a disadvantage, because of a
paper record, but they should aim for a richly textured appreciation of the particular case.
Outside the courts, other constitutional actors (judges) are themselves participants in
context of constitutional decisionmaking. But in all of these situations, the key to good
decisionmaking is fact sensitivity. Good constitutional doctrine emerges from perception
of the particular case. Constitutional particularism rejects the Constraint Principle as a
binding principle, because judgments about particular cases always trump general
principles. Of course, in particular cases, a particularist judge might follow the text, but
the judge’s reasons would be tied to the particular case and not a general rule.
(2) Constitutional Pragmatism
Constitutional pragmatism embraces fact-sensitivity, but it does not embrace the
priority of the particular as a general approach to constitutional practice. As articulated
by Richard Posner, pragmatism adopts “reasonableness” as its guiding principle.250
Posnerian constitutional pragmatism is forwarding looking, but it is not purely
consequentialist:
[D]espite the emphasis on consequences, legal pragmatism is not a form of
consequentialism, the set of philosophical doctrines (most prominently
utilitarianism) that evaluates actions by the value of their consequences: the best
action is the one with the best consequences. There are bound to be formalist
pockets in a pragmatic system of adjudication, notably decision by rules rather than
by standards. Moreover, for both practical and jurisdictional reasons the judge is
not required or even permitted to take account of all the possible consequences of
his decisions.251
Despite his formidable intellect, Posner is not adept at the precise articulation of his own
theoretical framework. Let us stipulate that constitutional pragmatism is the view that
constitutional practice should be guided by the principle of reasonableness with a
forward-looking emphasis on legally salient consequences. Judges should choose
between rules, standards, and other forms of legal norms252 by determining which form of
norm is most reasonable in the circumstances. Constitutional pragmatism rejects the
Constraint Principle as a general rule, although a pragmatic judge may follow the text for
pragmatic reasons on a case-by-case basis.
There may be forms of pragmatism that avoid the rule of law and legitimacy problems
that clearly attend Judge Posner’s version. Posnerian pragmatism suffers from the all of
the rule of law problems identified above.253 Indeed, the point of this form of
pragmatism is to free judges from the constraints of the rule of law and enable them to
make all-things-considered judgments when deciding particular cases. Likewise, this
250
Richard A. Posner, Legal Pragmatism Defended, 71 U. CHI. L. REV. 683 (2004) (“The ultimate
criterion of pragmatic adjudication is reasonableness.”).
251
Id. at 684.
252
See Lawrence B. Solum, Legal Theory Lexicon 026: Rules, Standards, Principles, Catalogs, and
Discretion, http://lsolum.typepad.com/legal_theory_lexicon/2004/03/legal_theory_le_3.html (last modified
on January 4, 2015).
253
See supra, Part IV.A, p. 55.
122
The Constraint Principle
form of pragmatism raises all of the legitimacy problems raised in the earlier
discussion.254 Pragmatic decisionmaking lacks democratic legitimacy, because a
pragmatic judge is not bound by democratic decisions. Likewise, the all-thingsconsidered approach is clearly inconsistent with a judicial role that differentiates
constitution making from constrained constitutional adjudication. In Judge Posner’s case,
the pragmatism may have the virtue of transparency; Judge Posner has been unusually
frank about his approach to judging and his disdain for the constitutional text, but even
Posner may have felt constrained to disavow his rejection the Constraint Principle: as the
Wall Street Journal reported, “The iconoclastic federal appeals judge from Chicago on
Friday walked back his assertion in a Slate column that the nation’s founding document
has no use for judges today.”255
Of course, a modified version of pragmatism might acknowledge that the
constitutional text should play some role but deny that there are any metaprinciples (other
than pragmatism itself) that define what that role should be. This version of pragmatism
does a better job than Posner’s in accounting for the appearances, but it is not clear that it
provides any basis for intersubjective agreement on what counts as a correct
constitutional decision.
(3) Constitutional Eclecticism
Constitutional Eclecticism is the view that constitutional practice should be guided by
an eclectic mix of methods. In the judicial sphere, eclecticism has both interjudicial and
intrajudicial dimensions. Across different judges, eclecticism embraces methodological
diversity: thus, it affirms the goodness of some judges being originalists, others embrace
common law constitutionalism, and others who are constitutional pragmatists. Even a
single judge might be eclectic, rendering originalist decisions on Monday and strictly
adhering to precedent on Tuesday. Eclecticism differs from pragmatism, because it does
not offer a metaprinciple (like reasonableness) that would provide guidance as to which
methods should be used by different judges or at different times. But eclecticism does
not need to embrace the absurd notion that “anything goes.” Eclectics can and should
impose a filtering principle that sorts constitutional methods into the group of reasonable
theories that are eligible for deployment in constitutional practice and unreasonable
theories that should play no role. Eclecticism rejects the Constraint Principle as a general
rule, but an eclectic judge might deploy the idea of constraint in some cases, while
ignoring it in others.
Constitutional eclecticism might be held as an explicit theoretical stance and selfconsciously articulated as such: call this variant “Sophisticated Constitutional
Eclecticism.” The sophisticated version of eclecticism explicitly denies the Constraint
Principle, because it embraces the option of ad hoc reliance on nonoriginalist
constitutional theories. But eclecticism need not be consciously embraced; one can at
least imagine a judge who is disposed to act in a way that is descriptively captured by
eclecticism but who does so intuitively (without explicit awareness): call this variant
“Naïve Constitutional Eclecticism.” The naïve version of eclecticism might be combined
254
See supra, Part IV.B, p. 74.
Jacob Gershman, Posner Apologizes for Saying the Constitution Is Useless, July 1, 2016,
http://blogs.wsj.com/law/2016/07/01/posner-says-he-didnt-mean-to-suggest-the-constitution-is-useless/.
255
123
Lawrence B. Solum
with constitutional self-deception: naïvely eclectic judges might believe that they do
affirm the Constraint Principle, when in fact they are disposed to act in ways that
systematically inconsistent with constraint.
(4) Constitutional Opportunism
Constitutional Opportunism is the view that constitutional practitioners can and should
pick the theory that leads to the best consequences (or the ideologically preferred
outcome) on any particular occasion. Today, I might affirm originalism because it
enhances the right of plaintiffs to a jury trial, but tomorrow I might argue for
Thayerianism because it would support deference to Congress on the Affordable Care
Act. Or today, I affirm originalism because it leads to a strong individual right to own
and carry a gun, but tomorrow I might argue for Thayerianism because it supports
overruling Roe v. Wade.256 The argument that Constitutional Opportunism does not
respect the Constraint Principle is so obvious that it need not be stated.
Related to the notion of constitutional opportunism at any given moment is the idea
that the time-frame for theoretical opportunism is fixed by the current composition of the
Supreme Court. Thus, when progressives command a majority on the Supreme Court,
progressive constitutional theorists will be attracted to constitutional theories that enable
ideology to play a role in judging: multiple-modalities, moral readings, and common-law
constitutionalism are examples. But when conservatives hold the court, progressive
constitutional theorists will be attracted to theories of judicial restraint and constraint. It
should go without saying that the same considerations apply to conservative
constitutional theorists: they may be attracted to restraint and constraint when
progressives dominate the court and to views that enable a conservative court to change
the legal content of constitutional doctrine and to invalidate executive and legislative
action. This can give rise to the interesting phenomenon of theoretical flip-flopping—
where the very same theorist changes views shortly after there is a dramatic shift in the
composition of the court that effects the position of the median justice in ideological
space.257
Constitutional Opportunism is not identical to a best-outcomes theory, because it
requires the judge to follow some constitutional theory on each occasion and some
outcomes may not be justifiable by any theory. Opportunism rejects the idea that
constitutional decisionmaking must be principled, but it accepts some weaker idea—for
example, the idea that that the positive legitimacy of judicial decisionmaking in
constitutional cases requires that some principled explanation for constitutional decisions
be offered (or available).
In sum, all four forms of antitheory reject originalism—which is a normative theory
that provides a single method of constitutional interpretation. Antitheorists can endorse
originalists results reached by particularist, pragmatic, eclectic, or opportunistic methods.
And antitheorists can even endorse reasoning that includes originalist ideas. But no
256
410 U.S. 113 (1973).
See Jack Balkin, The Constitution in 2020, er.... 2016, Bakinization,
http://balkin.blogspot.com/2016/02/the-constitution-in-2020-er-2016.html (February 19, 2016); Eric
Posner, The Coming Flip-Flop in Constitutional Theory, EricPosner.com, http://ericposner.com/thecoming-flip-flop-in-constitutional-theory/ (February 19, 2016).
257
124
The Constraint Principle
antitheorists can endorse originalism (and hence the Constraint Principle) as a general
method for constitutional practice to be applied by all judges in all cases.
b) Pairwise Comparison of Constitutional Antitheory with Originalism and the
Constraint Principle
Pairwise comparison of originalism with constraint and the four forms of
constitutional antitheory suggests favor originalism with respect to both the rule of law
and legitimacy. Each of the antitheoretical positions permits judicial tyranny;
constitutional opportunism embraces rule by decree with relish. Each of the antitheories
undermines the rule of law values of publicity, stability, and certainty. Each of the
antitheories encourages the politicization of constitutional law. And the same is true of
legitimacy. Antitheory authorizes juristocracy and undermines democratic legitimacy.
Some forms of antitheory can be transparent: particularism and pragmatism could be
practiced in a transparent way. Eclecticism seems less transparent, because it offers no
explanation for the choice of one approach over another other than a results orientation
that is unlikely to be disclosed. Opportunism is the opposite of transparency. None of
the antitheories respects limits on the judicial role.
9. Rejectionism: Anticonstitutionalism & Constitutional Replacement
Finally, we have two views that reject a premise shared by the other forms of
nonoriginalism. Most constitutional theorists accept the normative premise that the
communicative content of the constitutional text should make a significant contribution
to the legal content of constitutional doctrine.258 We can use “rejectionism” as an
umbrella term for theories that reject the idea that the communicative content of the
constitutional text should make a significant contribution to the legal content of
constitutional doctrine.
a) Forms of Rejectionism
We will consider two forms of rejectionism: anticonstitutionalism and constitutional
replacement.
(1) Anticonstitutionalism
Anticonstitutionalism is the view that constitutional texts should play no substantial
role in constitutional practice. It is rare to see this view articulated explicitly, but Louis
Michael Seidman has articulated a view like this in his monograph Constitutional
Disobedience.259 Strict anticonstitutionalism rejects the idea of a constitution altogether,
and thus would reject both written and unwritten constitutions—irrespective of their
content. Anticonstitutionalism explicitly rejects the Constraint Principle.
258
The idea of contribution is defined more precisely below in the form of the “Contribution Principle.”
See infra Part II.A, p. 1919.
259
LOUIS MICHAEL SEIDMAN, CONSTITUTIONAL DISOBEDIENCE (2013).
125
Lawrence B. Solum
(2) Constitutional Replacement
Constitutional replacement theories reject the current Constitution of the United
States, but would accept some form of contribution or constraint for a better constitution.
Again, it is rare to see this view articulated explicitly, but my formulation of this
possibility was inspired by Sandy Levinson’s book, Our Undemocratic Constitution:
Where the Constitution Goes Wrong (And How We the People Can Correct It).260
Levinson argues for constitutional replacement, but his view could be extended to include
an explicit rejection of the current constitution pending the adoption of a new one. For
the purposes of this article, we will use the phrase “constitutional replacement” to
designate theories that maintain that contribution and constraint, while appropriate for a
morally acceptable constitution, should be rejected for the current Constitution of the
United States. For example, a constitutional replacement theorist might affirm the
Constraint Principle if the United States Constitution were amended to be more like the
current South African Constitution, but reject the Constraint Principle given the
communicative content of the actual constitutional text.
It goes without saying that both anticonstitutionalism and constitutional replacement
theories reject the United States Constitution in general and the Constraint Principle in
particular.
b) Pairwise Comparison of Constitutional Rejectionism with Originalism and the
Constraint Principle
Finally, we come to rejectionism in both its anticonstitutionalism and constitutional
replacement forms. Both views are radical alternatives to originalism.
Anticonstitutionalism in one form is simply a rejection of the rule of law; that form of
constitutional antitheory raises issues that are far beyond the scope of this paper and far
outside the mainstream of contemporary constitutional thought. This is not to say that
this view does not deserve engagement; rather, it is to say that the terms of engagement
require examination of a set of ideas that are different in kind than those that provide the
organizing focus on this Article.
Constitutional replacement theory certainly rejects the idea that the original meaning
of the current constitutional text should govern constitutional practice, but it is not clear
that replacement theorists reject originalism at a higher level of generality. A new
constitution would require interpretation and construction. Each constitutional
replacement theories must adopt a stance on the question whether the new provisions of
the replacement constitution should constrain judges and other officials. Again, this is a
topic for another day. To the extent that replacement theorists believe that the current
constitution is nonbinding, and therefore that the Supreme Court can decide cases without
reference to the constitutional text, their view converges with the superlegislature theory.
But they might hold some other position, e.g., that the Supreme Court should practice
Thayerian deference until an adequate constitution comes into effect.
260
SANFORD LEVINSON, OUR UNDEMOCRATIC CONSTITUTION: WHERE THE CONSTITUTION GOES
WRONG (AND HOW WE THE PEOPLE CAN CORRECT IT) (2008).
126
The Constraint Principle
***
The complexity of debates about originalism and constraint is in part a function of the
fact that the conceptual space of contemporary constitutional theory is multidimensional.
Many positions in this conceptual space are occupied: witness the nine families of
nonoriginalist theories that reject the Constraint Principle! And there are many possible
variations within each of the nine families: common law constitutionalism, for example,
could have many variations, corresponding to different theories of the common law
combined with various views of constitutional stare decisis.
This complexity gives rise to a practical problem for
constitutional theory. There are too many variations for
each to be given full consideration when advancing the case
for any one theory. The attempt to do so gives rise to what
we might call “Constitutional Theory Whack-a-Mole.”
Every time a particular theory is shown to have a
substantial problem, another theory arises to take its place.
The best that can be done under these circumstances is to identify those families of
theories that seem most salient—either because they have well-developed representatives
or because they occupy important nodes in conceptual space or both. Thus, Dworkin’s
theory is well developed and influential. The “superlegislature” approach does not have
a well-developed representative, but it occupies an important position in the conceptual
space of constitutional theory. Theories can be organized into families around leading
representatives: thus, David Strauss’s theory can serve as a representative of common
law constitutionalism. This seems to be the only feasible way of limiting the list of
alternatives to originalism to a manageable number.
Focusing on families of theories is the only feasible course of action, but it has a
serious disadvantage. Each advocate of a particular theory has a reasonable complaint
that their particular version of nonoriginalism has not been discussed in detail. To the
extent that a particular theory is fully articulated and well defended, the it should be
considered in due course. Nonetheless, it seems reasonable to begin with the theories
that are widely recognized as strong and then proceed to less prominent views.
***
B. The Constitutional Status Quo
At this stage of the argument, some nonoriginalists may object to the method of
pairwise comparison of originalism to its rivals in theoretical theory. Consider the
following statement of the objection:
Originalists are arguing for a change to the constitutional status quo. Pairwise
comparison of originalism to nonoriginalists theories is really beside the point. The
status quo is not theoretical; it is antitheoretical. The relevant comparison should
begin with a recognition that the status quo is Constitutional Eclecticism. Making
out the case for originalism requires a demonstration that adoption of originalism
would be better than what we have now. Once the issue is framed this way, we can
recognize that a move to originalism would be difficult and costly. Pro tanto
127
Lawrence B. Solum
justifications for originalism are all well and good, but they are not sufficient to
justify the many costs that would be imposed by originalism in practice.
Notice that this objection is rooted in a particular conception of constitutional time. The
relevant framework for evaluation of originalism is rooted in the constitutional present,
and the questions that should be asked are about the constitutional future. This
framework entails that constitutional thought experiments about a hypothetical
constitutional past are mostly irrelevant to the question at hand: what should we do now
and in the foreseeable future?
This question is entirely fair, and advocates of the Constraint Principle must provide
an answer. Notice that systematic comparisons between unconstrained eclecticism and
the Constraint Principle are inherently difficulty, because eclecticism can lead to any
approach in any particular case—including originalism (albeit originalism by a particular
constitutional actor or set of actors on a particular occasion). This makes is difficult for a
true constitutional eclectic to object to originalism in general.
This point can be brought out in greater detail by explicit considerations of the
implications of constitutional possibility and constitutional time for comparison for
future-to-future comparisons of unconstrained eclecticism with constrained originalism.
Given the malleability of eclecticism, almost any constitutional future we can imagine is
consistent with eclecticism. As a practical matter, it may well be the case that the effects
of continue eclecticism would depend on who is appointed to the federal bench. And it
seems unlikely that the future of American politics can be predicted with any certainty
once the time frame extends out for more than several months (or a few years at best).
Rule of law arguments apply with special force when comparing originalism with
eclecticism. An eclectic Supreme Court truly does rule by decree creating an especially
vicious version of the judicial tyranny problem. Eclecticism runs into problems with
each of the rule of law values. Indeed, eclecticism introduces a meta-level rule of law
problem that many other versions of nonoriginalism avoids. Eclecticism makes the law
uncertain at the meta-level: judges pick and choose their method of constitutional
construction, making the first-order content of law even more difficult to ascertain in
advance. Eclecticism introduces a particularly vicious form of legal instability at the
meta-level: eclecticism provides no guarantee of continuity in the methods of
constitutional construction, opening a space for radical shifts in constitutional law in
response to whatever motivations cause meta-level shifts in practice at the intra-judicial
and inter-judicial levels. If eclecticism has very serious problems with predictability and
certainty, it is an utter failure when it comes to publicity: indeed, the whole point of
eclecticism is that judicial methods of construction should not be made publicly known at
the meta-level. Finally, when it comes to the risk of politicization, eclecticism leaves the
barn door wide open: eclectic judges can reach almost any result that politics or ideology
would dictate.
Eclecticism fares badly with respect to legitimacy as well. Some forms of
nonoriginalism fare well with respect to democratic legitimacy—the clearest example is
the Thayerian family of nonoriginalist theories. Eclecticism embraces democratically
legitimate forms of nonoriginalism on some occasion, but rejects them on others. The
option to act in a democratically illegitimate fashion is part of what makes eclecticism
truly eclectic. Eclecticism is nontransparent for reasons that have already been discussed
with respect to the rule-of-law value of publicity: the ultimate basis for decision by an
128
The Constraint Principle
eclectic judge is not “out in the open.” But because the judge affirms some set of
principles as the basis for decision in the particular case, the decision has a false
appearance to transparency. The surface level reasons conceal an underlying basis for
choice that is rarely disclosed. With respect to judicial role, eclectic judges sometimes
are “in role” and sometimes “out of role.” When they adopt the constraint principle, as
they may from time to time, their behavior is legitimate. But when the override the text,
as they will on some occasions, their decisions are illegitimate given the appropriate
conception of the proper role of judges.
At this stage, some critics of originalism may complain that eclecticism is not the
status quo, arguing that their preferred version of nonoriginalism is actually the status
quo. This argument has been advanced by advocates of the multiple-modalities view—
particularly by Stephen Griffin in his important essay, Pluralism in Constitutional
Interpretation.261 It is certainly true that judges use multiple modalities of constitutional
argument: no one should dispute that. What is less clear is that they employ the
theoretical framework of the multiple-modalities approach. It is difficult to prove a
negative. Judge Reavely of the Fifth Circuit made explicit reference to the multiplemodalities approach in a 1992 case.262 There is one other occasion upon which a judge
has explicitly relied on the theory. Judge Reinhardt made the following statement in a
denial of a petition for rehearing en banc:
A necessary implication of the constitutional prohibition on congressional
tinkering with the manner of judicial decisionmaking is that Congress may not
eliminate wholesale any tool of judicial reasoning from the judicial toolbox. Courts
in common law systems have developed a panoply of tools to guide them in the
interpretive process, among them inferring rules from text or structure, reasoning
from analogy, and applying rules from precedent. Cf. Philip Bobbitt, Constitutional
Interpretation 12-22 (1991). No one tool takes precedence in any situation, and in
any given case several of the tools may work in tension with one another.263
This passage has never been cited by another judge, and it is the only passage of which I
am aware in which the theoretical structure of the multiple-modalities has been explicitly
endorsed. This one statement does constitute some evidence that the multiple-modalities
view is accepted by the Court, but given the vagaries of opinion writing and the relatively
free hand that some law clerks are given in drafting opinions, this evidence is surely not
decisive. Similar points could be made about common-law constitutionalism—the only
other theory that has a credible claim to being the “status quo.”264
261
See Griffin, supra note 211.
Jones v. Clear Creek Indep. Sch. Dist., 977 F.2d 963, 965 (5th Cir. 1992) (“Of the six forms of
argument recognized in constitutional interpretation, it is the doctrinal arguments that control Establishment
Clause cases.”).
262
263
Crater v. Galaza, 508 F.3d 1261, 1265 (9th Cir. 2007) (Rheinhardt, J., dissenting from denial of
petition for rehearing en banc).
264
So far as I can tell, no court has ever explicitly endorsed common-law constitutionalism.
129
Lawrence B. Solum
***
I read a lot of constitutional cases. If you are reading this article, I bet you do too. My
impression is that eclecticism best describes the constitutional status quo. And I would
guess that some readers will agree. Other readers will not. Some may think that the
multiple-modalities theory is the status quo. If David Strauss reads this article, I am
fairly sure he will think that common-law constitutionalism is the status quo! And some
may think that representation-reinforcement Thayerianism or Dworkin’s theory or the
multiple-meanings view is roughly the status quo—although the emphasis would have to
be on the “roughly.” But they can’t all be right. This disagreement about what
constitutes the status quo is encompassed by eclecticism—it is exactly the situation that
eclecticism would predict. None of the other theories has an easy time explaining the
disagreement among theorists about the nature of the status quo. Eclecticism fits the
primary evidence—what we know about the way opinions are written—and it fits the
secondary evidence—the pattern of disagreement about how to interpret the primary
evidence. No other theory has this virtue.
***
C. Constitutional Compromise
Consider another set of alternatives to the Constraint Principle, which I shall call
“constitutional compromises.” The basic idea of a compromise between originalism and
nonoriginalism is to divide constitutional issues into two sets—one set to be governed by
the Constraint Principle and the other set to be governed by some other principle or
theory. I do not know of any explicit articulation of this kind of theory, but it is “in the
air” so to speak. This idea is best explained by giving examples, three of which are set
out in the next subsection.
1. Three Proposals for Constitutional Compromise
There could be many forms of Compromise with respect to the Constraint Principle. It
simply isn’t practicable to consider them all, but the following three seem to me to be
natural expressions of the idea that are particularly salient given the landscape of
contemporary constitutional theory.
a) Originalism Plus Grandfathered Canonical Cases
Here is one possible compromise. We might accept originalism as the theory that
should guide constitutional practice in general, but make a stipulated exception for the
canonical cases like Brown v. Board of Education. Something like this view was
accepted by Justice Scalia and played a role in his so-called “faint hearted
originalism.”265 One can imagine an originalist like Scalia who advocates a compromise
version of originalism that accepts the Constraint Principle, except for a set of canonical
cases like Brown v. Board of Education.
265
Randy Barnett, Scalia's Infidelity: A Critique of “Faint-Hearted” Originalism, 75 U. CIN. L. REV. 7
(2006).
130
The Constraint Principle
It is conceivable that the compromise might limit the set of canonical cases to a list,
but a more principled form of this position would specify criteria for canonicity. For
example, a canonical case might be one with the following criteria: (1) deep and wide
acceptance by officials and the public at large, (2) the decision is foundational for a
substantial set of legal practice such that reversal of the decision would require wholesale
revision of precedent, statutes, or other legal norms, (3) reversal of the decision would
involve serious moral wrong.
b) Originalism for the Hard-Wired Constitution, Common Law Constitutionalism
for Constitutional Software
Another version of compromise might limit originalism to the so-called “hard-wired
constitution”—those portions of the constitution that specify basic institutional
arrangements, such as bicameralism, the rules for representation in the House and the
Senate, the veto power, and so forth. The original meaning of the hard-wired constitution
would be subject to the Constraint Principle, but common-law constitutionalism would
provide the regime for “constitutional software”—those provisions of the Constitution
that do not provide bright-line unambiguous rules.
c) Short-Term Originalism, Long-Term Common Law Constitutionalism
Another compromise position is represented by David Strauss’s idea that
constitutional amendments should be enforced for some period of time after they are
adopted, but then be supplanted by the common-law method. Strauss is more or less
forced to adopt this position by the reductio of his general theory: it leads to the
conclusion that a Supreme Court decision cannot be overruled by a clear and explicit
constitutional amendment. To that extent, this compromise theory seems more like a
legitimating fig leaf than a serious position. All of the objections to common law
constitutionalism apply to the compromise position, but it the fig leaf is either
disingenuous or an admission that many of the arguments against originalism are
outweighed in the short run. In either case, the position seems unstable.
d) Originalism and Thayerianism
A final compromise solution might combine the Constraint Principle for cases where
the communicative content of the constitutional text is clear (not ambiguous, not vague,
and so forth) with a principle of Thayerian deference for cases where the communicative
content of the text was underdeterminate. In practice, this would mean that the hardwired structural constitution would be governed by originalism, but the vague and opentextured provisions would not be judicially enforceable outside a minimal core of
determinate meaning.
Another version of a Thayerian compromise might involve Representation
Reinforcement Thayerianism. Footnote Four of Carolene products included the rights
enumerated in the Bill of Rights in addition to the protection of democratic procedures of
discreet and insular minorities in its set of exceptions to a general rule of deference
131
Lawrence B. Solum
(implemented by rational basis review).266 One might take an originalist approach to the
first eight amendments, but use Representation Reinforcement Thayerianism for other
constitutional issues. Variations on this approach might extend the domain of originalism
to other constitutional issues, including the hard-wired provisions.
2. The Case Against Constitutional Compromise
The case for the Constraint Principle and hence against the nonoriginalist component
of particular constitutional compromises has already been explicated, but there are some
additional issues raised by compromise positions.
One possible problem the compromise is instability. Constitutional compromise
between originalism and nonoriginalist form of living constitutionalism puts
constitutional actors in the position of employing two incompatible sets of ideas about
constitutional interpretation and construction. This incompatibility results from the fact
that the normative grounds for each half of the compromise: the originalist half is
supported by rule of law and legitimacy concerns, the nonoriginalist half by rejection of
these concerns and by other values. It seems reasonable to believe that such conceptual
incoherence might lead constitutional actors to drift away from comprise and towards a
more consistent position. Drift might occur in either direction, moving the system from
compromise towards originalism and more consistent application of the constraint
principle—or moving in the opposite direction, towards some form of nonoriginalist
living constitutionalism.
3. Constitutional Compromise as an Originalist Second Best
Despite the problems with constitutional compromise, originalists might come to think
of a compromise position as an originalist second best. For example, it might be the case
that the canonical cases are politically entrenched such that their reversal is outside the
feasible choice set. Given the unavailability of the originalist first best—strict adherence
to the Constraint Principle for all constitutional issues, compromise might well represent
the second best alternative.
D. Completing the Project of Pairwise Comparison
The project of pairwise comparison is a large one. Each pairwise comparison would
seem to require a sustained effort—a full length law review article or monograph. But
completing this projects would require many years—decades at the pace at which I work.
This suggests that completing of the project of pairwise comparison will require a
division of labor, with different scholars undertaking different parts of the project at
different times.
VIII. A RESTATEMENT OF THE CASE FOR THE CONSTRAINT PRINCIPLE
At this point, we can summarize the case for the Constraint Principle and against the
rivals of originalism. That summary can begin with the core of the case for constraint.
266
United States v. Carolene Products Co., 304 U.S. 144, 152 n. 4 (1938).
132
The Constraint Principle
Second, another set of nonoriginalist theories are constructed so that they are more
democratic than originalism. Popular Constitutionalism and Thayerianism belong in this
group. Moreover, some of these theories do not involve rule by judicial decree and hence
escape the judicial tyranny argument: government by legislative supremacy, the
institutional form of Unconstrained Thayerianism, fits in this category. These theories
have the property that they are substantial departures from the status quo—more radical
than originalism and subject to many of the same objections that are levied at originalism.
They raise an entirely different set of issues than does the first group, and a deep
consideration of these alternatives requires an inquiry into the question whether the
project of American constitutionalism requires radical revision.
Third, a final set of nonoriginalist theories involves an even more radical departure
from the status quo. Antitheoretical views such as eclecticism and opportunism reject the
idea that constitutional practice should answer to a consistent set of principles with
normative grounding. These views suffer from serious problems with respect to the rule
of law and legitimacy, but they are based on premises that call into question the very
ideas of democracy and legitimacy.
The relationship between the core justifications for the Constraint Principle and the
nine pairwise comparisons is summarized in the following table:
Table 2: Summary of Pairwise Comparisons for Pro Tanto Reasons for Constraint
Rule of Law Cluster
Judicial Tyranny
Rule of Law Values
Politicization
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Democratic Legitimacy
Transparency
Judicial Role
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Legitimacy Cluster
Meaning of the Colors
Red = Favors the Constraint Principle
Green = Favors Neither
Purple = Favors Alternative
Blue = Contestable
135
D R A F T February 15, 2017
THE CONSTRAINT PRINCIPLE:
ORIGINAL MEANING AND CONSTITUTIONAL PRACTICE*
LAWRENCE B. SOLUM**
TABLE OF CONTENTS
Introduction .................................................................................................................... 2
I. The Role of the Constraint Principle in Contemporary Constitutional Theory .......... 5
A. Originalism as a Family of Constitutional Theories ............................................. 5
B. Constraint and the Debate Over Originalism and Living Constitutionalism ........ 9
C. What Hangs on Acceptance or Rejection of the Constraint Principle? ............... 17
II. Contribution, Constraint, and Restraint ................................................................... 18
A. The Distinction Between Contribution and Constraint ....................................... 19
B. The Distinction Between Constraint and Restraint ............................................. 20
C. Three Versions of the Constraint Principle .......................................................... 22
D. Defeasibility Conditions and Mending Constructions ........................................ 26
III. Framing the Debate Over the Constraint Principle ................................................ 28
A. The Types of Justification ................................................................................... 28
B. The Methods of Justification ............................................................................... 32
C. The Structure of Argumentation .......................................................................... 41
D. The Role of Pairwise Comparison ...................................................................... 53
IV. Justifying the Constraint Principle ......................................................................... 54
A. Constraint and the Rule of Law .......................................................................... 55
B. Constraint and Legitimacy .................................................................................. 74
C. The Relationship Between the Rule of Law and Legitimacy .............................. 79
V. Alternative Justifications for the Constraint Principle? ........................................... 79
A. Is the Constraint Principle the Law? ................................................................... 80
B. Is the Constraint Principle Implied by Writtenness? ........................................... 81
C. Does the Constraint Principle Lead to Good Consequences? ............................. 81
D. Is the Constraint Principle the Only Alternative to Judicial Lawlessness? ......... 82
VI. Objections and Answers......................................................................................... 82
*
© 2017 by the author. Permission is hereby granted to reproduce this work in whole or in part
(including distribution via the Internet) for educational or scholarly purposes. The author requests that any
reproduction provide a citation to the work that includes the author’s name, the title, and the journal or
website from which the article was obtained.
**
Carmack Waterhouse Professor of Law, Georgetown University Law Center. I owe thanks to
participants in a Faculty Workshop at Georgetown Law and at the Originalism Works in Progress
Conference for helpful comments and suggestions. I am especially grateful to Richard Fallon Greg Klass,
Marty Lederman, and Michael Seidman.
Lawrence B. Solum
A. The Canonical Cases Objection .......................................................................... 83
B. The Disruptive Force of an Originalist Big Bang ............................................... 86
C. Three Versions of the Injustice Objection ........................................................... 87
D. The Systemic Bias Objection .............................................................................. 95
E. The Objection from Majoritarianism ................................................................... 96
F. The Objection from Exclusion and Illegitimacy .................................................. 97
G. The Dead Hand Problem ................................................................................... 100
H. Objections that Do Not Clash with the Constraint Principle ............................ 101
VII. Pairwise Comparisons: Rivals, the Status Quo, and Compromise ..................... 104
A. Pairwise Comparison of Originalism with Nine Rivals .................................... 105
B. The Constitutional Status Quo........................................................................... 127
C. Constitutional Compromise ............................................................................... 130
D. Completing the Project of Pairwise Comparison .............................................. 132
VIII. A Restatement of the Case for the Constraint Principle .................................... 132
A. Summing Up: The Core Case for Constraint .................................................... 133
B. Reflective Equilibrium: Considered Judgments and Outcomes ........................ 134
C. Pairwise Comparisons Revisited ....................................................................... 134
D. Does It Matter Whether We Call It Originalism? .............................................. 136
Conclusion ................................................................................................................. 136
INTRODUCTION
Originalism is a family of constitutional theories that agree in principle on the
proposition that the original meaning of the constitutional text should constrain
constitutional practice. Call this idea the “Constraint Principle.”1 Provisionally,2 the
Constraint Principle is the claim that (at a minimum) the content of constitutional
doctrine and the decision of constitutional cases should be consistent with the original
meaning of the constitutional text. The aim of this article is to explicate and justify the
Constraint Principle. Originalists also agree on a second idea, that the linguistic meaning
(or more precisely, “communicative content”) of the constitutional text is fixed when
each provision is framed and ratified. We can call this second idea the “Fixation Thesis;”
that thesis is defended elsewhere.3 Together, constraint and fixation form the core of
contemporary originalist theory.
Originalists agree on fixation and constraint, but they disagree about other things. The
dominant strain of contemporary originalism emphasizes the public meaning of the
constitutional text, but other versions of originalism focus on the original intentions of the
framers or the original methods of constitutional interpretation and construction. The
1
The Constraint Principle is capitalized to indicate that it is being used as a proper noun phrase that
names the principle of constraint formulated in this Article. A similar convention will be used other
elements of the theory on offer here and in related work.
2
For a more complete statement of the Constraint Principle that is explicated in this Article, see infra
Part II.C.2, p. 23 (articulating a minimalist version of the Constraint Principle).
3
Lawrence B. Solum, The Fixation Thesis: The Original Meaning of the Constitutional Text, 91 NOTRE
DAME L. REV. 1 (2015).
2
The Constraint Principle
core of agreement and the differences lead to two distinct modes of originalist theorizing,
which we can call “ecumenical” and “sectarian.”4 For the most part, this Article will
elucidate and defend the Constraint Principle from an ecumenical perspective,
emphasizing common ground and considering the implications of the variations among
originalists for the articulation and justification of constraint.5 At some points, the Article
will take up the sectarian perspective of “Public Meaning Originalism”—the view that
the original meaning of the constitutional text is it’s public meaning.
This Article has two principal aims. The first aim is to explicate the Constraint
Principle. That task has two parts. The initial part of the explication task is to identify
the range of possible views about the contribution that the constitutional text should make
to constitutional practice. At one extreme is the view that the text should play no role at
all—except as a symbolic representation of national unity similar to the role played by
the American Flag. Another position, held by many living constitutionalists is that
original meaning plays an important role in constitutional practice, but that role is not
constraining: the second position is exemplified by the pluralist approaches to
constitutional theory that conceive of constitutional law as a complex argumentative
practice constituted by multiple modalities of justification. At the other extreme is the
view that all of the content of constitutional doctrine must be derived from the
communicative content of the constitutional text.
The other part of the explication task is to work out a precise explication of what I call
“Constraint as Consistency, which I shall argue is the “Minimalist Version of the
Constraint Principle”—the form of constraint that can serve as the “least common
denominator”6 among originalist views and hence marks the divide between originalist
and nonoriginalist constitutional theories.
The second aim of this Article is to enter into debates about normative justifications
for (and against) the Constraint Principle. This second aim proceeds in two steps. The
first step involves questions of framing. What would constitute a good and sufficient
reason to affirm the constraint principle? What is the role of legal arguments from the
internal perspective of lawyers and judges? And what is the role of arguments of political
morality? Can constraint be justified in isolation or must it be compared to particular
forms of nonoriginalism? These questions and many others need to be explored before
we can proceed to the second step, examining the arguments for and against the
constraint principle, initially as a general matter and then in relation to the rival
constitutional theories that reject constraint.
There are three road maps—of these, the first is the conventional listing of the Parts.
Part I explores the role of the Constraint Principle in contemporary constitutional theory,
answering the question “what is originalism?” and laying out the most important forms of
4
Id. at 9.
My own sectarian view is a form of what we can call “Public Meaning Originalism.” I will defend the
core component of this view in “The Public Meaning Thesis,” a work in progress.
6
I am using this term in a colloquial and metaphorical sense. In common usage, “least common
denominator” is frequently used for what mathematicians would call the “greatest common divisor.” And
by way of analogy, the expression is then used to indicate what is common or shared by some group or set.
For this understanding of “common denominator,” see
http://www.finedictionary.com/common%20denominator.html (“common denominator an attribute that is
common to all members of a category”). By using the phrase “least common denominator” I mean to refer
to the least-restrictive principle that can serve as common ground among originalists.
5
3
Lawrence B. Solum
nonoriginalism and living constitutionalism. Part II excavates the conceptual foundations
of the idea of constraint and examines the various forms that constraint could take. Part
III frames the debate over the Constraint Principle by exploring the various forms that
justification could take and considering the role of the burden of persuasion in the debate.
Part IV sets two clusters of justifications for the Constraint Principle, the rule-of-law
cluster and the legitimacy cluster. Part V considers some alternative justifications for
constraint. Part VI considers various objections to the Constraint Principle. Part VII
reconsiders the rivals of originalism. Part VIII summarizes the case for constraint. The
Article ends with a Conclusion.
The second road map is for readers who are looking for the gist of the argument, I
suggest that they read the following parts and subparts:
(1) Part I.A, "Originalism as a Family of Constitutional Theories (pages 5-9),
(2) Part II.C.2, “A Minimalist Version of the Constraint Principle: Consistency
of Constitutional Doctrine,” (pages 23-25),
(3) Part IV, "Justifying the Constraint Principle,” (pages 54-79), and
(4) Part VII, “Pairwise Comparisons: Rivals, the Status Quo, and Compromise”
(104-132).
For an even shorter version, read Part VIII, “A Restatement of the Case for the Constraint
Principle” (pages 132-136).
The third road map situates this article in the larger project. The conceptual
foundations of the project are found in Communicative Content and Legal Content7 and
The Interpretation-Construction Distinction.8 The three key arguments for public
meaning originalism as a theory of constitutional interpretation and construction are
found in The Fixation Thesis,9 The Public Meaning Thesis10 and this Article. The topic
of constitutional construction is explored in Originalism and Constitutional
Construction11 and a work-in-progress tentatively titled The Construction Zone.12
Originalist methodology is briefly outlined in a short article, Originalist Methodology13
and will be the focus of future work, including a piece on the relationship of originalism
as a legal theory to the methods of academic historians, tentatively titled Originalism and
History.14 This Article (despite its length) is only one part of a large and complex
ongoing project.
We begin with a bird’s eye view of the idea of constitutional constraint in the context
of the landscape of contemporary constitutional theory and practice.
7
Lawrence B. Solum, Communicative Content and Legal Content, 89 NOTRE DAME L. REV. 480
(2013).
8
Lawrence B. Solum, The Interpretation-Construction Distinction, 27 Const. Comment. 95 (2011).
9
Solum, The Fixation Thesis, supra note 3
10
Lawrence B. Solum, The Public Meaning Thesis (unpublished manuscript on file with the author).
11
Lawrence B. Solum, Originalism and Constitutional Construction, 82 FORDHAM L. REV. 453 (2013).
12
This project is best described as halfway between the idea stage and a first draft.
13
Lawrence B. Solum, Originalist Methodology, U. CHI. L. REV. (forthcoming 2016-17).
14
Lawrence B. Solum, Originalism and History (unpublished manuscript on file with the author).
4
The Constraint Principle
I. THE ROLE OF THE CONSTRAINT PRINCIPLE IN CONTEMPORARY CONSTITUTIONAL
THEORY
Perhaps the most important debates in contemporary constitutional theory cluster
around the disputes between originalists and living constitutionalists. Sometimes the
disagreement sometimes seems to focus on the idea of fixation, but the most important
disagreements are about other issues. One important issue concerns determinacy, with
some nonoriginalists taking the position that meaning of the constitutional text cannot
constrain because it is indeterminate; other nonoriginalists admit that the text is only
moderately underdeterminate, but suggest that the most important issues of constitutional
law are within the zones of underdeterminacy created by open texture or vagueness.
Ultimately, questions about the degree to which the meaning underdetermines
constitutional practice are empirical or theoretical—and not normative.
The extent of constitutional underdeterminacy is an important question, but there is an
even more fundamental normative disagreement among contemporary constitutional
theorists.15 To the extent that the constitutional text is clear, originalists believe that it is
binding: we owe a duty of fidelity to the original meaning of the Constitution.
Nonoriginalists think otherwise. Although they may believe that the constitutional text is
worthy of respect and consideration, they reject the claim that the meaning of the text
provides hard limits on constitutional practice. Somewhat contentiously, we can say that
nonoriginalists believe that judges have the power to override the linguistic meaning of
the text. That is, nonoriginalists reject the Constraint Principle.
The function of this Part of the Article is to investigate the role of the Constraint
Principle in contemporary constitutional theory. That investigation can begin by
elucidating the nature of “originalism,” starting with the observation that originalism is a
family of constitutional theories.
A. Originalism as a Family of Constitutional Theories
What is originalism? In prior work, I have argued that originalism is a family of
constitutional theories organized around two core ideas, fixation and constraint.16 Almost
all originalists agree with both the idea that the meaning of constitutional text is fixed at
the time each provision is framed and ratified and the idea that this fixed original
15
For an informative discussion of the importance of indeterminacy in debates about originalism, see
Heidi Kitrosser, Interpretive Modesty, 104 GEO. L. J. (forthcoming 2016).
16
Lawrence B. Solum, What Is Originalism? in THE CHALLENGE OF ORIGINALISM: ESSAYS UN
CONSTITUTIONAL THEORY (Grant Huscroft and Bradley W. Miller eds., Cambridge University Press,
2011). The idea that originalism is a family of theories organized around the core ideas of fixation and
constraint is widely accepted. See, e.g., William Baude, Is Originalism Our Law?, 115 COLUM. L. REV.
2349, 2362 (2015) (stating “originalism is a family of theories united by principles of fixation and
constraint); New Originalism combines the “fixation thesis” with the “constraint principle”; Leslie C.
Griffin, Hobby Lobby: The Crafty Case That Threatens Women's Rights and Religious Freedom, 42
HASTINGS CONST. L.Q. 641, 655-56 (2015); Ian Bartrum, Two Dogmas of Originalism, 7 WASH. U.
JURISPRUDENCE REV. 157 (2015) (describing fixation and constraint as the two dogmas of originalism); see
also Solum, supra note 3, at 6 n. 38 (collecting sources that accept the notion that originalism is a family of
theories organized around the Fixation Thesis and the Constraint Principle or advancing similar ideas about
the nature of originalism).
5
Lawrence B. Solum
meaning should constrain constitutional practice. We can express these two ideas in a
preliminary way as follows:
The Fixation Thesis: The meaning (or more precisely communicative content) of the
constitutional text is fixed at the time each provision is framed and/or ratified.17
The Constraint Principle: Constitutional practice, including the elaboration of
constitutional doctrine and the decision of constitutional cases, should be constrained
by the original meaning of the constitutional text. At a minimum, constraint requires
that constitutional practice be consistent with original meaning (as specified below).
Originalists agree on fixation and constraint, but they disagree on other questions. The
most important area of disagreement concerns the nature of original meaning. Among the
important variations are the following:
Public Meaning: The original meaning is the public meaning of the constitutional
text.
Framers’ Intentions: The original meaning is provided by the framer’s intentions.
Intentionalism has further variants, including forms that focus on purposive
intentions and communicative intentions.
Ratifiers’ Understandings: The original meaning is given by the understandings of
the ratifiers of each provision.
Original Methods: The original meaning is the meaning that would be given to the
text by the original methods of constitutional interpretation and construction.
Any particular originalist theory will combine fixation and constraint with an
understanding of the nature of original meaning. For example, Public Meaning
Originalism combines the Fixation Thesis and Constraint Principle with the Public
Meaning Thesis18—the claim that the original meaning of the constitutional text is a
function of the conventional semantic meaning of the words and phrases in the public
available context of constitutional communication. Theories of the nature of original
meaning might be thought of as modules that can plug into the theoretical structure of
originalism as a whole.
Originalists also disagree about the extent of constitutional underdeterminacy and the
role of constitutional construction. Some originalists may believe that the original
meaning is “thick” (rich in communicative content) and hence that constitutional
underdeterminacy is rare or nonexistent. Other originalists may believe that the answers
to some constitutional questions are underdetermined by the constitutional text:
underdetermination can result from language that is vague, open-textured, or irreducibly
17
For a more precise explication of the Fixation Thesis, see Solum, supra note 3, at 15-178. For
qualifications and clarifications, see id. at 30-42. The version explicated in that article is: “The object of
constitutional interpretation is the communicative content of the constitutional text, and that content was
fixed when each provision was framed and/or ratified.” See id. at 15. This formulation is substantially the
same as that in text, which is formulated to be more accessible to readers unfamiliar with theoretical
framework articulated in the prior article.
18
See Lawrence B. Solum, The Public Meaning Thesis (unpublished work-in-progress, August 20,
2105) (articulating and defending the thesis that the original meaning of the constitutional text is its public
meaning).
6
The Constraint Principle
ambiguous, or from gaps or contradictions within the text. And yet another group of
originalists may believe that specific provisions of the text are unclear but that the text as
a whole is determinate because it implicitly contains default rules—for example, a default
rule of deference to elected officials where the text does not otherwise determine the
outcome.
Ecumenical originalism acknowledges the range of reasonable disagreement among
originalists about the extent of underdeterminacy. But from a sectarian perspective, each
originalist takes a stand on this question. For the purpose of this Article, my position on
this question is an undefended assumption: the constitutional text is neither fully
determinate nor radically indeterminate: many constitutional questions are answered by
the communicative content but others are not. Let us call the thesis that significant
constitutional underdeterminacy exists the “Fact of Constitutional Underdeterminacy.”19
Finally, some originalists do (and all originalists should) embrace a distinction
between “interpretation” and “construction.” The notion of a distinction between
interpretation and construction goes back at least as far as 1839 when it was articulated
(in a different form) by Franz Lieber in his Legal and Political Hermeneutics.20 The
modern version of the distinction appears (perhaps for the first time, perhaps not) in
twentieth-century treatises on contract law by Williston and especially Corbin;21 and has
been deployed in many judicial decisions.22 It is important to note that some originalists
reject the interpretation-construction distinction, while others may accept the distinction
as a matter of abstract theory but reject its significance on the ground that the original
meaning of the constitutional text is sufficiently thick to settle all or almost all
constitutional questions.23
For the purposes of this Article, I will use the words “interpretation” and
“construction” in stipulated technical senses, as follows:
Constitutional Interpretation: The phrase “constitutional interpretation” is stipulated
to refer to the activity of that discerns the communicative content (linguistic
meaning) of the constitutional text.
19
See Solum, , supra note 3, at 10-13.
FRANCIS LIEBER, LEGAL AND POLITICAL HERMENEUTICS 43–44, 111 n.2 (Roy M. Mersky & J. Myron
Jacobstein eds., Wm. S. Hein & Co. 1970) (1839). Lieber’s version of the distinction does not explicitly
differentiate communicative content and legal content. See Greg Klass, Interpretation and Construction 1:
Francis Lieber, New Private Law: Project on the Foundations of Private Law,
http://blogs.harvard.edu/nplblog/2015/11/19/interpretation-and-construction-1-francis-lieber-greg-klass/;
Greg Klass, Interpretation and Construction 2: Samuel Williston, New Private Law: Project on the
Foundations of Private Law, https://blogs.harvard.edu/nplblog/2015/11/23/interpretation-and-construction2-samuel-williston-greg-klass/; Greg Klass, Interpretation and construction 3: Arthur Linton Corbin,
http://blogs.harvard.edu/nplblog/2015/11/25/interpretation-and-construction-3-arthur-linton-corbin-gregklass/; see also Ralf Poscher, The Hermeneutical Character of Legal Construction, (November 28, 2015).
Simone Glanert and Fabien Girard (eds), Law’s Hermeneutics: Other Investigations (London: Routledge,
2016). Available at SSRN: http://ssrn.com/abstract=2696486.
21
4 Williston, Contracts §§ 600-02 (3d ed.1961); 3 Corbin, Contracts §§ 532-35 (1960 & Supp.1980).
22
See Fausel v. JRJ Enters., Inc., 603 N.W.2d 612, 618 (Iowa 1999); In re XTI Xonix Technologies
Inc., 156 B.R. 821, 829 n. 6 (D.Ore.1993); Berg v. Hudesman, 115 Wash.2d 657, 663, 801 P.2d 222, 226
(1990). More examples are collected in Solum, supra note 11, at 486-87.
23
See Solum, Originalism and Constitutional Construction, supra note 11 (discussing originalist
positions on the interpretation-construction distinction).
20
7
Lawrence B. Solum
Constitutional Construction: The phrase “constitutional construction” is stipulated
to refer to the activity that determines the content of constitutional doctrine and the
legal effect of the constitutional text (including the decision of constitutional cases
by the courts).24
Interpretation-Construction Distinction: The phrase “Interpretation-Construction
Distinction” will be used to designate the distinction as articulated via the stipulated
definitions immediately above.
The phrase “new originalism”25 has no definite content, but it will be used in this
Article to name a version of originalism that embraces the Fixation Thesis and the
Constraint Principle. Some new originalists emphasize public meaning; others are
focused on original methods or the original law. Given the Fact of Constitutional
Underdeterminacy and the Interpretation-Construction Distinction, many (but not all)
new originalists believe in the existence of what we can call “construction zones.” Such
zones encompass the cases and issues with respect to which the communicative content
of the constitutional text does not fully constrain constitutional practice.
Different versions of originalism might adopt different approaches to the construction
zone. One possibility is that decisions in the construction zone should be guided by a
principle of deference: for example, the courts could defer to Congress (or other political
institutions) where the meaning of the constitutional text was unclear.26 Or construction
might be guided by a “presumption of liberty.”27 There are many other possibilities,
including the use of moral readings, original purposes, or common law methods in the
construction zone. This issue is important, but will be set aside for the remainder of this
Article.
Finally, one recent strand of originalist theory emphasizes the idea that originalism can
be viewed as a theory of constitutional change. This has been implicit in the originalist
claim that the constitutional actors (especially judges) lack the power to adopt
constitutional constructions that are the function equivalent of constitutional
amendments.28 But there are both theoretical and empirical questions implicated by this
originalist observation. Stephen Sachs articulates the idea that originalism is a theory of
constitutional change as follows:
What originalism requires of legal change is that it be, well, legal; that it be
lawful, that it be done according to law .... The originalist claim is that each change
in our law since the Founding needs a justification framed in legal terms, and not
24
These definitions were presented in Solum, supra note 11, at 457 (2013).
See Evan S. Nadel, The Amended Federal Rule of Civil Procedure 11 on Appeal: Reconsidering
Cooter & Gell v. Hartmarx Corporation, 1996 ANN. SURV. AM. L. 665, 691 n. 191 (“An example of the
‘textualism’ to which I refer is the ‘new originalism’ theory often associated with Justice Scalia.”); Randy
E. Barnett, An Originalism for Nonoriginalists, 5 Loy. L. Rev. 611, 620 (1999); Keith Whittington, The
New Originalism, 2 GEO. J.L. & PUB. POL’Y 599 (2004).
26
See Solum, The Fixation Thesis, supra note 3, at 5-6; Solum, Originalism and Constitutional
Construction, supra note 11, 511–23 (discussing Paulsen and Lawson’s default rules approach).
27
See RANDY E. BARNETT, RESTORING THE LOST CONSTITUTION (2d ed. 2013).
28
See Randy E. Barnett, An Originalism for Nonoriginalists, 45 LOY. L. REV. 611, 635 (1999) (stating,
“a proper respect for the writtenness of the text means that those committed to this Constitution have no
choice but to respect the original meaning of its text until it is formally amended in writing.”).
25
8
The Constraint Principle
just social or political ones. To put it another way, originalists believe that the
American legal system hasn't yet departed (even a little bit) from the Founders' law
in the way that the colonies threw off the British yoke or the states got rid of the
Articles of Confederation.29
This position could entail a modified understanding of the constraint principle: if the
original methods of constitutional interpretation and construction can be used to authorize
still other methods and those methods authorize amending constitutional constructions,
then originalism could, in theory, be consistent with some violations of the constraint
principle. Whether it does so or not involves complex questions about originalist
theory—are original methods of construction part of the Constitution?—and history—did
the original methods authorize amending constructions as a matter of historical fact?
These questions will be set to the side for the remainder of this article.
B. Constraint and the Debate Over Originalism and Living Constitutionalism
What role does the Constraint Principle play in debates about originalism and living
constitutionalism? Before we answer that question, we need to recognize that in this area
of constitutional theory, the terminology itself is contested. Because terms like
“originalism” are contested, debates about originalism can give rise to confusion and
misunderstanding. Why is it the case that some of the most basic terminology in
constitutional theory is the subject ongoing contestation? And how should we proceed
given the unstable nature of the terminology? Answering these questions requires us to
step back from the terminological disputes and examine the reasons why the meaning of
the word “originalism” is contested.
1. Metalinguistic Negotiation: What Should Count as “Originalism,”
“Nonoriginalism,” and “Living Constitutionalism”?
Like the term “originalism,” the word “nonoriginalism” and the phrase “living
constitutionalism” do not have precise and universally accepted definitions. Both the
words and the concepts for which they stand are disputed. Sometimes these disputes
about the categories of constitutional theory are just as sharp and rancorous as the
substantive disagreements with which they are associated. For example, in a recent
essay, Martin Redish and Matthew Arnould characterize the use of the term “originalism”
to describe new originalist constitutional theorizing as “Orwellian.”30 This
characterization is misleading given the transparency of recent originalist theorizing,31
but Redish and Arnould’s use of a powerful epithet is strong evidence that at least some
constitutional theorists are heavily invested in particular characterizations of
“originalism.”
29
Stephen E. Sachs, Originalism as a Theory of Legal Change, 38 HARV. J.L. & PUB. POL'Y 817, 82021 (2015); see also William Baude, Is Originalism Our Law?, 115 COLUM. L. REV. 2349, 2367 (2015)
(paraphrasing Sachs as arguing, “Originalism . . . can be recast as an account of lawful change.”).
30
Martin H. Redish & Matthew B. Arnould, Judicial Review, Constitutional Interpretation, and the
Democratic Dilemma: Proposing a “Controlled Activism” Alternative, 64 FLA. L. REV. 1485, 1509 (2012)
31
See Solum, Originalism and Constitutional Construction, supra note 11, at 527-529.
9
Lawrence B. Solum
Borrowing from the philosophy of language, we can use the idea of “metalinguistic
negotiation”32 to refer to the process by which the meaning of words like “originalism”
and phrases like “living constitutionalism” are contested (adversarially) or negotiated
(cooperatively). These words and phrases represent theoretical concepts that are shaped
differently by different theorists. On the one hand, we want our theoretical vocabulary to
be theoretically precise and conceptually clear. On the other hand, we would like our
terms to reflect actual usage. When the usage itself is inconsistent and when even
individual theorists do not formulate their concepts precisely, the result can easily be a
muddle of miscommunication.
There are undoubtedly a variety of reasons for persistent disagreement over the
meaning of “originalism” and the dividing line that separates originalism and
nonoriginalism. The word “originalism” is ideologically charged. Conservative
originalists may reject the idea that there could be such a thing as “progressive
originalism.” Progressive textualists may be uncomfortable with “originalism” as a label
for their position, fearing guilt by association. Some nonoriginalists may wish to reserve
the label “originalism” for the most extreme and implausible versions of that theory;
some originalists may want to define terms in a way that includes the reasonable terrain
on the originalist side of the theoretical divide. We will return to this topic below.33
We can deal with this problem in a variety of ways. Constitutional theorists might
cooperatively seek a common, conceptually clear, and precise vocabulary. Or they might
engage in contestation, advancing reasons for and against various ways of cleaving
conceptual space on the basis of premises drawn from conceptual ethics.34
Ideally, theorists would reach agreement on the meaning of “originalism,” but both
metalinguistic negotiation and contestation require engagement by the community of
constitutional theorists over time. Given the current state of terminological disorder, we
can proceed in a different way by stipulating definitions and explicitly recognizing that
stipulations suspend but do not end the process of metalinguistic negotiation.
2. Stipulated Definitions of Originalism, Nonoriginalism, and Living
Constitutionalism
Let us stipulate the following definitions for the purposes of this Article:
Originalism: Stipulate that a constitutional theory is a form of originalism if that
theory affirms (1) the Fixation Thesis and (2) a version of the Constraint Principle
that is at least as strong as the minimalist version—and if the theory includes (3)
some plausible account of how original meaning is determined (e.g., original
intentions, public meaning, ratifiers’ understandings, or original methods) as well as
32
Plunkett, David, and Timothy Sundell, Disagreement and the Semantics of Normative and Evaluative
Terms, 13 PHILOSOPHERS' IMPRINT 23 (2013); Plunkett, David, and Timothy Sundell, Dworkin's
Interpretivism and the Pragmatics of Legal Disputes, 19 LEGAL THEORY 3 (2013); Plunkett, David, and
Timothy Sundell, Antipositivist Arguments from Legal Thought and Talk: The Metalinguistic Response in
PRAGMATISM, LAW, AND LANGUAGE 56-75. (G. Hubb and D. Lind eds. 2014).
33
See infra Part I.B.4, p. 13.
34
On conceptual ethics, see Alexis Burgess & David Plunkett, Conceptual Ethics I, 8 PHILOSOPHY
COMPASS 1091 (2013).
10
The Constraint Principle
(4) some account of constitutional determinacy such that adoption Constraint
Principle would have significant effects on constitutional practice.
Nonoriginalism: Stipulate that a constitutional theory is a form of nonoriginalism if
it denies either the Fixation Thesis or the Minimalist Version of the Constraint
Principle or both.
Living Constitutionalism: Stipulate that a constitutional theory is a form of living
constitutionalism if it affirms that the content of constitutional doctrine should
change in response to changing circumstances and values.
Given these definitions, it is possible that some forms of originalism are compatible with
some forms of living constitutionalism.35 For example, a living constitutionalist might
accept the Constraint Principle but affirm the Fact of Constitutional Underdeterminacy: a
New Originalist of this sort would allow for changing constitutional doctrine within
constructions zones whose outer limits were set by the communicative content of the
constitutional text. But other living constitutionalists will reject the Constraint Principle
and affirm that constitutional actors may act in ways that are inconsistent with the text:
that is, some living constitutionalists are nonoriginalists. And there are some originalists
who may reject all forms of living constitutionalism: originalists of this kind may believe
that the communicative content of the constitutional text is sufficiently rich or thick so as
to fully determine the content of constitutional doctrine. Other originalists may believe
that there are constitutional default rules (such as a default rule of Thayerian deference)
that make the resolution of constitutional cases.
3. Forms of Nonoriginalism that Deny the Constraint Principle
Given our stipulated definitions, we can now identify the contending views in
contemporary constitutional theory that count as nonoriginalist because they reject the
Constraint Principle. This is a crucial step in the development of the case for the
Constraint Principle—because the justification for constraint necessarily depends on a
comparison with the alternatives. There are several different forms of nonoriginalism,
and for this reason the complete argument for the Constraint Principle requires that we
compare a particular version of the Constraint Principle with a representative version of
the most important forms of nonoriginalism. At this stage of the Article, the aim is to
identify and articulate the major alternatives (or families of alternatives). At a later stage,
we will then consider the case for constraint as it applies to each of the alternatives
identified here.36
For now, I will simply provide a list of the relevant forms of nonoriginalism. Note
that this list is limited to forms of nonoriginalism that deny the Constraint Principle, and
hence excludes versions of nonoriginalism that accept constraint but deny the Fixation
Thesis. Here is the list:
35
The most prominent example would be Jack Balkin’s theory, living originalism (or framework
originalism). See JACK BALKIN, LIVING ORIGINALISM 149 (2011).
36
See infra Part VII, p. 104.
11
Lawrence B. Solum
Multiple Modalities: This is the view that law is a complex argumentative practice
with multiple modalities of constitutional argument.
Constructive Interpretation: This is Dworkin’s theory, also called “law as integrity”
or the “moral readings” theory.
Common Law Constitutionalism: This is the view that the content of constitutional
law should be determined by a common-law process.
Popular Constitutionalism: This is the view that “We the People” can legitimately
change the constitution through processes such as transformative appointments that
do not formally amend the text.
Multiple Meanings: This is the view that the constitutional text has multiple
linguistic meanings and that constitutional practice should choose between these
meanings on a case by case basis.
The Supreme Court as Superlegislature: This is the view that the Supreme Court
should act as an ongoing committee of constitutional revision with the power to
adopt amending constructions of the constitutional text on the basis of the same
kinds of reasons that would be admissible in a constitutional convention.
Thayerianism: This is a family of views that require courts to defer to Congress, with
three variants:
Constrained Thayerianism is the view that courts should defer to Congress but
that Congress itself should be constrained by the original meaning of the
constitutional text.
Unconstrained Thayerianism is the view that courts should defer to Congress and
that Congress should have the constitutional power to revise the constitutional
text, either by adopting amending legislation or by creating implicit
amendments through ordinary statutes.
Representation Reinforcement Thayerianism is the view that courts should defer
to Congress except when judicial review is necessary to preserve democracy,
including protection of discreet and insular minorities and protection of
democratic processes.
Constitutional Antitheory: There are four views that are “antitheoretical” in the
sense that they deny that constitutional practice should be guided by any normative
theory, whether that theory be originalist or nonoriginalist:
Particularism is the view that constitutional practice should be guided by salient
situation-specific normative considerations in particular constitutional
situations.
Pragmatism is the similar view, associated with Judge Richard Posner (and in a
different form with Daniel Farber and Suzanna Sherry) that constitutional
decisions should be made pragmatically on the basis of various normative
considerations.
Eclecticism is the view that different judges should embrace different approaches
to constitutional interpretation and construction, and that even a single judge
should adopt different approaches on different occasions.
12
The Constraint Principle
Opportunism is the view that theoretical stances should be deployed strategically
to achieve ideological or partisan goals.
Rejectionism: Anticonstitutionalism & Constitutional Replacement:
Anticonstitutionalism is the view that the communicative content of constitutions
in general should play no role in constitutional practice.
Constitutional replacement theories would allow the text of a normatively
attractive replacement constitution to play a role in constitutional practice, but
reject any constraining role for the current Constitution of the United States.
There are undoubtedly other forms of nonoriginalism, but these nine are among the most
salient, in part because they play a prominent role in theoretical discourse about the
constitution and in part because they are representative of families of possible theories.
The aim of this Article is to consider the best versions of nonoriginalism—and the
argument would need to be expanded if an important family of alternatives has been left
off the list.
At this stage of the argument, I will simply assume that there are versions of each of
these nine approaches that reject the Constraint Principle. In Part VII, we will return to
each of the nine theories and engage in pairwise comparison of each theory with
originalism and the Constraint Principle. At that stage, I will explicate each theory,
explain why the theory (or some plausible version of the theory) rejects constraint, and
then evaluate the theory in light of the general arguments for and against the Constraint
Principle that are discussed in Part IV and Part VI of this Article.
***
I have left one view off the list, but it is closely related to what I call “constitutional
opportunism.” If one believes that the current Constitution of the United States is not
just flawed but rather is fundamentally evil, then many conclusions would follow.
Fundamental evil would justify revolution, rebellion, or secession, but other measures
might be more appropriate. For example, one might be justified in engaging in
intentional deception—masking arguments that call for judicial amendment of the
constitution in the cloak of constitutional fidelity. Thus, one might affirm allegiance to
“true meaning” of the Constitution, while in fact engaging in a program of constitutional
subversion. This sounds unprincipled and manipulative, but it would actually be justified
if the current Constitution of the United States were truly evil and if the current
amendment process is unworkable. Under these circumstances, “constitutional
deception” (the name we can give this view) could be morally justified on the grounds
that deception, although prima facie wrong, is preferable to constitutional change
through violent means.
A sophisticated constitutional opportunist might adopt the constitutional theory that
creates the maximum opportunity for judicial constitutional constructions that
approximate justice. And such opportunists might affirm that they actually accept that
the text is binding by pointing to examples where judicial construction is not a politically
realistic means of effecting constitutional change. For example, opportunists might
affirm that they accept that the two-senators-per-state provision is binding, even though it
is unjust—with the silent reservation that they would change their mind if they thought
13
Lawrence B. Solum
that there was a realistic chance that the Supreme Court would require reapportionment
of the Senate on the basis of the one-person-one-vote rule implicit in the Equal Protection
Clause as incorporated in the Fifth Amendment Due Process Clause by Bolling v.
Sharpe.37
Moreover, the sociology and psychology of constitutional discourse might enable
constitutional deception to be erased from the conscious beliefs of constitutional theorists
whose dispositions are best explained by unconscious opportunism. Consider the
following simplified model of how this might happen. Imagine three successive
generations of constitutional actors. Generation one consists of judges, lawyers, and
scholar who consciously affirm constitutional opportunism because they believe the
Constitution is fundamentally flawed or even evil. They are able to change constitutional
doctrine in many areas—such that constitutional doctrine more closely approximates a
reasonably just constitution. Generation two consists of theorists and practitioners who
have an ambivalent attitude—partially recognizing patterns of constitutional deception
but coming to believe that some of the amending constructions actually are the true
meaning of the constitutional text. Generation three no longer recognizes the deception.
They believe that the amending constructions are obviously the true meaning of the text
and the key decisions announcing such amending constructions become, for them, the
canonical cases—whose validity cannot be challenged. In other words, constitutional
deception can, through a gradual process, become constitutional self-deception.
***
4. Some Motivations for and Consequences of Metalinguistic Negotiation Over
“Originalism” and “Living Constitutionalism”
We have now defined “originalism” and “nonoriginalism,” outlined the originalist
family of constitutional theories, and identified nine forms of nonoriginalism that reject
the constraint principle. Some readers may object to this way of carving constitutional
theory space and hence resist the way in which the role of the Constraint Principle in
contemporary constitutional theory has been framed. And constitutional theorists are
certainly entitled to resist stipulated definitions. At this point, however, we can step back
and ask about the source of this resistance.
The word “originalism” and the phrase “living constitutionalism” are both
ideologically charged. “Originalism” has been associated with constitutional
conservatism and “living constitutionalism” with progressive or liberal constitutionalism.
“Originalism” is associated with the Reagan administration and especially with Attorney
General Ed Meese and the Office of Legal Counsel during Reagan’s second term. And
contemporary originalism is associated with Justices Antonin Scalia and Clarence
Thomas. “Living Constitutionalism” is associated with pre-New Deal progressive
constitutional theory and with the Warren Court in general and Justice William Brennan
in particular. On both sides, there is a tendency to associate the labels with associated
37
347 U.S. 497 (1954).
14
The Constraint Principle
political and ideological positions. It is my belief that this results in a tendency to
“choose sides” in the debates over originalism and living constitutionalism.38
If originalists are the “good guys” and living constitutionalists are the “bad guys” (or
vice versa), then ideological consequences follow. Progressives may believe that
originalism is obviously false and pernicious, and conservatives may believe the same
thing about living constitutionalism. Both sides may believe that it simply impossible to
affirm what we can call “compatibilism”—the view that some forms of living
constitutionalism are compatible with some forms of originalism. Compatibilism would
imply that one could be simultaneously conservative and progressive, but these positions
are conceived oppositionally and are demarcated on the basis of differences and not
agreements. For this reason, metalinguistic negotiation over the meaning of
“originalism” and “living constitutionalism” is likely to be contested and not cooperative.
If this diagnosis is correct, then it will have consequences for the process of
metalinguistic negotiation over the terms “originalism” and “living constitutionalism.”
Some progressives may insist that the criteria for what counts as “originalism” must
include an ideological component; a related view would be that “originalism” must reject
certain progressive outcomes or a set of cases that are praised by progressives and
criticized by conservatives. Someone who argues that the original meaning of the
constitutional text is consistent with left-leaning outcomes (such as a right to abortion, a
right to same sex marriage, or expansive national legislative power) are not really
“originalists”—because such a person fails to satisfy the ideological component of the
conception of “originalism.” Some conservatives will be similarly motivated: anyone
who advocates for progressive outcomes must, by definition, be a “living
constitutionalist” who rejects fidelity to the constitutional text as a fundamental political
value.
One way to think about the ideological dimension of metalinguistic negotiation over
terminology in constitutional theory borrows from Bernard Williams’s notion of a “thick
moral concept”39 for which descriptive and evaluative content are entwined. Similarly, it
is possible that one of the positions taken in metalinguistic negotiation over the words
“originalism” and “living constitutionalism” advocates for the position that these words
should be used as “thick ideological concepts”: in other words, the content of the terms
should be conceived as combining ideological and descriptive-theoretical elements.
Contestation over the proper conceptual shape of “originalism” and “living
constitutionalism” is likely to be particularly sharp in connection with the Constraint
Principle. Some progressive and liberal constitutional theorists may wish to resist the
conclusion that their position should properly be labeled as “originalist” if they accept the
Fixation Thesis and the Constraint Principle. For this reason, they may wish to draw the
line between originalism and nonoriginalism at a point that allows them to claim that they
do not support a judicial power to override the meaning of the constitutional text.
In this connection, it is useful to consider Mitchell Berman’s suggestion that
originalist theories can be classified according to the degree to which they are “hard” and
“soft.” He writes:
38
The assertions in this paragraph are based on my own sense of the contemporary scene. They
certainly are not the product of extensive research in the intellectual history of constitutional theory.
39
BERNARD WILLIAMS, ETHICS AND THE LIMITS OF PHILOSOPHY 1985). For additional commentary, see
the essays collected in THICK CONCEPTS (Simon Kirchin ed. 2013).
15
Lawrence B. Solum
At the weakest end of the spectrum lies the view that the originalist focus
(framers’ intent, ratifiers’ understanding, original public meaning, or what-haveyou) ought not to be excluded from the interpretive endeavor. This view–what we
might call “weak originalism”–maintains merely that the proper originalist object
(whatever it may be) should count among the data that interpreters treat as relevant.
At a polar extreme from weak originalism rest views that collectively I will label
“strong originalism.”
Strong originalism, as I will use the term, comprises two distinct subsets.
Probably the most immediately recognizable originalist thesis holds that, whatever
may be put forth as the proper focus of interpretive inquiry (framers’ intent,
ratifiers’ understanding, or public meaning), that object should be the sole
interpretive target or touchstone. Call this subtype of strong originalism, “exclusive
originalism.” It can be distinguished from a sibling view a shade less strong–viz.,
that interpreters must accord original meaning (or intent or understanding) lexical
priority when interpreting the Constitution, but may search for other forms of
meaning (contemporary meaning, best meaning, etc.) when the original meaning
cannot be ascertained with sufficient confidence. Call this marginally more modest
variant of strong originalism “lexical originalism.”40
Berman then argues that the term “Originalism” should be reserved for what he calls
“strong originalism”:
As Dennis Goldford put it in his recent book-length examination of the
originalism debate, what distinguishes originalism from non-originalism is the
claim “that the original understanding of the constitutional text always trumps any
contrary understanding of that text in succeeding generations.” Self-described
originalists differ regarding countless details–whether the proper interpretive focus
is framers’ intent, ratifiers’ understanding, or original public meaning; whether the
best reasons for originalism concern what it means to interpret a text, or what must
be presupposed in treating a Constitution as binding, or how best to constrain
judges and provide stability and predictability; whether extra-judicial constitutional
interpretation is subject to the same constraints as is judicial constitutional
interpretation; and so on. The contention urged consistently – from originalist icons
Raoul Berger and Robert Bork to standard bearers from the younger generation like
Gary Lawson and Michael Paulsen – is that judges should interpret the Constitution
solely in accordance with some feature of the original character of the
constitutional provision at issue.41
Berman’s conclusion is “Translated into my proposed terminology and shorthand:
Originalism is strong originalism. The Great Divide, to complete Scalia’s observation,
lies between those who attend exclusively to the original object and those who attend to
changed meanings too.”42
40
Mitchell N. Berman, Originalism Is Bunk, 84 N.Y.U. L. REV. 1, 10 (2009).
Berman, supra note 40, at 19-20 (citing DENNIS J. GOLDFORD, THE AMERICAN CONSTITUTION AND
THE DEBATE OVER ORIGINALISM 139(2005).
– it is the only method that is suited to discovering the actual meaning of the relevant text”).
42
Berman, supra no note 40, at 20.
41
16
The Constraint Principle
Notice that Berman’s definition of “originalism” is closely connected with the
Constraint Principle. Berman’s strong originalism requires acceptance of a maximalist
version of the Constraint Principle. Berman defines forms of originalism that accept the
Minimalist Version of the Constraint Principle as “weak” and hence as forms of
nonoriginalism.
Berman’s definition is attractive to what we might call “moderate living
constitutionalists,” because it defines the moderate territory (which endorses changes in
constitutional doctrine that are consistent with original meaning) as within the bounds of
nonoriginalism. And perhaps more importantly, it allows progressive constitutional
theorists to define their position in opposition to originalism. Shaping the concept in this
way is consistent with the notion that “originalism” is a thick ideological concept for
which the theoretical content cannot be disentangled from the ideological associations. If
one believes that moderate living constitutionalism should be endorsed by progressives,
then one might be attracted to the view that this theory is (by definition) not true
“originalism.”
The stipulated definitions in this article are motivated by a desire to capture the terrain
of contemporary constitutional theory. My sense is that many self-identified originalists
embrace Constraint as Consistency (or a more robust form of constraint) but reject what
Berman calls “strong originalism.” For example, Justice Scalia would not count as an
originalist for Berman, since he affirms a role for precedent. Indeed, it is difficult to
think of any self-identified originalist who endorses what Berman calls “strong
originalism” without qualification. The stipulations offered above provide a working
definition of originalism that reflects what I believe are the linguistic practices of selfidentified originalists.
There are important questions of conceptual ethics implicated by the role of ideology
in metalinguistic negotiation over “originalism” and “living constitutionalism.” In other
work, I argue against the idea that these concepts should be understood as thick
ideological concepts and for the values of conceptual clarity, precision, and descriptive
accuracy: my claim is that these are the most important values for conceptual ethics in the
realm of constitutional theory.43 But on this occasion, my aim is simply clarity. Clarity
can be achieved through stipulation. Readers should feel free to bracket the stipulations
by adding “as defined by Solum in The Constraint Principle” whenever “originalism,”
“nonoriginalism,” and living constitutionalism” are used in this Article. I will now put the
enterprise of metalinguistic negotiation to the side and proceed on the basis of the
stipulations.
C. What Hangs on Acceptance or Rejection of the Constraint Principle?
Elsewhere, I have argued that the Fixation Thesis is and should be relatively
uncontroversial.44 When we are interpreting an eighteenth century text, we should appeal
to the meanings that words and phrases had then; the alternative is to allow the accidents
43
See Solum, Originalism and the Landscape of Contemporary Constitutional Theory (work in
progress).
44
See Solum, supra note 3, at 30 (“In broad outline, the thrust of the discussion that follows is
that the Fixation Thesis should not be controversial.”).
17
Lawrence B. Solum
of linguistic drift (the process by which meanings change over time) to determine the
legal content of constitutional law.45 Assuming this diagnosis is correct, then the
Constraint Principle plays a crucial role in contemporary constitutional theorists.
Constitutional actors and scholars who affirm the Constraint Principle are almost all
originalists—in the sense of originalism stipulated in this article. And those who reject
the Constraint Principle almost always reject originalism as well. In other words, the
Constraint Principle demarcates an important dividing line between originalism and
nonoriginalism. If there are good reasons to accept constraint, then originalism is likely
to follow.
***
Why should we care about the Constraint Principle? Of course, there are many
important issues in constitutional theory that are independent of acceptance or rejection
of the Constraint Principle. But if the Constraint Principle and the Fixation Thesis (and
hence originalism) are accepted, then much follows. Many substantive debates
concerning the proper shape of constitutional doctrine take on a whole new shape if
constitutional practice should be constrained by the original meaning of the
constitutional text. Some constitutional debates are settled once it is agreed that the
original meaning should govern.
If we accept the Fact of Constitutional Underdeterminacy and hence the existence of
construction zones, there will be other constitutional debates that will continue unabated.
Some provisions of the constitution seem vague or open textured, at least on their face.
With respect to these provisions, moderate forms of living constitutionalism could provide
theories of constitutional construction. And there will continue to be disagreement
among living constitutionalists as to what theory should govern in the construction zones.
Critics of living constitutionalism might reformulate their positions as theories of
constitutional construction; one such approach might be to advocate a set of default
rules, roughly modeled on Thayerian deference, that would produce judicial restraint
where the meaning of the constitutional text is not clear.
***
II. CONTRIBUTION, CONSTRAINT, AND RESTRAINT
The aim of this Part is to explicate the constraint principle. The first two steps involve
the drawing of important distinctions: (1) the distinction between “contribution” and
“constraint” and (2) the distinction between “constraint” and “restraint.” The third step
involves formulation of three versions of the Constraint Principle, one of which,
“Constraint as Consistency,” can serve as the dividing line between originalism and
nonoriginalism. The fourth step is the introduction of the idea of “defeasibility
conditions” and the formulation of the minimum content of defeasibility.
45
See Solum, supra note 3, at 17. For a description of linguistic drift, see SOL STEINMETZ, SEMANTIC
ANTICS: HOW AND WHY WORDS CHANGE MEANING 49–50 (Random House 2008).
18
The Constraint Principle
A. The Distinction Between Contribution and Constraint
So far, our discussion of the Constraint Principle has been formulated using the word
“constraint” and this way of talking might give rise to an inference that the role of
original meaning must either be characterized as “fully constrained” or “completely
unconstrained.” But that inference would be incorrect, because original meaning can
play an important role in constitutional practice that falls short of constraint. Let us use
the term “contribution” to designate the most general and inclusive type of influence that
the original meaning of the constitutional text could have on the content of constitutional
doctrine and on the decisions made by constitutional actors (including both judicial and
nonjudicial officials).46 To make this point clear and precise, we need to unpack the
distinction between “contribution” and “constraint.”
The communicative content (or roughly, linguistic meaning) of the constitutional text
could contribute to the legal content of constitutional doctrine in various ways. Very few
legal theorists argue that the original meaning of the constitutional text should have no
influence on constitutional doctrine. For example, the Multiple Modalities Theory would
allow for originalist arguments as inputs to the complex argumentative practice that
constitutes constitutional law. And Dworkin’s theory requires that the theory that best fits
and justifies the content of constitutional law would take the original meaning of the text
into account, as one of the many features of the constitutional practice that are the object
of an overall moral reading. Even the most extreme nonoriginalist view, the
Superlegislature Theory, does not forbid consideration of original meaning in the
superlegislative process: indeed, the idea that the Supreme Court amends some provisions
of the Constitution assumes that the Court continues to enforce others.
There is, however, conceptual space for a theory that does deny that the original
meaning of the constitutional text should play any role in constitutional practice. Louis
Michael Seidman’s work on constitutional disobedience and constitutional skepticism
could be interpreted as taking this position.47 Clearly, any view that takes the position
that original meaning should make no contribution to constitutional doctrine and practice
should be called “nonoriginalist,” but for the purposes of this Article, contribution
without constraint is stipulated to be a form of nonoriginalism.
***
This Article articulates and defends the Constraint Principle, which serves as an
important dividing line between originalists and nonoriginalists. But we should note that
one could argue for a much weaker principle, which we might call “The Contribution
Principle”—the claim that the original meaning of the constitutional text should make
some contribution to constitutional doctrine. The Contribution Principle is accepted (in
different ways) by several nonoriginalist constitutional theories. For example, the
original meaning of the constitutional text would be a valid form of constitutional
argument in a Multiple Modalities approach. Likewise, the original meaning of the
constitutional text is one of the facts that Dworkin’s theory of constructive interpretation
46
For discussion of the key ideas, “contribution,” “communicative content,” and “legal content,” see
Solum, supra note 7.
47
LOUIS MICHAEL SEIDMAN, ON CONSTITUTIONAL DISOBEDIENCE (Oxford: Oxford University Press
2012).
19
Lawrence B. Solum
would require Hercules to take into account in developing the theory that best fits and
justifies the law as a whole.
It is only by rejecting the Contribution Principle that constitutional theorists can
establish the irrelevance of original meaning for constitutional practice. This is
important. Many of the criticisms of the Constraint Principle would seem to apply to the
Contribution Principle as well. Nonoriginalists might jettison the Contribution
Principle, but this would require them to pay a price—sacrificing much of the normative
and descriptive plausibility of their positions. Anticonstitutionalism is theoretically
consistent but it is hard to market: even Don Draper would have found it challenging to
design a successful ad campaign! The alternative is to jettison the criticisms that
undermine the Contribution Principle, but again, there is a price to be paid, because
many arguments against originalist will fall to the wayside.
***
B. The Distinction Between Constraint and Restraint
There is a second important distinction that requires clarification, which we can
articulate by stipulating a distinction between “constraint” and “restraint” in a form
similar to that first articulated by Thomas Colby,48 as follows:
•
•
Restraint: A judicial decision is stipulated to be “restrained” if and only if the
decision defers to decisions made by executive officials or statutes enacted by
a legislative body.49
Constraint: Constitutional practice (by judges or other officials) is stipulated to
be “constrained” if and only if the actions taken by officials are consistent with
the communicative content of the constitutional text.
Notice that only judicial decisions can be restrained in the stipulated sense, but that the
actions of any official can be constrained by a requirement of consistency with the
communicative content of the constitutional text. We can represent the possible
interactions between constraint and restraint via the following table:
48
See Thomas B. Colby, The Sacrifice of the New Originalism, 99 GEO. L.J. 713, 751 (2011)
(“[A]lthough originalism in its New incarnation no longer emphasizes judicial restraint--in the sense of
deference to legislative majorities--it continues to a substantial degree to emphasize judicial constraint--in
the sense of promising to narrow the discretion of judges. New Originalists believe that the courts should
sometimes be quite active in preserving (or restoring) the original constitutional meaning, but they do not
believe that the courts are unconstrained in that activism. They are constrained by their obligation to remain
faithful to the original meaning.”). For an interesting discussion, see Jack M. Balkin, Judicial Constraint,
Judicial Restraint, and the New Originalism, Balkinization, May 12, 1016,
http://balkin.blogspot.com/2016/05/judicial-constraint-judicial-restraint.html.
49
One important form of restraint is limited to deference to democratic officials. Let us use the phrase
“democratically restraint” to designate this subspecies of restraint.
20
The Constraint Principle
Table 1: Restraint and Constraint
Restrained
Constrained Judicial decision defers to
executive or legislative
action and is consistent with
the constitutional text.
Not Judicial decision defers to
Constrained executive or legislative
action, but such deference is
inconsistent with the
constitutional text.
Not Restrained
Judicial decision conflicts
with executive or
legislative action. but this
decision is required by
consistency with the
constitutional text.
Judicial decision conflicts
with executive or
legislative action, and such
deference is inconsistent
with the constitutional text.
Let us call this understanding of the relationship of constraint and restraint, the
“Constraint-Restraint Distinction.” We can now define the notion of “judicial activism,”
which we shall stipulate to mean judicial decisionmaking that does is not restrained in the
sense specified by the distinction just made.50
Once we marked this distinction, it follows that the Constraint Principle is not a
principle of judicial restraint (in the stipulated sense). Depending on the context and the
actual original meaning of the relevant constitutional provisions, the Constraint Principle
may require judicial activism and may not instantiate judicial restraint. But we should
note that the Constraint Principle can be combined with complimentary principles of
judicial restraint that operate in the construction zones—those areas of constitutional
doctrine that are underdetermined by the communicative content of the constitutional
text.
***
The path to contemporary originalist constitutional theory does not have a single
starting point. But one of pathways seems to have begun with criticism of the Warren
Court that focused on “judicial activism” and departures from the original intentions of
the framers. These critics wanted both “constraint” (adherence to original meaning) and
“restraint” (noninterference with the political branches). Originalism has moved on.
Original intent has mostly given way to original public meaning. Some supporters of
originalism have replaced the phrase “judicial activism” with “judicial engagement.”51
Nonetheless, the once unnoticed tension between constraint and restraint remains. For
some originalists, the holy grail is a version of originalism that brings constraint by the
pubic meaning of the text into perfect alignment with a robust idea of judicial restraint
and deference to democratic politics. Other originalists are skeptical about the chances
50
On the phrase “judicial activism,” see Lawrence B. Solum, Legal Theory Lexicon: Strict
Construction and Judicial Activism; Keenan D. Kmiec, The Origin and Current Meanings of Judicial
Activism, 92 CAL. L. REV. 1441 (2004).
51
Randy Barnett, “Judicial engagement” is not the same as “judicial activism,” The Volokh Conspiracy,
https://www.washingtonpost.com/news/volokh-conspiracy/wp/2014/01/28/judicial-engagement-is-not-thesame-as-judicial-activism/ (January 28, 2014).
21
Lawrence B. Solum
of this project for ultimate success. My view is that constraint is essential to
originalism—one of its defining characteristics. The relationship of restraint to
originalism is contingent: whether originalism results in judicial restraint depends on the
communicative content of the constitutional text.
***
C. Three Versions of the Constraint Principle
We are now in a position to provide precise formulations of the Constraint Principle.
There are many possible formulations, but for the purposes of this Article, we will
explore three versions of constraint. We can begin with a very strong version of the
Constraint Principle, which we can call “maximalist.”
1. A Maximalist Version of the Constraint Principle: Constraint as Derivability
It turns out that a precise formulation of the “Maximalist Version of the Constraint
Principle” is not easy to state in positive terms. The general idea can be stated
negatively: maximum constraint eliminates the role of anything other than the
communicative content of the constitutional text in the determination of the legal content
of constitutional doctrine. If we focus on constitutional practice, then the idea is that
every official action must comply with constitutional doctrines determined by the text.
One way to capture this relationship might be one of identity. Thus, one might try the
following formulation:
Constraint as Identity: Proposition P is a legally correct proposition of
constitutional doctrine if and only if the content of P is identical to the
communicative content of the constitutional text.
If this formulation were used to define the Constraint Principle, legal content of
constitutional doctrine and communicative content of the constitutional text would be
identical.
But this formulation faces intractable problems. For example, some correct
propositions of constitutional doctrine are logically implied by the constitutional text
itself: we might call these propositions “constitutional deductions” or “constitutional
implications.” Others are the logical consequences of doctrines corresponding to the text
and facts about the world: we might call these “constitutional applications.” For
example, the First Amendment freedom of press might apply to newspapers if, as a matter
of fact, newspapers are produced by printing presses. The proposition “Newspapers are
protected by the freedom of press” is not identical to the communicative content of the
First Amendment’s freedom of press clause, but originalists will want to be able to say
encompass rules like this.
We can remedy this defect by adding derivability to identity. Let us call the resultant
view “Constraint as Derivability” and stipulate following formulation:
Constraint as Derivability: Proposition P is (or should be) a legally correct
proposition of constitutional doctrine only if it satisfies one of the following
conditions:
22
The Constraint Principle
(1) Identity: the legal content of the doctrine is identical to the communicative
content of some provisions of the constitutional text;
(2) Deduction: the legal content of the doctrine can be deduced from legal
content that satisfies condition (1).
(3) Application: the legal content of the doctrine results from application of
legal content that satisfies condition (1) or condition (2).
Constraint as Derivability has the following corollary, which states the same idea in
negative form:
Corollary to Constraint as Derivability: Proposition P is (or should not be) not a
legally correct proposition of constitutional doctrine if its content is not derivable
from the communicative content of the constitutional text by identity, deduction,
or application.
And this corollary implies that there can be no constitutional doctrine that is created by
supplementation of the constitutional text to resolve irreducible ambiguities, to precisify
vague or open textured provisions, to fill in the content of constitutional gaps, or to
resolve constitutional contradictions. All of the content of constitutional doctrine must be
derived from the text or from the application of the text to the facts.
Recall that living constitutionalism is stipulated to be the view that the content of
constitutional doctrine should change in response to changing values and circumstances.
The third provision allows the derivation of new legal content from the application of
valid legal content to new facts. Because new facts can come into being, it follows that a
very limited form of living constitutionalism is consistent with Constraint as Derivability.
2. A Minimalist Version of the Constraint Principle: Consistency of Constitutional
Doctrine
Constraint as Derivability, the Maximalist Version of the Constraint Principle, is the
most restrictive form of originalism that is plausible given the normative commitments of
originalism. Given the variations among originalists, the Maximalist Version cannot
serve as a least common denominator among originalists.52
To identify the least common denominator, we need to identify the version of the
Constraint Principle upon which all or almost all originalists could agree. When we say
“agree,” I do not mean that all originalists could agree that a minimalist version of the
52
Some originalists might object to the notion that Constraint as Consistency can serve as a least
common denominator. In this regard it is important to remember that the claim is that consistency is the
minimal form of constraint that can serve as the dividing line between originalists and nonoriginalists. The
claim is not that Constraint as Consistency is somehow the better formulation because it can serve as the
least common denominator. Constraint as Consistency is a normative principle and it would need to be
defended (against more robust forms of consistency and against less robust forms of contribution) on
normative grounds.
Of course, originalists who affirm a more rigorous form of constraint are free to define “originalism” in
a way that excludes self-identified originalists who do not share their understanding of the Constraint
Principle. And they can also engage in metalinguistic negotiation over the the meaning of “originalism”—
arguing that Constraint and Consistency should not serve as the dividing line. In this Article, I operate on
the basis of stipulations and set these questions to the side.
23
Lawrence B. Solum
Constraint Principle is the best version of constraint or even that it is sufficient for
legitimacy (or for some other normative goal). Rather, I mean that all or almost all
originalists would agree that the satisfaction of the Minimalist Version provides a level of
constraint that is normatively justified (better than less constrained alternatives)—even if
some originalists believe that more constraint is justified as well. In other words, the
Minimalist Version acts as a floor but not a ceiling on the level of constraint from an
ecumenical originalist perspective.
The form of constraint that can best play the role of the Minimalist Version of the
Constraint Principle is “Constraint as Consistency,” which can be define as follows:
Constraint as Consistency. Constraint as Consistency is the conjunction of three
requirements and three qualifications as follows:
Requirement One: Constitutional doctrines and the decisions of
constitutional cases must be consistent with the “translation set.” The
translation set consists of the set of doctrines that themselves directly translate
the communicative content of the text into doctrine and the set of doctrines that
are the logical implications of that set.
Requirement Two: All of the communicative content of the constitutional
text and its logical implications must be reflected in the legal content of
constitutional doctrine.
Requirement Three: All of the content of constitutional doctrine must be
fairly traceable to the direct translation set, with traceable content including
precisifications, implementation rules, and default rules presupposed (or
otherwise supported) by the text.
Qualification One: Requirements One, Two, and Three operate only to the
extent that the communicative content of the constitutional text is epistemically
accessible; they are not violated by departures from unknown communicative
content.
Qualification Two: If Requirements One, Two, and Three are not satisfied,
then constitutional practice should be brought into compliance with constraint
over time, giving due regard to the effects of constitutional change on the rule
of law.
Qualification Three: Requirements One, Two, and Three are defeasible in
limited and extraordinary circumstances, as specified by the best theory of
defeasibility.
For the sake of simplicity in exposition, restatements of Constraint as Consistency will
include the two requirements but not the qualifications—except as the qualifications are
relevant to a particular issue or concern. The third qualification, defeasibility, is explored
in depth at the end of this Part of the Article.53
Constraint as Consistency allows for the creation of constitutional doctrines that
cannot be derived from the text by logical implication or application—so long as these
additional doctrines are consistent with and fairly traceable to the communicative content
of the constitutional text. Constraint as Consistency thus provides a mechanism for
dealing with the Fact of Constitutional Underdeterminacy and hence for creating rules of
constitutional law in the case of irreducible ambiguity, vagueness or open texture, gaps,
53
See infra Part II.D, p. 26.
24
The Constraint Principle
or contradictions. Any violation of Constraint as Consistency will also be a violation of
Constraint as Derivation, but the reverse relationship does not hold. Relatedly, any
violation of the weaker Contribution Principle will also violate Constraint as
Consistency—and again, the reverse does not obtain.
3. A Mixed Version of the Constraint Principle: Constraint: Constraint by Core
Structural Provisions and Deference Otherwise
There are many possible versions of the Constraint Principle. For ease of explication,
most of the discussion that follows will be limited to Constraint as Derivability (the
Maximalist Version) and Constraint as Consistency (the Minimalist Version), but a
complete account of the Constraint Principle would need to consider all of the possible
variations.
Because a full account of the possible forms of constraint is impracticable, we can use
one example for the purposes of illustration. Consider a version of constraint that
exemplifies what I call a “mixed” theory. I will use the term “mixed” to refer to theories
that adopt at least the Minimalist Version of the Constraint Principle for some subdomain
of constitutional questions. Consider a mixed theory that adopts the Minimalist Version
for what are sometimes called the “hardwired” provisions of the Constitution (the clauses
that deal with the structure of the national government), but which adopts a form of
nonoriginalism for other constitutional questions. Let us formulate this mixed theory (the
“Hardwired Only Constraint Principle”) as follows:
The Hardwired Only Constraint Principle. The content of constitutional doctrine
should reflect the following requirements and permissions:
(1) in cases in which a constitutional provision governing the structure of the
national government provides a clear rule (understood as a rule that is not
significantly vague or open textured), constitutional practice must be consistent
with the clear rule,
(2) in cases in which a constitutional provision governing the structure of the
national government is vague or open textured, constitutional practice may be
inconsistent with the original meaning of the constitutional text, and
(3) in cases which a constitutional provision governs individual rights, the
content of constitutional doctrine must incorporate the clear and undisputed core
(if there is one) of the enumerated individual rights provisions of the constitution,
but it may go beyond the core.54
We can call this theory “mixed” because it does contain an originalist element. The first
element respects the Constraint Principle with respect to clear rules of constitutional
structure. Many nonoriginalists seem to admit that originalism is a plausible view in
cases governed by clear structural provisions, such as the requirement that the President
be at least 35 years of age, the division of Congress into two houses, the provision
54
This formulation does not deal with the question of unenumerated individual rights. Because the
Hardwired Only Constraint Principle is introduced solely for the purpose of illustrating the idea of a mixed
version of the constraint principle, it is not fully formulated. A complete version of the principle would
need to deal with the unenumerated rights question and many other questions as well.
25
Lawrence B. Solum
providing that each state shall be represented by two senators, and so forth. And other
originalists may agree that there is an undisputed core of original meaning with respect to
some of the individual rights provisions of the Constitution; that core provides a “floor”
below which the protection of individual rights cannot fall. All other questions would
then be guided by some version of nonoriginalist constitutional theory, such as the
Multiple Modalities Theory, Common Law Constitutionalism, or Dworkin’s Constructive
Interpretation.
***
It should be obvious that we have not exhausted the possible forms of the Constraint
Principle. Many other versions could be proposed, and each version would have to be
justified. For the purposes of this Article, the enterprise of justification will focus on
Constraint as Consistency which serves as the Minimalist Version of the Constraint
Principle. If Constraint as Consistency can be justified and if the Fixation Thesis can
also be justified, then some form of originalism is true or correct. But if Constraint as
Consistency cannot be justified, then more demanding forms of originalism are also likely
to fall and nonoriginalism will be true or correct. Similarly, if Constraint as Consistency
can be shown to be more attractive than the major forms of nonoriginalism, we will have
good reason to believe that mixed theories are likely to be defective—although a
demonstration of their defects would require that each mixed form of constraint be
considered separately.
***
D. Defeasibility Conditions and Mending Constructions
Up to this point, we have implicitly assumed that the Constraint Principle is
nondefeasible. That is, we have assumed that the Constraint Principle always holds. But
this seems unlikely. As the third qualification of Constraint as Consistency suggests, the
best formulation of constraint should include a defeasibility condition.
At a minimum, it seems likely that the Constraint Principle is defeasible under
conditions that are (a) anticipatable, (b) truly exceptional, (c) uncontroversially decisive
from the point of view of all or almost all theories of political morality, and (d) such that
there is no feasible opportunity for constitutional amendment to respond to the
conditions. Let us call this “Minimal Defeasibility.” There may be a case for more
robust forms of defeasibility, but the point of this section is to show that the minimum
form of defeasibility is obviously required by any normatively defensible version of
originalism.
Consider a familiar example. Imagine that a virus were to wipe out every adult person
who was older than 30. It seems clear that the original meaning of the constitutional text
requires the President to be 35 years of age.55 Of course, from an originalist perspective,
the preferable course would be a constitutional amendment. But we can imagine
circumstances in which amendments could not be enacted: the virus has wiped out state
legislatures as well, and it will simply take too long to reconstitute them. In
55
U.S. CONST. Art. I.
26
The Constraint Principle
circumstances like these, we can think of the Constraint Principle as defeasible: it
becomes inoperable when exceptional and unanticipated circumstances make it
impossible for the constitutional order to continue otherwise.
This idea can be generalized. We can call the constitutional doctrines that are
necessary to preserve the constitution “mending constructions” by way of contrast to
“amending constructions” that violate the Constraint Principle. The notion of a mending
construction might be adapted to a variety of situations in which the communicative
content of the constitutional text is in need of repair, not for improvement, but for the
preservation of the constitutional scheme. For example, if there are gaps or
contradictions in the constitutional text that cannot be resolved by interpretation, then
mending constructions could resolve the contradiction or fill the gap.56
It is likely that originalists will disagree about the proper limits on defeasibility
conditions and the propriety of mending constructions. Hard liners will want to limit
defeasibility to the most extreme and unlikely circumstances; for them, mending
constructions would be a last resort. Moderates may argue for a more expansive set of
defeasibility conditions and greater acceptance of mending constructions. But all
originalists should agree that defeasibility conditions should not be so expansive as to
negate the practical operative force of the Constraint Principle.
Disagreements among originalists about defeasibility are likely to be sharp in
connection with the transition from the current constitutional regime to an originalist
regime. Of course, much depends on what the original meaning of key constitutional
provisions actually is. But let us suppose that the original meaning of the Commerce
Clause and of the Necessary and Proper Clause is much narrower than post-New Deal
constitutional practice presupposes. And suppose further that the original meaning of the
grants of executive, legislative, and judicial power imply that many features of the
modern administrative state are unconstitutional, e.g., independent regulatory agencies
are unconstitutional. Given these assumptions, it seems likely that a transition to
originalism would disruptive and costly.
Moderate originalists are likely to endorse a set of defeasibility conditions that can
serve as transition rules. Rather than an originalist “big bang” in which the original
meaning is restored in a single “superterm” of the Supreme Court, moderates might argue
for transition rules that call for an initial phase in which new violations of the original
meaning are forbidden but old violations are “grandfathered in” for a relatively long
transitional period. Over time, the original meaning might be restored, initially by
cancelling the grandfathered status of the most egregious and obvious constitutional
violations. As the direction of constitutional change became clear, some provisions of the
Constitution might be amended to square constitutional practice with the constitutional
text.
A full account of the defeasibility conditions for the Constraint Principle is outside the
scope of this article. This is a large topic, and it requires sustained and in-depth
treatment. For present purposes, we simply note that the most plausible version of the
Constraint Principle will incorporate an account of defeasibility. The third qualification
of the Constraint Principle makes this explicit by incorporated the best version of
defeasibility without specify its content. It is important to recognize that this is a very
large promissory note that must be redeemed in due course.
56
27
Lawrence B. Solum
III. FRAMING THE DEBATE OVER THE CONSTRAINT PRINCIPLE
This Article has two principal aims. The first aim was to elucidate the nature and
content of the Constraint Principle; this aim was the subject of the prior two Parts of this
Article. The second aim, justification of constraint, is taken up in this Part and those that
follow. In this Part, we will explore a set of preliminary questions that serve as a
prolegomenon to presentation of the arguments for the Constraint Principle.
This section is concerned with various questions about the appropriate methods for
assessing normative constitutional theories in general and theories of constitutional
interpretation and construction in particular. What kinds of arguments are appropriate in
the constitutional context? What role should the method of reflective equilibrium play?
Which side has the burden of proof and does the allocation of the burden make a
difference? How can we decide whether a constitutional theory is feasible? Should
constitutional theories be assessed on the basis of their implications for the future? Or
should we instead ask what effects a theory would have had if it had been implemented at
some point in the constitutional past?
A. The Types of Justification
What kinds of arguments can be advanced for and against constraint? We can begin to
address these issues by marking a distinction between internal and external justifications
for propositions of constitutional theory.
1. Internal (Legal) or External (Moral) Justifications
There are two distinct ways that justification can proceed in normative constitutional
theory. We can approach justification from within the practice of law. From the internal
perspective, the appropriated justifications will be legal. Legal justifications are
normative—because the law is a system of norms. So, one approach to justifying the
Constraint Principle would be to argue that it is a legal norm that is currently in force.
For example, if one accepted Hartian positivism, one might argue that the Constraint
Principle is a legal norm that follows from the rule of recognition that serves as the
ground for identifying all legal norms in the United States. Or one might argue that the
Constraint Principle is part of the rule of recognition; that argument would require a
showing that the Constraint Principle is part of the social rule that enables officials to
identify what is (and is not) law in the United States. Or one could dispense with the
theoretical apparatus, and argue in a lawyerly way that the constitutional text is binding
on all officials. We will return to this approach later in the article,57 and Will Baude has
explored this argument in depth in his magisterial article, “Is Originalism Our Law?”58
For now, I will simply note that in this Article, the argument that originalism is the law
does not play a central role—although elements of that argument are borrowed and
incorporated into other arguments.
57
58
See infra Part V.A.
See William Baude, supra note 29.
28
The Constraint Principle
The alternative to the internal perspective is an external one. Whether or not
originalism is the law, we can ask whether our constitutional practice should incorporate
the Constraint Principle. In other words, we can ask whether the Constraint Principle
should be adopted from the perspective of political morality. Viewed in this way, the
debate over constraint is a debate in political philosophy or political theory—and it is
connected to wider debates in general normative ethics (as are other questions of political
morality).
Many originalists believe that our current constitutional practice is only partially
originalist and that a return to the original meaning of the constitutional text would
require substantial revision of constitutional doctrine. Originalists with this view can
frankly acknowledge that originalism is a program of law reform—and hence that it must
be justified on the basis of moral (rather than legal) arguments. Other originalists may
see the relationship as more complicated. They may believe that the status quo
incorporates a good deal of originalism at the level of constitutional principle but also
includes substantial deviations from originalism at the level of constitutional practice.
These originalists may believe that originalism is supported by the deep structure of
constitutional law but that it lacks support in substantial areas of the surface structure.
Originalists of this variety might argue that an originalist deep structure provides pro
tanto reasons for reform of the surface structure.
At this stage in the development of the argument for the Constraint Principle, the most
important thing is to get clear on the distinction between internal legal arguments for
originalism and external moral arguments. And we also need to recognize that marking
this distinction is itself a controversial move. Interpretivist theories of the nature of law,
such as Dworkin’s theory, “Law as Integrity,” may deny there is a sharp distinction
between moral and legal justifications. We should recognize that this metatheoretical
disagreement exits and must be taken into account (in due course) in developing the case
for originalism, but in this Article, some of the metatheoretical issues are bracketed due to
their complexity.
The remainder of this Article will advance a case for originalism that does not depend
on the proposition that originalism is the law. That is, the case will rest on arguments of
political morality that aim to justify originalism as a program of law reform, assuming
(perhaps arguendo) that originalism is not yet “our law.”
2. Deep or Shallow Justifications
If the Constraint Principle is to be justified by arguments of political morality, then the
further question arises whether these reasons should be deep or shallow. Deep reasons of
political morality draw on foundational views in political philosophy, normative ethics,
and metaethics. Shallow reasons are formulated in terms avoid reliance on deep premises
and instead rely on reasons that can be shared by citizens who affirm divergent views on
deep matters.59
Consider three strategies for justifying the Constraint principle, which I will call the
“deep strategy,” the “comprehensive strategy,” and the “shallow strategy.” The first
strategy for justifying the Constraint Principle starts with the deep reasons provided by
the true or correct foundational views in political philosophy, normative ethics, and/or
59
See Lawrence B. Solum, Public Legal Reasons, 92 VA. L. REV. 1449 (2006).
29
Lawrence B. Solum
metaethics. For example, one might to argue for the Constraint Principle by assuming (or
arguing that) welfarism is the correct view in normative ethics and then argue that
welfarism should also serve as the basis for political morality. One could then argue that
the Constraint Principle is justified on welfarist grounds. A strategy like this forms the
basis of John McGinnis and Michael Rappaport’s book, Originalism and the Good
Constitution.60 One can imagine similar strategies that take Kantian deontology, virtue
ethics, or social contract theory as their starting points.
The difficulty with these strategies is that the starting points are deeply controversial
given the pluralistic nature of the political, moral, and religious cultures in the United
States. Welfarism may be a coherent and attractive view, but it is not accepted by most
Americans or by the majority of official constitutional actors or by most constitutional
theorists. Disagreements at the foundational level have persisted for centuries, and there
is no reason to believe that any foundational view will achieve consensus status in the
foreseeable future.61
This suggests a second strategy—a comprehensive approach utilizes all of the
plausible deep normative theories. One might argue that the Constraint Principle is
justified by each and every plausible view in moral philosophy—and by every plausible
view in theological morality and political theory as well. But this strategy would be
difficult to execute—even if viewed as the project of an entire lifetime of scholarly
activity. There are too many plausible views in moral and political philosophy and too
many theological views. And the work of connecting the deep premises that constitute
any one of these views to constitutional theory is itself complex, with many layers of
connection and many possible branches in the argument to be considered. If the second
strategy could be executed, it would provide a very strong argument for the Constraint
Principle, but as a practical matter, the second strategy is unavailable.
These considerations lead to the third strategy. We can attempt to justify the
Constraint Principle on the basis of (relatively) shallow reasons that can be shared by the
public in a pluralist society.62 This shallow strategy is closely related to the Rawlsian
idea of justification through public reasons63 and Cass Sunstein’s notion of midlevel
principles.64 Shallow reasons can be supported using the method of wide reflective
60
See JOHN O. MCGINNIS & MICHAEL B. RAPPAPORT, ORIGINALISM AND THE GOOD CONSTITUTION
(2013).
61
I am not arguing against the presentation of deep arguments for views in constitutional theory. Such
arguments may play an important role, both in constitutional theory and in the internal development of
comprehensive moral and political theories. My claim is only that such arguments have a limited audience,
and therefore that shallow arguments also have an important role to play. In this regard, a division of
intellectual labor may be appropriate. Adherents of particular comprehensive theories of political morality
might offer justifications for originalism (or one of its rivals) from within their theory. There is no a priori
guarantee that any comprehensive theory can provide grounds for supporting (or opposing) originalism.
And it should go without saying, that it seems unlikely that any comprehensive theory will come to be seen
as objectively correct by all or almost all political philosophers, much less all reasonable citizens. Work on
the deep questions is valuable even if it seems likely that the outcome of such work will be intersubjective
agreement. But the task of connecting deep theories to normative debates over the Constraint Principle is
not the task of this Article.
62
For a generalized version of the strategy sketched here, see Solum, supra note 59.
63
JOHN RAWLS, POLITICAL LIBERALISM 212 (expanded ed. 2005).
64
CASS R. SUNSTEIN, LEGAL REASONING AND POLITICAL CONFLICT (1996).
30
The Constraint Principle
equilibrium65 which is explored in greater depth below.66 To preview that discussion,
reflective equilibrium begins with our existing opinions about particular cases, intuitions
about hypothetical cases, and beliefs about general principles of constitutional theory.
We ask whether the Constraint Principle itself and the justifications upon which it rests
are consistent with these opinions, intuitions, and beliefs. If there are inconsistencies, we
adjust our considered judgements.
In this Article, I will pursue the third strategy, relying on shallow arguments that can
appeal to a wide variety of constitutional theorists, officials, and citizens. The third
strategy imposes limits on relevancy of reasons advanced for and against the Constraint
Principle. Opponents of originalism may be tempted to argue against shallow reasons
with deep ones. So, a consequentialist might argue that the shallow pro tanto reasons for
originalism are trumped by the consequentialist case for nonoriginalism. Of course,
consequentialist reasons would trump all other normative reasons if it could be shown
that consequentialism is the best normative theory and that it directly applies to the realm
of normative constitutional theory. But (and this is a big “but) this deep argument
requires deep foundations. One cannot simply assume that normative constitutional
consequentialism is true or correct, this must be shown. And the required showing will
require an extensive investigation of metaethics, moral philosophy, political philosophy,
and normative legal theory. Absent such a showing, the deployment of the first strategy
(relying on the true deep theory) against the third strategy (shallow reasons) simply will
not even get off the ground as a source of public justifications in a pluralist society.
Pursuing the third strategy entails that the debate over the Constraint Principle must be
conducted at the level of shallow reasons (Rawlsian public reasons or Sunsteinian
midlevel principles). It takes a shallow argument to beat a shallow argument—unless
you can persuade your fellow citizens that your deep argument is true: such persuasion
may require conversion. If your deep reasons are rooted in a particular religious or
secular conception of the good, you may need to convince your fellow citizens that they
should convert to your religion or to abandon their religion and accept your secular
perspective. Daunting task!
***
Given the difficulties attendant on the enterprise of constitutional theory, it is unlikely
that any position on the Constraint Principle (either pro or con) can be justified by
“knock down” arguments. But this does not entail that we cannot make progress in
constitutional theory. If the range of disagreement is narrowed and critical issues are
identified with precision, we will have made substantial progress. And if we can identify
what arguments are “live” in the sense that they have not yet been answered or
conceded, that is another form of progress. “Progress” might seem a very modest goal,
but in the context of contemporary constitutional theory, making real progress in debates
65
For an overview of the method of reflective equilibrium, see Norman Daniels, Reflective Equilibrium,
Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/entries/reflective-equilibrium/ (2011). For
discussion of reflective equilibrium in the context of legal theory, see Lawrence B. Solum, Situating
Political Liberalism, 69 CHI.-KENT L. REV. 549, 558 (1994). For Rawls’s explication, see JOHN RAWLS, A
THEORY OF JUSTICE 19-22, 48-51, 120, 432, 434, 579 (1971); John Rawls, Kantian Constructivism in
Moral Theory, 77 J. PHIL. 515, 534 (1980).
66
See infra Part III.B.1, p. 32.
31
Lawrence B. Solum
about originalism and living constitutional is actually an ambitious task. The realistic
alternative to progress is stalemate followed by the retreat of nonoriginalists and
originalists to their own camps—where all the debates are squabbles among true
believers.
***
B. The Methods of Justification
What general method is appropriate for the justification of a normative constitutional
theory? What role (if any) does the burden of persuasion play? We can begin our
investigation of these questions with the idea of reflective equilibrium.
1. Reflective Equilibrium
The notion of “reflective equilibrium,” familiar from the work of John Rawls, 67 has
been invoked as providing an appropriate method for resolving theoretical debates in
normative constitutional theory.68 How would this method apply to the question whether
to affirm the constraint principle in particular and originalism in general? We can begin
to answer this question by explicating the idea of reflective equilibrium.
a) The Idea of Reflective Equilibrium
Given the nature of the problems of constitutional theory, we should not expect that
the claims made about constitutional meaning will usually be justified by deductive
proof. Of course, deductive proof is likely to play a role at the level of supporting detail.
Some positions in constitutional theory may involve contradictions, and these positions
are demonstrably false. But in other cases, our starting points will be our prereflective
beliefs about various matters, ranging from the very particular and concrete to the general
and abstract. Such starting points will include relatively particular beliefs like “Brown v.
Board was rightly decided” and relatively abstract beliefs like “The rule of law values of
predictability, certainty, consistency, and publicity are an important component of
political morality.” On this picture, the method of constitutional theory starts with an
examination our prereflective beliefs and their relationships. Some beliefs may be
inconsistent. In that case, one or more of the beliefs may need to be reexamined and
revised. Gradually, our prereflective beliefs will become more refined and coherent. At
some stage, the theorist will begin to regard some of these beliefs as considered
judgments. A wholly successful constitutional theory will bring all of our considered
judgments into reflective equilibrium, a relationship of consistency and mutual support.
So far, the description of the method of reflective equilibrium has treated
constitutional theory as operating within the discourse of constitutional practice. But it is
67
JOHN RAWLS, A THEORY OF JUSTICE (1971).
See Richard H. Fallon, Arguing in Good Faith about the Constitution: Ideology, Methodology, and
Reflective Equilibrium, U. CHI. L. REV. (forthcoming 2016-17); Mitchell N. Berman, Reflective Equilibrium
and Constitutional Method: Lessons from John McCain and the Natural-Born Citizenship Clause in THE
CHALLENGE OF ORIGINALISM : THEORIES OF CONSTITUTIONAL INTERPRETATION 246 (Grant Huscroft &
Bradley Miller, eds., 2011).
68
32
The Constraint Principle
not the case that our beliefs about constitutional theory fall or stand independently of our
beliefs about other matters. Consider an analogy to tort law. Normative tort theory may
require recourse to general normative legal theory which in turn must be reconciled with
moral philosophy and political theory. Given a consequentialist approach to tort theory,
the normative evaluation of particular tort rules may best be accomplished by utilization
of the tools of economics, which themselves may involve formal techniques. Likewise,
originalist constitutional theory uses tools drawn from the philosophy of language and
theoretical linguistics as the basis for claims about meaning (e.g., the Fixation Thesis and
the Public Meaning Thesis). The Constraint Principle is a thesis in normative
constitutional theory, but its validity may depend in part on our beliefs about
nonconstitutional matters within law and a variety of other matters outside of law.
b) Intuitions and Considered Judgments
The method of reflective equilibrium may begin with our prereflective beliefs about
various matters. Let us all these initial beliefs “intuitions” and in the case of
constitutional matters “constitutional intuitions.” By using the term “intuition” I do not
mean to imply anything deep about the nature or epistemic basis of these beliefs. For
example, I do not mean to imply that our moral beliefs are produced by a faculty of moral
intuition.
We have beliefs about various matters (e.g., about facts, physics, and aesthetics), but
let us stipulate for present purposes that our constitutional intuitions are normative beliefs
that are constitutionally salient. Thus, our constitutional intuitions include beliefs about
the rightness or wrongness of particular cases (both real and hypothetical), about the
moral soundness of constitutional doctrines and principles, and about more general
normative matters, such as the normative attractiveness of democratic self-government,
the importance of the rule of law, and so forth.
The method of reflective equilibrium assumes that there is a contrast between our
intuitions (before reflection) and our “considered judgments.” Let us use this latter
phrase to designate beliefs that have been subject to reflection and evaluation. The
method of reflective equilibrium assumes that our intuitions are subject to revision for
various reasons. Thus, one might have an intuitive belief that Griswold v. Connecticut69
was correctly decided (as a matter of normative constitutional theory), but after reflection
one might conclude that this intuition was incorrect and that Griswold was an error. The
initial belief might be based on a first-order normative judgment that the decision of
married couples to use contraception should not be regulated by the state, and the
revision might be motivated by the second-order consideration that the courts should not
have the power to create constitutional rights absent authorization to do so by the
constitutional text. Given the complexity of constitutional law and theory, it would be
quite surprising if all of our prereflective beliefs were already in a state of perfect
consistency and mutual support—some of our beliefs will have to give way.
Our normative beliefs about constitutional matters exist at a variety of levels of
generality, ranging from beliefs about particular cases (actual and hypothetical) to
midlevel principles and constitutional doctrines, and highly abstract propositions about
political morality. Thus, we may have a belief that Brown v. Board was decided correctly,
69
381 U.S. 479 (1965).
33
Lawrence B. Solum
that the Equal Protection Clause forbids discrimination on the basis of race, and that a
principle of equal citizenship should govern the basic structure of society. To reach
reflective equilibrium, we seek to make our beliefs at all of these levels of generality
consistent and mutually supportive. Reflective equilibrium does not assume that beliefs
at one point on the continuum between specific considered judgments about particular
matters and general considered judgments about abstract matters have priority. We might
begin with a high level of confidence in our intuitions about a particular case and a lower
level of confidence in our intuition about an abstract principle, or vice versa.
c) Wide and Narrow Reflective Equilibrium
We can distinguish between wide and narrow reflective equilibrium, and this
distinction is particularly important with respect to constitutional matters in a pluralist
society. The distinction between wide and narrow reflective equilibrium can be
articulated in various ways, but for present purposes, let us stipulate to the following
definitions that are tailored to the constitutional context:
Narrow Reflective Equilibrium: The considered judgments of an individual on
constitutional theory are in narrow reflective equilibrium when they are consistent
and mutually supportive with each other.
Wide Reflective Equilibrium: The considered judgments of a political community are
in wide reflective when there is overlapping consensus among reasonable citizens on
considered judgments that support a family of constitutional theories that agree on
essential principles that are sufficiently similar to provide adequate guidance for
constitutional practice.
Should constitutional theory aim at wide or narrow reflective equilibrium? And what
difference does this distinction make to the justification of the Constraint Principle?
These questions are important and deep. The aim of the discussion that follows is to
provide reasons for believing that constitutional theory ought to employ the method of
wide reflective equilibrium.
Let us begin by examining the contrary position—that narrow reflective equilibrium
provides the correct method of justification for normative constitutional theory. From the
point of view of an individual (a judge or a constitutional theorist), narrow reflective
equilibrium will result in a constitutional theory that is coherent. The individual
theorist’s views will be consistent and mutually supporting. Narrow reflective
equilibrium will insure that the individual’s constitutional views are consistent with that
individual’s general views about political morality. Narrow reflective equilibrium begins
with individual intuitions and ends with reflective equilibrium among considered
judgments—from the point of view of the individual.
But it is clear that narrow reflective equilibrium does not provide the kind of
justification that is appropriate to a constitutional theory for a pluralist society in which
there is disagreement about deep matters—what Rawls called comprehensive religious
and philosophical conceptions of the good. Our discussion of deep and shallow
justifications shows why this is the case. If each individual seeks internal consistency,
then different individuals will reach reflective equilibrium on different constitutional
theories. But the primary role of a normative constitutional theory is not to provide
34
The Constraint Principle
internal consistency, but is instead to provide a shared basis for agreement on a
framework for the decision of constitutional cases. Narrow reflective equilibrium for
each individual will produce a plurality of inconsistent views corresponding to the
plurality of views about deep matters.
This point can be illustrated (albeit simplistically) by considering five hypothetical
justices, each of whom seeks narrow reflective equilibrium for their own constitutional
theory:
Justice Immanuel holds a comprehensive deontological theory of the good. that the
plain meaning of the constitution be observed strictly and without exception.
Justice Jeremy holds a comprehensive welfarist theory of the good. His
constitutional theory requires that each constitutional case be decided in the way that
produces the greatest sum of preference-satisfaction.
Justice Rosalind holds a comprehensive virtue-centered theory of the good. Her
constitutional theory requires that each constitutional case be decided in accord with
the virtue of practical wisdom so as to promote human flourishing.
Justice Frances holds a comprehensive religious conception of the good. Her
constitutional theory emphasizes the promotion of the true faith as the central aim of
constitutional decisionmaking.
Justice Gerald holds a theory of political morality in which equality of income and
resources is the highest political value. His constitutional theory emphasizes the
promotion of economic equality as the central aim of constitutional decisionmaking.
The justices are each in narrow reflective equilibrium with respect to their own deep
views, but none of the justices can affirm the method of any the others. Moreover,
members of the public who affirm a different comprehensive conception than any of the
justices will view that justice’s constitutional theory as both wrong and illegitimate:
internal consistency is not a sufficient basis for a shared agreement among citizens given
the fact of pluralism.
Now consider the contrasting case of wide reflective equilibrium. Wide reflective
equilibrium aims for consistency and mutual support among considered judgments that
can be stated as public reasons. Each Justice would take into account the fact of
pluralism and seek agreement on constitutional principles that can be affirmed on the
basis of considered judgments that can be shared by an overlapping consensus of
reasonable citizens. Narrow reflective equilibrium is structured so as to produce
constitutional dissensus—with different individuals and groups affirming different
constitutional theories. Wide reflective equilibrium aims at constitutional consensus;
more precisely, wide reflective equilibrium should aim at the greatest convergence among
constitutional views that is practicable. Practicability could be theorized in various ways;
for example, we might define the practicable by reference to Rawls’s notion of the
burdens of judgment.70
To reach wide reflective equilibrium, each of the hypothetical justices will need to
avoid direct reliance on their own comprehensive views and instead to seek for public
reasons or midlevel agreements. Given the fact of pluralism, constitutional theory
70
Rawls, Political Liberalism, supra note 70, at 36-37, 55-57.
35
Lawrence B. Solum
requires principled compromise (wide reflective equilibrium) and not internal consistency
(narrow reflective equilibrium).
2. The Burden of Persuasion and the Status Quo in Constitutional Theory
Many constitutional theorists are lawyers, and lawyers are trained to identify and
exploit burdens of persuasion. So, it is not surprising that ink has been spilled on the
question as to who bears the burden of persuasion in arguments about originalism.71 The
discussion that follows addresses the factors relevant to the burden of persuasion,
concluding that this issues is of little ultimate significance in debates about originalism
and living constitutionalism.
a) The Constraint Principle (in the Abstract) Is Intuitively Plausible and Rarely
Contested in Constitutional Practice
One might argue that the Constraint Principle should be viewed as the theoretical
status quo.72 The argument would be premised on the observation that as an abstract
matter it is very unusual for constitutional actors to explicitly deny the Constraint
Principle in words that are clear and unequivocal. No nominee for the Supreme Court is
likely to say, “Yes, I will view myself has having the power to override the original
meaning of the constitutional text.” It is much more likely that they will instead say
something like, “We are all originalists now."73
To the extent that the Supreme Court does make statements that are in tension with the
Constraint Principle, they are likely to be worded ambiguously—so that they can be
interpreted in a way that is consistent with the fidelity to the original meaning of the
constitutional text. Consider Home Building & Loan Ass'n v. Blaisdell—widely
considered to be a nonoriginalist opinion. Yet the Opinion of the Court states:
71
See, e.g., Jamal Greene, Originalism's Race Problem, 88 DENV. U. L. REV. 517 (2011) (“The
question, rather, is whether African-Americans have especially good reason to reject originalism, such that
selling African-Americans on originalism carries, and reasonably should carry, an unusually high burden of
persuasion.”); Randy E. Barnett, The Gravitational Force of Originalism, 82 FORDHAM L. REV. 411, 418
(2013) (“Let me conclude this section by noting that, to the extent we are engaging in a normative debate
about how we ought to interpret the Constitution, the burden of persuasion does not rest solely on
originalists. Nonoriginalists should also be able to articulate how they think “this Constitution” should be
interpreted and why.”); Ingrid Wuerth, An Originalism for Foreign Affairs?, 53 ST. LOUIS U. L.J. 5, 27
(2008) (stating, “originalists bear the burden of persuasion”); Frank Guliuzza III, The Practical Perils of an
Original Intent-Based Judicial Philosophy: Originalism and the Church-State Test Case, 42 DRAKE L.
REV. 343, 355 (1993) (“Even if originalists successfully make their case that, whenever possible, the
original intent should control interpretation, they have a substantial burden of proof to show a sweeping
original-intent-based jurisprudence is possible.”).
72
For an elegant and important discussion of these issues from a different perspective, see Baude, supra
note 29, at 2391 (“I am not saying that everything that has ever happened, or even everything that happens
today, is consistent with what an originalist would do. But I am saying that when you look at our current
legal commitments, as a whole, they can be reconciled with originalism. Indeed, not only can they be
reconciled, but originalism seems to best describe our current law. Our higher-order practices point toward
textualism and originalism. Our lower-order practices are messier, but once originalism is understood
inclusively, they actually seem to point toward inclusive originalism as well.”).
73
Cf. Confirmation Hearings, supra note 241 (statement of Elena Kagan).
36
The Constraint Principle
Nor is it helpful to attempt to draw a fine distinction between the intended
meaning of the words of the Constitution and their intended application. When we
consider the contract clause and the decisions which have expounded it in harmony
with the essential reserved power of the States to protect the security of their
peoples, we find no warrant for the conclusion that the clause has been warped by
these decisions from its proper significance, or that the founders of our Government
would have interpreted the clause differently had they had occasion to assume that
responsibility in the conditions of the later day. The vast body of law which has
been developed was unknown to the fathers, but it is believed to have preserved the
essential content and the spirit of the Constitution.74
Of course, sophisticated readers may conclude that such statements are disingenuous, but
for present purposes that possibility reinforces the argument: the Court is committed as a
matter of articulated principle to original meaning.
This commitment is consistent with avowals by the Court that factors other than
original meaning are relevant to the process of constitutional construction if one assumes
the Fact of Constitutional Underdeterminacy and the Interpretation-Construction
Distinction. In this regard, it is helpful to juxtapose originalism with the Multiple
Modalities theory—which is discussed in detail below.75 Recall that the theory posits the
existence of multiple modes of constitutional argumentation. In Bobbitt’s version of the
theory, the modalities are flat: there is no hierarchical structure or lexical ordering among
the modalities. It is this feature that renders the Multiple Modalities view a form of
nonoriginalism. But we can imagine a modified form of pluralism that incorporates the
Minimalist Version of the Constraint Principle. The modified version would reconcile
the observation that the Supreme Court does not disavow the Constraint Principle in the
abstract, with the fact that the Court does explicitly rely on alternative modalities of
constitutional argumentation.
Explicit rejection of the Constraint Principle is not the constitutional status quo. But
this is not the end of the matter, because constraint may be rejected implicitly rather than
explicitly
b) Deviations from the Constraint Principle May Be Common in Constitutional
Practice
If originalism is affirmed in principle, many scholars believe that the Court does not
observe the Constraint Principle in practice. Indeed, one of the motivations for the
development of originalist theory was the belief that the Warren Court (and its extension,
the Berger Court) had rendered constitutional decisions that reached outcomes that could
not be reconciled with the Constraint Principle (although they would have expressed their
motivations in a different vocabulary).
Determining whether and to what extent the Court has deviated from original meaning
in practice is a large task. To accomplish this task, we would need to determine the
original meaning of many of the particular provisions of the constitutional text and then
consider the Supreme Court’s jurisprudence in relationship to that meaning. There is a
range of opinion on this question.
74
75
Home Building & Loan Assn. v. Blaisdell, 290 U.S. 398, 443 (1934).
See infra Part VII.A.1, p. 105.
37
Lawrence B. Solum
We can illustrate the disagreement about the consistency of original meaning with
doctrine by considering the Commerce Clause as an example. On the one hand, some
originalists believe that large swaths of constitutional doctrine cannot be squared with
original meaning. If the meaning of “commerce” is limited to trade in tangible goods and
the meaning of “necessary and proper” is limited to ancillary powers, then it is possible
that substantial portions of the United States Code are beyond congressional power given
constraint by the original meaning.76 Likewise, it is possible that the original
understandings of the phrases “executive power,” “legislative power,” and “judicial
power” are inconsistent with much of the contemporary administrative state.
On the other hand, it is possible that the original meaning of “commerce” is quite
sparse, equivalent to something like “social interaction”77—an interpretation that would
result in a very expansive conception of national legislative power. Likewise, one might
believe that the phrases the original meaning of “executive,” “legislative,” and “judicial”
were substantially vague or open textured—and hence, that most of the modern
administrative state is within the outer limits of original meaning.
Despite the possibility that most or almost all of modern constitutional practice is
consistent with the Constraint Principle, let us accept, arguendo, the proposition that
substantial portions of contemporary constitutional practice are inconsistent with original
meaning. One might then argue that nonoriginalist practice is the status quo, and hence
that the burden of persuasion is on originalists to justify a substantial revision in
constitutional practice.
c) The Relationship of Principle and Practice with the Burden of Persuasion
Suppose for the sake of argument that as a matter of abstract constitutional principle,
constitutional actors accept something like Constraint as Consistency, but that substantial
portions of contemporary constitutional practice violate that principle. How would that
state of affairs affect the burden of persuasion with respect to the Constraint Principle?
This question is an instantiation of a very general problem in intellectual and practical
life. One can be committed to principles in the abstract that one breaches in practice.
Consider some prosaic examples:
•
•
•
One is committed to the abstract principle that lying is almost always wrong, but
one tells white lies to avoid hurting others feelings and sometimes makes lying
excuses to extract oneself from social occasions that would be unpleasant or
boring.
One is committed to the abstract principle that one should recycle to the
maximum extent feasible, but one frequently puts recyclable material in the
nonrecycling bin for no good reason other than inattention and sloth.
One is committed to the abstract principle that one should not gossip about one’s
colleagues, but one does in fact repeat gossip about colleagues, especially when
76
See Randy E. Barnett, New Evidence of the Original Meaning of the Commerce Clause,55 ARK. L.
REV. 847-899 (2003); Randy E. Barnett, The Original Meaning of the Commerce Clause, 68 U. CHI. L.
REV. 101-147 (2001); Randy E. Barnett, Jack Balkin's Interaction Theory of "Commerce", 2012 U. ILL. L.
REV. 623-667.
77
See Balkin, supra note 35.
38
The Constraint Principle
the gossiped-about conduct is humorous or can serve as an occasion for
schadenfreude.
In cases like these, we can account for the variance between principle and practice in
many ways. One possibility is akrasia or weakness of will, and the response might be to
resolve to act in accord with one’s principles and the adoption of mechanisms of selfcontrol that will advance that resolve. Another possibility is that one is mistaken about
the abstract principle: lying is morally permissible in many circumstances, recycling is
not truly important, and even malicious gossip is permissible much of the time.
But there is another possibility. Perhaps, one never believed in the abstract principles
in the first place. One says, “lying is wrong,” but actually believes it is permissible. One
professes a commitment to recycling, but does not believe recycling is worth all the
trouble. One publicly opposes gossip, but actually thinks gossip is ok so long as it is
discreet. In the public context, there is a more high-minded version of these stances, the
so-called “noble lie.” In other words, the gap between principle and practice can be
explained by hypocrisy.
The same kind options exist with respect to our beliefs about constitutional theory.
We may believe in our abstract commitment to constraint, but fail to live up to it when it
conflicts with our political preference. Or we might be mistaken about the correctness of
the Constraint Principle. Or the idea of constraint might be hypocritical (or more
generously, a noble lie78).
It is not clear that the idea of a burden of persuasion or a presumption in favor of the
status quo is particularly helpful in sorting out the conflict between principle and practice.
Rather than wrangle about where the burden should be placed, the normative significance
of practice and principle can and should be addressed directly. The method of reflective
equilibrium provides a method for reconciling considered judgments about practice and
principle. To the extent that they are inconsistent, revision is required to achieve
consistency.
d) The Significance of Pro tanto Reasons for the Constraint Principle
There is one more aspect of the burden of persuasion that requires clarification. As
will become clear, it actually should not be controversial that there are pro tanto reasons
for affirming the Constraint Principle. For those unfamiliar with the terminology, here is
a standard explication of the notion of a pro tanto reason:
Pro tanto Reason: A pro tanto reason to φ is a reason that genuinely speaks in
favour of φ-ing, but, while a pro tanto reason to φ favours φ-ing, it may not do
so decisively: the overall balance of reasons may direct one to do otherwise.79
In the context of debates about originalism, one believes that there are pro tanto reasons
for affirming the Constraint Principle so long as there are genuine reasons to follow the
constitutional text and not to act inconsistently with the text—even if one does not accept
these reasons as decisive. In the discussion that follows, I will argue that the real locus of
78
The idea of the “noble lie” is attributed to Plato. See Catalin Partenie, Plato's Myths, Stanford
Encyclopedia of Philosophy, http://plato.stanford.edu/entries/plato-myths/ (2014).
79
James Lenman, Reasons for Action: Justification vs. Explanation, Stanford Encyclopedia of
Philosophy, http://plato.stanford.edu/entries/reasons-just-vs-expl/ (2009).
39
Lawrence B. Solum
controversy is not the question whether there are pro tanto reasons for originalism, but
whether the balance of reasons (pro and con) favors the Constraint Principle. For this
reason, assessment of the countervailing reasons is likely to be the crucial area of
controversy.
e) A Final Word on the Significance of Burdens and Presumptions
In the end, it is not clear that the burden of persuasion should play any important role
in the debate about the Constraint Principle. If one conceives of the burden of persuasion
as a tie-breaker, it seems extraordinarily unlikely that anyone will see the debate over the
Constraint Principle as a tie. If the burden of persuasion is supposed to be more than a
tie-breaker and reflects a “weighty presumption” one way or the other, then some
argument must be offered that provides the weight to the presumption. But that argument
would then be a substantive reason for or against the Constraint Principle—and calling it
a presumption that creates an elevated burden of persuasion seems to add complexity and
subtract clarity from the debate.
This abstract point can be made more concrete by considering a particular substantive
argument for an elevated burden of persuasion. Opponents of the Constraint Principle
might argue that the Constraint Principle must be justified by especially compelling
reasons, because implementation of the principle would disrupt existing constitutional
practice: “If you want to roll back the New Deal, you are going to have provide awfully
good reasons.” But this is simply an argument against rolling back the New Deal on the
basis of the Constraint Principle. Recasting the argument in terms of the burden of
persuasion obfuscates rather than illuminates the nature of the argument.
***
What is the “status quo” with respect to methods of constitutional interpretation and
construction in constitutional practice? Honestly, I think the answer to this question is
that current practice is an indefensible and incoherent mishmash. Some judges are
originalists some of the time. Some judges pay a lot of attention to precedent, but many
are very loose about the doctrine of precedent and many do not have a coherent theory of
stare decisis. Hardly any judges would endorse Dworkin’s theory, but at least some
judges some of the time employ a method that balances considerations of fit and
justification. Some judges are pragmatists, but very few would endorse pragmatism as a
general approach. In the lower courts, the interaction the luck of the draw (e.g. the
assignment of cases to trial judges and the assignment of appeals to randomly created
panels of intermediate appellate judges) means that the method of constitutional
interpretation and construction is almost entirely unpredictable ex ante and may well be
incomprehensible ex post.
That sounds bad, but constitutional practice is not limited to the judicial branch. If
judicial methods of constitutional interpretation and construction are disordered, the
situation is even worse outside the courts. Executive officials and legislators are
responsible for constitutional interpretation and construction in a wide variety of cases
that do not or cannot reach the courts, but in many cases, they lack the institutional
resources to seriously consider constitutional cases—much less to reflect on
constitutional method.
40
The Constraint Principle
My label for the theory corresponding to this state of affairs is “Constitutional
Eclecticism”—a form of constitutional antitheory. If there is a status quo method of
constitutional interpretation and construction it is eclecticism. No one should be pleased
with the status quo with respect to methods of constitutional interpretation and
construction. Even constitutional pragmatists should be dismayed. If there is a
“presumption” in favor of the current system, it is at most a “bursting bubble”—and the
bubble may well be “self-bursting.” I will have more to say about the status quo below.80
***
C. The Structure of Argumentation
This section addresses a variety of issues regarding the structure of argumentation in
constitutional theory. We begin with the role of consistency.
1. The Implications of Consistency
Inconsistency in normative argumentation is obviously problematic. The case for the
Constraint Principle must be consistent to command our assent—and the same standard
applies to arguments against constraint. In particular, the opponents of constraint are at
risk of two serious forms of inconsistency. The first danger of inconsistency arises for
nonoriginalists who reject the Constraint Principle but endorse the Contribution
Principle—that is, they affirm the weaker (but still substantial) requirement that
constitutional practice should take original meaning into account and it give it significant
weight. The second danger of inconsistency arises for nonoriginalists who reject the
Constraint Principle as applied to the Constitution, but accept analogous forms of
constraint for other texts, such as statutes, cases, rules, or regulations.
Consistency is required by the method of reflective equilibrium, but one might reject
that method and still affirm the value of consistency. For example, one might believe that
constitutional theory can proceed by the method of geometry, starting with first principles
and then deducing conclusions about constitutional interpretation and construction—
obviously, this method requires consistency. Even particularists seek to avoid
inconsistency, although they do this by foreswearing generality rather than affirming a
consistent set of abstract principles.
a) Arguments for the Rejection of the Constraint Principle Must Be Consistent
with Arguments for the Contribution Principle (For Those Who Affirm That
Principle)
Some nonoriginalists may believe that the original meaning of the constitutional text
should play no role in constitutional practice. This position may not be popular, but it
offers the advantage of theoretical consistency.
Other originalists embrace some role for original meaning and hence accept what we
have called the Contribution Principle—the claim that the original meaning of the
constitutional text should play an important role in constitutional practice and that
departures from original meaning should be given substantial (or weighty) justifications.
80
See infra Part VII.B, p. 127.
41
Lawrence B. Solum
Accepting contribution but rejecting constraint requires a bit of a high wire act—
embracing the normative justifications for contribution but rejecting the case for the
Constraint Principle, even it is minimalist form, Constraint as Consistency. If you lean
too far to the left, you end up without any justification for contribution and hence are at
risk of embracing the unpopular and intuitively implausible position that the
constitutional text should play no role in constitutional law. If you lean too far to the
right, you end up embracing constraint and hence become an originalist.
In particular, nonoriginalists are at risk of embracing arguments against the Constraint
Principle that are inconsistent with the normative claim that original meaning should play
a role in constitutional practice. If the original meaning of the constitution is the product
of a racist and sexist dead hand, utterly lacking in legitimacy, and if that meaning is
unknowable to boot, then the logical implication would seem to be the rejection of the
written constitution as a source of law. And if there are very good reasons for the text to
play an important role in constitutional practice, why do they stop short of constraint?
These questions suggest discussion of the Constraint Principle would be clarified if
nonoriginalists were to address them head on. For example, opponents of constraint
might answer the following questions:
1. What role should the original meaning of the constitutional text play in
constitutional practice?
2. If the answer to question 1 is not “none,” then what are the affirmative reasons
that justify the specified role?
3. Do the reasons provided in your answer to 2 provide pro tanto justification for
the Constraint Principle? If not, why not?
4. What are your normative reasons for rejecting the Constraint Principle?
5. Do the reasons specified in 4 also apply to the role for original meaning
specified in your answer to 1? If not, why not?
It may well be possible for nonoriginalists to affirm contribution and deny constraint on
the basis of a wholly consistent set of reasons, but if these questions are not addressed,
then nonoriginalists are at risk of inconsistency of which they may not even be aware.
b) Arguments for the Rejection of Constitutional Constraint Must Be Consistent
with the Reasons for Rejecting Antitextualist Anarchism (For Those Who Do
Reject It)
Some nonoriginalists may reject originalism on the basis of a much more general
position in general legal theory. Let us stipulate a definition for “antitextualist
anarchism” as follows:
Antitextualist Anarchism: The reasons for action provided by the communicative
content of legal texts are not peremptory reasons for action, and therefore, one
should act on the basis of all the reasons that are salient.
Antitextualist Anarchists reject constitutional constraint, but they also reject constraint on
the basis of all legal texts, including statutes, judicial opinions, rules, regulations,
contracts, wills, and so forth. Antitextualist Anarchy is unlikely to run into problems of
inconsistency with respect to the Constraint Principle. If you reject the authority of
42
The Constraint Principle
statutes, rules, regulations, written judicial decisions, contracts, and wills, then your
rejection of the Constraint Principle is likely to be part of a consistent normative theory.
But it seems likely that many nonoriginalists reject Antitextualist Anarchism, because
they do affirm the idea that the communicative content of statutes, cases, and other legal
texts provides either peremptory or weighty reasons for action. For example, some
nonoriginalists might argue against originalism on the ground that the original meaning is
inconsistent with the Supreme Court’s opinion in a canonical case (such as Brown v.
Board of Education81): this argument assumes that the Supreme Court’s opinion should
have constraining force. Similarly, Thayerian objectors to originalism may believe that
statutes should have constraining force. Unlike Antitextualist Anarchists, these objectors
to the Constraint Principle face consistency problems.
Consider the following example: if one objected to originalism on the ground that the
Constitution of 1789 was ratified by a process that disenfranchised women and members
of many ethnic minority groups, but affirmed that the McCulloch v. Maryland82 or the
Sherman Act83 should have constraining force, one would face an obvious problem of
inconsistency—since neither McCulloch nor the Sherman Act was the outcome of an
inclusive process. These positions might be made consistent through careful qualification
of the claims, but those who take these positions owe us an explanation that makes the
basis for consistency clear.
2. The Reciprocity Principle (No Double Standards)
Arguments for positions in constitutional theory should satisfy what we can call the
“Reciprocity Principle.” Before we state the principle, it may be helpful to give an
informal explanation. Given the high normative stakes in constitutional law (and hence
theory), there may be a tendency to embrace double standards—applying one standard to
your ideological friends and a different standard to your ideological enemies.
The following examples illustrate the general idea. Suppose that you are an
ideological progressive. You might be tempted to praise living constitutionalism when it
is practiced by progressive justices, but apply a different standard to conservative
members of the court. Likewise, conservatives might criticize progressives for departing
from the original meaning of the constitutional text, but praise conservatives when they
do the same thing.
We can explicate this kind of double standard by imagining the following two
statements:
Statement One: “The conservative decision in this case is awful, because five
Justices advanced their own social vision and ignored both precedent and
constitutional text.”
Statement Two: “The progressive decision in this case is wonderful, because five
Justices advanced their own social vision and overcame precedent and the
constitutional text.”
81
347 U.S. 483 (1954).
17 U.S. 316 (1819).
83
26 Stat. 209 (1890).
82
43
Lawrence B. Solum
These two statements evince a double standard, because of the reasons offered for praise
and blame. The very same reason “advancing their social vision” is taken as illegitimate
for conservative decisions but as praiseworthy for progressive decisions. The double
standard disappears if we make the following substitution for Statement One:
Statement One Prime: “The decision in this case is awful, because five Justices
advanced conservative values and not progressive values.”
Statement One Prime does not employ a double standard. It accepts the legitimacy of the
method (disregarding precedent and text) but criticizes the substance of the values. Of
course, One Prime will draw the following reply: “Your argument only provides reasons
for those who affirm progressive values; for those who affirm conservative values, the
argument actually provides a justification at the level of principle for acting in accord
with those values.” In other words, Statement One Prime might be affirmed from the
perspective of narrow reflective equilibrium but it fails the test of wide reflective
equilibrium.
The Reciprocity Principle forbids double standards. We can formulate the principle as
follows:
The Reciprocity Principle: Normative arguments in constitutional theory should not
impose standards for the legitimacy of constitutional methods that are inconsistent
and vary with the ideological orientation of the constitutional actor or the ideological
valence of the constitutional action at issue.
Arguments for and against the Constraint Principle should not violate the Reciprocity
Principle. The Reciprocity Principle is a natural complement to the method of wide
reflective equilibrium, but one could affirm this principle while rejecting reflective
equilibrium as a general method for constitutional theory.
Some scholars may reject the Reciprocity Principle on the ground that reciprocity is
only warranted when all or almost all of the relevant constitutional actors are willing to
act on the basis of the principle.84 Thus, progressives might argue that they are excused
from reciprocity on the ground that conservatives cannot be trusted to observe the
principle—and vice versa. The problem of trust in constitutional theory is an important
and underexplored topic. The problem arises because there are respects in which
constitutional theory can be viewed as a prisoners’ dilemma. Suppose that both left and
right wing scholars would prefer to observe the Reciprocity Principle, but neither group
would be willing to observe reciprocity unilaterally. Unless trust can be established, both
groups would be rationale to defect from reciprocity unilaterally.
If constitutional theorizing is viewed as a political contest with ideologically defined
winners and losers, it is difficult to resist the logic of the prisoners’ dilemma. But it is not
clear that it makes sense to view constitutional theory in this way. Constitutional
theorists should aim at the truth—and this aim might be taken as a fundamental defining
characteristic of the activity of theorizing about the constitution. On this alternative view,
rejection of the Reciprocity Principle in the service of ideological ends would be viewed
as inconsistent with the values of constitutional scholarship. The view that scholarship is
84
This position was suggested to me by a somewhat different point made by Mark Tushnet. See Mark
Tushnet, Consistency and Aggressive Liberal Constitutionalism, Balkinization, May 14, 2016,
http://balkin.blogspot.com/2016/05/consistency-and-aggressive-liberal.html.
44
The Constraint Principle
the continuation of politics by other means has a deeply corrosive effect on the search for
truth. If truth is the aim of scholarship, then the proper response to the breaches of the
Reciprocity Principle by others is to identify and criticize the violations.
3. Hard Cases and Easy Cases
Constitutional theories are sometimes evaluated on the basis of their ability to handle
so-called “hard cases.” It is not entirely clear what “hard cases” are, but one way that this
phrase is used focuses on cases that are legally hard—where there is more than one
outcome that is supported by good legal reasons—this seems to be the sense in which
Dworkin used the phrase in his famous essay, Hard Cases.85 But in the saying “hard
cases make bad law,” the phrase seems to be used in a different sense. As Judge Rolfe
put it in the first recorded use of the phrase, “This is one of those unfortunate cases... in
which, it is, no doubt, a hardship upon the plaintiff to be without a remedy, but by that
consideration we ought not to be influenced. Hard cases, it has frequently been observed,
are apt to introduce bad law.”86 And Holmes echoed this sentiment in Northern
Securities Co. v. United States, “Great cases like hard cases make bad law. For great cases
are called great, not by reason of their importance... but because of some accident of
immediate overwhelming interest which appeals to the feelings and distorts the
judgment.”87 Rolfe and Holmes seem to be referring to cases that are morally hard,
where the legally correct outcome is normatively problematic. Correspondingly, cases
can be legally easy or morally easy.
Sometimes constitutional theory proceeds as if the only important cases are hard cases.
There is something to this idea. We want our theories provide “guidance where guidance
is needed”88 and guidance is needed in hard cases.
Given the method of reflective equilibrium,89 however, it would be a serious mistake
to focus only on hard cases. Some cases are easy. The constitutional text provides that
each state shall have two and only two senators. Bicameralism requires that both the
House and the Senate enact legislation before it can become law. Although there may be
a very good case for changing these requirements, almost all nonoriginalists (who are not
anticonstitutionalists) seem to agree that these hardwired provisions should not be
overridden by an amending construction.90 It is not enough for nonoriginalists to simply
stipulate that the Constraint Principle is fine and dandy when it comes to provisions like
this. They need an explanation as to why this is so. And that explanation must then be
consistent with their analysis of hard cases. Saying “easy cases are easy, so no
explanation is required” is an avoidance technique—and by its own terms, this avoidance
technique does not provide an explanation.
The argument in the prior paragraph needs to be qualified. Some nonoriginalists
embrace constitutional antitheory. And some forms of antitheory allow for inconsistency
at the level of theory, because of a metaprinciple (the priority of the particular) or a
85
Ronald Dworkin, Hard Cases, 88 Harvard Law Review 1057 (1975) reprinted in RONALD DWORKIN,
TAKING RIGHTS SERIOUSLY (Harvard University Press, 1977).
86
Winterbottom v Wright, 10 M&W 109 (Exchequer of Pleas 1842).
87
Northern Securities Co. v. United States, 193 U.S. 197 (1904),
88
JOHN RAWLS, A THEORY OF JUSTICE 19 (rev. ed. 1999).
89
See supra Part III.A.2.
90
See Strauss, supra note 221, at 56-57.
45
Lawrence B. Solum
metatheory (legal pragmatism). But these escape routes are not available to all or even
most nonoriginalists. The important point is that inconsistency demands explanation at
some level—theoretical or metatheoretical.
4. Formalism and Realism
Because of the pervasive influence of legal realism on contemporary legal thought,
many legal scholars affirm a cluster of beliefs that may seem “obviously true” but in fact
rely on controversial premises about the nature and justification of legal rules. Here are
some of the assumptions and their impact on thinking about originalism and the
Constraint Principle.
Indeterminacy: The realist assumption that law is radically indeterminate
undermines the possibility of constraint; very substantial underdeterminacy renders
constraint largely irrelevant.
Realist Justification of Legal Norms: The realist assumption that substantive legal
norms must be justified on the basis on arguments of policy (consequentialism) or
principle (fairness) rules out the possibility that the content of constitutional doctrine
could be justified on the basis of the communicative content of the constitutional
text.
Fact Sensitivity: The realist assumption that cases are decided in response to the
normative characteristics particular facts is inconsistent with the idea that cases
should be decided on the basis of general rules derived from the constitutional text.
Debates about these realist assumptions are important and deep; those debates cannot be
resolved as a side issue in the context of making the case for the Constraint Principle.
But neither can these assumptions be taken as self-evident truths. In the context of
debates about originalism, the most sensible way to proceed is to consider the merits of
the assumptions in the constitutional context. Thus, we can ask the questions whether
radical indeterminacy holds in the constitutional realm, whether is it possible to justify
constitutional norms by appeal to the constitutional text, and whether constitutional
adjudication can avoid the ad hoc particularism entailed by fact sensitivity.
5. “It Takes a Theory to Beat a Theory”
The evaluation of the case for originalism necessarily requires that it be compared to
some alternative. We can summarize this notion using the familiar aphorism, “It takes a
theory to beat a theory,” but the aphorism is too simple to capture fully the position that I
will outline here.91
The idea that originalism must be evaluated comparatively was first (to my
knowledge) articulated by Justice Scalia in the following passage:
Apart from the frailty of its theoretical underpinning, nonoriginalism confronts a
practical difficulty reminiscent of the truism of elective politics that “You can't beat
91
Lawrence B. Solum, It Takes a Theory to Beat a Theory, Legal Theory Lexicon,
http://lsolum.typepad.com/legal_theory_lexicon/2006/08/legal_theory_le_1.html (Last modified on July 12,
2015.).
46
The Constraint Principle
somebody with nobody.”' It is not enough to demonstrate that the other fellow's
candidate (originalism) is no good; one must also agree upon another candidate to
replace him. Just as it is not very meaningful for a voter to vote “non-Reagan,”' it is
not very helpful to tell a judge to be a “non-originalist.”' If the law is to make any
attempt at consistency and predictability, surely there must be general agreement
not only that judges reject one exegetical approach (originalism), but that they
adopt another. And it is hard to discern any emerging consensus among the
nonoriginalists as to what this might be. Are the “fundamental values”' that replace
original meaning to be derived from the philosophy of Plato, or of Locke, or Mills,
or Rawls, or perhaps from the latest Gallup poll? This is not to say that originalists
are in entire agreement as to what the nature of their methodology is; as I shall
mention shortly, there are some significant differences. But as its name suggests, it
by and large represents a coherent approach, or at least an agreed-upon point of
departure. As the name “‘nonoriginalism”’ suggests (and I know no other, more
precise term by which this school of exegesis can be described), it represents
agreement on nothing except what is the wrong approach.92
This passage has been criticized in several ways. One criticism attempts to show that
originalism lacks a conceptual core,93 but this argument is mistaken on the facts:
originalism is a family of constitutional theories united around the core ideas of fixation
and constraint.
Another criticism focuses on the notion that you can rationally affirm a nonoriginalist
construction of a particular provision of the Constitution without offering a full-blown
constitutional theory.94 It is difficult to know what to make of this criticism as it might be
applied to the Constraint Principle. In my opinion this criticism is most plausibly
understood as an argument for particularism as an antitheoretical position: if read in this
way, it is correct in the following sense: our aphorism needs to be modified to take into
account antitheoretical alternatives to originalism, including legal particularism and legal
pragmatism. The modified aphorism might read as follows: “It takes a theory or an
antitheory to beat a theory.”
Constitutional actors have many choices. The could adopt originalism including the
Constraint Principle as their approach to constitutional interpretation and construction.
Or they could reject the need for principle and theory altogether, by embracing
92
Antonin Scalia, Originalism: The Lesser Evil, 57 U. CIN. L. REV. 849, 855 (1989).
Thomas B. Colby & Peter J. Smith, Living Originalism, 59 DUKE L. J. 239, 269-72 (2009). See also
James E. Fleming, Living Originalism and Living Constitutionalism As Moral Readings of the American
Constitution, 92 B.U. L. REV. 1171, 1174 (2012).
94
See Mitchell N. Berman, Originalism Is Bunk, 84 N.Y.U. L. REV. 1, 85 (2009) (“In short, those who
would defend Originalism on the ground that opponents haven't proposed something determinate in its
stead and that “it takes a theory” might be confusing the need to select a meaning with the need to select a
protocol for selecting that meaning. To hope for the latter is human nature; to demand it is simply
unreasonable. Understanding what law is hard. So too is articulating a fully satisfactory normative account
of judicial constitutional interpretation.”); James E. Ryan, Does It Take A Theory? Originalism, Active
Liberty, and Minimalism, 58 STAN. L. REV. 1623, 1653 (2006) (“Perhaps only by ignoring the complexities
of actual cases and the inevitable clash of contesting principles they occasion, and by fudging about the
guidance offered by general theories, could one claim that a general theory of interpretation is truly useful
in deciding cases.”).
93
47
Lawrence B. Solum
particularism or legal pragmatism. Or they could adopt a rival theoretical view, such as
the Multiple Modalities Theory, Common Law Constitutionalism, or Dworkin’s theory.
But one cannot simply reject originalism as one’s approach to constitutional practice.
Judges and officials are necessarily in the business of making constitutional choice.
Every possible way of proceeding, even flipping a coin or acting instinctively, involves
some method of constitutional choice that can be compared to originalism. Critics of
originalism owe us an explanation of their approach to constitutional practice in general.
The case for rejecting the Constraint Principle will necessarily be incomplete unless some
alternative is offered.
Consider one final objection to our revised aphorism: “It takes a theory or an
antitheory to beat a theory”—but it doesn’t take anything to beat a really bad theory.95
That’s right. If someone proposed that constitutional practice should be guided by
astrology, numerology, or a three-year old wielding an Ouija board, one would be
justified in saying something like this:
I don’t know how the Supreme Court decides cases now, but I am sure it is
better than decision by a three-year old wielding an Ouija board. That’s just batshit
crazy. Anything remotely reasonable would be better than that.
This example suggests a test for the threshold level of plausibility that a theory must pass
before we ask its critics for their alternative. Let us call this the “Batshit Crazy Test.”96
The Constraint Principle is not batshit crazy.
One last point about Scalia’s argument. Scalia made two points. The first point is that
critics of originalism must offer an alternative—about that Scalia is right. The second
point is that the failure of nonoriginalists to agree on a single position somehow shows
that nonoriginalism must be incorrect: that point is wrong—as explained below.97
Nonoriginalism wins (as a matter of constitutional theory) if any of the nonoriginalist
theories (or antitheories) in the feasible choice set is better than the Constraint Principle.
***
There is a tendency in debates about legal theory to make a move that can be
summarized by the phrase, “I don’t buy it.” This is phrase serves as a conversation
stopping move. As I understand it, “I don’t buy it” is shorthand for something like the
following: “The burden of persuasion is on you to persuade me. Otherwise, I am entitled
to retain my opinion and to dismiss your theory as ‘interesting’ but ‘unconvincing.’ I do
not need to respond further unless and until you offer an argument that I find
convincing.” If I can sincerely respond to your argument with “I don’t buy it, then the
debate between us is over.”
95
See Lynn Stout, The Troubling Question of Corporate Purpose, 3 ACCOUNTING, ECONOMICS, AND
LAW: AEL; A CONVIVIUM 61, 64 (2013) ("It doesn’t take a theory to beat a really bad theory. A really bad
theory can beat itself, because following a really bad theory leads to bad results.”).
96
See “Batshit Crazy” in the Urban Dictionary,
http://www.urbandictionary.com/define.php?term=batshit+crazy (“A person who is batshit crazy is so nuts
that not only is their belfry full of bats, but so many bats have been there for so long that the belfry is
coated in batshit. Hence, the craziest of crazy people are BATSHIT CRAZY.”).
97
See infra Part V.D, p. 82.
48
The Constraint Principle
Of course, there is something that is absolutely correct about this move. One should
not adopt any position in constitutional theory unless one is at least provisionally
convinced that the position is true or correct. But that does not entail the further
conclusion that arguments that are not subjectively persuasive do not require answers.
Persuasion is a psychological concept (in this context). In the face of strong arguments,
it is possible to say something like, “I do not know how to answer this argument, but I am
(subjectively) sure that my prior position is correct, so there must be some answer—even
if I don’t know what that answer might be.”
It is a serious mistake to use the psychological standard of personal persuasion as the
objective standard for which arguments in constitutional theory require answers before
one has good reasons for denying that the arguments are true or correct. If constitutional
theorists all adopt the psychological standard of subjective persuasion, there is little
chance for progress in constitutional theory. Indeed, constitutional theory would become
a matter of rhetorical posturing rather than serous scholarship.
Well-reasoned arguments invite acceptance or answer, withholding the former does
not excuse a refusal to the provide the latter.
***
6. Time and Possibility in Constitutional Theory
Constitutional theory can proceed in a very abstract manner that focuses on ideal
constitutions for near perfect societies. The debate about originalism and living
constitutionalism is not that kind of constitutional theory. The case for the Constraint
Principle is made in the constitutional here and now; participants in the debate make
explicit or implicit assumptions about what is feasible in our historical circumstances.
Even though the debate takes place in the constitutional present, many of the arguments
appeal to the constitutional past or the constitutional future. How would Brown v. Board
have been decided by public meaning originalism? What effect would originalism have
on the ability of Congress to respond to global climate change in the future?
How can constitutional theory create a framework for analyzing the roles of possibility
and time in theoretical discourse? Two theoretical constructs can make these problems
more tractable: call the first of these constructs “Constitutional Possibility” and the
second “Constitutional Time.”
a) Constitutional Possibility: Possible Worlds, Feasibility, and the Role of Ideal
Theory
Some proposals for the reform of constitutional practice are feasible; others seem like
pie in the sky. Constitutional theorists may have intuitive judgments about feasibility vel
non, but how can those intuitions be made rigorous and transparent? Borrowing from
49
Lawrence B. Solum
philosophy, we can use the idea of possible world semantics98 to provide an explicit and
precise structure for talk about Constitutional Possibility. 99
The basic idea of possible worlds semantics is that talk about possibility and necessity
can be translated into talk about possible worlds. A possible world is not a planet; it is a
state of the whole universe, but for practical purposes, possible worlds talk usually
focuses on particular parts of a possible world. Thus, we can talk about the possible
world in which Donald Trump did not run for the presidency in 2016 or the possible
world in which Hurricane Katrina missed New Orleans.
A proposition, P, is necessarily true, if it is true in all possible worlds; similarly, P is
possibly true if P is true in at least one possible world. Different senses of possibility can
be specified by sets of possible worlds. Thus, logical possibility ranges across all worlds
in which the conditions specified by a given logic hold: thus, on standard logics, there are
no logically possible worlds in which a logical contradiction holds.
In possible worlds semantics, the world “actual” designates this world (the one that we
are in now). Worlds that share the same laws of physics with the actual world are
nomologically accessible from the actual world. Worlds that share the history of the
actual up until now are historically accessible from the actual world. Thus, historically
and nomologically accessible possible worlds are those worlds that share the history of
the actual world up to now and in which the same laws of physics, chemistry, and so forth
apply.
Some possible worlds are closer to the actual world than others. For example, a
possible world that shares the history of the actual world through sometime in 2015, but
in which Donald Trump did not enter the Republican presidential primaries is closer to
the actual world than one in which Martians invaded on September 10, 2011 and
subsequently enslaved all humans. We can use distance as a metaphor for relationships
between possible worlds: more similar worlds are closer, less similar worlds are farther
apart.
Constitutional theories do not need to range across all possible worlds. There is a
logically possible world in which the Supreme Court adopts Unconstrained Thayerianism
and because of a bizarre (but logically possible) law of physics, this results in the entire
universe disappearing. This bizarre world is logically possible, but this kind possibility is
simply not relevant to constitutional theory. The set of worlds that is relevant to
98
. The idea of possible worlds was introduced by Leibniz. See Gottfried Wilhelm Freiherr von
Leibniz, The Theodicy: Abridgement of the Argument Reduced to Syllogistic Form, in LEIBNIZ: SELECTIONS
509, 509–11 (Philip P. Weiner ed., 1951). Leibniz used the idea of a possible world in answer to the
argument against the existence of good from the problem of evil. See id. at 511. The argument is not
proven, Leibniz maintained, until it is shown that the actual world is not the best of all possible worlds. Id.;
see generally JOHN DIVERS, POSSIBLE WORLDS (2002) (providing a comprehensive introduction to possible
worlds semantics and the metaphysics of modality); SAUL A. KRIPKE, NAMING AND NECESSITY (1981)
(discussing model theoretic study of modal logic “possible worlds” semantics); DAVID LEWIS, ON THE
PLURALITY OF WORLDS (1986) (defending modal realism’s view that our world is one of many, each with
its own inhabitants); ALVIN PLANTINGA, ESSAYS IN THE METAPHYSICS OF MODALITY (Matthew Davidson
ed., 2003).
99
I am grateful to Christopher Green for focusing me on the importance of possible worlds semantics as
a tool for untangling the complex issues raised by the role of possibility in debates about constitutional
theory.
50
The Constraint Principle
theorizing about methods of constitutional interpretation and construction is limited to
worlds that are like ours in most important respects.
One set of worlds that is particularly important consists of the worlds that are very
close to our own through the present moment and that share the laws of physics,
psychology, and sociology that govern the actual world. These worlds enable us to define
what we can call the feasible choice set for constitutional theory. Constitutional theories
that would require massive changes in human psychology are not feasible. Constitutional
theories that could be implemented if a critical mass of constitutional actors were
convinced of their desirability would be inside the feasible choice set.
One more distinction completes our discussion of constitutional possibility. We can
distinguish between ideal, nonideal, and partially ideal political theory. This distinction
can be approached by borrowing the distinction between “ideal” and “nonideal” theory
from John Rawls.100 By “ideal theory,” Rawls means to refer to a moral or political
theory that satisfies a condition of “full compliance” or “strict compliance.” In the
constitutional context, ideal theory addresses the following question:
What constitution ought a society to adopt for the purposes of designing its basic
legal structure on the condition that all of the institutions in society conform to the
constitution?
In other words, we can ask what constitution we ought to adopt, assuming that perfect
compliance—each branch of government always respecting the limits on its power and
the rights (if any) that the constitution confers on individuals. By way of contrast, we can
ask questions of nonideal constitutional theory:
What constitution ought a society to adopt for the purposes of designing its basic
legal structure on the condition that the institutions of society will violate the
constitution to the extent, and under the circumstances, that are predicted on the
best understandings of human psychology and political science?
That is, we might assume that constitutional actors will sometimes fail to comply with
their constitutional duties by exceeding their allocated powers or violating the
constitutional rights of individuals. Of course, the conditions for departure from perfect
compliance can themselves be varied by making different assumptions about human
psychology and institutional behavior or in some other way. Nonideal constitutional
theory deals with the unconstitutional; the institution of judicial review might be seen as a
paradigmatic topic for nonideal constitutional theory.101
Debates about constitutional theory usually sit at the boundary of ideal and nonideal
theory. We can call such theories, “partially ideal.” Constitutional theory is not in the
realm of fully ideal theory. This is because violations of the constitution are one of the
most important topics in constitutional theory; the institution of judicial review and many
other topics assume less than full compliance. But constitutional theory does not operate
in the realm of what we might call “constitutional determinism.” If we ask what will
100
. See JOHN RAWLS, A THEORY OF JUSTICE 7–8, 215–16, 308–09 (rev. ed. 1999); Michael Phillips,
Reflections on the Transition from Ideal to Non-Ideal Theory, 19 NOÛS 551 (1985).
101
. Of course, other institutions (such as congressional committees or executive offices) may deal with
unconstitutionality outside the courts. See generally James E. Fleming, The Constitution Outside the
Courts, 86 CORNELL L. REV. 215 (2000); Laurence Gene Sager, Fair Measure: The Legal Status of
Underenforced Constitutional Norms, 91 HARV. L. REV. 1212 (1978).
51
Lawrence B. Solum
happen in the future of the actual world, given all the causal forces that affect
constitutional practice, normative constitutional theory becomes irrelevant.
Constitutional theory requires that we assume (perhaps counterfactually) that
constitutional agents are open to rational persuasion on the basis of argument.
b) Constitutional Time: Past, Present, and Future
Debates about constitutional theory frequently make unstated assumptions about
constitutional time. Of course, we are always in the constitutional present in the actual
world. But the practice of constitutional argument frequently takes us back in
constitutional time. What would have happened if Unconstrained Thayerianism had
guided the Supreme Court’s decision in Brown v. Board of Education? Or we may ask
questions about the constitutional future. What would happen to the Supreme Court’s
gender equality jurisprudence if the Supreme Court were to adopt public meaning
originalism as its guiding principal?
These questions involve possible worlds. When we ask what effect a given
constitutional theory would have had on the decision of a case from our constitutional
past, we are positing a possible world that is close to our own, except that a counterpart
of the actual historical Supreme Court has adopted the constitutional theory that is being
evaluated. Likewise, when we talk about the effects of a constitutional theory on the
constitutional future, we imagine a possible world that is adjacent to the actual world but
in which a counterpart Supreme Court (and other relevant political actors) have embraced
whatever constitutional theory we are considering.
Of course, this kind of counterfactual reasoning is not an exact science. The best that
we can do is to make reasonable assumptions about how things would be likely to go in
the possible world. It may well be the case that different assumptions will lead to very
different outcomes, but so long as the assumptions are stated clearly, we can enter into
debates about which set of assumptions is the most reasonable—given the purpose of the
thought experiment (or hypothetical).
c) A Cautionary Note About Constitutional Thought Experiments
Constitutional thought experiments have come to play an important role in
argumentation about constitutional theory. For example, opponents of originalism
frequently make what can be called the “Canonical Cases Objection”: for example, it
might be argued that originalism is not an acceptable constitutional theory, because in the
possible world in which the Supreme Court applied originalism to Brown v. Board of
Education, the outcome would have been different. Some theorists may believe that this
is a decisive objection to originalism on its face, without any further elaboration. But this
thought is obviously wrong. Consider the following set of questions that might be asked
about the thought experiment:
•
Does it matter that originalism would have resulted in a different outcome in
Plessy v. Ferguson and that in a possible world in which Plessy had come out
otherwise, school segregation would have ended long before Brown was
decided?
52
The Constraint Principle
•
•
Does it matter that if Brown v. Board had come out the other way, the actual
practice of school segregation would have ended at approximately the same
time, because it was the civil rights and voting rights legislation of the 1960s
that actually caused the demise of de jure segregation?
Does it matter that Brown v. Board could have come out differently under
many of the nonoriginalist theories of constitutional interpretation?
For constitutional thought experiments to produce meaningful results, they must be fully
specified in such a way to make the assumptions of the thought experiment transparent
and open to question. These questions outline above show that an underspecified thought
experiment can lead to misleading results.
D. The Role of Pairwise Comparison
Our discussion of the framing of the debate over the constraint principle is almost at
an end. One final topic requires some brief discussion. The method of argument
employed in this paper uses the device of pairwise comparison to bring out the
underlying structure of the debate.
The basic justification for the method of pairwise has already been discussed above:
“It takes a theory to beat a theory.”102 The case for and against the constraint principle
will vary with the particular pairwise comparison that is being discussed. Consider the
following examples:
•
•
Pairwise comparison of originalism with the Moral Readings approach raises a
much different set of issues than does comparison of originalism with
Unconstrained Thayerianism. Complete deference to Congress by judges is
obviously different in almost every important respect than an approach that
requires judges to engage directly in moral reasoning to determine the correct
“constructive interpretation.”
Pairwise comparison of originalism with Common Law Constitutionalism
raises a substantially different set of issues than does comparison of
originalism with Popular Constitutionalism. The former comparison will focus
on how judges should approach constitutional adjudication; the latter
comparison will focus on the role of popular movements in constitutional
revision outside of the formal amendment process.
Pairwise comparison also requires attention to the distinction between contribution
and constraint. Some of the rivals to originalism affirm a contribution principle that
requires that constitutional actors give due weight (in some cases, very substantial
weight) to the communicative content of the constitutional text. Other rivals of
originalism deny the contribution principle (e.g., Unconstrained Thayerianism and the
Superlegislature theory). But the arguments for and against originalism frequently will
differ with respect to the role that contribution plays in the particular rival of originalism
that is being discussed. The argument that originalism is unjustified, because the
102
See supra Part III.C.5, p. 46.
53
Lawrence B. Solum
constitution itself is illegitimate because the ratification process excluded a variety of
groups is consistent with theories that deny that contribution principle, but difficult to
square with theories that give great weight to original meaning but allow overrides in
limited circumstances.
In the discussion that follows, the focus will be on comparison of Public Meaning
Originalism with Constraint as Consistency to each of the nine most important forms of
nonoriginalism. The comparison begins in next Part, which examines two clusters of
arguments that favor originalism over many of its rivals. The first set of arguments
clusters around the idea of the rule of law; the second set clusters around the notion of
legitimacy. Generic objections to originalism are considered in Part VI. The generic
arguments for and against originalism are then reframed in the context of particular
pairwise comparisons in Part VII.103
***
Sometimes critics of originalism complain about the lack of normative justifications.
And, truth be told, there was a time when some originalists seemed to argue in a way that
failed to recognize the obvious: originalism is a theory that seeks to guide constitutional
practice. Like all theories about how individuals and institutions should act, originalism
stands in need of some kind of normative foundation.
But we should not confuse the failure of early originalists to address the topic with the
quite different notion that it is difficult to provide the normative foundations for
originalism. The idea that there are good reasons to act consistently with the text of the
United States Constitution is shared by most constitutional actors. It is not that pro tanto
justifications for the constraint principle are hard to find. There are so many of them and
they are so obvious that they may be seen by many nonoriginalists as trivially true but
obviously not decisive.
This suggests that the real action in the normative debate about originalism is
elsewhere. Nonoriginalists sometimes hide behind a smokescreen of theoretical
objections to the general idea of constraint by any legal text, but as we shall see, many of
these, if taken seriously, would lead to absurd consequences and radical alterations in
legal practice generally. The real action arises from the fact that constraint by the text
will lead to some results that nonoriginalists (and many originalists as well) find
normatively unattractive. The question then becomes whether this results-oriented
approach to constitutional theory provides appropriate reasons to reject originalism.
But before we deal with the objections, we need to understand the case for the
constraint principle.
***
IV. JUSTIFYING THE CONSTRAINT PRINCIPLE
This Part identifies and defends two clusters of pro tanto justifications for adherence
to the Constraint Principle. The first cluster focuses on the idea of the rule of law:
constraint supports the rule of law, whereas lack of constraint undermines the rule of law.
103
See infra Part VII, p. 104.
54
The Constraint Principle
The second cluster focuses on legitimacy; constraint enhances constitutional legitimacy.
Each of the two clusters consists of a set of overlapping but partially independent
reasons. The rule of law and legitimacy provide pro tanto justifications for constraint,
but they are not, by themselves, decisive reasons. To determine whether the balance of
reasons supports the Constraint Principle, we will examine the objections to originalism
in Part VI104 and engage in pairwise comparison of originalism with its rivals in Part
VII.105 We can begin the examination of the case for constraint with the idea of the rule
of law.
A. Constraint and the Rule of Law
The first cluster of arguments for constraint focuses on the idea of the rule of law. The
ideal of the rule of law is part of political morality. Sometimes the ideal is articulated as
a contrast, made famous by John Adams, and summarized in the notion of “the rule of
law and not of men.”106 Another formulation emphasizes the rule of law values, such as
predictability, certainty, stability, consistency, generality, and publicity of the laws.107
And yet another approach articulates the rule of law as a set of principles.108 Each of
these ideas about the rule of law plays some role in the cluster of rule-of-law
justifications for originalism; rather than beginning with a general investigation, let us
proceed to the particular arguments.
Sometimes the rule of law is discussed as if adherence to the rule of law is all or
nothing, but this is not the case. The rule of law is best understood as a scalar and not a
binary.109 A system of governance can be more or less compliant with the rule of law. At
one extreme is a system in which a tyrant governs by whim; at the other extreme, would
be a system in which an eternal, rigid, and highly particularized code of law provides a
preordained decision in every case. In the more fortunate parts of the actual world,
governance will lie in between these extremes. The arguments in the rule of law cluster
aim to show that observance of the Constraint Principle is a reasonable way to realize the
ideal of the rule of law to a greater degree than nonoriginalist practice that violates
constraint.
The first argument in the cluster plays off the idea that the constrained rule of law of
can be distinguished from the unconstrained rule of individual persons. I shall call this
the “Judicial Tyranny Argument” for reasons that will become apparent.
104
See infra Part VI, p. 82.
See infra Part VII, p. 104.
106
See John Adams, Novanglus Papers, BOSTON GAZETTE, no. 7 (1774); see also Steven G. Calabresi,
A Critical Introduction to the Originalism Debate, 31 Harv. J.L. & Pub. Pol'y 875, 882 (2008) (“At the end
of the day, Justice Brennan's primary concern was that the text of the Constitution be construed to produce
what he deemed to be good consequences. Doing this makes “the rule of law and not of men” impossible,
which leads to very bad long-term consequences.”).
107
Lawrence B. Solum, The Supreme Court in Bondage: Constitutional Stare Decisis, Legal
Formalism, and the Future of Unenumerated Rights, 9 U. PA. J. CONST. L. 155, 165 (2006).
108
Lawrence B. Solum, Equity and the Rule of Law, in NOMOS XXXVI: THE RULE OF LAW 120 (1994).
109
See Lawrence B. Solum, Legal Theory Lexicon: Scalars and Binaries,
http://lsolum.typepad.com/legal_theory_lexicon/2014/09/legal-theory-lexicon-072-scalars-andbinaries.html (Last revised on November 15, 2015).
105
55
Lawrence B. Solum
1. The Judicial Tyranny Argument
The first argument in the rule-of-law cluster is that the Constraint Principle provides
the best feasible alternative to judicial tyranny. This argument is based on a philosophical
conception of tyranny that distinguishes tyranny from both injustice and violations of
democratic legitimacy. The philosophical conception of tyranny is focused on the
distinction between rule by decree and the rule of law. Violations of the Constraint
Principle by the Justices of the Supreme Court constitute rule by decree and hence are a
form of judicial tyranny.
a) Tyranny as Rule by Decree
In the Nicomachean Ethics, Aristotle develops an account of a “law” (nomos in the
ancient Greek) and its relationship to a “decree” (psēphismata). Richard Kraut, the
distinguished Aristotle scholar, explained the difference as follows. Kraut’s exposition
begins with the idea of person who is “lawful” (nominos):
[W]hen [Aristotle] says that a just person, speaking in the broadest sense is
nominos, he is attributing to such a person a certain relationship to the laws, norms,
and customs generally accepted by some existing community. Justice has to do not
merely with the written enactments of a community’s lawmakers, but with the
wider set of norms that govern the members of that community. Similarly, the
unjust person’s character is expressed not only in his violations of the written code
of laws, but more broadly in his transgression of the rules accepted by the society
in which he lives.
There is another important way in which Aristotle’s use of the term nomos
differs from our word ‘law’: he makes a distinction between nomoi and what the
Greeks of his time called psēphismata—conventionally translated as ‘decrees’. A
decree is a legal enactment addressed solely to present circumstances, and sets no
precedent that applies to similar cases in the future. By contrast a nomos is meant
to have general scope: it applies not only to cases at hand but to a general category
of cases that can be expected to occur in the future.110
Aristotle’s conception of tyranny is rule by decree (psēphismata); the ideal of the rule of
law is governance by laws (nomoi).
b) Two Distinctions About Judgments: First versus Second Order and Public
versus Private
We can restate this last point by introducing a set of distinctions between types of
normative judgments. Begin with the distinction between what we can call “first order
judgments” and “second order judgments”:
•
110
First Order Judgments: A first order normative judgment is a judgment that
attributes a normative characteristic to some state of affairs, person, or action.
Thus, the proposition expressed by “lying is wrong” is a first order moral
RICHARD KRAUT, ARISTOTLE 105-06 (2002).
56
The Constraint Principle
•
judgment. The proposition expressed by “flag burning is protected by the First
Amendment” is a first order legal judgment.
Second Order Judgments: A second order normative judgment is a judgment
about how first order questions should be settled. Typically, second order
judgments have two parts: (1) a judgment about which institution, person, or
other entity should have authority with respect to normative first order
judgments in a given domain, and (2) a judgment about how the favored actor
should make its decision. The proposition expressed by “the constitutionality
of state statutes that forbid same sex marriage should be determined by the
Supreme Court and not by state elected officials” is a second order legal
judgment. Likewise, proposition, “the Supreme Court should adopt only those
constitutional doctrines that are constitutional text,” expresses as second order
judgment.
To this distinction we can add the distinction between public and private normative
judgments:
•
•
Public Judgments: A normative judgment is public if the judgment is based on
a norm (such as a standard or rule) that is publicly accessible. For example, a
normative judgment that an action is unlawful on the basis of a statute that is
public (accessible to citizens) is a public legal judgment.
Private Judgments: A normative judgment is private if the judgment is based
on a belief or opinion that is not publicly accessible. For example, if a judge
bases a decision on her personal belief that one outcome is preferable, the
judge has made a private normative judgment.
If judges rely on their own private, first-order judgments of fairness as the basis for the
resolution of disputes, then it follows inexorably that their judgments will be decrees
(psēphismata) and not decisions on the basis of a second order, public judgment—in
other words, not on the basis of a nomos. In other words, a judge who decides on the
basis of her own private judgments about which outcome is fair in particular cases is
making decisions that are tyrannical in Aristotle’s sense.
c) The Argument from Tyranny Distinguished from Arguments from Injustice and
Democracy
It is not clear that the word “tyranny” has a clearly defined meaning in ordinary
discourse.111 In public discourse, the charge of judicial tyranny could be shorthand for
arguments that judges are acting contrary to justice or that they are violating a principle
of democratic legitimacy.112 There is nothing wrong with using the word “tyranny” in
these ways, but we are using the word “tyranny” in a precise and technical sense:
111
The OED gives several definitions, including “oppressive or unjustly severe government” with
similar definitions for tyrant. (http://0www.oed.com.gull.georgetown.edu/search?searchType=dictionary&q=tyranny&_searchBtn=Search).
112
See, e.g., James E. Wright, III, Bush v. Orleans Parish School Board: The Second Battle of New
Orleans, Chronicles of the Case and the Judge, 61 Loy. L. Rev. 135, 161 (2015) (describing “judicial
tyranny” argument that focused on decision by unelected judges).
57
Lawrence B. Solum
“tyranny” is rule by decree and not something else. For this reason, tyrants may be just
and democracies can be tyrannical.
(1) Tyranny May Be Consistent with Substantial Justice (Except Insofar as Justice
Includes the Rule of Law)
“How can this be?” you may ask. “Aren’t decisions that are motivated by fairness the
very opposite of tyranny?” This objection misunderstands the nature of tyranny in the
philosophical sense used here. Whether a decision is tyrannical or lawful is a different
question than the question whether the decision is just or unjust. There can be unjust
laws and just tyrants. In a perfect world, we could have the rule of law and the content of
the law would be perfectly just. And it is frequently the case that tyrannical rule will also
be unjust. But in the actual world, it is unlikely that the rule of law will be perfectly just
and it almost impossible to achieve a state of affairs where the rule of law is perfectly just
(according to the best or true theory of justice) and in which there is universal agreement
among citizens that this is the case.
The impossibility of achieving near universal agreement or even wide consensus that
any set of laws is perfectly just is a function of the pluralism that has characterized the
United States from the beginning—a topic we have already discussed in connection with
the distinction between deep and shallow reasons.113
Our private, first-order judgments about the all-things-considered requirements of
fairness do not agree. So, in any given case, a decision that the judge believes is required
by fairness will be seen by others quite differently. At best, the decision will be viewed
as a good faith error of private judgment about fairness. More likely, those who disagree
will describe the decision as a product of ideology, personal preference, or bias. At
worst, the decision will be perceived as the product of arbitrary will or self-interest. In no
event, will a decision based on a controversial first order private judgment of fairness be
viewed as outcome of a nomos—a publicly available legal norm.
(2) Rule by Decree May Be Consistent with Majoritarian Democratic Legitimacy
(Except Insofar as a Thick Conception of Democracy Incorporates the Rule of
Law)
The Argument from Judicial Tyranny is not based on democratic legitimacy for two
reasons, one relating to the internal structure of the argument and the other concerning the
possibility of democratic tyranny.
First, the internal structure of the argument does rely on any premise about democratic
legitimacy. The key premises are (1) rule by decree is tyrannical, (2) violations of the
Constraint Principle by judicial decision constitute rule by decree, and (3) tyranny
undermines the rule of law and is for that reason normatively undesirable. None of these
premises refers to democratic legitimacy.
Second, decisions that satisfy the requirement of democratic legitimacy can be
tyrannical. This is obviously true in the case of executive tyranny. Suppose, for example,
that a democratically elected President were to rule by decree, disregarding both judicial
decisions and statutes, and instead governing through ad hoc executive orders. The
113
See supra Part III.A.2, p. 29.
58
The Constraint Principle
decisions of the President could have democratic legitimacy; for example, suppose that
the President was elected by a clear majority on a platform of rule by decree. Likewise,
legislative tyranny is also possible in theory. If Congress were to enact decrees (ex post
facto laws governing particular situations) rather than laws, then it would engage in
legislative tyranny. Again, assuming that a majority of the members of Congress were
elected on a platform of rule by decree, their actions would have been simultaneously
democratic and tyrannical.
It might be argued that the conception of democratic legitimacy is thick enough to rule
out the argument made in the immediately prior paragraph. This would certainly be true
if our concept of democratic legitimacy were formulated so as to exclude rule by decree.
But this point does not deny the point that has just been made—which could be
reformulated so as to make it clear that the argument only applies to thin conceptions of
democratic legitimacy.
d) Constraint Results in the Rule of Law and not Rule by Decree
Decisions that accord with the Constraint Principle are lawful because they are made
on the basis of a legal norm (e.g., a rule or standard) that governs a general category of
cases for the future (until the constitutional text is amended). To the extent that the
Constraint Principle governs constitutional decisionmaking, constitutional practice cannot
be characterized as a form of tyranny.
It is hard to see how there could be a serious objection to the argument that the
Constraint Principle obviates judicial tyranny in constitutional practice. Of course, there
are rhetorical moves that could be made. One could argue that the Constitution itself is
tyrannical because it lacks democratic legitimacy, but this argument is using the word
“tyranny” as a synonym for “nondemocratic” and hence is misplaced as an objection with
respect to the narrow claim made in the prior paragraph. Likewise, one might argue that
purportedly originalist judges actually decide on the basis of ideology and hence that
their decisions are tyrannical (in the relevant sense). If this were true, it would be a
problem for originalism114, but it is misplaced as an objection to the Judicial Tyranny
argument as a justification for the Constraint Principle.
e) Violations of Constraint by the Supreme Court Constitute Rule by Decree and
Hence Judicial Tyranny
But from the fact that the Constraint Principle requires lawful decisionmaking, it does
not follow that the alternatives are tyrannical. That requires a separate argument. We can
begin by noting that there are possible arrangements that avoid tyranny through means
other than the Constraint Principle. Consider, for example, a parliamentary system—
where the parliament itself governs only through laws (and not decrees) and where the
courts decide cases on the basis of laws and do not override the laws by decree. This
system avoids judicial tyranny, but lacks a written constitution and hence the Constraint
Principle.
114
See infra Part VI.H.3.
59
Lawrence B. Solum
This paper is not about the question whether a regime in which there is a written
constitution combined with the Constraint Principle is superior to a lawful parliamentary
system properly authorized by a constitution.115 Instead, the topic at hand is whether the
Constraint Principle should be affirmed given the written constitution that we actually
have and the institution of judicial review (which might be better described as the
combination of constitutional supremacy and a judicial duty of lawfulness). In order to
answer that question, we need to examine each of the major alternatives to originalism
and see how they fare with respect to judicial tyranny.
The comparison of originalism (with the Constraint Principle) and the alternatives
(without the Constraint Principle) will proceed in two stages. In this Subsection, we will
take a preliminary look at some of the alternatives and develop a generic argument that
combining constitutionalism with judicial supremacy in the absence of constraint will
produce judicial tyranny. In a subsequent Part, we will take a look at each of the major
alternatives in greater depth.116
Nonoriginalists who embrace judicial supremacy may reply to the objection in the
following way:
The rule of law can be achieved without the Constraint Principle, because there
are alternative methods of achieving lawfulness and avoiding rule by decree.
Indeed, constitutional practice currently does not comply with the Constraint
Principle but it does realize the rule of law. And to the extent that the status quo
does not fully comply with the rule of law, it could be brought into compliance by
means other than adoption of the Constraint Principle.
But is this reply correct? Originalists disagree with the contention that current
constitutional practice instantiates the rule of law and they will argue that the alternatives
to originalism cannot avoid the problem of tyranny.
Of course, our legal system as a whole does realize the rule of law to a substantial
degree. At any given point in time, the system of constitutional doctrine is likely to be
relatively stable. The decrees of the Supreme Court are elaborated by the lower federal
Courts of Appeal and the highest courts of the several states and these elaborations enable
lawyers and trial court judges to operate on the basis of a relatively stable set of legal
rules—so long as lower courts adhere to the doctrine of vertical stare decisis and public
officials and citizens engage in a practice of acquiescence the decisions reached by the
Supreme Court in particular cases. The Argument from Judicial Tyranny can concede
this point: judicial tyranny by the Supreme Court can coexist with substantial realization
of the rule of law at other levels of the system.
Nonoriginalists might concede this point but press for the normative significance of
Supreme Court tyranny. The argument might be articulated as follows:
Even if we concede that the Supreme Court is tyrannical, this level of tyranny is
tolerable. The label “tyranny” has negative connotations, but in practice the
Supreme Court is a benevolent tyrant. So long as the decisions of the Court are
tolerably just and the values of the rule of law (predictability, certainty, uniformity,
115
Parliamentary democracy could be authorized by a written constitution or by an unwritten
constitution (as in the case of the United Kingdom). The important point is that the current United States
Constitution clearly does not authorize government by parliamentary democracy.
116
See infra Part VII, p. 104.
60
The Constraint Principle
even-handedness, and publicity) are realized to a substantial degree in the system as
a whole, judicial tyranny is better than originalism, which would lead to other
problems.
This line of reply shifts the main focus on debate to the disadvantages of originalism.
Fair enough: it is possible that judicial tyranny can be justified. But this line of reply
accepts that the Argument from Judicial Tyranny provides a pro tanto reason for the
Constraint Principle. The affirmative objections to the Constraint Principle are
considered separate below.117 Perhaps, some defenders of nonoriginalism would make
the more radical argument that there is nothing wrong, even pro tanto, with tyranny, but
this position seems implausible.
A more promising line of reply would aim to show that the alternatives to constraint
do not involve tyranny. Consider, for example, the Multiple Modalities theory. One
might argue that the complex practice of constitutional argument itself constrains judicial
decisionmaking. But is this really the case? Of course, there will be cases in which all of
the modalities point in the same direction, and in these cases, judges will be constrained.
But by hypothesis, these are cases in which the modalities reach the same result as that
required by originalism (and the Constraint Principle). The test cases for the Multiple
Modalities view are those in which the modalities conflict. Recall that in such cases, the
Multiple Modalities account rejects the idea that here is a second-order public standard
for the resolution of conflicts among the modalities: there is no hierarchy or lexical
ordering of the various modes of constitutional argument. This suggests that what is
doing the work in such cases is the private first-order judgments of the Justices—but that
is rule by decree and hence tyranny. Another line of reply might appeal to the standards
of argumentation that are shared by competent legal practitioners—these standards might
be argued to discipline the court and hence to avoid judicial tyranny. But this argument
seems quite implausible, because the standards of the legal community are very
latitudinarian. Indeed, the community of judges, lawyers, and scholars does not even
agree that compliance with the Multiple Modalities theory is required.
Each of the major alternatives to originalism must be evaluated in this way, but we
will reserve that task for the penultimate Part of this Article.118 At this stage, however,
we can preview the more detailed arguments to come. Almost all of the major rivals of
originalism relies, in the end, on private first-order judgments by the Justices. And if this
is correct, then only a few of the alternatives to originalism avoids the judicial tyranny
objection.
The most important rival to originalism that avoids judicial tyranny is Thayerianism—
in both its unconstrained and representation-reinforcement forms. If one accepts a very
strong principle of Thayerian deference, combined with an ordering rule for resolving
conflicts between branches of the national government and between the national
government and the states, then judicial tyranny can be avoided. Of course, if the
principle of deference privileged the President, the result would be executive tyranny—
which is arguably worse than the judicial variety. But this problem could be avoided by
limiting deference to duly enacted statues. A system of legislative supremacy limited by
117
118
See infra Part VI, p. 82.
See Part VIII, p. 104.
61
Lawrence B. Solum
the requirement to legislate by law and not decree avoids the problem of tyranny, but it
does this by rejecting the idea of a judicially enforceable written constitution.
f) The Normative Significance of Tyranny
What is the normative significance of judicial tyranny? Why should we care that the
fundamental structures of government and individual rights are determined on the basis
of ad hoc decrees issued by a committee of nine unelected officials who are appointed for
life turns? To some, the answers to these questions may seem obvious.
***
Consider the following thought experiment:
You are a member of the constitutional convention of a newly formed nation. You
are asked to vote on two proposals regarding the fundamental institutional
arrangements, including the structure of government and individual rights. The
proposals are:
Rule by Decree of an Unelected Committee: The structure of government
and the rights of persons will be decided on a case-by-case basis by a
committee of nine persons, selected by the executive and confirmed by the
legislature. Members of the committee shall serve for life. They shall
have the ultimate power to decide all fundamental questions, including
what questions are fundamental.
Rule by a Written Constitution: The structure of government and the rights
of persons shall be decided in advance in a written constitution. The
constitution shall be interpreted and applied by a court that shall be
constrained by the communicative content of the text. The judges shall
serve life terms but shall be removed from office by the legislature if their
interpretations are clearly inconsistent with the communicative content of
the text. The constitution may be amended by supermajority vote of the
national legislature and the regional legislatures.
The constitutional convention is at the final stage and you must choose between
these two proposals—the other alternatives have been voted down and may not be
reconsidered at this time. One of the considerations in your deliberations is
which proposal best realizes the rule of law. How would you rank the proposals
on that dimension? Would the rule of law ranking be a pro tanto reason to favor
one proposal over the other?
Before you fight the hypothetical, pause for a moment to answer the questions. It is
obvious that rule by a written constitution does a better job of instantiating the rule of
law. And it is obvious that this would be at a minimum a pro tanto reason to favor rule by
a written constitution over rule by decree of an unelected committee.
“Okay,” you say, “But I do want to fight the thought experiment. I don’t favor rule by
decree. There are many alternatives to constraint by a written constitution. What about
Thayerian deference? What about moral readings? What about constraint by common
law methods?” Fair enough. The thought experiment is designed for a limited purpose,
to highlight the relationship of the rule of law to the choice between rule by decree and
62
The Constraint Principle
rule by a written constitution. Each of the alternatives to constraint must be considered
on its own merits. Those comparisons are coming.
***
Tyranny is rule by decree, but what is wrong with that? One answer to that question
lies in the idea of the rule of law and the values with which it is associated, including
predictability, certainty, stability, and consistency. Rule by decree undermines
predictability and certainty for obvious reasons: ad hoc decision of particular cases makes
the law less predictable and certain as compared to the alternative: in the case at hand, the
alternative is the Constraint Principle, which creates predictability and certainty by
conforming constitutional practice to the communicative content of the constitutional
text. Likewise, rule by decree undermines the stability of law, because it enables
constitutional change based on the preferences of judges and especially by the
preferences of a majority of the Supreme Court. Finally, rule by decree undermines
consistency (treating like cases alike), because it enables differential treatment based on
the unconstrained decisions made by judges and officials. The relationship between
constraint and the rule of law values will be considered in greater depth below.119
But the judicial tyranny argument directs are attention to another defect of rule by
decree. Rule by decree is rule by a tyrant—someone with unconstrained power who rules
by deciding particular cases or issuing particularized directives. As bad as a
constitutional monarchy might be, the rule of laws issued by a monarch would be
preferable to rule by decree—so long as the monarch was bound by the constitution and
governed through the promulgation of general laws that applied to all persons, including
the monarch.
Rule by decree entails the fundamental rights and powers of individuals are subject to
the whim of the tyrant. Living under these conditions necessary undermines freedom, as
that concept is understood in its republican sense. Individuals are dependent on the
tyrant.120
The republican conception of political liberty . . . defines freedom as a sort of
structural independence—as the condition of not being subject to the arbitrary or
uncontrolled power of a master. Pettit, who has done more than anyone else to
develop this republican conception of freedom philosophically, puts it thus: a
person or group enjoys freedom to the extent that no other person or group has “the
capacity to interfere in their affairs on an arbitrary basis”.121
The connection between the rule of law and the republican conception of freedom from
domination and dependence illuminates a basic reason for valuing the rule of law. The
rule of law and not of individual persons does more than serve the rule-of-law values
(publicity, stability, certainty, and the rest). The rule of law is an essential safeguard of
119
See infra Part IV.A.2, p. 67 (presenting the rule-of-law values argument).
See PHILIP PETTIT, REPUBLICANISM: A THEORY OF FREEDOM AND GOVERNMENT (1999).
121
See Frank Lovett, Republicanism, Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/entries/republicanism/#LibNonDom (April 15, 2014) (quoting Pettit, supra note
120, at 165).
120
63
Lawrence B. Solum
one of the most important aspects of human freedom—the independence of the individual
from domination by others.
The psychological impact of the loss of freedom may depend in part on the substance
of the decrees. If you applaud the outcome in a particular case (whether it be Obergefell
v. Hodges122 or Citizens United v. FEC123) you might be less bothered by the fact that it is
the product of rule by decree than those who are opposed to the outcome. But you might
still be bothered. After all, the case that you like does not bind the Supreme Court, and a
change in the court’s personnel would result in a reversal of the outcome. If the next
election goes against you, a swing justice retires, and the votes align the other way, your
most cherished substantive right could vanish—at the whim of the Supreme Court. If the
right in question is one you view as essential to your life plan or your standing as an
equal citizen, then you are reduced to dependence on the Court and its decrees.
One of the most insidious aspects of an unconstrained Supreme Court is that it
disguises rule by decree as the rule of law. It takes for itself the power to make the
fundamental decisions regarding the constitution of society, including the scope of
individual rights and liberties, but then wraps its rule by decree in the false flag of
constitutional constraint. The decisions of the Court are presented as derived from the
Constitution and hence as settled, pending constitutional amendment. But absent
constraint, the presentation is illusion and not reality. This state of affairs threatens
republican freedom in an especially pernicious way—and brings to mind the dystopias of
speculative fiction.
g) Objections to the Argument from Judicial Tyranny
How might a nonoriginalist answer the argument from judicial tyranny? Here are two
possible objections.
(1) Objection One: The Supreme Court’s Decisions Are Laws Not Decrees
The first objection goes to the question whether violations of the Constraint Principle
constitute “decrees” on the one hand or whether they are more properly described as
“laws” on the other hand. It is important to remember that this question is framed in a
technical vocabulary: “decrees” are ad hoc decisions of particular matters that do not
create general rules, whereas “laws” do create rules (or rule-like legal norms).
One answer to the judicial tyranny objection might focus on the doctrine of stare
decisis. It might be argued that the Supreme Court does, in fact, legislate (make “laws”),
because its holdings create binding rules. This argument has a superficial appeal. Some
Supreme Court decisions do alter the legal landscape in ways that create law-like
consequences. If a statute is struck down by the Supreme Court and the political
branches acquiesce in this decision, then the consequence of the Supreme Court’s action
is very much like a repealing statute. To the extent that the rule of law prevailed before
the statute was enacted, repeal by judicial decree would not undermine the rule of law.
So, there are cases where rule by decree will not undermine the rule of law values of
predictability, stability, certainty, and consistency. But can we generalize from these
122
123
576 U.S. ___ (2015).
558 U.S. 310 (2010).
64
The Constraint Principle
cases to the conclusion that Supreme Court never or rarely governs by decree. Of course,
there are examples where the Supreme Court’s decisions clearly do undermine the rule of
law. The current disorder with respect to state laws restricting abortion is a fairly clear
example. The pattern of Supreme Court decisions has created substantial uncertainty
regarding the score of the power of states to regulate abortion and that uncertainty invites
restrictive legislation and subsequent litigation, with inconsistent outcomes as different
judges with different ideological priors reach inconsistent outcomes.
One of the underlying causes of this disorder is a function of the chaotic state of the
doctrine of stare decisis. There are at least three different operative conceptions of the
doctrine of stare decisis in contemporary American jurisprudence:
Fact Bound Holdings. One view is that the holding of a case is limited to the legally
salient facts. This view entails that judicial violations of the Constraint Principle
constitute decrees—as it denies that holdings can generate rules.124
Ratio Decidendi. Another view is that the holding of case is the narrowest rule
implied the reasoning necessary to the decision of the case given the legally salient
facts and the arguments made by the parties.125 This view results in holdings that are
more like laws, but allows for considerable variations in future cases based on
factual differences and new arguments raised by future parties.
Legislative Holdings. Yet another view is that courts have the power to create
legislative holdings by clearly stating the rule that they intend to create. This view
results in holdings that are laws, not decrees—assuming that such holdings are truly
binding.126
It is not clear that American courts, including the Supreme Court, follow any consistent
practice with respect to the doctrine of stare decisis. Courts seem to use all three
theories—perhaps choosing the theory of stare decisis to justify outcomes reached on
other grounds. If the practice of stare decisis is disordered in this way, the result is that a
superficial appearance of rule-like holdings in particular cases is undermined by a metalevel practice that results in rule by decree.
There is another problem with the idea that the doctrine of stare decisis results
obviates judicial tyranny. The Supreme Court does not consider itself bound by the
doctrine of stare decisis: it is free to overrule its prior decisions and to decide new cases
in ways that fail to adhere to its prior holdings. The Court does give some weight to its
prior decisions, but that weight is light and variable. The most influential modern
statement of the Court’s practices is found in Planned Parenthood of Southeastern
Pennsylvania v. Casey, quoted in full in the accompanying footnote.127 Of course, this is
a very large topic, but I think it is fair to say that it would be difficult to argue that Casey
124
See Arthur L. Goodhart, Determining the Ratio Decidendi of a Case, 40 YALE L. J. 161 (1930).
See Lawrence B. Solum, The Supreme Court in Bondage, supra note 107, at 155-208.
126
Id.
127
505 U.S. 833 (1992). The court stated, “Even when the decision to overrule a prior case is not, as in
the rare, latter instance, virtually foreordained, it is common wisdom that the rule of stare decisis is not an
“inexorable command,” and certainly it is not such in every constitutional case,” and “when this Court
reexamines a prior holding, its judgment is customarily informed by a series of prudential and pragmatic
considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of
law, and to gauge the respective costs of reaffirming and overruling a prior case.” Id. at 854–55.
125
65
Lawrence B. Solum
creates a rule-like approach to horizontal stare decisis. And because the Supreme Court
does not bind itself, lower courts and officials are permitted to guess whether the current
Justices will (or will not) follow past precedents. Open flouting of prior precedents might
result in summary reversal, but given the fluidity in vertical stare decisis, it seems likely
that lower court judges (and lawyers for extrajudicial constitutional actors) will be able to
create plausible arguments that the actual holding of the Supreme Court is
distinguishable.
In sum, it is far from clear that the doctrine of stare decisis is sufficient to transform
the Supreme Court’s decisions from decrees to laws. Indeed, the disordered state of the
current doctrine of stare decisis suggests that it may have the opposite effect—creating
even more uncertainty, instability, unpredictability, and inconsistency.
(2) Objection Two: The Rule by Decree Objection Applies to All Common-Law
Decisionmaking
The second objection is that the judicial tyranny argument proves too much—because
it suggests that all common-law decisionmaking involves judicial tyranny. Articulating
this objection will be a large task—because it will require an account of the common-law
process. Nonetheless, the gist of the objection is clear enough. Common-law
decisionmaking is like the Supreme Court’s constitutional jurisprudence. If the latter is
rule by decree, then so is the former. So, if the rule by decree objection justifies the
Constraint Principle, it will also invalidate all of the common law.
Of course, there is something to this objection. Critics of the common law have long
complained that it undermines the rule of law. If they are right, then the proper solution
would seem to be codification and not the extension of the evil to the constitutional
realm. Two wrongs do not make a right—especially when they are just two different
wrongs!
There are, however, ways in which common law decisionmaking could avoid the rule
by decree objection. For example, the declaratory theory of the common law conceives
of the common-law process as declaratory and not legislative. Common-law judges
recognize existing norms and customs. So long as the common law is limited to judicial
decision on the basis of widely shared and deeply held social norms and customs, the
common law will not constitute rule by decree. It is only if judges view themselves as
having the power to make common law (through legislative decrees) that the problem
arises.
Common-law judging could avoid the judicial tyranny objection in a second way, by
adhering to a strong version of the doctrine of stare decisis or precedent. So long as
common law courts view themselves as bound by prior decisions, common law making is
constrained. But if common law courts view themselves as permitted to overrule prior
decisions on the basis of private first-order judgments about what the law ought to be,
then common-law judging will be inconsistent with the rule of law and will constitute a
form of judicial tyranny.
Finally, notice that the second objection to the Judicial Tyranny Argument does not
actually defend theories that authorize violations of the constraint principle from the
charge they are tyrannical; rather, it argues that the status quo includes other practices that
many be tyrannical. If the common law can be made consistent with the rule of law and
66
The Constraint Principle
would thereby be improved, the argument that the common law now has tyrannical
elements does not seem very persuasive as an objection to constraint.
***
Experience suggests that the judicial tyranny argument is “difficult to swallow.” I
image a reader asking the following question: “How could it possibly be the case that I
live in a judicial tyranny? Surely I would know it already if that were the case. I
certainly didn’t learn that the Court was a tyrant in law school; my colleagues don’t say
that violations of constraint are tyrannical.” This hypothetical reader seems to lack
access to the intuitions that motivate the objection.
But perhaps you already have intuitive access to the notion that rule by decree is
tyrannical. Call to mind a Supreme Court decision to which you object on normative
grounds—a decision you find both wrong and unjustified. Perhaps for you that is
Citizens United or maybe it is Obergefell v. Hodges. Does this decision conjure the idea
of judicial tyranny? Now imagine that the composition of the Supreme Court changes
and the Court reverses this decision. You are very pleased! The rule of law has been
restored! And now imagine that unexpectedly, the composition of the Court changes
again. A younger justice resigns due to ill health, and unexpectedly a President of your
party is not reelected. The new Justice votes to reinstate the despised decision. Would
the label “tyrannical” be out of place?
Here is another way to access the force of the judicial tyranny argument. Imagine that
the President rather than the Supreme Court assumes the power to adopt amending
constructions of the Constitution. A very popular President, a demagogue, abrogates to
himself the power to overrule the Supreme Court and the constitutional text using the
newly created device of “Executive Constitutional Decrees.” The academy howls,
Senators protest, but political forces are such that the President prevails. With the
precedent established, the next President continues to exercise this power. Would you be
inclined to call the exercise of this power “tyrannical”? If you would, then does it really
make a difference that the power of rule by decree is currently exercised by the Supreme
Court (a committee of nine unelected judges appointed for life terms) and not the
President (hopefully elected by a majority of voters)?
***
2. The Rule-of-Law Values Argument
The second argument in the rule of law cluster is that violations of the Constraint
Principle undermine the rule of law values. There is no definitive list of these values, but
publicity, stability, and consistency are frequently counted among them. The rule of law
values provide a set of overlapping and partially redundant pro tanto reasons to adopt the
Constraint Principle. Consider first the value of publicity.
67
Lawrence B. Solum
a) Constraint Enhances the Publicity of Law
Theories of the rule of law frequently articulate “publicity” as a requirement of the
rule of law.128 Putting it another way, publicity is one of the rule-of-law values. The
thesis of this subsection is that observing the constraint principle enhances the publicity
of constitutional law. This argument is closely related to another argument, which
focuses on transparency and its relationship to the legitimacy of constitutional law. The
publicity argument focuses on our ability to know what the law is in advance and to act
on the basis of that knowledge.
In what sense does the failure to observe the Constraint Principle undermine the ruleof-law value of publicity? There are two aspects of an unconstrained judiciary that
undermine the publicity of law: first, the true basis of unconstrained decisions is not
reliable provided by judicial opinions, and second, the legal norms created by judicial
decisions are frequently ambiguous or obscure.
The first aspect of the publicity argument stems from the reluctance of judges to admit
that their decisions violate the Constraint Principle. This reluctance may lead to
circumlocution and avoidance. Most obviously, when the actual basis for a judicial
decision lies in the first-order private judgments of the judge, there will be a temptation to
attribute the decision instead to the “meaning” of the constitutional provision or to
precedent. Given the importance of stare decisis to the system, the failure to disclose true
reasons makes it difficult to determine the content of legal doctrine. By way of contrast,
judges who observe the Constraint Principle can clearly state the basis of their decisions;
for them, legal content is based on the communicative content of the text.
The second aspect of the publicity argument is closely related to the first. In an
unconstrained system, the precise legal content of the law may be unnecessarily
ambiguous, obscure, or open textured. This lack of clarity can arise in many ways. For
example, judges may conceal the role of their own first order private judgements by use
of a flexible balancing test or an all-things-considered standard expressed in terms of thin
legal concepts like “reasonableness” or “just.” Because there may be intersubjective
disagreement about how such standards apply to particular cases, they obscure the real
basis for the decision, substituting the appearance of publicity for the real thing.
These general considerations are illustrated by the Supreme Court’s decision in
International Shoe Co. v. Washington.129 Famously, Shoe initiated a process of
transformation, replacing the highly formalist scheme of categorical rules from Pennoyer
v. Neff130 with a highly abstract and indeterminate formula: “due process requires only
that in order to subject a defendant to a judgment in personam, if he be not present within
the territory of the forum, he have certain minimum contacts with it such that the
maintenance of the suit does not offend 'traditional notions of fair play and substantial
justice.”131 Subsequent cases provided a baroque elaboration of this formula, including
distinctions between general and specific jurisdiction, an obscure purposeful availment
128
LON L. FULLER, THE MORALITY OF LAW 42, 44 (rev. ed. 1964); Jeremy Waldron, The Concept and
the Rule of Law, 43 GA. L. REV. 1, 24 (2008).
129
326 U.S. 310 (1945).
130
95 U.S. 714 (1878).
131
326 U.S. at 316.
68
The Constraint Principle
test, and a five factor interest-balancing test.132 The alternative originalist approach,
suggested but not adopted by Justice Scalia, would involve a return to relatively clear and
determinant standards that prevailed at the time the Fourteenth Amendment was
adopted.133 This is a complex area of law, and a full treatment certainly requires an
extended treatment, but for present purposes, the Supreme Court’s personal jurisdiction
jurisprudence illustrates the mechanisms by which violations of constraint can create
publicity problems that are solved by adherence to original meaning.
At this point, defenders of originalism might argue that the doctrine of stare decisis
provides a sufficient degree of publicity to the law. Of course, the decisions of the
Supreme Court are public and binding on the lower federal courts via the doctrine of
vertical stare decisis. They are not binding on the Court itself. 134 This results in both
formal overrulings and what is sometimes called “stealth overrulings”135—decisions that
vitiate the gravitational force of a precedent that formally remains “on the books.” The
Supreme Court says that it gives its prior opinions substantial weight and has articulated
the factors that it purports to apply when deciding whether to overrule,136 but there are
reasons to doubt that the factors stand in the way of Justices who wishes to overrule a
decision that they oppose on the basis of their first-order private judgments of political
morality. On the great issues of the day, it surely not unreasonable to believe that the
outcome of the next election and the timing of new appointments to the Supreme Court
mean that the content of the law is “up for grabs.”
This problem is compounded by substantial uncertainty about the true scope of the
doctrine of vertical stare decisis. That doctrine requires a distinction between holdings
and dicta—a murky topic at best. There are at least three competing theories of vertical
stare decisis. The first theory, associated with Arthur Goodhart, is that that the holding of
a case is limited by legally salient facts.137 This theory results in extraordinarily narrow
holdings—as the ability of lawyers to find legally relevant factual differences between
cases is very substantial indeed.138
The second theory, the classic theory of the ratio decidendi, limits the holding of a
case to the rule implied by the actually articulated reasoning necessary to the decision on
132
There is a vast literature on the Supreme Court’s personal jurisdiction jurisprudence. See, e.g.,
Howard B. Stravitz, Sayonara to Fair Play and Substantial Justice?, 63 S.C. L. REV. 745 (2012); Ryne H.
Ballou, Civil Procedure-Be More Specific: Vague Precedents and the Differing Standards by Which to
Apply "Arises Out of or Relates to" in the Test for Specific Personal Jurisdiction, 35 U. ARK. LITTLE ROCK
L. REV. 663 (2013); Bernadette Bollas Genetin, The Supreme Court's New Approach to Personal
Jurisdiction, 68 SMU L. REV. 107 (2015).
133
Lawrence Rosenthal, Does Due Process Have an Original Meaning? On Originalism, Due Process,
Procedural Innovation . . . and Parking Tickets, 60 OKLA. L. REV. 1, 24-25 (2007).
134
The question whether nonoriginalist Supreme Court decisions should have binding vertical stare
decisis effect on state court judges is explored by Lee Strange. See Lee J. Strang, State Court Judges are
Not Bound by Nonoriginalist Supreme Court Interpretations, 11 FIU LAW REVIEW 327 (2016).
135
See Barry Friedman, The Wages of Stealth Overruling (With Particular Attention to Miranda v.
Arizona), 99 Geo. L.J. 1 (2010).
136
Planned Parenthood v. Casey, 505 U.S. 833 (1992).
137
See Arthur L. Goodhart, Determining the Ratio Decidendi of a Case, 40 YALE L.J. 161 (1930)
(elaborating theory of the ratio decidendi).
138
Over time, a long succession of cases might result in the emergence of a broad rule, but the opposite
is frequently the case, with inconsistencies among cases resulting in rules becoming more complex and
uncertain as time goes on.
69
Lawrence B. Solum
the basis of the legally salient facts and the arguments presented by the parties.139 This
theory results in both narrow holdings (because the necessary reason is limited by salient
facts and the parties’ arguments) and substantial uncertainty (because both the content of
the reasoning and the rule that is logically implied are subject to contestation and unlikely
to be the subject of intersubjective agreement in ideologically charged disputes).
The third theory affords great weight to we-hold-that statements, essentially giving
such statements what Ronald Dworkin called “enactment force”140 because such
statements provide the best prediction likely future behavior of the Court.141 But even if
the prediction theory of precedent is correct, it does not follow that propositions
introduced by “we hold that” are sufficient to provide the kind of publicity required by
the rule of law. Notice first that the predictive theory implicitly accepts the idea that the
Supreme Court engages in rule by decree: the job of lawyers and lower court is to predict
the decrees of the tyrant. Moreover, the predictive value of we-hold-that statements
degrades over time as the views of the Justices change and old Justices retire and new
ones are appointed.
No one theory of precedent dominates the other. Indeed, one might suspect that
different judges hold different theories and that the same judge is likely to pick and
choose among the theories depending on how judges wants the case before them to come
out. Perhaps someone can make the case that rule by Supreme Court decree provides
substantial publicity, but given the absence of binding horizontal stare decisis and the
complexity and uncertainty of vertical stare decisis, making the case is difficult.
Moreover, it should be noted that the claim here is not that rule by decree provides no
publicity at all. Much of the content of constitutional law is clear and settled. The
question is whether adoption of the Constraint Principle would enhance the publicity of
law. The case that it would is compelling—so long as the communicative content of the
constitutional text is not radically indeterminate. Of course, the extent to which the
content is determinate is itself a complex empirical question, which is bracketed for the
purpose of this article. But even at this preliminary stage, there are good reasons to doubt
claims of radically indeterminacy.142
Finally, the it should be noted that the publicity argument favors constraint over some
but not all of the alternatives. Consider, for example, Unconstrained Thayerianism (the
view that courts should always defer to Congress and that Congress should govern
through general legislation but should not consider itself constrained by the Constitution).
Because Unconstrained Thayerianism substitutes Congress for the Constitution but
requires Congress to govern through legislation that itself is public, it does not share the
publicity problems associated with a Supreme Court that promulgates the constitution
through an unconstrained process of case-by-case decisionmaking. Once again, the
139
See Lawrence B. Solum, How NFIB v. Sebelius Affects the Constitutional Gestalt, 91 WASH. U.L.
REV. 1, 22 (2013) (There is, to be sure, a formalist version of the doctrine that is rooted in the idea of the
ratio decidendi: the holding of a case is the rule that is logically implied by the stated reasons necessary to
the resolution of the case on the facts before the appellate court and the legal arguments presented by the
parties. But there is another tradition of thinking about stare decisis that views the holding of a case as the
rule that best predicts the future behavior of a court from the opinions expressed by the judges.)
140
See RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 111 (1977).
141
See Solum, supra note 139, at 22.
142
See Lawrence B. Solum, On the Indeterminacy Crisis: Critiquing Critical Dogma, 54 U. CHI. L.
REV. 462 (1987); see also Heidi Kitrosser, Interpretive Modesty, 104 GEO. L. J. (forthcoming 2016).
70
The Constraint Principle
question whether the Constraint Principle is best depends on pairwise comparisons with
particular alternatives—the subject of Part VII of this Article.
b) Constraint Supports the Stability of Law
The stability of law is usually counted as a virtue, but too much stability counts as a
vice. Nonetheless, it should be clear that ceteris paribus the stability of law is a good
thing, whereas instability of law has no distinctive virtue that is independent of the
content of the law. Why is this so? One of the reasons that we value the stability of law
is that it enables individuals and institutions to formulate long-run plans. When the law is
unstable, planning may be impossible. And even if long-range planning is possible under
conditions of legal instability, individuals and institutions may need to invest resources in
order to prepare for multiple contingencies: stability of law reduces the costs of long run
planning.
These abstract points can be made more concrete by considering the various ways in
which the content of constitutional law affects planning. Consider, for example, the
ability of Congress to legislate concerning complex social and economic problems. The
Constraint Principle entails that the basic rules of the road are fixed, subject only to
constitutional amendments which are likely to be infrequent given the amendment
process. There has only been one significant example of constitutional instability
associated with the amendment process: the ratification of the Eighteenth Amendment
authorizing prohibition in 1919 followed by its repeal by the Twenty-First Amendment in
1933. Moreover, because change by constitutional amendment requires a supermajority,
the costs of instability are likely to taken into consideration during deliberation over
constitutional change.
The Constraint Principle promotes constitutional stability, which provides a pro tanto
reason for constraint as opposed to many (but not all) of its rivals.
c) Constraint Reinforces Legal Certainty
Consider finally the value of legal certainty. Certainty of the law is, ceteris paribus, a
good thing, but uncertainty about the content of the law is only very rarely desirable.
Whereas publicity refers to accessibility of the public to the content of constitutional
doctrine and stability refers to the persistence of doctrine over time, the value of certainty
is focused on the question whether the content of constitutional doctrine at any given time
is certain or uncertain. In this regard, it is helpful to distinguish two different kinds of
uncertainty, which we shall call “risk” and “ignorance” (in lieu of the alternative
vocabulary which distinguishes uncertainty as “risk” from “Knightian uncertainty”).143
Uncertainty as risk is a scalar: the content of the law can be more or less uncertain.
The Constraint Principle increases certainty about the content of constitutional law as
compared to most of the important alternatives. At any given time, many questions of
constitutional law will not have been addressed by the Supreme Court. Some of these
questions give rise to circuit splits or splits among the state appellate courts. Other
questions are only addressed by the federal trials courts, the decisions of which lack stare
decisis effect. And yet other questions have never been addressed by any court. The
143
See FRANK H. KNIGHT, RISK, UNCERTAINTY, AND PROFIT (1921).
71
Lawrence B. Solum
Constraint Principle generates certainty with respect to many of these open questions.
Many of the rivals of constraint do not, for reasons that have already been explored. This
kind of uncertainty is particularly acute during periods when the future composition of
the Supreme Court is difficult to ascertain and with respect to issues where the
preferences of the Justices are unclear.
3. The Politicization Argument
The rule of law is not the rule of politics: this idea needs to be clarified and qualified,
but it captures an important insight about political morality. Of course, law and politics
are closely related. Politics shapes the law, and widely held ideas about democratic
legitimacy suggest that the shaping of law by democratic politics is a good thing.
Moreover, the word “politics” is sometimes understood as referring to statecraft, the
activity that aims to shape communities and nation-states for the common good: in this
sense, the “politicization” of constitutional law would seem to be a virtue and not a vice.
But there is another sense of the word “politics” that associates that term with partisan
and ideological conflict. Understood in this sense, the politicization of constitutional law
is a vice and not a virtue. Hardly anyone is willing to publicly state that the Supreme
Court should vote along partisan lines on constitutional issues—and there is a widely
shared sense that constitutional deliberation should be something other than ideological
strife. Originalism makes this explicit: originalists believe that constitutional practice
should be guided by the communicative content of the constitutional text and not by
partisan politics or ideology. But originalism is not alone in this regard. The multiple
modalities view does not admit of a partisan or ideological modality. Common law
constitutionalism posits a distinctive common-law method that contrasts with partisan
and ideological approaches to governance. The moral readings theory argues that moral
philosophy, not politics, should guide the process of constitutional interpretation and
construction.
Nonetheless, the rejection of the Constraint Principle may, in fact, lead to the
politicization of constitutional law. Take the moral readings theory as an example. Firstorder private judgments about political morality are likely to be divergent given the fact
of pluralism. Even if morality is objective as a matter of metaphysics, human experience
suggests that there is no method for the discovery of moral truths that leads to reliable
intersubjective agreement: metaphysical objectivity (assuming the objective view is
correct) and epistemology subjectivity seem to coexist in the realm of political morality.
Moreover, many moral disagreements are political and ideological disagreements as well.
Contemporary constitutional law is replete with examples, from Roe v. Wade144 to
Citizens United.145
The Supreme Court is subject to political pressures and can become thoroughly
politicized. It is true that the Supreme Court consists of Justices appointed for life terms,
subject to removal through a supermajoritarian process, which requires a majority vote of
impeachment by the House146 and a two-thirds vote for conviction by the Senate, but this
is not sufficient to isolate the Court from politics, because the appointments process
144
410 U.S. 113 (1973).
558 U.S. 310 (2010).
146
U.S. Const. Art I, sec. 2, cl. 5.
145
72
The Constraint Principle
allows political actors, especially the President to influence the composition of the Court.
Under these circumstances, a Court that decides on the basis of the first-order private
judgments of the Justices on matters of political morality can become politicized—even if
the individual members of the Court view their own decisions are based on true moral
readings.
Similar points can be made about many of the other rivals of originalism. The
common-law process may provide some discipline, but given the power of the Supreme
Court to overrule its own prior decisions, it seems likely that a common-law court can
become thoroughly politicized while couching its decisions in the language of the
common-law process. Moreover, contemporary views of the common law frequently
equate “common law” with “judge made law”—leading to a collapse of common law
constitutionalism into the Supreme Court as a superlegislature. Likewise, the multiplemodalities view makes it difficult, if not impossible, to provide intersubjective agreement
on the question whether a given contested Supreme Court decision is wrong, again
opening the door to the politicization of the Court.
The politicization of constitutional law is clearly inconsistent with the rule of law.
And the more serious the politicization, the more the rule of law will be eroded. In times
of intense partisan conflict there is the possibility of a downward spiral of politicization,
with each party attempting to staff the Court with politically reliable judicial operatives.
The bottom of such a downward spiral could involve very serious erosion of the rule of
law. The high politics of principled disagreement about matters of political morality
could be replaced by the low politics of rigging elections and deciding cases on the basis
of an explicit calculus of political costs and benefits.
At this point, it might be objected that the Constraint Principle cannot insulate the
Court from politicization. Original meaning can be disputed, and politics can operate in
the construction zones, where the original meaning does not fully determine doctrine and
outcomes. This is a fair criticism, but we need to recall that the realization of the rule of
law is a scalar and not binary. The relevant question is not whether an originalist court
would be completely apolitical or a common-law court would be thoroughly politicized.
The relevant question is whether the Constraint Principle can reduce the level of
politicization by providing an objective standard for legally correct decisions that can
serve as the focus for intersubjective agreement in a wide variety of cases. That depends,
of course, on epistemological questions about the discoverability of original meaning and
on further questions about constitutional construction. A rule of law approach to the
construction zone may be a necessary complement to the Constraint Principle. There are
a variety of such approaches ranging from default rules (like a principle of Thayerian
deference when constitutional meaning is unclear) to procedural options, like a strong
doctrine of constitutional stare decisis. Those questions are outside the scope of this
Article, but my judgment is that the comparative superiority of originalism with respect
to politicization of constitutional law is likely to be established once those questions are
addressed in due course.
Let us step back and take stock of the cluster of rule of law arguments. Three
arguments have been advanced: (1) the argument from judicial tyranny, (2) the rule-oflaw values argument, and (3) the politicization argument. These arguments all make the
same basic point, that the Constraint Principle advances the rule of law, but each of the
three arguments approaches the question from a different angle. The arguments are not
73
Lawrence B. Solum
fully independent—they interlock in a variety of ways by sharing assumptions and value
premises. But each argument illuminates the connection between the Constraint Principle
and the rule of law in a distinct way.
B. Constraint and Legitimacy
The second cluster of arguments focuses on constitutional legitimacy. Legitimacy is a
complex concept.147 Legitimacy should be distinguished from justice or rightness. It is
possible for a just law to lack legitimacy (because it was imposed by an unelected
dictator), but nonetheless be substantively just. Likewise, a legitimate law (made by an
elected legislature employing the proper procedures) might be unjust. We can think of
legitimacy as a process value: that a law is legitimate is a reason to consider it
authoritative, providing a pro tanto reason for action that stems from characteristics of
the law other than the moral rightness of its substantive content.
There are at least three reasons to affirm the Constraint Principle that connect with the
idea of legitimacy. The first is the familiar argument that the communicative content of
the constitutional text possesses democratic legitimacy, but that violations of constraint
by unelected judges have this characteristic to a much lesser degree. The second
legitimacy argument focuses on the idea of transparency: judicial decisions complying
with constraint can be made transparently, but violations of the Constraint Principle are
frequently disguised through dissimulation, obfuscation, or deception. The third
argument focuses on judicial role; it argues that the role of judges (and most other
officials almost all of the time) does not include an authorized power of constitutional
revision.
1. The Democratic Legitimacy Argument
The democratic legitimacy argument is one of the most familiar in debates over
originalism. The gist of the argument is based on the idea that the Constitution (and
especially the public meaning of the constitutional text) was enacted by “We the People”
and therefore the original public meaning is legitimate, because it has a democratic
pedigree. Moreover, the Constitution is subject to amendment through democratic
processes—and hence to democratic control. These facts provide a pro tanto reason for
complying with the Constraint Principle. If (1) the polity has through democratic
processes ratified a text and retains the power to modify the text, and (2) the text is a
constitution the communicative content of which creates constitutional law, then the
combination of these two facts constitutes a reason for constitutional actors (judges and
other officials) to act in compliance with the text.
The argument itself is simple, but things become more complex once we consider
objections. One familiar objection is that most of the provisions of the Constitution are
old and were enacted by persons now dead. This argument has some superficial appeal,
but if it were accepted it would deprive almost all law of democratic legitimacy. Huge
numbers of statutes were enacted by legislators who were not themselves elected by
147
For an introduction to the concept of legitimacy, see Fabienne Peter, Political Legitimacy, Stanford
Encyclopedia of Philosophy (2016).
74
The Constraint Principle
contemporary majorities. We believe that legislation retains democratic legitimacy until
it is properly repealed. This objection is discussed further below.148
Another objection focuses on the fact almost every provision of the Constitution now
in force was enacted before the franchise was fully extended by five amendments: (1) the
Reconstruction Amendments (race and former condition of servitude)), (2) the Nineteenth
Amendment (women’s suffrage), (3) the Twenty-Third Amendment (limited voting rights
for residents of the District of Columbia), (4) the Twenty-Fourth Amendment (prohibiting
poll taxes), and (5) the Twenty-Sixth Amendment (extending the right to vote to 18 year
olds). The only amendment to complete ratification after the Twenty-Sixth Amendment
is the Twenty Seventh and a majority of the ratification process for that amendment took
place before the Twenty Sixth. Arguably, all of the Constitution lacks full democratic
legitimacy.
Moreover, massive number of Americans are disenfranchised because they have been
convicted of a felony and there is no immediate prospect that this will be remedied.
Hence, it might be argued that even future constitutional amendments would
democratically legitimate, unless and until this problem of disenfranchisement is
addressed. And this objection would appear to apply to many statutes as well.
Progressive legislation from the turn of the century was enacted before women had the
franchise. New Deal legislation was enacted at a point during a period when most
African Americans were disenfranchised. And the same is true of contemporary
legislation, the regulations promulgated various agencies and executive departments, and
the decisions of the Supreme Court.
At this point, most readers will have recognized the reductio, the pattern of argument
that denies the Constitution of 1789 democratic legitimacy leads to the conclusion that
almost none of the law now in effect is democratically legitimate. The problem that leads
to the reductio is the false assumption that democratic legitimacy is a binary, when in fact
it is a scalar. Moreover, in a society in which the franchise has expanded over time,
democratic legitimacy is relative to the standards of the time at which a given law (hence
constitutional provision) is enacted. The original constitution possesses a substantial
degree of democratic legitimacy; constitutional provisions enacted after the extension of
suffrage to women are more legitimate. There is further discussion of this problem
below.149
What about the rivals of constraint? Here it is important to recognize that particular
alternatives differ with respect to democratic legitimacy. One group of theories (e.g.,
multiple modalities, moral readings, common law constitutionalism) are committed to
juristocracy—rule by unelected judges appointed for life terms. The Constraint Principle
is clearly superior to them with respect to democratic legitimacy. Another group of
theories (unconstrained Thayerianism and representation-reinforcement Thayerianism,
for example) are actually superior to the Constraint Principle: this group is organized by a
principle of democratic legitimacy. The democratic legitimacy argument provides a pro
tanto reason to favor the originalism incorporating constraint over some but not all of the
rivals of originalism that reject the constraint principle.
148
149
See infra Part VI.G, p. 100 (discussing dead hand problem).
See infra Part VI.F, p. 97 (discussing exclusion objection).
75
Lawrence B. Solum
2. The Transparency Argument
The second argument in the legitimacy cluster focuses on the relationship between
transparency and legitimacy. The core of the argument is that abidance with the
Constraint Principle is the best alternative to serious violations of the requirement of
transparency that is required for the legitimacy of law. This argument is closely related to
the publicity argument within the rule-of-law cluster. The difference is that transparency
ties to legitimacy, whereas publicity is bound to the rule of law. The two arguments rely
on common premises, but differ with respect to the value they invoke.
The transparency argument is based in part on the observation that nonoriginalists are
reluctant to frankly acknowledge that they have rejected the Constraint Principle; we
have already taken not of this phenomenon in our discussion of the view that the
Supreme Court is a superlegislature.150 The opinions of the Supreme Court have not
explicitly claimed a power to violate the Constraint Principle, and when the issue arises
the Court almost always claims that its decision is allowed or compelled by the original
meaning of the constitutional text. One suspects that individuals who wish to be
nominated for the Supreme Court are well advised to be avow their allegiance to the
Constraint Principle (or something close to it) and to disavow the Supreme Court’s power
to adopt amending constructions.
The Constraint Principle can be affirmed publicly and transparently by constitutional
actors, but denial of the Constraint Principle by constitutional actors is almost always
done privately or expressed publicly in ways that obfuscate the fact that the Constraint
Principle is being denied. It is widely assumed that political transparency is required for
the legitimacy of law and that a secret decisionmaking procedure is illegitimate—except
in special contexts where the need for secrecy is great (such as certain national security
matters). Action on the basis of principles that cannot be made public lack an important
form of legitimacy.
Nonoriginalists might reply to the argument from legitimacy by arguing that false
allegiance to the Constraint Principle is a noble lie. Consider the following version of the
argument:
The Constraint Principle is superficially attractive to citizens because they are
naïve. This naïveté is illustrated by law students, who enter law school as
formalists (assuming that the communicative content of legal texts does and should
constrain judges) and leave as legal realists (who realize that judges do and should
engage in policymaking). It takes intensive training to grasp the truth of legal
realism, and it is simply not practical for ordinary citizens to acquire such training.
For this reason, judges are warranted when they affirm the Constraint Principle as
an abstract truth but act contrary to the constitutional text in practice. This is a
noble lie, and it is justified because it is necessary for the common good.
Notice once again, that this argument actually accepts that that the Argument from
Transparency offers a valid pro tanto reason for acceptance of the Constraint Principle.
The noble lie argument does not defeat the pro tanto force of the transparency argument:
Lies are only “noble” if they are supported by very good reasons.
150
See supra Part I.B.3.f), p. 17.
76
The Constraint Principle
Once again, different rivals of constraint fare differently with respect to transparency.
For example, the superlegislature theory is completely transparent, but for that reason is
likely outside the feasible choice set.151 Common law constitutionalism fares poorly on
the transparency scale—if one believes that the common-law method disguises judging
on the basis of private first-order judgments of political morality (or even worse, politics
and ideology) by cloaking the real basis for judicial decisions in legalistic language of the
common-law method. The moral readings theory is transparent in theory, but in its
transparent form, it seems unlikely that it will be openly avowed in any case in which it
clashes with the constraint principle.
3. The Judicial Role Argument
The gist of the argument is that the judicial role does not include the power to amend
the constitution, and therefore amending judicial constructions have a legitimacy
problem. By contrast, no one would think that a judge who complies with the
constitution is acting illegitimately for that reason.
Although the primary target of the judicial role argument is the judiciary, a similar
argument applies to other officials when they act outside the Article V amendment
process. As applied to the President, the idea is that the President has executive power,
but lacks the power to amend the Constitution. The role of President does not include the
power to amend, circumvent, or nullify the constitution. As presented on this occasion,
the focus of the official roles argument will be on judges—but we will briefly consider
other officials, including the President.
What evidence do we have for the notion that the role of officials such as the President
and Justices of the Supreme Court does not include the power to amend the Constitution?
One powerful piece of evidence is the oath that these officials are required to swear or
affirm. For example, the President of the United States must swear the following oath:
I do solemnly swear (or affirm) that I will faithfully execute the Office of
President of the United States, and will to the best of my Ability, preserve, protect
and defend the Constitution of the United States.152
And the remaining officials of the government, including the Justices of the Supreme
Court are bound by oath as well.
The Senators and Representatives before mentioned, and the Members of the
several State Legislatures, and all executive and judicial Officers, both of the
United States and of the several States, shall be bound by Oath or Affirmation, to
support this Constitution; but no religious Test shall ever be required as a
Qualification to any Office or public Trust under the United States.153
The rules that constitute the government of the United States are provided in the
Constitution. Those rules specify roles. And the plan that established these roles
151
See Lawrence B. Solum, Constitutional Possibilities, 83 IND. L. J. 307 (2008).
U.S. Const. Art. II, Sec. 1.
153
U.S. Const. Art. VI.
152
77
Lawrence B. Solum
provides that individuals empowered to serve in official roles must are bound to support
the rules—that is “this Constitution.”
The phrase “this Constitution” is significant. The first word in the phrase is “this” and
that word is an indexical.154 It specifies that the relevant “Constitution” is “this”
Constitution; in other words, the constitutional text refers to itself. Officials are not
bound to obey constitutional decisions of the Supreme Court or the constitutional
conventions created by Congress or the President. They are bound by the constitutional
text. And this does not mean that they are somehow bound to mere marks on a piece of
parchment. In context, they are required by oath or affirmation to bind themselves to
communicative content (or linguistic meaning) of the constitutional text. If the Fixation
Thesis is true, then means they are bound by the original meaning of the constitutional
text.
Justices of the Supreme Court are required to bind themselves to “this Constitution”
by oath or affirmation. They swear an oath that commits them to what amounts in
substance to the Constraint Principle. This oath is a prerequisite to their assumption of
the office, and the oath defines the limits of the role of judge. The same is true of other
officials, including all other state and federal judges. This means that the officials of the
United States, including Supreme Court justices, have at least a pro tanto reason to
perform their official actions in a way that is constrained by the original meaning of the
constitutional text.
Of course, it is possible for officials to disregard their oath or affirmation and to act in
ways that are inconsistent with the rules that constitute the United States government.
Thus, the President might assume the powers of a dictator and dissolve Congress and
suspend the operation of the Supreme Court. But if the President did this, he would be
acting contrary to his commitment to the scheme of agency established by the
Constitution. If the President succeeded (through military force or demagogic
persuasion), he would have effectively dissolved the Government of the United States
and established a new constitutional regime. Such an action is ultra vires—it is not an
action that is within the scope of his role as a constituent component of the group agent.
And hence he action would be illegitimate.
Similarly, Justices of the Supreme Court can act in ways that are outside the scope of
their roles in the government of the United States. Given the system established by the
Constitution, the United States Supreme Court is given the power to resolve cases and
controversies by offering authoritative interpretations and constructions of the
constitutional text. Given the power of the lower courts to enforce their judgments
through binding coercive orders (including injunctions and writs) and given the rules of
vertical stare decisis, this entails that the Justices can attempt to evade the restrictions on
their official role that bind them to the original meaning of the constitutional text in a
particularly insidious way. They can attempt to convert ultra vires actions into authorized
actions by judicial fiat. But the Supreme Court does not have the power to convert falsity
into truth or faithlessness into fidelity. The Justices can claim their actions are authorized
and they may be able to coerce cooperation by other officials—but when they do this, the
act contrary to the reasons that apply to them in their official capacity.
154
See Christopher R. Green, 'This Constitution': Constitutional Indexicals as a Basis for Textualist
Semi-Originalism, 84 NOTRE DAME L. REV. 1607 (2009).
78
The Constraint Principle
***
At this point some readers may object that the oath does not require that judges
respect the constitutional text, but instead is a more abstract and flexible commitment.
Perhaps the oath is to the constitutional doctrine enunciated by the Supreme Court
through a process of common law decisionmaking? Maybe the oath is to the best moral
reading of our practices as a whole and thus authorizes constructions that override the
constitutional text? Or perhaps the oath requires judges to engage in a complex
argumentative practice that allows them to nullify the constitutional text on the basis of
historical practice or constitutional values? These possibilities might be advanced by
sophisticated constitutional theorists, but to me they seem silly. The oath says none of
these things: parsing “this Constitution” as “my favorite constitutional theory that
allows overrides of the constitutional text” might be charitably construed as the result of
an unconscious process of motivated reasoning. Less charitably, this move might be
characterized as deliberate obfuscation or deception.
***
C. The Relationship Between the Rule of Law and Legitimacy
There are ways that the Supreme Court could ameliorate the effect of violations of the
Constraint Principle on the rule of law. For example, the Court might adopt a rigid
doctrine of vertical and horizontal stare decisis that approximates legislation. Rather than
deciding “one case at a time,” the Court could include a statute-like legislative holding in
each opinion. The Court could then police the lower federal courts for violations of its
holdings. The Court could adopt a notice-and-comment procedure, inviting amicus briefs
in cases in which it announces that it plans to “hold hearings” with an aim to revising or
repealing a legislative holding. This way of proceeding would implement the view that
the Court is a superlegislature.155
Turning the Supreme Court into a superlegislature that holds hearings and issues
legislative holdings would minimize the rule-of-law problems that attend its current
practices, but it would maximize the legitimacy problems. No one should think that an
unelected superlegislature that explicitly governs as such has democratic legitimacy. And
the superlegislature version of the Court abandons any pretense of proper judicial role.
The Constraint Principle fares better than the status quo with respect to both
legitimacy and the rule of law. It is difficult for nonoriginalist theories to achieve both
simultaneously. Some forms of nonoriginalism overcome rule of law problems at the
expense of legitimacy; others achieve legitimacy, but undermine the rule of law. These
tradeoffs are discussed in more detail below when we turn to pairwise comparisons.156
V. ALTERNATIVE JUSTIFICATIONS FOR THE CONSTRAINT PRINCIPLE?
Originalism is a family of constitutional theories unified by the Fixation Thesis and
the Constraint Principle. Originalists differ on a variety of matters, including the
justifications for originalism. In this Part of the Article, we will briefly investigate some
155
156
See infra Part VII.A.6, p. 115.
See infra Part VII, p. 104.
79
Lawrence B. Solum
of the alternative justifications for originalism and their relationship to the Constraint
Principle.
A. Is the Constraint Principle the Law?
One possible justification for originalism is the argument that originalism is the law,
and hence that violations of the Constraint Principle are unlawful. This argument has
been developed in a powerful and sophisticated form by William Baude in his magisterial
Is Originalism Our Law?157 Baude identifies what he calls “inclusive originalism” as
follows:
A moderate position would be “inclusive originalism.” Under inclusive
originalism, the original meaning of the Constitution is the ultimate criterion for
constitutional law, including of the validity of other methods of interpretation or
decision. This means that judges can look to precedent, policy, or practice, but only
to the extent that the original meaning incorporates or permits them.158
Baude’s “inclusive originalism” is roughly equivalent to what I call “Constraint as
Consistency.” Baude’s claim is that a good case can be made that inclusive originalism is
“our law,” assuming a positivist theory of law (such as H.L.A. Hart’s theory). The
difficulty for this claim, as Baude recognizes, is that decisions of the Supreme Court that
seem to violate the Constraint Principle are recognized as law by the practices of
officials, and hence would seem to be part of the positive law. Baude’s strategy is to
differentiate between higher order practices (which are committed to constraint) and
lower order practices (which are not).159 In an unpublished work in progress, Semantic
Originalism, I made a similar argument.160
For the purposes of this Article, I will remain agnostic about the ultimate correctness
of the claim that the Constraint Principle is the law. Suppose that it is. That fact would
not resolve the debate over whether constitutional practice should accept the Constraint
Principle—because the advocates of nonoriginalism could still affirm their views as a
proposal for reform of our constitutional practice. Now suppose that the Constraint
Principle is not the law. The tables are turned and it would be the originalists who would
argue that their theory ought to be adopted as a reform of our higher order constitutional
practices. Perhaps this all goes without saying: the fact that originalism is (or is not) the
law does not settle the question whether it should be the law.
Of course, if originalism (including constraint) is already the law and if there is a
general obligation to follow the law, then we have a reason to comply with the Constraint
Principle until and unless the law is changed—but this argumentative strategy is not
pursued on this occasion. The questions whether there is a general obligation to obey the
law and whether there is a virtue of lawfulness are deep and for that reason, beyond the
scope of this Article.
157
See Baude, supra note 16.
Id. at 2355.
159
Id. at 2365.
160
Lawrence B. Solum, Semantic Originalism, p. 140.
158
80
The Constraint Principle
B. Is the Constraint Principle Implied by Writtenness?
Does the Constraint Principle follow from the fact that the Constitution is written?
Andrew Coan has explored this question in depth in his The Irrelevance of Writtenness in
Constitutional Interpretation.161 But the question is not well formulated. Of course,
there can be a written constitution that is not binding: the Constitution of the Confederate
States of America is written but should not constrain our constitutional practice. And
there could be an oral constitution that would be binding. Moreover, there could be a
written constitution that was in force (unlike the confederate constitution) but which
authorized violations of the Constraint Principle: imagine a Twenty-Eighth Amendment
that read: “The Supreme Court shall be authorized to decide cases and controversies on
the basis of amending constructions of this Constitution when such constructions are
required to bring its provisions into conformity with changing circumstances and values.”
Our constitution does not include such a provision, but it could and this possibility is
sufficient to show that writtenness does not entail the Constraint Principle.
Suppose that Coan is correct and that the argument from writtenness fails. What
follows? Coan suggests the following answer to this question:
The failure of the argument from writtenness has profound implications for
originalism. Without it, the New Originalism--at least in its normative aspect--looks
strikingly similar to the old originalism of the 1970s and 1980s. The shift from
original intent to original public meaning may sidestep some difficult questions
about the possibility of aggregating collective intentions. But stripped of the
argument from writtenness, the affirmative case for the New Originalism largely
boils down to two very old arguments: popular sovereignty and the need to
constrain government officials, especially judges.162
Stripped of polemics, Coan’s point is that originalists must make normative arguments
for originalism. And so they must. Coan is wrong that the only arguments are the old
ones. This Article demonstrates that “popular sovereignty” and “the need to constrain
government officials” do not provide the only justifications for originalism, and perhaps
more importantly, it shows that closely related ideas do provide pro tanto justifications
for the Constraint Principle. Nonetheless, Coan was on to something important: concern
for the rule of law and for legitimacy do provide the strongest reasons of political
morality for originalism.
C. Does the Constraint Principle Lead to Good Consequences?
In The Good Constitution, John McGinnis and Michael Rappaport make a powerful
case that the supermajoritarian procedures for the adoption and amendment of the
Constitution make it likely that the compliance with the Constraint Principle is more
likely to produce good consequences than some alternatives, especially constitutional
amendment by a committee of nine persons serving life terms.163 On the other hand, it
161
Andrew B. Coan, The Irrelevance of Writtenness in Constitutional Interpretation, 158 U. Pa. L. Rev.
1025 (2010).
162
Id. at 1088.
163
See McGinnis & Rappaport, supra note 60
81
Lawrence B. Solum
might be argued that judicial updating is required for consequentialist reasons, especially
for those provisions of the Constitution that are very old: one might argue that these
provisions become less and less likely to be justified on consequentialist grounds as time
goes by.164 Or more directly, one might argue that the regulatory state produces good
consequences, but that originalism would require striking down the delegation of powers
to independent agencies and regulation of matters outside the original meaning of
“commerce among the several States.”165
These are complex issues and their resolution would require an extended investigation.
Arguendo, let us assume that an on-balance consequentialist assessment of the Constraint
Principle is very difficult or impossible. If this is the case, then we must turn to other
considerations, including the concerns about the rule of law and legitimacy explored in
this Article.
D. Is the Constraint Principle the Only Alternative to Judicial Lawlessness?
Finally, reconsider Justice Scalia’s argument for originalism, quoted in full above.
Scalia argues that (1) it takes a theory to beat a theory, (2) consistency and predictability
require a consistent methodology, (3) there is no “emerging consensus among the
nonoriginalists” on an alternative to originalism.166 How do these ideas relate to the case
for constraint offered here?
Consider the idea that originalism must be compared with an alternative. We can
express this as the idea that it takes a normative theory to beat a normative theory.167
Surely this is correct. But Scalia seems to imply that originalism has no serious rivals,
other than the moral readings theory. On this score, he is surely wrong. There are at least
nine alternatives to originalism, and not all of them lead to judicial lawlessness. For
example, the Thayerian family of alternatives substitutes judicial restraint (defer to
Congress) for the Constraint Principle. Hence, if originalism is to be justified as the best
theory of constitutional interpretation and construction, it must be compared pairwise to
each of its significant rivals. It is a mistake to claim that originalism can only be defeated
if its opponents agree on a single alternative. The requirement of agreement is arbitrary
and unwarranted: if one alternative to originalism is superior and feasible, then that
alternative should be adopted. Making a compelling case for originalism requires that
each alternative be examined and compared to originalism.
VI. OBJECTIONS AND ANSWERS
In this Article, we cannot hope to canvas all of the objections to the Constraint
Principle and provide a thorough and complete answer to each one. Indeed, each of the
important objections to originalism should be the subject of an extended treatment—an
Article or monograph of its own. So, the strategy here will be to focus on just a few of
164
See Adam M. Samaha, Dead Hand Arguments and Constitutional Interpretation, 108 COLUM. L.
REV. 606 (2008).
165
U.S. Const. Art. I, sec. 8.
166
Antonin Scalia, Originalism: The Lesser Evil, 57 U. CIN. L. REV. 849, 855 (1989).
167
Lawrence B. Solum, Legal Theory Lexicon 053: It Takes a Theory to Beat a Theory, Legal Theory
Lexicon (July 12, 2015), http://lsolum.typepad.com/legal_theory_lexicon/2006/08/legal_theory_le_1.html.
82
The Constraint Principle
the most important objections and to sketch lines of reply that draw on the framework
established in Part III above.168
A. The Canonical Cases Objection
It is commonly assumed that originalism is in trouble if it cannot be squared with the
canonical constitutional cases, decisions like Brown v. Board or Marbury v. Madison
which are assumed to be normatively correct and beyond legal challenge.169 If
originalism is inconsistent with a canonical case, so the objection goes, then originalism
must be wrong.170
Although the canonical cases objection is in the air, I believe that it is radically
undertheorized. The objection is sometimes presented as if it were a conversation
stopper: “Originalism is inconsistent with Brown. Game over.”171 The idea is that it
would be unthinkable for Brown to be reversed or to have come out the other way in the
first place. Any theory that undermines Brown must be wrong.
One line of reply to the canonical cases objection is the attempt to show that these
cases are consistent with originalism. Thus, William Baude writes, “Indeed, the
canonical cases that are most frequently invoked as examples of anti-originalism are
actually reconcilable with originalism.”172 Famously, Michael McConnell has argued
that Brown v. Board is supported by the original understanding of the Privileges or
168
See supra Part III, p. 28.
Jack M. Balkin, The Distribution of Political Faith, 71 Md. L. Rev. 1144, 1157 (2012) (“A far more
promising solution for conservative originalists is to rework their understanding of originalism's
commitments and entailments so that the current set of canonical cases were always correct.”); Dan T.
Coenen, The Pros and Cons of Politically Reversible "Semisubstantive" Constitutional Rules, 77 FORDHAM
L. REV. 2835, 2891 (2009) (“Critics of originalism say, for example, that originalists must reject the Court's
canonical decision in Brown v. Board of Education because the ratifiers of the Fourteenth Amendment had
no intention to end school segregation.”); Michael C. Dorf, Living Originalism. by Jack M. Balkin.
Cambridge, Mass.: The Belknap Press of Harvard University Press. 2011. Pp. VI, 474. $35.00. the Living
Constitution. by David A. Strauss. New York, N.Y.: Oxford University Press. 20, 125 Harv. L. Rev. 2011,
2046 (2012) (“In The Living Constitution, Strauss aims to establish the superiority of the common law
method of constitutional interpretation over originalism by showing how the beloved modern case law of
free speech and the now-canonical decision in Brown v. Board arose through the common law method.”).
170
For an extended blog discussion of the argument, see Michael Dorf, Stare Decisis and Originalism,
Dorf on Law, October 23, 2015, http://www.dorfonlaw.org/2015/10/stare-decisis-and-originalism.html;
Lawrence B. Solum, Three Questions for Professor Dorf: Stare Decisis and Originalism, Legal Theory
Blog, October 23, 2015, http://lsolum.typepad.com/legaltheory/2015/10/some-questions-for-professor-dorforiginalism-stare-decisis-and-brown.html; Michael Dorf, Further Thoughts on Originalism and Stare
Decisis--In Response to Prof. Solum, Dorf on Law, October 26, 2015,
http://www.dorfonlaw.org/2015/10/further-thoughts-on-originalism-and.html; Lawrence B. Solum, Dorf on
Public Meaning Originalism: Clarifying the Nature of the Debate, October 26, 2015
http://lsolum.typepad.com/legaltheory/2015/10/dorf-on-semantic-originalism-clarifying-the-nature-of-thedebate.html.; Michael Dorf, Canonical Case Skepticism and the Cartoon Version of Nonoriginalism,
October 28, 2015, http://www.dorfonlaw.org/2015/10/canonical-case-skepticism-and-cartoon.html.
171
See Eric A. Posner & Adrian Vermeule, Transitional Justice as Ordinary Justice, 117 HARV. L. REV.
761, 782 (2004) (“By about the mid-1980s, however, Brown had become a canonical case in American
constitutional law--a fixed point whose validity had to be presumed by any viable account of constitutional
theory.”).
172
Baude, supra note 16, at 2371.
169
83
Lawrence B. Solum
Immunities Clause of the Fourteenth Amendment.173 Assuming, arguendo, that they are
correct, this would pose a problem for some nonoriginalists who accept the canonical
cases argument, because many nonoriginalist theories seem to undermine the outcome in
Brown v. Board. This is especially true of common-law constitutionalism and the
multiple-modalities approach for obvious reasons given Plessy v. Ferguson174 and longstanding segregationist practice. Unconstrained Thayerianism would not have led to
Brown. Other theories are capable of generating Brown but are equally capable of
supporting a different outcome: a superlegislature could have produced Brown but
depending on the preferences of the nine Justices, the opposite could be true.
For all of these reasons, it is not clear which way the Canonical Cases Objection cuts,
assuming that the objection is even sound. Part of the difficulty is that the objection
seems to require us to inquire into complex counterfactual history, asking questions like,
“What if the Constraint Principle had been observed throughout American history?” If
that is the relevant question, then the makers of the Canonical Cases Objection have a lot
of work to do. Perhaps, they believe that there are entitled to cherry pick and ask the
counterfactual questions about some subset of cases—just the cases that are least
favorable for originalism, for example. But it is not clear whether this version of the
objection makes sense, and it is obvious that cherry picking the unfavorable cases suffers
from selection bias.
Moreover, the use of Brown as an unexplained and untheorized trump card in
arguments against originalism oversimplifies issues that are complex. What if the
Constraint Principle is inconsistent with Plessy and without Plessy there would have been
no Jim Crow? Plessy might be thought of as an Anti-canonical case.175 But Plessy is
consistent with a variety of the theories that deny the Constraint Principle. The judges
who decided Plessy likely saw that case as supported by a moral reading of the
constitution; Thayerian deference supports Plessy; and so forth.
In addition, it is not clear why canonical cases count as theoretical trump cards. What
if Brown v. Board played no or little role in ending Jim Crow—as might be suggested by
Gerald Rosenberg’s The Hollow Hope.176 The psychological force of the Brown version
of the Canonical Cases Objection may rest on the notion Brown was the key causal factor
in ending de jure segregation, but that is a complex empirical question that is rarely
addressed when Brown is invoked as a decisive objection to originalism. It would seem
reasonable to ask the proponent of the canonical cases argument to spell out their
argument: surely hand waving is not sufficient.
Perhaps the canonical cases argument should be understood as a positivist objection to
originalism:
Premise: Brown v. Board is legally correct, because everyone agrees that it is.
173
See Michael W. McConnell, Originalism and the Desegregation Decisions, 81 VA. L. REV. 947
(1995); see also John Harrison, Reconstructing the Privileges or Immunities Clause, 101 YALE L.J. 1385
(1992); Michael B. Rappaport, Originalism and the Colorblind Constitution,
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2244610 (2013).
174
163 U.S. 537 (1896).
175
Jamal Greene, The Anticanon, 125 HARV. L. REV. 379 (2011).
176
GERALD N. ROSENBERG, THE HOLLOW HOPE 70–71 (2d ed. 2008).
84
The Constraint Principle
Premise: If the Constraint Principle were true as a matter of law, then Brown v.
Board would be legally incorrect.
Conclusion: Therefore, the Constraint Principle is not true as a matter of law.
(Modus Tollens)
Of course, this argument would only show that the Constraint Principle is not the positive
law—not that we should reject the Constraint Principle as a matter of political morality.
Moreover, the argument is quite odd, since it is obvious that Brown itself was not the law
until the Supreme Court made it so. Moreover, many nonoriginalist theories are
consistent with a different outcome in Brown—most obviously, Common Law
Constitutionalism and the Multiple Modalities Theory, either of which could have been
used to justify a different outcome in Brown.
And why should we decide which constitutional theory to adopt on the basis of a
handful of canonical cases? If we employ the method of reflective equilibrium (either
wide or narrow), then it is very odd indeed to limit our gaze to just a few canonical cases
and to ignore the rest. Moreover, the Canonical Cases Objection seems to assume that we
should ignore our more abstract and general beliefs and privilege our intuitions about the
cases that are called “canonical,” but again this is inconsistent with the method of
reflective equilibrium.
There is another way to understand the Canonical Cases Objection: perhaps, the
argument is that morality requires that the canonical cases be decided as they were. That
objection (the Injustice Objection) will be considered below.177 The Injustice Objection
version of the Canonical Cases Objection is not really about “canonicity” but is instead
about the alleged substantive injustice of originalist outcomes.
One final note about the Canonical Cases Objection. To the extent that the objection is
based on cases where a contrary result would be unthinkable given the current status of
legal and social norms, it is not clear whether the adoption of the Constraint Principle
would undermine the outcome or effect of these cases. Take Brown v. Board as an
example. Considering the worst case for originalism, let us assume that Brown was
contrary to the original meaning of the Fourteenth Amendment. No one thinks that
realization that Brown was legally incorrect would result in the reinstatement of Jim
Crown and the overruling of Brown. If the Canonical Cases Objection is not about the
future, it must be about the past—and more precisely, about an alternative version of the
past in which the Constraint Principle prevented Brown from being decided in the first
place.
Once we focus on the counterfactual nature of the Canonical Cases Objection, it
becomes unclear what impact the objection is supposed to have. To my knowledge, no
one who has pressed the objection has tried to show that Brown v. Board was a necessary
precondition for ending school segregation (that involves complex counterfactuals)—and
there is doubt that Brown actually played a significant causal role.
177
See infra Part VI.C, p. 87.
85
Lawrence B. Solum
B. The Disruptive Force of an Originalist Big Bang
Another objection goes to the disruptive force of an originalist “big bang.”178 Imagine
that there were five originalist Justices on the United States Supreme Court. Further
imagine a “superterm” in which the Court reformulated all of constitutional doctrine to
conform to the original meaning of the constitutional text. An originalist big bang might
have substantial disruptive effect, invalidating a large number of statutes and requiring
substantial restructuring of the administrative state. This would be especially true if the
original meaning of the Constitution with both the delegation of legislative power to
administrative agencies and with the expansion of national legislative power.
But there is no reason that a reasonable implementation of the Constraint Principle
must embrace the big-bang model of constitutional change. Instead we can imagine a
multiphase approach to the implementation of originalism:
Phase One: In phase one, the Supreme Court might adopt a principle of “this far but
no farther,” allowing existing violations of the Constraint Principle to stand, but
striking down new legislation and executive actions that are inconstant with original
meaning.
Phase Two: In phase two, the Court could selectively strike down the most egregious
violations—taking disruption and reliance into account.
Phase Three: In phase three, the Court could begin to dismantle the remaining
constitutional violations on a case-by-case basis.
As this process preceded, the political system would be likely to react—especially as the
transition from Phase Two to Phase Three began. For example, a constitutional
amendment giving Congress the power to regulate the environment might be enacted if
the Supreme Court signaled the possibility of future invalidation of key environmental
legislation by striking down a new environmental statute. Similarly, there could a
constitutional amendment authorizing the delegation of legislative power to the
executive. The Court might postpone acting on some issues once it was clear that the
amendment process was underway and likely to succeed.
Of course, a gradual transition to originalism would be slow. The whole process
would likely take decades—perhaps several decades would be required. But nothing
about a gradual transition to originalism is inconsistent with a commitment to the
Constraint Principle in the long run.
In this regard, it is important to recall the implementation of progressive realism in
constitutional practice did not involve an “antioriginalist big bang.” James Thayer’s
influential writings appeared just a few years before the turn of the nineteenth to the
twentieth century.179 Holmes’s influence on constitutional law began a few years after
that.180 Although the period following 1937 involved a relatively rapid rate of
constitutional change, the full impact of progressive realism occurred during the Warren
178
See Lawrence B. Solum, The Supreme Court in Bondage, supra note 107, at 194.
See James Bradley Thayer, The Origin and Scope of American Constitutional Law, 7 HARV. L. REV.
129 (1893).
180
See THE OXFORD COMPANION GUIDE TO THE SUPREME COURT OF THE UNITED STATES 1132 app. B
(2d ed. 2005) (Justice Holmes was appointed by President Roosevelt August 11, 1902.).
179
86
The Constraint Principle
and Berger Court eras and continues today.181 There is no reason to believe that a
transition to originalism could proceed at a much faster pace than did the demise of
constitutional formalism. Surely, the relevant time frame is decades and not two or three
years.
C. Three Versions of the Injustice Objection
Another objection to the Constraint Principle might focus on the possibility that
constraint might prevent the Supreme Court from making decisions that will advance
substantive justice in ways that are inconsistent with the original meaning of the
constitutional text. Two versions of this objection will be considered—one that focuses
on constitutional imperfection and another that is limited to cases of grave constitutional
evil.
1. The Injustice Objection, Version One: Constitutional Imperfection
It seems extraordinarily unlikely that the Constraint Principle will lead to
constitutional perfection. It is surely the case that there are some constitutional outcomes
that are just or morally desirable but that are not consistent with the original meaning of
the constitutional text. This objection to the Constraint Principle also applies to many of
its rivals. Unconstrained Thayerianism leads to unjust outcomes whenever Congress acts
unjustly. The multiple modalities theory does not prioritize justice as the sole or decisive
modality of constitutional argument; this means that arguments from text, history,
structure, or doctrine might result in an unjust outcome. There is not guarantee that
common law constitutionalism will lead to perfectly just outcomes; indeed, the commonlaw process has a built-in bias towards the past that is quite different than a method that
aims at constitutional perfection.
The view that comes closest to providing a pathway to constitutional perfection is the
moral readings theory. As a matter of ideal theory, a moral readings approach might
approximate constitutional perfection, but even the advocates of this approach concede
that moral readings are constrained and maintain that we nonetheless owe a duty of
fidelity to “our imperfect constitution.”182 But as a matter of nonideal theory, the ability
of constructive interpretation to deliver a morally perfect constitution is dependent on the
first-order private judgments of political morality of the individual Justices and the
processes that structure their interactions. From the perspective of any one citizen, this
means that the Supreme Court could well be aiming at something other than moral
perfection. Indeed, it seems highly likely that the large majority of individual citizens
would deny that the Court aims at moral perfection if they were fully informed about all
of the moral beliefs of the Justices. Progressives are unlikely to want a conservative
181
This is a complex and long story, and it is simply impractical to provide fully adequate
documentation. For a variety of perspectives, see William D. (Bill) Graves, Evolution, the Supreme Court,
and the Destruction of Constitutional Jurisprudence, 13 REGENT U. L. REV. 513 (2001); Edward A.
Purcell, Jr., Democracy, the Constitution, and Legal Positivism in America: Lessons from A Winding and
Troubled History, 66 FLA. L. REV. 1457 (2014); Raoul Berger, The Activist Legacy of the New Deal Court,
59 WASH. L. REV. 751 (1984).
182
See JAMES FLEMING, FIDELITY TO OUR IMPERFECT CONSTITUTION: FOR MORAL READINGS AND
AGAINST ORIGINALISMS (2015).
87
Lawrence B. Solum
court to aim at the Justices’ vision of moral perfection, and vice versa. For this reason, a
progressive might prefer a constrained conservative court over an unconstrained
conservative court; conservatives might prefer a constrained progressive court over one
that is unconstrained.
Another way to see the difficulty with moral perfection as a goal for constitutional
theory focuses on the difference between wide and narrow reflective equilibrium.
Individuals or cohesive groups may be able to reach agreement on the content of a
morally perfect constitution, but this is only because the method of narrow reflective
equilibrium allows us to disregard the views of those with whom we disagree. Once we
employ the method of wide reflective equilibrium, it becomes apparent that a morally
perfect constitution is not a feasible goal for a pluralist society in which reasonable
citizens disagree about what would constitute perfection.
For all these reasons, constitutional imperfection fails to provide a good and sufficient
reason to reject the Constraint Principle.
2. The Injustice Objection, Version Two: Constitutional Evil
Consider the more difficult case of constitutional evil. Should we follow adopt the
Constraint Principle if it requires us to accept grave evils that could otherwise be avoided.
In this regard, consider the case of slavery. If we assume that the Constitution prior to the
Thirteenth Amendment was a pro-slavery document, does this provide us with a reason to
reject the Constraint Principle?
The problem of evil positive law is a very general one, not limited to the constitutional
context. The question whether there is an obligation to obey evil laws connects with
general debates over the obligation to obey the law. Let us suppose that our conclusion
about is that citizens and officials have no obligation to obey laws that require grave evil.
One plausible view is that even judges should disobey the law in such cases, and should
instead act make the best of a bad situation. That might entail resigning one’s position as
judge and joining a resistance movement. Or a judge might engage in dissimulation,
attempting to disguise a decision that is contrary to law with a sophistic cloak of legality.
The generality of the problem is illuminated by noting that it occurs for lower court
judges faced with a Supreme Court decision that would require grave evil. In such a
case, the lower court judge might conclude that there is no moral obligation to respect the
decision of the Supreme Court and attempt to reach a just outcome by deliberating
engaging in results oriented factfinding or manipulating procedural rules.
Citizens and officials who believe that the original meaning of the United States
Constitution today sanctions grave evil have good moral reasons to reject the Constraint
Principle as a matter of principal, although they might observe constraint in some cases if
that were the best method to minimize the effects of the evil provisions of the
Constitution. And those who believe that the decisions of the Supreme Court or the
statutes enacted by Congress are gravely evil have good moral reasons to reject constraint
by judicial decision or by statute.
It is possible that some legal scholars, lawyers, and judges believe that the
Constitution is gravely evil and that they are therefore warranted in violating the
Constraint Principle. But this rational does not apply to those who believe we have a
reasonably just constitution. The objection from constitutional evil is rarely articulated
by scholars, and is unlikely to be affirmed by judges or elected officials. Whether or not
88
The Constraint Principle
the argument is true, it is far outside the mainstream of public discourse in the United
States. No nominee who openly stated that the Constitution is gravely evil could be
confirmed, and similarly, no candidate for office who openly took this position would be
electable given current political conditions.
Nonetheless, originalists ought to articulate the case that the Constitution is reasonably
just. And such an argument ought to acknowledge the grave defects that attended the
Constitution before the adoption of the Thirteenth Amendment. Slavery was a grave evil,
and a constitution that allowed and partially supported slavery is an evil constitution.
Whether the current constitution is similarly evil is a different question, the answer to
which is simply beyond the scope of this Article.
Even if the Constitution itself is not gravely evil, it is possible that the document is
defective because it tolerates grave evil by elected officials or private citizens. Thus, one
might argue that the Constraint Principle stands in the way of amending constructions of
the Constitution that would address grave evils. Consider the implications of this
argument for the controversy over abortion. Opponents of abortion might argue that
permitting abortion is a grave evil, and hence that the Supreme Court should recognize a
constitutional right to life invalidates legislation permitting or encouraging abortion.
Proponents of abortion might argue that denying women the right to choose whether to
take a pregnancy to term is a grave evil, and hence that there should be a constitutional
right to choice. Both the opponents and the proponents of abortion might be able to reach
narrow reflective equilibrium on their preferred constitutional construction. Given that a
grave evil is involved, each side would seem to have good reason to attempt to stack the
Supreme Court with candidates who would reliably vote for a constitutional construction
that would avoid the grave evil. If both sides pursued this course, it would result in the
politicization of both the particular constitutional issue and potentially result in the
politicization of the Supreme Court itself—since Justices selected for ideological (or
moral) purity would be disposed to pursue preferred results and hence to subordinate the
rule of law.
It will be difficult to reach wide reflective equilibrium when reasonable citizens agree
that an issue involves grave evil, but disagree about which outcome is evil and which is
good. Could they reach agreement on the Constraint Principle under these
circumstances? Consider two scenarios. In the first scenario, the communicative content
of the constitutional text is neutral with respect to the issue of grave evil: in the case of
abortion, suppose that the text neither creates a constitutional right to choice nor a
constitutional right to life. This would leave the issue for resolution by democratic
politics. Either side might object that a political solution is unacceptable, because the
issue is one of gravel evil, but this objection fails to recognize that the constitutional
resolution of the controversy would itself be a political solution. Given this fact, both
sides might come to see that the Constraint Principle provides the most reasonable
solution—the lesser evil.
Now consider the second scenario. Suppose that the original communicative content
of the constitutional text does address the issue of grave evil. This would mean that
acceptance of the Constraint Principle would guarantee a loss for the side that the
Constitution disfavors—unless they can muster the supermajority required for a
constitutional amendment. Moreover, political conditions may be such that there is no
realistic prospect for a constitutional amendment in the short term, while at the same time
89
Lawrence B. Solum
there is a realistic chance that majority of the Justices would adopt an amending
construction that would override the original meaning and eliminate the grave evil. If the
moral importance of the grave evil outweighs the rule of law and legitimacy concerns that
support the Constraint Principle, then it will be impossible to reach an overlapping
consensus for or against the Constraint Principle as a matter of principal. Nonetheless,
the side that is disadvantaged by the original meaning might nonetheless accept the
Constraint Principle provisionally: violating the Constraint Principle would make sense if
one were assured of victory but would accomplish nothing if political conditions were
such that one could not obtain the necessary votes in the Supreme Court.
Under the conditions of the second scenario, there would be a good reason to reject the
Constraint Principle from the point of view of someone who believed (1) that the original
meaning of the Constitution perpetuated a gravel evil and (2) that violation of the
Constraint Principle was the only feasible alternative to allowing the grave evil to persist.
Whether the actual United States Constitution does require grave evil is an important
question. But if it does, the implications are much more radical than mere rejection of the
Constraint Principle. Suppose that we are in the conditions of second scenario, except
that there is no realistic hope of persuading a majority of the Supreme Court to adopt an
amending construction of the Constitution that averts the grave evil. These conditions
would warrant armed revolution, deliberate rigging of elections, and other extreme
measures. If you believe that the constitution sufficiently just such that neither
revolutionary violence nor a coup d’état is justifiable in the absence of judicial correction,
then you should also believe that it is sufficiently just for the Constraint Principle to hold.
3. The Injustice Objection, Version Three: Particular Wrongs
Some readers may find the discussion of the Injustice Objection up to this point to
overly abstract. They may focus instead on particular injustices that the Supreme Court
might correct. There are many possible versions of the Particular Wrongs variation on
the Injustice Objection. In the discussion that follows, we will focus on two, gender
discrimination and same-sex marriage. Much of the discussion above applies to these
particular wrongs, but here the focus on these particular issues and their relationship to
originalism.
The focus on particular wrongs naturally leads to consideration of the pairwise
comparison between the Constraint Principle and Popular Constitutionalism. This is
because social movements may coalesce around constitutional litigation targeted at
particular wrongs as the means by which to achieve their agenda for change. Professor
David Cole has explored this phenomenon in his monograph, Engines of Liberty.183 Both
the Supreme Court’s gender equality jurisprudence and its marriage equality decisions are
plausibly viewed as instances of popular constitutionalism. Pairwise comparison of
constraint with Popular Constitutionalism will be explored in greater depth below,184 but
at this stage of the analysis, the key point is that the partisans of popular constitutionalist
social movements focused on particular wrongs are likely to take themselves as having
good and sufficient reasons to reject the Constraint Principle.
183
184
DAVID COLE, ENGINES OF DESTRUCTION (2016).
See infra Part VII.A.4, p. 111.
90
The Constraint Principle
a) The Marriage Equality Example
This abstract point can be made more concrete by considering an example—same sex
marriage. From the point of view of an advocate of marriage equality, the Constraint
Principle might appear as an obstacle to the achievement of justice. This perception
might be reinforced by the interaction between partisan politics and constitutional
politics. Although originalist arguments have appeared in opinions joined by Justices
appointed by presidents of both political parties, constitutional originalism is most
strongly associated with the late Justice Scalia and with Justice Thomas—both appointed
by Republican presidents and associated with conservatism in constitutional politics. The
hermeneutics of suspicion may be activated by this correlation. Advocates of marriage
equality may believe that conservative Justices selectively observe the Constraint
Principle for instrumental reasons to achieve their ideological ends, and they may suspect
that “originalist” Justices will violate the Constraint Principle when it conflicts with their
ideological ends.
Moreover, partisans of the popular constitutionalist social movement for marriage
equality may believe that the Constraint Principle is biased against them and that Living
Constitutionalism is on their side. Living Constitutionalism explicitly endorses the
notion that constitutional doctrine should change in response to changing circumstances
and values: the marriage equality movement argument for constitutional change was
based in part on a changing values argument. When it comes to originalism, some of the
resistance may come from an outdated understanding of the nature of originalism. Social
movements are not academic theories, and for this reason the view of originalism within
a social movement is not likely to reflect originalist constitutional theory as developed in
the academy. Rather, the perception of originalism by partisans will be shaped by the
interaction between constitutional politics and partisan politics: this process is likely to
emphasize cartoon versions of originalism (and all other constitutional theories). Thus,
proponents of marriage equality might focus on a version of originalism that emphasizes
original intentions and original expected applications. Because a right to marriage
equality would not have been among the original expected applications of the drafters of
the Fourteenth Amendment, this version of originalism might be viewed as having a
built-in bias against marriage equality.
For these reasons, partisans of marriage equality might be reluctant to consider the
possibility that there might be a public meaning originalist argument for marriage
equality. For example, it might be argued that marriage is one of the privileges or
immunities of citizens of the United States protected by Section One of the Fourteenth
Amendment. This possibility does not seem outlandish if one imagines a reconstruction
era statue that prohibited marriage by the former slaves. Even if this principle had been
recognized, it is very unlikely (or even impossible) that the right to marry would have
been extended to same-sex couples in the Nineteenth Century: the social norms against
public same sex relationships were very strong. Although the language of the Privileges
or Immunities Clause is very general and abstract, the application of open-textured
language (a matter of constitutional construction for New Originalists) is subject to causal
influences. If widely shared and deeply held social norms make a constitutional
construction unthinkable, then it simply will not occur. But things will be different once
the social norms have changed, it might then be the case that application of a general
91
Lawrence B. Solum
right of citizens to marry would plausibly support a right to same-sex marriage so long as
one member of the couple was a citizen.
At this point, it is important to clarify the point of this analysis. I am not claiming that
originalism does, in fact, support a right to same sex marriage. That would require an
adequate theory of the meaning of the Privileges or Immunities Clause—one of the most
contested topics amongst originalists.185 Rather, the point is that the opposition of the
partisans of marriage equality to originalism in general and the Constraint Principle in
particular may not be fully informed. To the extent that the bias objection fails to
consider the implications the interpretation-construction distinction and the limited role
of original expected applications in public meaning originalism, the partisans of marriage
equality may wish to reconsider their position. Given a fuller understanding of
originalism, the bias objection to originalism might drop out, if not in whole, then in part.
Of course, this would not entail that the partisans of marriage equality would endorse
the Constraint Principle. Proponents of same-sex marriage equality might take the
position that they will only support constraint if it favors a constitutional right to marriage
equality. From the point of view of a committed partisan, a “single-issue” approach to
constitutional theory might make sense. It is certainly not irrational to attach
transcendental importance to a single issue—especially if that issue is deeply connected
to one’s sense of equal citizenship or equal moral worth.186 But from the point of view of
constitutional theorists (as opposed to partisans of a cause), single-issue theory selection
is clearly problematic. Narrow reflective equilibrium might be consistent with a singleissue approach to constitutional theorizing; if the view about the single issue is strong
enough, it might dominate all other considerations—although this seems doubtful. But
the same cannot be true of wide reflective equilibrium. There are surely many
constitutional issues of great importance, and a single-issue approach risks subordinating
matters of great importance to the chosen single issue.
Partisans of marriage equality would also need to take into account the alternatives to
constraint. Obviously, many of the rivals of originalism do not guarantee marriage
equality. Unconstrained Thayerianism guarantees that there will be no judicially
enforceable right to same-sex marriage; a similar result would be likely under
Representation Reinforcement Thayerianism. Common-law constitutionalism might
eventually lead to a marriage equality right, but only when the law changes in other ways
so as to support common-law reasoning in favor of the right. Many other rivals of
originalism are simply indeterminate with respect to the issue of marriage equality: for
example, Pragmatism, Multiple-Modalities, and the Multiple Meanings Theory are all
capable of supporting or opposing a constitutional right to marriage equality.
Even if the Constraint Principle is inconsistent with a constitutional right to marriage
equality, it does not follow that marriage equality itself is inconsistent with an originalist
constitutional regime. To the extent that marriage equality is endorsed by democratic
majorities, it can be adopted by normal democratic processes at the state level. To be
185
See, e.g., Barnett, supra note 27; KURT T. LASH, THE FOURTEENTH AMENDMENT AND THE
PRIVILEGES AND IMMUNITIES OF AMERICAN CITIZENSHIP (2014); CHRISTOPHER GREEN, EQUAL
CITIZENSHIP, CIVIL RIGHTS, AND THE CONSTITUTION (2015). The cited monographs are only a sample of
the very large literature.
186
Other terminology, such as “dignity,” “equal respect,” “human rights,” or “equal liberty” might be
employed to describe a very similar idea.
92
The Constraint Principle
sure, this would mean that marriage equality would likely have been implemented in
some states at a more rapid pace than in others. And if originalism were to gain the
ascendency in the Supreme Court in the near short term (an unlikely prospect), then it is
possible that the current right to marriage equality would be overturned and that some
states would reinstate their prohibitions on same-sex marriage. If originalism were
implemented in the medium to long term, then it seems unlikely that there would be any
practical effect on same-sex marriage as it would be unlikely that state legislatures would
turn back the clock once marriage equality was fully integrated into the social fabric.
b) The Gender Discrimination Example
Another example of a particular wrong that could be raised as an objective to
originalism is gender equality. The gist of the objection would be that the original
meaning of the Equal Protection Clause would not have produced the Supreme Court’s
gender equality decisions. Several of the points that are raised with respect to the
marriage equality example apply with equal force to gender discrimination. From the
point of view of constitutional theory, it is not clear that resolution of a single
constitutional question, no matter how important, should be a decisive factor in choosing
a general method of constitutional interpretation and construction. Of course, the effect
of originalism on the constitutional treatment of gender equality is one factor that would
be taken into account in determining whether the Constraint Principle is supported by
wide reflective equilibrium.
Should advocates of gender equality oppose originalism on the ground that it is biased,
because the original meaning of the Equal Protection Clause does not support the
Supreme Court’s gender equality jurisprudence? For contemporary originalists, this
question is complicated by the distortions in Fourteenth Amendment doctrine introduced
by the Supreme Court’s decisions in the Slaughterhouse Cases.187 Again, the full story is
too complex to treat adequately here, but one plausible view is that these cases essentially
nullified the Privileges or Immunities Clauses, causing a ripple effect of doctrinal
distortion. The original meaning of the Equal Protection Clause may have focused on
“protection of the law,” guaranteeing that legal protection of persons and property (and
other legal protections) would be provided to all persons. If this reading is correct, then
the whole edifice of modern Equal Protection Clause doctrine, with its tiers of scrutiny
and various classes of persons (suspect, quasi-suspect, nonsuspect) is a substantial
departure from original meaning. The role played today by the Equal Protection Clause
might well be assigned to the Privileges or Immunities Clause, which guarantees a basic
set of equal rights to all persons.
This is especially relevant to gender equality, because the Privileges or Immunities
Clause does not differentiate between men and women—and in this regard Section One
of the Fourteenth differs from Section Two, which extends protection only to male
persons. For this reason, the plaintiff in Bradwell v. Illinois188 argued that the right of
women to practice law was protected by the Privileges or Immunities Clause; the right to
engage in a lawful profession, it was argued, is one of the rights (privileges or
immunities) that must be extended to all citizens, irrespective of gender, race, or other
187
188
83 U.S. 36 (1873).
83 U.S. 130 (1873).
93
Lawrence B. Solum
arbitrary characteristics. Because the majority in Slaughterhouse had confined the
meaning of “privileges or immunities of citizens of the United States” to a very small set
of federal rights, this argument failed, but from an originalist perspective, there is a strong
argument that it should have succeeded—as it may have done for the dissenter, Chief
Justice Salmon Chase.189 The Justices who dissented in Slaughterhouse but joined the
majority in Bradwell did so on the grounds that women were unfit to practice law, but
that rationale is not part of the meaning (communicative content) of the Clause—it is the
application of the clause to Myra Bradwell’s case based on an erroneous view of the
facts. Once the Supreme Court became aware of the true state of the facts (women are
clearly fit for the practice of law), then Bradwell would have been reversed and the
Privileges or Immunities Clause could then have become an originalist foundation for an
alternative version of constitutional gender equality jurisprudence.190
Once again, my claim is not that the original meaning of the Privileges or Immunities
Clause supports contemporary constitutional gender equality doctrine. Rather, my point
is that advocates of gender equality have not, for the most part, engaged in a serious
inquiry into the original public meaning of the relevant provisions. If this is so, then their
belief that originalism “stacks the deck” against gender equality is not well founded.
Nonetheless, one can imagine that an advocate for gender equality seeking narrow
reflective equilibrium would reject the Constraint Principle on the ground that any
general principle that might impede gender equality is inconsistent with a very firm
(almost dogmatic) conviction that there must be a constitutional right to gender equality,
“come hell or high water.” But this perspective is not appropriate for constitutional
theory for the reasons developed above.
Advocates of constitutional gender equality might also consider the possible world in
which the Constraint Principle had been observed in the 1970s, when the Supreme Court
began to develop its gender equality jurisprudence. Let us assume arguendo that the
argument for gender equality rights under the Privileges or Immunities Clause is
incorrect, what would the consequences have been? One possibility is that Equal Rights
Amendment would have been ratified. Without the Supreme Court’s intervention, which
let off some of the political steam creating pressure for ratification, the very narrow
failure of the ERA might have turned into a narrow victory. Of course, had the Equal
Rights Amendment been ratified, it seems likely that the resultant constitutional right to
gender equality would be more robust than that provided by current doctrine. The
movement for gender equality would also have been channeled into efforts for legislation
providing gender equality rights. The ultimate effect on gender equality is uncertain,
because the counterfactual involves complex empirical questions. But it is hardly clear
that the net effect on gender equality would have been negative.
***
Not all popular constitutionalist movements are “progressive”—the campaign for an
individual constitutional right to bear arms is associated with conservatives and
189
Cf. Richard L. Aynes, Bradwell v. Illinois: Chief Justice's Dissent and the “Sphere of Women's
Work,” 59 La. L. Rev. 521, 537-38 (1999) (constructing a dissenting opinion that Chief Justice Salmon
Chase might have written in Bradwell in light of his values and the role that his daughter played in his life
and career).
190
This argument is based in part on an unpublished work-in-progress by Mary Francis Rooney.
94
The Constraint Principle
libertarians. Nonetheless, the perception that popular constitutionalism and other forms
of living constitutionalism are associated with the left and that originalism is associated
with the right seems widespread. Of course, these terms have different meanings in
different contexts and when used by different authors. In constitutional politics in the
public sphere, “originalism” surely has ideological overtones. But there is a difference
between “originalism” as a constitutional theory and “originalism” as a component of a
political ideology.
It may well be the case that popular perceptions of originalism are strongly influenced
by the ideological profile of particular originalists. Robert Bork and Antonin Scalia
surely count as conservatives. But from the fact that contemporary originalism has
conservative adherents, it does not follow that originalism itself is inherently
conservative. There are divisions within originalism, between conservative, libertarian,
and progressive versions. Some originalists are deeply committed to the idea of judicial
restraint, while others reject restraint when it conflicts with the constitutional text and
instead embrace what has been called “judicial engagement.” Attributing the ideology
of the proponent of a theory to the theory itself is simply to commit the genetic fallacy.
***
D. The Systemic Bias Objection
The Injustice Objection in its various versions is related to another possible objection
to originalism—that the originalism meaning of the constitutional text is too conservative
(or too libertarian) and therefore should not constraint contemporary constitutional
practice. This objection has at least some plausibility because of the age of most of the
constitutional text. All of the constitution through the Seventeenth Amendment is more
than one-hundred years old and the majority of the text (measured in characters) is more
than two-hundred years old.
Contemporary political “progressivism” or “liberalism” is complex, but it is at least
arguable that many of the basic features of the Constitution are either conservative or
libertarian in their function and effect. Consider the following features:
•
•
•
The system of separation of powers and check and balances, when combined
with a limited set of enumerated powers at the national level, is arguably
inconsistent with the ideal of plenary and virtually unlimited national power
that may be necessary to achieve progressive political ideals.
The set of constitutional rights (including the Contracts Clause and the Takings
Clause) are arguably biased in favor of the protection of private property and
against the redistribution of wealth.
The text of the Constitution does not establish economic equality as a goal of
the national government, nor does it create a set of economic rights such as the
right to employment, health care, housing, and adequate nutrition.
Thus, it might be argued that the Constraint Principle should be rejected because it
commits public officials including the courts to a constitutional text that is systematically
biased against progressivism and in favor of conservativism or libertarianism. Call this
the “Systematic Bias Objection.”
95
Lawrence B. Solum
Of course, it is not the case that all of the Constitution is biased in this way. Arguably,
the rights to freedom of speech, press, and assembly are consistent with progressive
ideals. The Reconstruction Amendments include explicit commitments to various
protections of equality. The Sixteenth Amendment gives Congress the power to levy and
income tax, and the Nineteenth Amendment forbids denial of the right to vote on the
basis of sex. Separation of powers, checks and balances, and federalism may create
obstacles to a progressive agenda, but they also operate to frustrate the agendas of social
conservatives, economic libertarians, and extreme political movements.
But suppose that the factual premise of the Systematic Bias Objection is correct, and
that the original meaning of the constitutional text does have a conservative or libertarian
tilt. And suppose further that progressives reject the Constraint Principle for that reason
and adopt instead the principle that the Supreme Court should pursue the political agenda
of the Justices even when that requires amending constructions. The Principle of
Reciprocity then suggests that conservative or libertarian justices are entitled to do the
same. Assuming that Lochner191 was incorrect as a matter of the original meaning of the
Due Process Clause, it would nonetheless be a legitimate decision if consistency of
outcomes with ideology is a valid choice procedure for constitutional methodology.
Similarly, a social conservative court might conclude adopt a constitutional rule
forbidding the states from enacting marriage equality statutes.
This reasoning might lead opponents of the Constraint Principle to reject both the
Principle of Reciprocity and the method of wide reflective equilibrium and to embrace
Constitutional Opportunism as their preferred form of living constitutionalism. When we
control the Court, we will reject the Constraint Principle and adopt the Moral Readings
Theory, but when conservatives control the Court we will argue for a combination of a
strong principle of stare decisis with supplemental principles of restraint. It seems
unlikely that this stance would fool anyone. It is arguable that the result of this approach
would be to increase politicization of the courts. Once it becomes clear that there is no
feasible alternative to ideological judging, then the political actors will have every
incentive to appoint reliable ideologues to the bench.
E. The Objection from Majoritarianism
Those who believe in a strong principle of majoritarianism have a good reason to
reject the Constraint Principle—because the United States Constitution (unlike some
parliamentary systems) does not provide for majoritarian governance. But this is a reason
to reject a judicially-enforceable nonmajoritarian Constitution altogether and hence a
reason to favor unconstrained Thayerianism or anticonstitutionalism over originalism.
Strict majoritarianism is a controversial view. Resolving the arguments surrounding
this view would require an excursus into the fundamentals of political philosophy that
would take us far afield from the mainstream of contemporary constitutional discourse.
Many of the objections to originalism are offered as defenses of Supreme Court decisions
that are inconsistent with both the original meaning of the constitutional text and with
majoritarianism.
In particular, strict majoritarianism (Unconstrained Thayerianism) is inconsistent with
most of the other objections made to originalism. To take one example, the Canonical
191
198 U.S. 45 (1905),
96
The Constraint Principle
Cases Objection assumes the legitimacy of countermajoritarian decisionmaking by the
Supreme Court: one cannot simultaneously argue against originalism on the basis of its
inconsistency with cases like Brown or Griswold while simultaneously arguing that
originalism should be rejected because the original meaning of the constitution as a
whole does not implement majoritarian democracy.
Moreover, the objection from majoritarianism is clearly not a reason to favor common
law constitutionalism or most of the other rivals of originalism that reject the constraint
principle but accept some form of contribution. To state the obvious, the Supreme Court
is not a majoritarian institution, but common law constitutionalism makes decisions of
the Supreme Court superior to constitutional amendments adopted by supermajorities. A
similar point could be made about the multiple modalities and moral readings views.
The countermajoritarian nature of the Supreme Court poses a famous problem for
nonoriginalists, the so-called “counter-majoritarian difficulty.” This suggests that the
Constraint Principle might come to be viewed as a majoritarian second-best. If pure
majoritarianism is not a practical possibility (because there is no realistic prospect of
adopting a parliamentary system of government), then constraint by the constitutional
text that was approved by supermajoritarian procedures might be better than an
unconstrained Supreme Court. This second-best solution is obviously imperfect—the
Constitution is old and portions were approved by supermajority procedures at times
when many were excluded from the franchise. But as compared with an unconstrained
and unelected committee of nine, a constrained Court would seem to fare better on the
criterion of democratic legitimacy.
There is a more modest version of majoritarianism that is consistent with some very
significant aspects of contemporary constitutional practice. Representation
Reinforcement Thayerianism would support a modified version of Carolene Products
footnote four192—although it would require jettison those provisions of the Bill of Right
that are cannot be justified as protections of the political process, including the Fourth,
Fifth, and Eight Amendments. Of course, this theory would likely require overruling
much of the Court’s substantive due process jurisdiction, including Roe v. Wade193 and
Obergefell v. Hodges.194 This view is one of the strongest rivals of originalism; it will be
considered in greater detail in Part VII below.195
F. The Objection from Exclusion and Illegitimacy
The unamended Constitution drafted in 1787 and the first fourteen amendments were
all enacted before the Fifteenth Amendment, which provided, “The right of citizens of the
United States to vote shall not be denied or abridged by the United States or by any State
on account of race, color, or previous condition of servitude.”196 The first eighteen
amendments were enacted before the Nineteenth Amendment which extended the
franchise to women.197 It was not until 1964 that the Twenty-Fourth Amendment
192
United States v. Carolene Products, 304 U.S. 144, 153 n. 4 (1938).
410 U.S. 113 (1973).
194
135 S. Ct. 2071 (2015).
195
See infra Part VII.A.7.c), p. 120.
196
U.S. Const. amend. XV, § 1.
197
U.S. Const. amend. XIX.
193
97
Lawrence B. Solum
prohibited poll taxes.198 In 1971, disenfranchisement on the basis of age for persons
older than twenty-one was prohibited.199 As of 2016, there is no constitutional provision
that prohibits the disenfranchisement of prisoners or convicted felons. Moreover,
constitutional guarantees may not translate into an effective right to vote. Thus,
disenfranchisement on the basis of race continued long after the ratification of the
Fifteenth Amendment. Even after the Voting Rights Act of 1965,200 disenfranchisement
may be accomplished by a variety of mechanisms, including voter identification laws.201
These facts give rise to an objection to originalism that is focused on the illegitimacy
of a constitution that was adopted under conditions of exclusion. Professor James Fox
summarizes the objection as follows:
[O]riginalism is anti-democratic at its very root because the political culture and
process that produced both the text and the original meanings were themselves not
democratic. At the founding, suffrage was largely limited to white male propertyholders. The group that wrote, approved, and commented on the original
Constitution comprised a small subset of people living in America, and many of
them were slaveholders. . . . [T]he “original exclusions” of women, AfricanAmericans, Native-Americans, and the non-propertied lower classes present
originalism with a foundational justificatory problem: whether from a moral or a
democratic viewpoint, ratification of constitutional provisions through a process
that excluded most people simply cannot claim the mantle of legitimacy.202
This is a serious and important objection to the normative foundations of the Constraint
Principle; it demands an answer.
The first step towards answering the objection from exclusion and illegitimacy
involves understanding the force and implications of the objection. This objection entails
the conclusion that the United States Constitution is illegitimate and therefore lacks
binding force. And the objection further entails that the decisions of the Supreme Court
and the statutes enacted by Congress are illegitimate—unless there is some date at which
the system became sufficiently democratic so as to confer legitimacy on decisions and
actions after that date. It is at least arguable that the current level of exclusion means that
the illegitimacy objection infects all law. And even if formal inclusion would be
sufficient to confer legitimacy, it may be that economic and social oppression of women
and persons of color negates formal enfranchisement.
The real force of the objection from exclusion and illegitimacy is that all or almost all
of the law in the United States is illegitimate and therefore lacks binding authority. This
leads to the conclusion that Antitextualist Anarchism: in other words, the reasons for
198
U.S. Const. amend. XXIV.
U.S. Const. amend. XXVI.
200
79 Stat. 437.
201
See RICHARD L. HASEN, THE VOTING WARS (2012); Richard L. Hasen, Race or Party?: How Courts
Should Think About Republican Efforts to Make it Harder to Vote in North Carolina and Elsewhere, 127
HARV. L. REV. F. 58 (January 7, 2014).
202
James W. Fox Jr., Counterpublic Originalism and the Exclusionary Critique, 67 ALA. L. REV. 675,
685 (2016) (citations omitted); see also ALEXANDER KEYSSAR, THE RIGHT TO VOTE: THE CONTESTED
HISTORY OF DEMOCRACY IN THE UNITED STATES 1-25 (2000); Reva B. Siegel, Heller and Originalism's
Dead Hand--In Theory and Practice, 56 UCLA L. REV. 1399 (2009); Mark S. Stein, Originalism and
Original Exclusions, 98 KY. L.J. 397 (2009-2010).
199
98
The Constraint Principle
action provided by the communicative content of legal texts are not peremptory reasons
for action, and therefore, one should act on the basis of all the reasons that are salient. In
other words, we should deny authoritative effect to the decisions of the Supreme Court,
statutes enacted by Congress, regulations issues by administrative agencies, rules of civil
procedure, and so forth. To put it bluntly, there is no legitimate law in the United States.
For this reason, the objection from exclusion and illegitimacy is not available to most
nonoriginalists. For example, many nonoriginalists are committed to the view that the
original meaning of the constitutional text should make a contribution to the legal content
of constitutional doctrine, but this position is difficult to square with the fundamental
illegitimacy of the Constitution. Other nonoriginalists believe in the binding force of
statutes and judicial decisions, but it is difficult to reconcile this position with the view
that all or almost all of American law is defective because of problems of exclusion.
Nonetheless, there are forms of nonoriginalism that overcome the illegitimacy
objection. Representation-reinforcement Thayerianism is the clearest example. How can
originalists answer the objection was pressed from this direction? One important line of
reply begins with the observation that representation-reinforcement theories depend on
controversial views about an ideal theory of representation. This means that the decisions
of such a court will vary with the ideological views of the Justices about the proper
theory of representation. Thus, a conservative representation-reinforcement Court would
likely render a very different set of decisions than a liberal one. Citizens United is seen
by some as required to protect majoritarian democracy and seen by others as a grave
threat. There is no guarantee that the abstract principles of RepresentationReinforcement Thayerianism would actually solve the problem of exclusion, precisely
because it requires the Justices to bring their first-order normative views about
democracy to bear in deciding constitutional cases.
Originalists might also argue that the principle of representation reinforcement faces
intractable problems once it confronts the question about what set of substantive rights is
necessary to create the background conditions for democratic legitimacy. One approach
would deny that substantive rights are required, limiting courts to the enforcement of
rights to majoritarian procedures. But this approach has well-known difficulties.
Majoritarian democracy can function to deny “discreet and insular” minorities
meaningful participation in the democratic process: let us call this scenario “exclusionary
majoritarianism.” To remedy exclusionary majoritarianism, courts might create a set of
substantive rights that create the preconditions for meaningful democratic inclusion. But
that alternative leads to a well-known slippery slope: for example, it can be argued that
meaningful democratic inclusion requires a robust social minimum, single-provider
national health care, and strong restrictions on expression that reinforces hierarchies of
exclusion. On the other side of the slippery slope, it can be argued that meaningful
democracy requires a strong set of liberty and property rights and a minimal role for the
state. Sliding down either slope would turn Representation-Reinforcement Thayerianism
into something like the Moral Readings view—with a functionally unimportant
democratic twist.
99
Lawrence B. Solum
G. The Dead Hand Problem
The dead hand objection to originalism can take many forms.203 If the emphasis is on
the dead hands of the framers and ratifiers, this objection is just another label for the
democratic legitimacy objection.204 Another version of the objection focuses on the
“dead hand of the framers” but does not object the dead hand of the constitutional text
itself.205 These versions of the dead hand objection have already been discussed in
connection with majoritarianism and exclusion. Yet another version of the objection
associates the “dead hand” with the bad consequences the flow from adhering to an old
and outdated constitution.206 This version of the objection is a process oriented version
of the objection from constitutional injustice, but it potentially wider in scope. For
example, one might argue that the framers had outdated views on the structure of
government; perhaps, the separation of powers is inconsistent with the functional need to
govern a complex modern society through an administrative state that combines
legislative, executive, and judicial powers in a single agency staffed by experts. If
combined with an argument that the supermajoritarian amendment process is incapable of
dealing with problems of this sort, this version of the dead hand argument would provide
a plausible case for a judicial power to nullify the textual commitment to separation of
powers.
Implicitly, this objection seems to rest on the notion that there is a contemporary
consensus on the desirability of the post-New Deal administrative state. If there really
were such a consensus, then we could expect an originalist court that took a gradualist
approach to prompt a constitutional amendment that authorized current arrangements. If
203
For an important reply to the dead hand objection, see Michael W. McConnell, Textualism and the
Dead Hand of the Past, 66 GEO. WASH. L. REV. 1127 (1998).
204
Reva B. Siegel, Heller & Originalism's Dead Hand-in Theory and Practice, 56 UCLA L. REV. 1399
(2009) (characterizing dead hand objections as follows: “Originalism's critics objected that giving past
generations this kind of control over the living would vitiate the Constitution's democratic authority.”);
Louis W. Hensler III, The Recurring Constitutional Convention: Therapy for A Democratic Constitutional
Republic Paralyzed by Hypocrisy, 7 TEX. REV. L. & POL. 263, 290–91 (2003) (Therefore, legal scholars
incessantly wrestle with why we should be expected to satisfy ourselves with a Constitution that neither we
nor anyone like us created and to which neither we nor anyone like us consented. Nobody yet has provided
a fully satisfactory answer97 to this “dead hand problem.”).
205
Daniel A. Farber, The Dead Hand of the Architect, 19 HARV. J.L. & PUB. POL'Y 245, 249 (1996)
(Ultimately, the dead hand is inescapable. Originalism, however, seeks in some sense to bestow ultimate
constitutional authority to some ghost in the dead hand, rather than to the dead hand itself. Rather than
focusing on the tangible legal structure that the Framers bequeathed, originalism seeks to invest authority in
the departed creators of that structure, as if their minds still animated their creation. Their intentions and
understandings my be relevant, but should not be regarded as binding. Being governed by the dead hand
may be inevitable, but we need not submit to rule by ghosts.).
206
Gregory Bassham, How Not to Argue for Originalism: A Review of Mcginnis and Rappaport's
Originalism and the Good Constitution, 42 J.C. & U.L. 235, 240 (2016) (“A standard objection to
originalism is that it has bad consequences for society, because it binds current generations to the “dead
hand” of the past.”); Howard Gillman, The Collapse of Constitutional Originalism and the Rise of the
Notion of the “Living Constitution” in the Course of American State-Building, 11 STUD. IN AM. POL. DEV.
191, 192 (1997) (“(“Originalism... runs the risk that our institutional arrangements and conceptions of
justice will be inappropriately tied to the antiquated experiences and prejudices of generations long
gone....”).
100
The Constraint Principle
the critic of originalism argues that there is no such consensus, but nonetheless it is “true”
that we need the administrative state, then familiar problems arise. The critic has
acknowledged that this version of the dead hand argument would not be included in wide
reflective equilibrium. And the critic must also deal with the possibility that empowering
the Supreme Court to revise the basis structure of government on the basis of the Justice’s
beliefs about what arrangements are best could lead to a constitutional structure that is,
from their point of view, even worse than the arrangements that are consistent with the
original meaning of the constitutional text.
H. Objections that Do Not Clash with the Constraint Principle
This Section briefly considers and then brackets a variety of objections to originalism
that do not object to the Constraint Principle itself, but instead objection to some other
feature of originalism. The aim is simply to explain why these objections are set to the
side on this occasion.
1. Epistemic inaccessibility
What if the historical facts necessary to the discovery of original meaning are simply
inaccessible? Although it is very unlikely that we lack access to the relevant linguistic
facts necessary to establish conventional semantic meanings and syntax, it seems at least
plausible that some of the facts necessary to establish contextual enrichments have been
lost. For present purposes, this objection should be bracketed, because it does not clash
with the Constraint Principle. The Constraint Principle does not require that
constitutional practice be constrained by content that is epistemically inaccessible.
What should originalists say or do about epistemic inaccessibility? From an
ecumenical perspective, it seems likely that different versions of originalism will answer
this question differently. The possibilities include default rules (such as a rule of
Thayerian deference), adopting the construction that best serves the purpose implied by
the known communicative content, and so forth.
Pervasive epistemic inaccessibility would be a serious objection to originalism, but it
goes to issues that are separate and distinct from those raised by the Constraint Principle.
The case for inaccessibility must be advanced on a clause-by-clause basis; as it seems
clear that some constitutional provisions have meaning that is relatively clear.
2. Lack of Institutional Capacity Reliably to Discover Original Meaning
Even if original meaning is epistemically accessible, it might be the case that judges
lack the institutional capacity reliably to discover that meaning. For example, judges and
their law clerks might lack the necessary training to work with the relevant sources. The
same could be true of other constitutional actors, including the President and Congress.
Again, this objection is simply bracketed here; this objection does not provide a
normative reason to violate the Constraint Principle if judges do have the institutional
capacity to discover original meaning.
But originalists must answer this objection in due course. Originalism would be
utopian if there were no feasible means by which original meaning could be discovered
with a reasonable degree of reliability. So, the discussion offered on this occasion should
101
Lawrence B. Solum
be understood as a promissory note—to be redeemed at a later stage in the development
of originalism.
Although a fully developed version of an originalist methodology has yet to be
developed, a sketch is already on offer in Originalist Methodology.207 The gist of that
account can be summarized as follows:
•
•
•
•
•
The original public meaning of the constitutional text is the communicative
content—the content that the authors of the text intended the public to grasp
based on the public’s recognition of the authors communicative intentions.
The communicative content of the constitutional text is conveyed by both
semantic meaning and contextual enrichment.
The semantic meaning of the constitutional text is a function of the
conventional semantic meanings of the words and phrases as combined by
syntax and punctuation. Semantic meaning is best ascertained by using a
combination of corpus linguistics, period dictionaries, linguistic intuitions, and
immersion in the language of the period.
Semantic and syntactic ambiguities can be clarified by consideration of the
publically accessible context in which the words or syntactic structures were
written.
Additions or modifications to semantic meaning may be provided by
contextual enrichment, including implicatures, implicitures, presuppositions,
and modulations.
These methodological principles must be supplemented by a set of principles for
originalist research. Datasets and sources should be made available to scholars and legal
actors (e.g., lawyers and judges). Originalist scholarship should be peer reviewed and
verified by replication.
3. The Political Impossibility Objection
It might be argued that originalism is politically impossible. Several forms of this
objection are floating in the air of theoretical discourse about originalism. One version
would focus on individual judges—arguing that it is necessarily the case that judges will
consciously decide on the basis of their ideological preferences. Another version might
focus on unconscious processes of motivated reasoning—arguing that even if judges
attempt to comply with the Constraint Principle, their policy preferences will distort their
reasoning processes. Yet another version of the objection might focus on the processes
by which judges and Justices are selected—arguing that Presidents and Senators have
powerful incentives to select ideologically reliable judges.
Once again, the objection does directly clash with the Constraint Principle. If
constraint were possible, the objection would drop out entirely. And again, originalist
must answer this objection in due course, but on this occasion, this question is reserved
for the future.
207
See Solum, Originalist Methodology, supra note 13.
102
The Constraint Principle
4. Variations on the Theme of Indeterminacy
Another set of objections is based on the theme of indeterminacy, associated with
American Legal Realism and more recently with the Critical Legal Studies Movement.
Indeterminacy has been the subject of intense theoretical debate. The strong
indeterminacy (the communicative content of the constitutional text is consistent with
every conceivable outcome in every conceivable case) seems implausible. Modest
version of the indeterminacy objection might claim that (a) almost all, (b) most, (c) the
most important, (d) the most contested, or (e) a significant portion of the constitutional
text is underdeterminate.208 Some originalists may accept that there is a significant
degree of underdeterminacy; others may claim that the text is fully or almost fully
determinate. The Constraint Principle assumes that the strong indeterminacy thesis is
false: there cannot be constraint by communicative content that is radically indeterminate.
But the thesis can be true, so long as the text constraints in some cases.
If we set radical indeterminacy to the side, it becomes clear that the indeterminacy
objection does not clash with the Constraint Principle, but it does call the practical
important of the principle into question. For this reason, originalists must take a stance
on indeterminacy, and it is likely that almost all originalists will wish to argue that the
constitutional text does in fact have communicative content that would substantially
constraint constitutional practice in important ways. Making this argument is itself a
substantially project—postponed until another occasion.
5. Objections to the Fixation Thesis
If the communicative content of the constitutional text is not fixed at the time each
provision is framed and ratified, then acceptance of the Constraint Principle would not
require adherence to originalism. In an Article entitled The Fixation Thesis, I have
provided an extensive defense of fixation and readers are referred to that article for my
views—a very brief summary of the central argument is provided in the accompanying
footnote.209
***
I have a nagging feeling that many readers will find the answers to objections
unsatisfying, not because they reject the particular arguments, but because they believe in
their heart of hearts that “originalism just can’t be right.” I suspect that this reaction is
208
See Solum, On the Indeterminacy Crisis, supra note 142.
See Solum, The Fixation Thesis, supra note 3. The basic argument offered there relies on the
plausibility of the Fixation Thesis as a general account of the interpretation of old texts. If we were reading
an eighteenth century letter, we surely would want to know the conventional semantic meanings of words
at the time the letter was written and as much as we could find out about the context of the letter at the time
it was written. Because conventional semantic meanings change over time (linguistic drift), contemporary
semantics could easily lead to interpretive mistakes. Because the relevant context for interpreting the letter
is the context that would have been shared by the author and recipient, contextual disambiguation and
contextual enrichment of the letter’s content is time bound. It is implausible to think that the meaning of
the letter changes when linguistic drift results in new meanings for the words used in the letter. And it is
implausible to think that the hypothetical meaning the letter would have had if it had been written today is
the actual communicative content of the letter.
209
103
Lawrence B. Solum
especially prevalent among legal academics of a certain age—those who went to law
school and started their academic careers before the turn of the century and especially
those whose attitudes were shaped by constitutional debates in the early 1980s. There
are many possible explanations for this phenomenon. For one thing, most legal scholars
were told that originalism is obviously wrong in law school and this message was
reinforced so many times in some many ways that it became unshakeable dogma, a fixed
point in the web of belief.
But there may be another reason. At a very high level of generality, our
understandings of the law are shaped by what might be called the “jurisprudential
gestalt,” a set of ideas about the nature of law that organize our beliefs about the law at
a very high level of generality. The jurisprudential gestalt organizes our beliefs about
which theories are contenders versus those that are nonstarters. It shapes our beliefs
about which narratives about the development of legal thought versus the stories that
implausible. The dominant jurisprudential gestalt of the late twentieth century was
shaped by the way that legal thinkers reacted to, absorbed, and transformed the insights
of the American legal realists in light of the New Deal and the Warren Court. The
dominant jurisprudential gestalt attempted to reconcile the realist critique of legal
formalism with the idea that legal reasoning infused with policy and principle could still
be “law”—an ambition that was shared by the legal process school, Ronald Dworkin,
and many others.
The problem with claims that the central ideas of originalism are true or correct is
that these ideas do not fit the dominant jurisprudential gestalt. Originalism is a variety
of formalism, and if “we” know anything it is that formalism is wrong. Therefore, there
must be something wrong with originalism. If the Fixation Thesis is true, then the
Constraint Principle must be incorrect. If the Constraint Principle is correct, then the
communicative content of the constitutional text must be radically (or at least
substantially) indeterminate. If the radical indeterminacy thesis is false, then maybe we
should take another look at the Fixation Thesis. But there must be a flaw somewhere.
***
VII. PAIRWISE COMPARISONS: RIVALS, THE STATUS QUO, AND COMPROMISE
One of the themes of the Article is that an evaluation of the Constraint Principle
requires pairwise comparisons of originalism with the alternatives. But in this Article, we
cannot hope to accomplish an in-depth pairwise with each many alternatives to the
Constraint Principle. In the long run, each rival view requires its own article or
monograph. On this occasion, all that can be provided is a sketch that draws out the
implications of the preceding analysis for further investigation of the pairwise
comparisons.
Pairwise comparison with proceed in three stages. First, we will compare originalism
to each of the nine rivals identified above. Second, we will compare originalism to what
we can call the “constitutional status quo”—which I argue is a form of eclecticism.
Third, we can compare originalism to compromise theories that partially accept the
Constraint Principle (for a limited domain), but reject constraint for some set of
constitutional issues.
104
The Constraint Principle
A. Pairwise Comparison of Originalism with Nine Rivals
The discussion that follows is crucially important. We have already seen that “it takes
a theory to beat a theory” and hence that pairwise comparison of originalism with rival
theories is required. This is the only way to conduct the debate over originalism in a
rigorous way. To beat originalism, the proponents of a rival view need only show that
their view is superior to originalism. To produce a compelling argument for the
Constraint Principle, originalism with constraint must be compared to each and every
rival. Recall that there are nine rival views (some of which have subtypes):
1.
2.
3.
4.
5.
6.
7.
8.
9.
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
The Supreme Court as Superlegislature
Thayerianism
Constitutional Antitheory
Constitutional Rejectionism
Each of the nine theories will be discussed in turn.
1. Multiple Modalities
The view that constitutional law is a complex argumentative practice with multiple
modalities of constitutional argument is a powerful one because it captures important
features of the actual practice of constitutional law. We begin with a brief explication of
the approach, and then turn to the question whether the multiple-modalities approach
rejects the constraint principle, and end with a discussion of the normative comparison
between the multiple-modalities approach with originalism (and the Constraint Principle).
Various forms of the multiple-modalities approach have been articulated by articulated
by Phillip Bobbitt,210 Stephen Griffin, 211 and by Mitchell Berman and Kevin Toh.212
This view is closely related to Richard Fallon’s influential approach, which he labels as
“constructivist.”213
The gist of Bobbitt and Griffin’s version of this approach is that there are multiple
modalities or a plurality of methods for establishing the truth or validity of a proposition
of constitutional law. Bobbitt’s list of the modalities includes the following:
210
See PHILIP BOBBITT, CONSTITUTIONAL INTERPRETATION 12-13 (1991).
See Stephen M. Griffin, Pluralism in Constitutional Interpretation, 72 Tex. L. Rev. 1753, 1753
(1994) (“Pluralistic theories of constitutional interpretation hold that there are multiple legitimate methods
of interpreting the Constitution.”).
212
Mitchell N. Berman & Kevin Toh, Pluralistic Nonoriginalism and the Combinability Problem, 91
TEX. L. REV. 1739 (2013).
213
Richard H. Fallon, A Constructivist Coherence Theory of Constitutional Interpretation, 100 HARV.
L. REV. 1189 (1987).
211
105
Lawrence B. Solum
historical (relying on the intentions of the framers and ratifiers of the
Constitution);
textual (looking to the meaning of the words of the Constitution alone, as they
would be interpreted by the average contemporary “man on the street”);
structural (inferring rules from the relationships that the Constitution mandates
among the structures it sets up);
doctrinal (applying rules generated by precedent);
ethical (deriving rules from those moral commitments of the American ethos
that are reflected in the Constitution); and
prudential (seeking to balance the costs and benefits of a particular rule).214
The multiple modalities approach qualifies as a form of nonoriginalism to the extent that
it denies that the textualist modality operates as a constraint on the others. This
understanding of the multiple modalities approach can be illustrated graphically, as in
Figure 1: Nonoriginalist Multiple Modalities.
Structural
Doctrinal
Textual
Historical
Ethical
Constitutional
Interpretation
&
Construction
Prudential
Figure 1: Nonoriginalist Multiple Modalities
This understanding of the multiple modalities approach is premised on the idea that the
modalities are “flat”—they lack hierarchical structure. This premise leads to the
conclusion that the textualist modality (which we shall assume is equivalent to “original
meaning”) can be overridden by any of the other modalities on the basis of the process of
constitutional argumentation. This conclusion entails the further consequence that the
multiple modalities approach denies the Constraint Principle, and hence is a form of
nonoriginalism. Thus, the multiple modalities approach is a version of nonoriginalism.
This understanding of the multiple modalities approach is not inevitable. One can
imagine a variation in which the modalities are hierarchically structured with a lexical
ordering that prioritizes the textualist modality (and hence original meaning). This
variation would look something like the following diagram, which incorporates the
interpretation construction distinction:
214
Bobbitt, Constitutional Interpretation, supra note 210, at 12-13 (emphasis added and paragraph
structure altered for clarity).
106
The Constraint Principle
StepOne:Interpretation
• Textualist
StepTwo:Construction
(ConstrainedbyOriginalMeaning)
• Historical
Doctrinal
• Structural
• Ethical
• Prudential
Figure 2: Originalist Variant of Multiple Modalities
On this alternative understanding of the multiple modalities approach, we would have a
form of originalism, because the constitutional doctrines that emerge from Step Two
must be consistent with the original meaning that is identified in Step One. Bobbitt
himself affirms the nonoriginalist version and hence rejects the Constraint Principle.
Compatibilist forms of the multiple modalities approach may be important, but for the
purposes of this investigation of the Constraint Principle, our consideration is limited to
the incompatibilist forms of pluralism, multiple-modalities, and constructivism.
As a normative matter, the multiple-modalities view has serious problems with the
rule of law. Precisely because the modalities are not hierarchical ordered, the multiple
modalities view allows first-order private judgments to play a decisive role in
constitutional law as a practical matter. So long as the Justices are capable of making
arguments from the modalities for the outcome they desire, their decisions will have the
appearance of compliance with the disciplining force of the complex argumentative
practice of constitutional law.
Here is another way of putting it: in every constitutional case that comes before the
Supreme Court, the lawyers make a reasonable case on both sides using the modalities of
constitutional argument. For this reason, the multiple modalities approach does a poor
job providing the grounds for intersubjective agreement that decision for one side or the
other was legally incorrect. As a consequence, the three rule of law arguments for
constraint apply with great force in the context of pairwise comparison with pluralism.
The multiple modalities approach creates the danger of judicial tyranny; it is inconsistent
with the rule of law values of publicity, stability, and certainty; and the multiple
modalities approach creates does not provide a mechanism for preventing the
politicization of constitutional law.
Similarly, the multiple modalities approach does not compare well with the Constraint
Principle with respect to legitimacy. Law as a complex argumentative practice lacks
democratic legitimacy for obvious reasons: complex argumentation by the justices might
take democratic values into account but it does not prioritize consistency with
democratically legitimate legal norms. As a matter of ideal theory, the multiple
modalities approach might be transparent, but this approach lends itself to violations of
107
Lawrence B. Solum
transparency because of the manipulability of the modalities of constitutional argument.
But the multiple modalities argument does fare well with respect to judicial role: one of
the strengths of this approach is that it requires judges to act on the basis of a coherent
theory of what counts as a legal consideration.
2. Moral Readings
Ronald Dworkin calls his general interpretive method (including his view of
constitutional interpretation and construction) “constructive interpretation.”215 A similar
approach is found in the work of James Fleming, who uses Dworkin’s phrase “Moral
Readings” to describe his constitutional theory.216 The discussion that follows will focus
on Dworkin’s view. Because Dworkin’s theory is much discussed, I will assume that
readers are familiar with the core notion of fit and justification. On my reading of
Dworkin’s theory, the content of constitutional law is determined by the moral theory that
best fits and justifies the legal materials as a whole. Thus, the content of constitutional
doctrine is primarily determined by the moral theory that best fits constitutional texts,
institutions, practices, and judicial decisions on constitutional matters.
Dworkin is sometimes hard to pin down, and his theory was elaborated in many texts
over the course of five decades. One understanding of Dworkin is the following: because
constructive interpretation aims to make our practices “the best that they can be” in light
of the institutional history, his theory implies that meanings can change over time. Our
“moral readings” of the constitutional text are not fixed, but instead evolve in response to
changing circumstances and our evolving constitutional values.217 This would make
Dworkin a nonoriginalist, but it might allow him to accept the Constraint Principle. In a
footnote, I suggest that that Dworkin may actually accept fixation (in his own way),218
but for the purposes of this Article, the key question is whether he also accepts the
Constraint Principle.
In his most explicit discussion of the relationship between what we are calling
communicative content and legal content, Dworkin juxtaposes his view with that of
Laurence Tribe. Here is the full passage, which includes an internal (double-indented)
quotation from Tribe:
Tribe endorses a very strong form of textual fidelity. Tribe states:
I nonetheless share with Justice Scalia the belief that the Constitution’s
written text has primacy and must be deemed the ultimate point of
departure, that nothing irreconcilable with the text can properly be
215
RONALD DWORKIN, LAW’S EMPIRE 62–86 (1986).
See Fleming, supra note 182.
217
The phrase “moral reading” is Dworkin’s. See RONALD DWORKIN, FREEDOM’S LAW: THE MORAL
READING OF THE AMERICAN CONSTITUTION 2–3 (1996), but it is now strongly associated with James
Fleming’s Dworkinian theory of constitutional interpretation and construction. See James E. Fleming,
Fidelity, Change, and the Good Constitution, 62 AM. J. COMP. L. 515, 515 (2014).
218
In 1997, Dworkin seems to have embraced some version of the notion that semantic content is fixed
at the time a text is written. See Dworkin, The Arduous Virtue of Fidelity: Originalism, Scalia, Tribe, and
Nerve, 83 Fordham L. Rev. 1249, 1251–52 (1997).
216
108
The Constraint Principle
considered part of the Constitution; and that some parts of the Constitution
cannot plausibly be open to significantly different interpretations.219
That is a stronger statement of textual fidelity than I [Dworkin] would myself
endorse, because, as I said, precedent and practice over time can, in principle,
supersede even so basic a piece of interpretive data as the Constitution’s text when
no way of reconciling them all in an overall constructive interpretation can be
found. I agree with the Tribe of this statement, however, that the text must have a
very important role: We must aim at a set of constitutional principles that we can
defend as consistent with the most plausible interpretation we have of what the text
itself says, and be very reluctant to settle for anything else.220
Dworkin does not use the same conceptual vocabulary as we have been employing here,
but his point can be translated. Dworkin recognizes that the communicative content of a
text is fixed at the time the text is written. But in the case of the constitutional text, the
legal content of constitutional doctrine can change, because the “constructive
interpretation” of the law as a whole can override the communicative content. In other
words, Dworkin accepts fixation as a thesis about “interpretation” (communicative
content), but rejects the Constraint Principle.
Dworkin’s rejection of the constraint principle is an essential and not an accidental
feature of his theory. Dworkin’s theory is holistic: his fictional judge Hercules constructs
the theory that best fits and justifies institutional practice as a hole. And his theory is
dynamic: as institutional practice (including judicial decisions) changes over time, the
content of the theory that best fits and justifies the law as a whole will change. Because
his theory is holistic and dynamic, the content of constitutional law cannot be constrained
(in the sense of constraint as consistency) by the communicative content of any single
text. Using the typology developed above, we would classify Dworkin as a
nonoriginalist.
Pairwise comparison of originalism (with the Constraint Principle) and constructive
interpretation (or moral readings) suggests that both the rule of law and legitimacy
provide pro tanto reasons to favor constraint. It is built into the moral readings approach
that officials (especially judges) will need to rely on their first-order private judgments of
political morality when engaged in constitutional construction. This leads to rule by
decree and hence judicial tyranny, undermines the rule of law values of publicity,
stability, and certainty, and creates the risk of politicization, given interaction between the
demands of the theory and the political nature of the appointments process.
Likewise, constructive interpretation fares poorly with respect to some but not all of
the legitimacy. The notion of judges as philosopher kings is antithetical to democratic
legitimacy. If practiced in good faith, constructive interpretation would be transparent,
but given political realities, judges who practice this method might conceal the role their
values play in their decisions. Similarly, constructive interpretation could, as a matter of
ideal theory, avoid politicization altogether—with the Supreme Court engaged in
apolitical philosophical deliberation, but in nonideal theory, the focus on direct resort to
219
Laurence H. Tribe, Comment, in ANTONIN SCALIA, A MATTER OF INTERPRETATION: FEDERAL
COURTS AND THE LAW 23 (1997).
220
Dworkin, supra note 218, at 1259–60.
109
Lawrence B. Solum
the moral believes of the Justices might facilitate politicization of the appointments
process.
3. Common Law Constitutionalism
In his book entitled, The Living Constitution,221 David Strauss has argued for a view
that we might call, “Common Law Constitutionalism.” Again, we can consider two
versions of the theory. The nonoriginalist version of Common Law Constitutionalism
would affirm that constitutional law is best conceived as the outcome of a common-law
process that takes into account the original meaning of the constitutional text, but it not
constrained by that meaning: this would be a form of nonoriginalism, because it would
explicitly reject that Constraint Principle.
But we can also imagine an originalist version of Common Law Constitutionalism.
The originalist version would affirm that constitutional law is best conceived as the
outcome of a common-law process but includes some version of the Constraint Principle.
Assuming that Common Law Constitutionalism accepts the Fixation Thesis, the resulting
theory would qualify as a member of the originalist family of constitutional theories.
What is Strauss’s actual position? Consider the following passage from The Living
Constitution:
We cannot make an argument for any constitutional principle without purporting
to show, at some point, that the principle is consistent with the text of the
Constitution. That is an essential element of our constitutional culture. And no
provision of the Constitution can be overruled in the way a precedent can, or
disregarded in the way original understandings often are.222
Strauss’s book is lucid and elegant, but very compact—and so there are some ambiguities
here.223 On the one hand, this passage might be read as an affirmation of the minimalist
version of the Constraint Principle, that is Constraint as Consistency. Strauss seems to
affirm both elements of Constraint as Consistency. The first element requires consistent
of the content of constitutional doctrine with the communicative content of the
constitutional text: Strauss explicitly says that constitutional principles must be
“consistent with the text” and “no provision of the Constitution can be overruled.” The
second element requires that every element of the text be reflected in constitutional
doctrine: Strauss explicitly says that no provision can be “disregarded.”
But Strauss also says that “original understandings” can be disregarded, but it is not
clear what he means by “understandings.” To be fair to Strauss, we would need to
undertake a painstaking examination of all his writings and precisely reconstruct his
operative conceptions including “meaning,” “understanding,” and “text.” My impression
is that such a reconstruction would reveal that Strauss employs the term meaning in
multiple senses, that by understandings he means application beliefs, not communicative
content (or linguistic meaning), and that his theoretical apparatus collapses the
221
DAVID A. STRAUSS, THE LIVING CONSTITUTION (2010).
Id.
223
Strauss’s position is clarified in his recent introduction to the Harvard Law Review’s Supreme Court
issue. See David A. Strauss, Foreword: Does the Constitution Mean What It Says?, 129 HARV. L. REV. 1,
57 (2015) (implicitly rejecting the Constraint Principle by stating that “original understandings are binding
for a time but then lose their force”). We might call this position “Short-Run Originalism.”
222
110
The Constraint Principle
interpretation-construction distinction. Hence, Strauss’s theoretical writings
underdetermine the question whether his version of Common Law Constitutionalism is
originalist or nonoriginalist. As of the writing of The Living Constitution, he did not have
a theoretically precise answer to the question whether the legal content of constitutional
doctrine must be consistent with the communicative content of the constitutional text;
subsequent work suggests that he may have come down decisively as a nonoriginalist—at
least with respect to the long-run implementation of the Constitution.224
Whatever position Strauss would ultimately take on the Constraint Principle, we can
formulate a nonoriginalist version of common-law constitutionalism. This version of
common-law constitutionalism would allow judges to make decisions that are
inconsistent with the constitutional text so long as the decisions were made in accord with
common law methods.
Pairwise comparison of originalism with common law constitutionalism leads to
conclusions that are similar to those reached with respect to the modalities and
constructive interpretation. On the surface, common law constitutionalism looks legal,
but it suffers from the rule of law defects that ground criticisms of the common law in
general. Because the common-law process is so open ended (absent a strong doctrine of
stare decisis), it opens the door for judicial tyranny. The same feature of the common law
undermines the rule of law values of publicity, stability, and certainty. As a matter of
ideal theory, common law judges are apolitical, but in the actual world, the looseness of
common-law constraints invites politicization.
Common law constitutionalism does not fare well with respect to democratic
legitimacy for familiar reasons. In theory, it scores well on transparency, but in practice
there are substantial worries that common law reasoning can disguise the real motives for
judicial decisions. The same point can be made with respect to politicization.
4. Popular Constitutionalism
The phrase “popular constitutionalism”225 can be used to describe many different
ideas. In one sense, popular constitutionalism is a positive theory of constitutional
change: the people themselves determine the content of constitutional law. In another
sense, popular constitutionalism is a normative theory: the content of constitutional
doctrine should be determined by “We the People.” One version of normative popular
constitutionalism would be largely consistent with the Constraint Principle: if one argued
that “We the People” are the authors of the constitutional text and therefore it is higher
law that should constraint the people’s agents (including judges and other officials), that
would provide a normative justification for constraint.
But there is a form of normative popular constitutionalism that would be inconsistent
with the constraint principle. If “We the People” authorized the President to make
transformative appointments to the Supreme Court and the Justices then adopted
amending constructions of the constitutional text that were authorized by the people, then
a normative popular constitutionalist might argue that these violations of the Constraint
224
See Strauss, id.
See Larry Alexander & Lawrence B. Solum, Popular? Constitutionalism?, 118 HARV. L. REV. 1594
(2005).
225
111
Lawrence B. Solum
Principle are democratically legitimate and hence justified. Bruce Ackerman’s work
might be read along these lines, but this is not the occasion to offer a textually supported
interpretation of his large corpus of work on the Constitution.226
Let us call the version of normative popular constitutionalism that sanctions
constitutional change outside of the amendment process, “Extratextual Normative
Popular Constitutionalism.” For ease of expression, I shall use the phrase “popular
constitutionalism” to refer to this view—unless otherwise specified. As defined, popular
constitutionalism authorizes violations of the Constraint Principle and should be
classified as a form of nonoriginalism.
The pairwise comparisons of originalism with the first three rivals of originalism
(multiple modalities, constructive interpretation, and common law constitutionalism)
revealed a good deal of similarity among the three approaches. Popular constitutionalism
is distinctly different. Popular constitutionalism can be consistent with the rule of law—
so long as the mechanisms of popular constitutional change are such that new
constitutional regimes have clearly defined starting points and content. And popular
constitutionalism fares well on the criterion of democratic legitimacy—which is baked
into the very idea of constitutional change authorized by “We the People!”
But it is not clear the popular constitutionalism gets off the ground as a normative
account of constitutional change. Whereas the Article V process provides clear, public,
and transparent criteria for constitutional amendments, popular constitutionalism outside
the Article V process leads to questions about when constitutional change has been
authorized by the people. Consider, for example, the claim that the New Deal Settlement
was authorized by the election of Franklyn Delano Roosevelt to an unprecedented fourth
term. It may be true that Roosevelt ran, in part, against the Supreme Court, but it is not
clear that he ran against constitutional constraint; indeed, it seems more likely that he
presented New Deal constitutionalism as restoration of what we are calling the Constraint
Principle. Constitutional change through transformative appointments may result in
substantial uncertainty about the content of constitutional law, especially if the Justices
disagree among themselves about the “original meaning” of the transformative
moment.227
5. Multiple Meanings
A fifth version of nonoriginalism is based on the idea that texts have “multiple
meanings” rather than a single fixed meaning: call this the “Multiple Meanings Theory”
of constitutional interpretation.228 The gist of the argument would go something like this:
226
See Bruce A. Ackerman, Transformative Appointments, 101 HARV. L. REV. 1164, 1166 (1988) (“As
traditionalist Justices retired, however, the Senate gave enthusiastic support to a series of transformative
appointments, which had precisely the consequences that Roosevelt's opponents feared—and Reagan's
opponents fear today. Within a few short years, the New Deal majority had decisively repudiated the rightsoriented jurisprudence so elaborately developed by its Republican predecessors.”).
227
See Kurt T. Lash, The Constitutional Convention of 1937: The Original Meaning of the New
Jurisprudential Deal, 70 FORDHAM L. REV. 101 (2001) (describing disagreements among the Justices about
the meaning of the New Deal).
228
My reconstruction of the argument has been influenced by a work-in-progress by Cass Sunstein, see
Cass R. Sunstein, There Is Nothing that Interpretation Just Is,
112
The Constraint Principle
texts do not have a single meaning (in the linguistic sense); instead, they have multiple
meanings. Because there are multiple meanings, we must select between them, and this
process of selection must be guided by normative considerations. Originalism is false,
because it privileges one of the multiple meanings (such as original intent or original
public meaning) and therefore begs the crucial normative questions as to which meaning
should govern in any particular case.
Some of the premises of the Multiple Meanings Theory are correct, but from them it
does not follow that the Constraint Principle is false. Mark Greenberg makes the point
that there is more than type of linguistic content in the context of a discussion229 of Smith
v. United States230—the Supreme Court case in which the question was whether offering
to trade a gun for cocaine constituted use of a firearm for the purpose of a penalty
enhancement provision.231
As the contemporary study of language and communication has made clear,
there are multiple components and types of linguistic content. In Smith, there are at
least two types of linguistic content plausibly associated with the statutory text that
would yield opposite outcomes in the case. First, there is the semantic content of
the statutory text—roughly, what is conventionally encoded in the words. Second,
there is the communicative content—roughly, what the legislature intended to
communicate (or meant) by enacting the relevant text.232
So far, so good. Semantic content is distinct from communicative content. Moreover,
the Gricean speaker’s meaning of an utterance is not necessarily identical to the meaning
that the audience actually takes from the utterance.
We can translate Greenberg’s point into constitutional terms. For the sake of
simplicity, we can focus on four distinct “meanings”:
•
•
•
•
Framers’ Meaning: The content that the authors of a constitutional provision
intended to convey to the relevant readers (e.g., the public) through the
readers’ recognition of the framers’ communicative intentions.
Original Semantic Meaning: The content that competent readers of a
constitutional provision would have attributed to a constitutional provision,
given the conventional semantic meanings of the words and phrases and their
syntactic structure, but without consideration of context.
Ratifiers’ Meaning: The content that the ratifiers (or the subset of ratifiers who
were competent speakers of English and who actually read the text) actually
attributed to a constitutional provision.
Reasonable Contemporary Meaning: The meaning that a reasonable
contemporary reader would attribute to the text given contemporary semantics
and syntax and the counterfactual assumption that the text was written in the
contemporary period.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2489088 (August 29, 2014), and by exchanges with
Richard Fallon. My version of the argument should not be taken as representing their positions.
229
Mark Greenberg, The Moral Impact Theory of Law, 123 YALE L.J. 1288, 1291–92 (2014)
230
508 U.S. 223 (1993).
231
Id.
232
Greenberg, supra note 229, at 1291–92.
113
Lawrence B. Solum
The first three forms of meaning are all originalist in the sense that they accept the
fixation thesis. Framer’s intentions are fixed at the time each provision is authored—and
original clause meaning is similarly fixed. Ratifiers’ understandings are fixed at the time
of ratification. If the Multiple Meanings theory were limited to these three meaning
types, it would be a form of originalism. It would differ from Public Meaning
Originalism of Original Intentions originalism because it would allow judges to pick and
choose between the three forms of original meaning. And because Original Semantic
Meaning is acontextual, it creates substantial construction zones.233
The Multiple Meanings theory explicitly rejects the Constraint Principle when it adds
Reasonable Contemporary Meanings to the list from which constitutional actors may
choose. This is not to say that the Multiple Meanings views rejects the idea of constraint
altogether. It imposes a form of constraint, but that form of constraint is untethered from
the communicative content of the constitutional text. It would, for example, allow judges
to base their decisions on meanings created by linguistic drift—the phenomena by which
the conventional semantic meanings of words change over time. This opens the door to a
process of constitutional change that begins with constitutional actors arguing for a new
meaning of a constitutional word or phrase: “commerce” should be understood as “social
interaction” and not as “trade in goods.” Once the new usage becomes established,
constitutional actors are then entitled to rely on the new sense of the word. Because
shifts in meaning through linguistic drift are not themselves limited by the Constraint
Principle, this version of the Multiple Meanings theory authorizes the creation of
constitutional doctrines that are inconsistent with original meaning.
So, the question whether the Multiple Meanings theory violates the Constraint
Principle depends on the precise set of meanings that is authorized by a particular version
of the theory. Versions of Multiple Meanings that are limited to “original meanings” do
not violate constraint, but once a nonoriginal meaning is admitted to the list, the version
that allows the nonoriginal meaning should be classified as nonoriginalist (in the sense in
which that term is used in this Article): those forms of the Many Meanings theory violate
the Constraint Principle.
The multiple meanings theory embraces a loose version of the Constraint Principle,
but the ability of the Justices to pick and choose among the possible meanings of the text
creates problems of both the rule-of-law and legitimacy.
The rule of law problems will be especially acute if the Justices are allowed to pick
and choose among the possible meanings of the constitutional text on a case by cases
basis. And these problems will be further exacerbated if one of the allowable meanings is
contemporary acontextual semantic meaning: this is because bare semantic meaning is
sparse and therefore opens the door to a very large zone of underdeterminacy and hence
judicial discretion as a functional matter. Unless the space is filled with a restraining
default rule (such as deference to Congress), all of the rule of law and legitimacy
233
More precisely, acontextual semantic content is sparse. See Victoria F. Nourse, Elementary
Statutory Interpretation: Rethinking Legislative Intent and History, 55 B.C. L. REV. 1613, 1658 (2014)
(Semantic content is exceedingly sparse, as Scott Soames and other philosophers of language have shown.).
In terms more familiar to lawyers, the literal meaning of the constitutional text, if read without any
reference to context, could mean many different things, corresponding to the many different contextually
enriched meanings that the text would have in possible worlds with the same text but different contexts.
114
The Constraint Principle
problems for the other views we have considered so far will apply with equal force to the
multiple meanings view.
6. The Supreme Court as Superlegislature
Consider yet another possible view. A constitutional theorist might affirm the view
that the Supreme Court should frankly operate as a superlegislature and explicitly
embrace the power to override the constitutional text by promulgating amending
constructions (judicial doctrines that are inconsistent with the communicative content of
the constitution). This view is rarely embraced explicitly as a normative theory. The
label “superlegislature” (or “super-legislature”) is frequently used to express a
criticism—as it was when first introduced by Justice Brandeis (joined by Justice Holmes)
in 1931.234 Max Lerner made the criticism explicit in 1933:
Viewed thus the Court through its power to veto legislation has also the power to
channel economic activity. In that sense, it has been often called a super-legislature,
exercising powers tantamount to the legislative power, but more dangerously since
it is not subject to the same popular control.235
This critical usage of “superlegislature” has been echoed by subsequent scholars.236 And
the Supreme Court itself denied that it “sits a super-legislature” in Griswold v.
Connecticut.237
There is logical space for nonoriginalists to explicitly embrace the notion that the
Supreme Court acts a perpetual constitutional convention238 (with the power to adopt
amending constructions by a majority vote). And this view has been articulated by Brian
Leiter in his essay, Constitutional Law, Moral Judgment, and the Supreme Court as
Super-Legislature. 239 Leiter writes,
234
Jay Burns Baking Co. v. Bryan, 264 U.S. 504, 534, 44 S. Ct. 412, 421, 68 L. Ed. 813 (1924)
(Brandeis J., dissenting (To decide, as a fact, that the prohibition of excess weights ‘is not necessary for the
protection of the purchasers against imposition and fraud by short weights,’ that it ‘is not calculated to
effectuate that purpose,’ and that it ‘subjects bakers and sellers of bread’ to heavy burdens, is, in my
opinion, an exercise of the powers of a super-Legislature-not the performance of the constitutional function
of judicial review.).
235
Max Lerner, The Supreme Court and American Capitalism, 42 YALE L.J. 668, 696 (1933).
236
William W. Van Alstyne, The Demise of the Right-Privilege Distinction in Constitutional Law, 81
Harv. L. Rev. 1439, 1447 (1968) (“A court need not “weigh” or “balance”; it need simply apply the literal
mandate of a given constitutional provision flatly to forbid government from conditioning its largess on any
waiver of such a provision regardless of the circumstances. A court may thus avoid any unseemly
appearance of acting as a superlegislature.”).
237
381 U.S. 479, 482 (1965).
238
The idea of the Supreme Court as a perpetual constitutional convention was suggested to me by Kurt
Lash’s important article on the constitutional theory of the New Deal Court. See Kurt T. Lash, The
Constitutional Convention of 1937: The Original Meaning of the New Jurisprudential Deal, 70 FORDHAM
L. REV. 459, 462-63 (2001) (“Acting, in effect, as a constitutional convention, the New Deal Court had the
responsibility to draft the charter for post-New Deal judicial review. Unanimously rejecting the common
law method of Lochner and Swift, the members of this New Deal Convention declared that judicial
interference with the political process henceforth required, at the very least, some clear textual
justification.”).
239
Brian Leiter, Constitutional Law, Moral Judgment, and the Supreme Court as Super-Legislature, 66
HASTINGS LAW JOURNAL 1601 (2015).
115
Lawrence B. Solum
[A]ll political actors know that the U.S. Supreme Court often operates as a
super-legislature, and thus that the moral and political views of the Justices are
decisive criteria for their appointment. This almost banal truth is, however, rarely
discussed in the public confirmation process, but is common knowledge among
political and legal insiders. To be sure there is media speculation about the political
predilections of the nominees, but their actual moral and political views are treated
as off limits in the real confirmation process. This anti-democratic secrecy is, in my
view, deeply wrong and must be replaced with a realistic acknowledgment of the
role of the Supreme Court as a political actor of limited jurisdiction.240
Leiter is not explicitly advocating the proposition that the Supreme Court should act as a
superlegislature; rather his point is that its role as a superlegislature should be made
public and transparent.
But it is at least possible that some nonoriginalists privately hold the view that the
Supreme Court should act as a superlegislature—adopting amending constructions of the
constitutional text on the basis that these amendments are desirable. One problem with
this position is that it is difficult to affirm in public, because it seems likely that any
constitutional actor who affirmed this position transparently would meet substantial
political resistance. Presidents do not openly affirm that they select Justices who will
override the constitutional text. Judicial nominees are likely to identify as originalists,
even if those “in the know” doubt the sincerity of their protestations.241 But even if the
superlegislature view is not made public, it may in fact be the view affirmed in private by
Presidents, Senators, Justices, and constitutional theorists.
It almost goes without saying that the superlegislature view can be formulated so that
it is inconsistent with the Constraint Principle. Indeed, the use of the phrase
“superlegislature” implies that the Court goes beyond interpretation and is playing the
role of a constitutional convention that creates and amends constitutional provisions. For
the purposes of this essay, we will assume that the Superlegislature Theory has this
feature and hence that it is a form of nonoriginalism.
Pairwise comparison of the superlegislature view with the Constraint Principle is
stark. If the Supreme Court has the power to rewrite the Constitution on a case by case
basis, we have rule by judicial decree and hence judicial tyranny. Depending on the
circumstances, this may lead to problems with stability and certainty as well. On the
other hand, the superlegislature theory potentially satisfies the publicity requirement—so
long as the Supreme Court’s opinions themselves make the rules clear. Similarly, the
superlegislature view results in problems with democratic legitimacy and politicization,
but not with transparency.
7. Thayerianism: Constrained, Unconstrained, and Representation-Reinforcement
I will use the term “Thayerianism” to designate a family of theories that include a
strong principle of deference to democratic institutions, usually Congress but variations
of Thayerianism might include deference to the national executive and to elected officials
240
Id. at 1-2.
Confirmation Hearing on the Nomination of Elena Kagan to Be an Associate Justice of the Supreme
Court of the United States: Hearing Before the S. Comm. on the Judiciary, 111th Cong. 62 (2010)
(statement of Elena Kagan).
241
116
The Constraint Principle
at the state level. The Thayerian family includes at least three important views that reject
or partially reject the Constraint Principle. I will call three two theories “Constrained
Thayerianism,” “Unconstrained Thayerianism” and “Representation Reinforcement
Thayerianism.” Let us stipulate as follows:
Constrained Thayerianism: This view includes two principles: (1) courts should
defer to Congress and should refrain from exercising the power of judicial review to
strike down statutes duly enacted by Congress, and (2) Congress should act
consistently with the original public meaning of the constitutional text unless the first
principle authorizes inconsistent action.
Unconstrained Thayerianism: This view includes two features: (1) Courts should
defer to Congress and refrain from exercising the power of judicial review to strike
down statutes enacted by Congress; (2) Congress should not consider itself bound by
the original meaning of the constitutional text and hence should enact statutes that
violate the original meaning (when there are good reasons for so doing).
Representation Reinforcement Thayerianism: This view has two elements: (1) courts
should defer to Congress and should refrain from exercising the power of judicial
review to strike down statutes duly enacted by Congress unless the exceptions
specified in the second element apply, and (2) courts should exercise the power of
judicial reviews when required to protect discreet and insular minorities or to protect
the democratic process (for example, by protecting the right to vote and the right to
engage in political speech).
a) Constrained Thayerianism
Constrained Thayerianism is partially consistent with the Constraint Principle. The
core idea is that courts should defer to Congress, but that Congress should consider itself
constrained by the constitutional text. In essence, this view transfers the final authority
over the meaning and effect of the Constitution from Congress to the Supreme Court.
Constrained Thayerianism is almost a version of originalism. It accepts the Constraint
Principle with one important exception: the role of the Supreme Court. For the purposes
of this discussion, I will simply assume that the original meaning of the Constitution
requires the Supreme Court (and the lower federal courts) to apply the Constitution as the
supreme law of the law. Although the modern description of the Supreme Court’s role
refers to the concept of “judicial review,” the more traditional formulation conceptualizes
this role in terms of a duty to follow the law.242 To the extent that Constrained
Thayerianism authorizes the courts to depart from the original understanding of the
phrases “judicial power” and “Supreme Court,” it violates the Constraint Principle.
But otherwise, Constrained Thayerianism is a form of originalism. It requires that
Congress’s action be consistent with the original communicative content of the
constitutional text. Thus, the normative question in pairwise comparison of originalism
with Constrained Thayerianism is whether the balance of reasons favors a violation of
constraint with respect to the role of the courts in constitutional interpretation and
construction while also favoring constraint by original meaning in all other cases.
242
See PHILIP HAMBURGER, LAW AND JUDICIAL DUTY (2008).
117
Lawrence B. Solum
Many of the usual arguments for and against constraint drop out in this pairwise
comparison. For example, the rule of law arguments (from judicial tyranny, the rule of
law values, and politicization of the courts) do not favor complete constraint over
constrained Thayerianism. Democratic legitimacy actually favors Constrained
Thayerianism, and transparency is neutral in this context. The legitimacy argument from
judicial role does favor the Constraint Principle over Constrained Thayerianism.
Likewise, most of the arguments against constraint apply with equal force to Constrained
Thayerianism, including the arguments from canonical cases, injustice, the dead hand,
and exclusion.
This leaves the majoritarian objection to originalism. On the one hand,
majoritarianism seems to favor Congress as the ultimate authority in constitutional
interpretation and construction: after all, Congress is elected, whereas the Supreme Court
is not. On the other hand, Constrained Thayerianism is antidemocratic in an important
respect—because a constrained Congress is disabled from making decisions on the basis
of democratic politics whenever the will of the majority is contrary to the original
meaning of the constitutional text.
Nonetheless, democratic politics might play an important role in the construction
zone—the set of issues and cases in which the communicative content of the
constitutional text underdetermines legal effect. Democratic Thayerianism could adopt a
majoritarian approach to constitutional construction with respect to vague and opentextured constitutional provisions. Pairwise comparison on this issue is not possible
without specifying the theory of constitutional construction that originalism should
employ in the construction zone, but for the purposes of this article, that theory has not
yet been specified. As noted above,243 one approach to the construction zone would
involve deference to Congress and hence would be identical to Constrained Thayerianism
in the construction zone. But other approaches (such as a presumption of liberty) would
differ substantially.
A full pairwise comparison of the full Constraint Principle with Constrained
Thayerianism would require investigation of the relative institutional competence of the
Supreme Court and Congress as the final authority on constitutional interpretation and
construction. The conventional wisdom is that the judiciary is better suited to this role,
but the conventional wisdom may be wrong. A full investigation of this question is
beyond the scope of this Article, and for that reason, this aspect of pairwise comparison
cannot be completed here.
Constrained Thayerianism occupies an important position in the conceptual space of
constitutional theory, but because it accepts so much of the Constraint Principle, it can be
set aside for the present purposes. If Constrained Thayerianism provides the best theory
of constitutional interpretation and construction, then originalism will largely have won
the day.
b) Unconstrained Thayerianism
“Unconstrained Thayerianism” incorporates the idea of judicial deference to Congress
that is common to all forms of Thayerianism, but it rejects the Constraint Principle.
Thus, Unconstrained Thayerianism authorizes congressional actions that are inconsistent
243
See supra, text accompanying notes 26-27.
118
The Constraint Principle
with the original meaning of the constitutional text, and it gives Congress the power to
authorizes violations by other institutions (e.g., the states, the national executive, and the
courts). Unconstrained Thayerianism does incorporate a principle of constraint, but that
principle only requires that Congress act through generally applicable legislation—and
hence, it forbids Congress to take on the role of a court of last resort with the power hear
petitions for the enactment of statutes that overrule particular judicial decisions. One can
imagine other versions of Unconstrained Thayerianism with different features, but on this
occasion, pairwise comparison will focus on the stipulated version.
Unconstrained Thayerianism avoids the Judicial Tyranny Argument: so long as
Congress acts through prospective statutory enactments of general application, there can
be no rule by decree. Likewise, Unconstrained Thayerianism avoids the politicization of
the judiciary, although it does this by politicizing constitutional law via a more direct
mechanism, transfer of constitutional authority from a judicially-interpreted written text
to an unconstrained Congress.
Although Unconstrained Thayerianism fares better than many of the other rivals of
constraint with respect to the rule of law values, it is still at a disadvantage when
compared to the Constraint Principle. Constraint provides a substantial degree of
publicity, stability, and certainty—for reasons discussed above.244 Precisely because the
Constitution is relatively difficulty to amend, constrained constitutional law provides
stable, certain, and public rules governing the structure of government and fundamental
rights. An unconstrained Congress has the ability to rewrite the Constitution by majority
vote in each session.
Unconstrained Thayerianism fares beats the Constraint Principle with respect to
democratic legitimacy and ties constraint on transparency, but it involves a
transformation of judicial role by requiring courts to uphold statutes even when they
clearly violate the Constitution. Assuming that a duty to act in accordance with law is a
core component of the role of courts and judges, Unconstrained Thayerianism faces a
legitimacy problem on this score.
Unconstrained Thayerianism runs into many of the objections that are made to
originalism and it fares worse than originalism on some. Unconstrained Thayerianism
cannot be squared with any of the canonical cases and it authorizes injustice of any
kind—so long as the injustice is approved by Congress. But Unconstrained
Thayerianism fares better than constraint with respect to majoritarianism, exclusion, and
the dead hand.
Unconstrained Thayerianism requires that we reject the Constitution. And for this
reason, many readers may believe that it is simply outside the feasible choice set. Indeed,
some readers may see discussion of Unconstrained Thayerianism as a “red herring” or
“straw man”—a rival of originalism that is so far outside the mainstream of constitutional
discourse, both in public and in the academy, that its introduction into the debate draws
attention away from the real and important issues.
My own sense is that Unconstrained Thayerianism is, for now, outside the feasible
choice set. There are scenarios in which Unconstrained Thayerianism might become our
political future, but they involve external shocks that change fundamental attitudes
towards the Constitution—a crisis with the normative force of the Great Depression or
one of the World Wars. Nonetheless, Unconstrained Thayerianism occupies an important
244
See supra Part IV.A.2, p. 67
119
Lawrence B. Solum
role in conceptual space. Parliamentary democracy is an important option, and
Unconstrained Thayerianism represents the closest approximation to that option so long
as the Constitution remains a formal part of American law.
The most important questions that inform pairwise comparison of Unconstrained
Thayerianism with the Constraint Principle concern institutional design. Would
protection of individual rights be better served by Unconstrained Thayerianism or by
constraint? If Congress assumed the power to alter the basic structure of government,
how would it use that power? And what consequences would follow? These are complex
questions that can only be answer by the methods of the social sciences. And they
certainly cannot be resolved in this Article.
c) Representation Reinforcement Thayerianism
“Representation Reinforcement Thayerianism” differs from Unconstrained
Thayerianism because it incorporates a representation-reinforcement principle. The most
prominent advocate of this theory is John Hart Ely,245 and his articulation of the theory
has a strong connection with Footnote Four of the Caroline Products decision.246
Representation Reinforcement Thayerianism adopts an extratextual view about the role of
the judiciary in protecting the democratic process and discrete and insular minorities as
the basis for a set of exceptions to a general rule of deference to elected officials. To
some degree these exceptions may be supported by the text: freedom of speech and press
as well as Section One of the Fourteenth Amendment might support some aspects of the
representation-reinforcement principle. But it is not the original meaning of these
provisions that governs, rather the representation reinforcement principle determines the
legal content of constitutional doctrines that are associated with these provisions. For this
reason, Representation Reinforcement Thayerianism rejects the Constraint Principle.
We have already discussed most of the issues raised by Representation Reinforcement
Thayerianism. The chief advantages of this approach are captured by the arguments from
majoritarianism and the illegitimacy of exclusion; those arguments are discussed above.
Representation Reinforcement Thayerianism is subject to rule of law problems, especially
once we recognize the subjectivity of judgments about the preconditions for adequate
representation.247 As with each of the important rivals of originalism, a full discussion of
these issues would require an extended treatment.
d) Pairwise Comparison with Originalism and the Constraint Principle
The strength of Thayerianism is democratic legitimacy. Moreover, this theory scores
well on publicity and transparency as well—so long as the legislature governs through
general laws that are made public and applied in accord with their public meaning.
Thayerianism also avoids the problem of judicial tyranny, and rule by generally
applicable laws is not rule by decree and hence is not tyranny in any form. Thayerianism
just is politicization, but so long as the politics are democratic, this may be a virtue and
not a vice.
245
JOHN HART ELY, DEMOCRACY AND DISTRUST (1980).
United States v. Carolene Products Co., 304 U.S. 144, 152 n. 4 (1938).
247
See supra Part VI.E, p. 96 & Part VI.F, p. 97.
246
120
The Constraint Principle
This suggests that the pairwise comparison of Thayerianism with originalism raises
issues that are quite different than those discussed thus far. There is a reason for that.
The case for originalism advanced in this Article focuses on the issues that are relevant to
a comparison of the jurisprudential status quo (in some sense) with originalism. The
status quo is judicial supremacy and Congressional supremacy. For this reason, this
particular pairwise comparison requires consideration of a set of issues that are mostly
outside the scope of this Article—a separate and extended treatment is required.
8. Antitheories: Particularism, Pragmatism, Eclecticism, and Opportunism
The first seven forms of nonoriginalism are all theories of constitutional interpretation
and construction. As such, they share the common assumption that constitutional theory
can offer a method that should guide constitutional practice. There is another family of
views about constitutional theory that rejects this assumption. Let us call the family of
views that rejects this assumption “Constitutional Antitheory.” Constitutional Antitheory
is related to antitheoretical views in other branches of moral and political philosophy.
Notably, there are a variety of antitheoretical views in normative ethics.248 One way of
thinking about antitheoretical positions focuses on the notion that there cannot be a
decision procedure for ethics. Hence, antitheory stands in opposition to utilitarianism
(act so that your action maximizes utility) or certain interpretations of Kant (act so that
the maxim of your action could be willed as a universal law of nature). “Moral
particularism” is one form of antitheoretical thinking in ethics; as the name implies, this
view emphasizes the priority of the particular,249 but there are other forms of
constitutional antitheory.
a) Forms of Constitutional Antitheory
Let us use the phrases “constitutional particularism,” “constitutional pragmatism,”
“constitutional eclecticism,” and “constitutional opportunism” to name four
antitheoretical approaches to constitutional practice. These four antitheories have
something in common: they reject the notion that constitutional theory can provide a
systematic method (or decision procedure) for constitutional practice. But each of the
four antitheories has distinctive characteristics.
(1) Constitutional Particularism
Constitutional particularism is the view that constitutional practice should be guided
by the fact-sensitive assessment by constitutional actors of particular cases (or decisions).
Paradigmatically, particularism lauds the assessment of constitutional by trial judges who
decide the constitutional cases with the benefit of a rich factual context involving live
248
See ANTI-THEORY IN ETHICS AND MORAL CONSERVATISM (Stanley G. Clarke & Evan Simpson eds.
1989); SOPHIE GRACE CHAPPELL, INTUITION, THEORY, AND ANTI-THEORY IN ETHICS (2015); NICK FOTION,
THEORY VS. ANTI-THEORY IN ETHICS: A MISCONCEIVED CONFLICT (2014).
249
For an introduction, see Jonathan Dancy, Moral Particularism, Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/entries/moral-particularism/ (2013); see also JONATHAN DANCY, ETHICS WITHOUT
PRINCIPLEs (2004); Brad Hooker, Moral Particularism—Wrong and Bad in MORAL PARTICULARISM
(Hooker and Little eds. 2000).
121
Lawrence B. Solum
testimony experienced in real time. Appellate judges are at a disadvantage, because of a
paper record, but they should aim for a richly textured appreciation of the particular case.
Outside the courts, other constitutional actors (judges) are themselves participants in
context of constitutional decisionmaking. But in all of these situations, the key to good
decisionmaking is fact sensitivity. Good constitutional doctrine emerges from perception
of the particular case. Constitutional particularism rejects the Constraint Principle as a
binding principle, because judgments about particular cases always trump general
principles. Of course, in particular cases, a particularist judge might follow the text, but
the judge’s reasons would be tied to the particular case and not a general rule.
(2) Constitutional Pragmatism
Constitutional pragmatism embraces fact-sensitivity, but it does not embrace the
priority of the particular as a general approach to constitutional practice. As articulated
by Richard Posner, pragmatism adopts “reasonableness” as its guiding principle.250
Posnerian constitutional pragmatism is forwarding looking, but it is not purely
consequentialist:
[D]espite the emphasis on consequences, legal pragmatism is not a form of
consequentialism, the set of philosophical doctrines (most prominently
utilitarianism) that evaluates actions by the value of their consequences: the best
action is the one with the best consequences. There are bound to be formalist
pockets in a pragmatic system of adjudication, notably decision by rules rather than
by standards. Moreover, for both practical and jurisdictional reasons the judge is
not required or even permitted to take account of all the possible consequences of
his decisions.251
Despite his formidable intellect, Posner is not adept at the precise articulation of his own
theoretical framework. Let us stipulate that constitutional pragmatism is the view that
constitutional practice should be guided by the principle of reasonableness with a
forward-looking emphasis on legally salient consequences. Judges should choose
between rules, standards, and other forms of legal norms252 by determining which form of
norm is most reasonable in the circumstances. Constitutional pragmatism rejects the
Constraint Principle as a general rule, although a pragmatic judge may follow the text for
pragmatic reasons on a case-by-case basis.
There may be forms of pragmatism that avoid the rule of law and legitimacy problems
that clearly attend Judge Posner’s version. Posnerian pragmatism suffers from the all of
the rule of law problems identified above.253 Indeed, the point of this form of
pragmatism is to free judges from the constraints of the rule of law and enable them to
make all-things-considered judgments when deciding particular cases. Likewise, this
250
Richard A. Posner, Legal Pragmatism Defended, 71 U. CHI. L. REV. 683 (2004) (“The ultimate
criterion of pragmatic adjudication is reasonableness.”).
251
Id. at 684.
252
See Lawrence B. Solum, Legal Theory Lexicon 026: Rules, Standards, Principles, Catalogs, and
Discretion, http://lsolum.typepad.com/legal_theory_lexicon/2004/03/legal_theory_le_3.html (last modified
on January 4, 2015).
253
See supra, Part IV.A, p. 55.
122
The Constraint Principle
form of pragmatism raises all of the legitimacy problems raised in the earlier
discussion.254 Pragmatic decisionmaking lacks democratic legitimacy, because a
pragmatic judge is not bound by democratic decisions. Likewise, the all-thingsconsidered approach is clearly inconsistent with a judicial role that differentiates
constitution making from constrained constitutional adjudication. In Judge Posner’s case,
the pragmatism may have the virtue of transparency; Judge Posner has been unusually
frank about his approach to judging and his disdain for the constitutional text, but even
Posner may have felt constrained to disavow his rejection the Constraint Principle: as the
Wall Street Journal reported, “The iconoclastic federal appeals judge from Chicago on
Friday walked back his assertion in a Slate column that the nation’s founding document
has no use for judges today.”255
Of course, a modified version of pragmatism might acknowledge that the
constitutional text should play some role but deny that there are any metaprinciples (other
than pragmatism itself) that define what that role should be. This version of pragmatism
does a better job than Posner’s in accounting for the appearances, but it is not clear that it
provides any basis for intersubjective agreement on what counts as a correct
constitutional decision.
(3) Constitutional Eclecticism
Constitutional Eclecticism is the view that constitutional practice should be guided by
an eclectic mix of methods. In the judicial sphere, eclecticism has both interjudicial and
intrajudicial dimensions. Across different judges, eclecticism embraces methodological
diversity: thus, it affirms the goodness of some judges being originalists, others embrace
common law constitutionalism, and others who are constitutional pragmatists. Even a
single judge might be eclectic, rendering originalist decisions on Monday and strictly
adhering to precedent on Tuesday. Eclecticism differs from pragmatism, because it does
not offer a metaprinciple (like reasonableness) that would provide guidance as to which
methods should be used by different judges or at different times. But eclecticism does
not need to embrace the absurd notion that “anything goes.” Eclectics can and should
impose a filtering principle that sorts constitutional methods into the group of reasonable
theories that are eligible for deployment in constitutional practice and unreasonable
theories that should play no role. Eclecticism rejects the Constraint Principle as a general
rule, but an eclectic judge might deploy the idea of constraint in some cases, while
ignoring it in others.
Constitutional eclecticism might be held as an explicit theoretical stance and selfconsciously articulated as such: call this variant “Sophisticated Constitutional
Eclecticism.” The sophisticated version of eclecticism explicitly denies the Constraint
Principle, because it embraces the option of ad hoc reliance on nonoriginalist
constitutional theories. But eclecticism need not be consciously embraced; one can at
least imagine a judge who is disposed to act in a way that is descriptively captured by
eclecticism but who does so intuitively (without explicit awareness): call this variant
“Naïve Constitutional Eclecticism.” The naïve version of eclecticism might be combined
254
See supra, Part IV.B, p. 74.
Jacob Gershman, Posner Apologizes for Saying the Constitution Is Useless, July 1, 2016,
http://blogs.wsj.com/law/2016/07/01/posner-says-he-didnt-mean-to-suggest-the-constitution-is-useless/.
255
123
Lawrence B. Solum
with constitutional self-deception: naïvely eclectic judges might believe that they do
affirm the Constraint Principle, when in fact they are disposed to act in ways that
systematically inconsistent with constraint.
(4) Constitutional Opportunism
Constitutional Opportunism is the view that constitutional practitioners can and should
pick the theory that leads to the best consequences (or the ideologically preferred
outcome) on any particular occasion. Today, I might affirm originalism because it
enhances the right of plaintiffs to a jury trial, but tomorrow I might argue for
Thayerianism because it would support deference to Congress on the Affordable Care
Act. Or today, I affirm originalism because it leads to a strong individual right to own
and carry a gun, but tomorrow I might argue for Thayerianism because it supports
overruling Roe v. Wade.256 The argument that Constitutional Opportunism does not
respect the Constraint Principle is so obvious that it need not be stated.
Related to the notion of constitutional opportunism at any given moment is the idea
that the time-frame for theoretical opportunism is fixed by the current composition of the
Supreme Court. Thus, when progressives command a majority on the Supreme Court,
progressive constitutional theorists will be attracted to constitutional theories that enable
ideology to play a role in judging: multiple-modalities, moral readings, and common-law
constitutionalism are examples. But when conservatives hold the court, progressive
constitutional theorists will be attracted to theories of judicial restraint and constraint. It
should go without saying that the same considerations apply to conservative
constitutional theorists: they may be attracted to restraint and constraint when
progressives dominate the court and to views that enable a conservative court to change
the legal content of constitutional doctrine and to invalidate executive and legislative
action. This can give rise to the interesting phenomenon of theoretical flip-flopping—
where the very same theorist changes views shortly after there is a dramatic shift in the
composition of the court that effects the position of the median justice in ideological
space.257
Constitutional Opportunism is not identical to a best-outcomes theory, because it
requires the judge to follow some constitutional theory on each occasion and some
outcomes may not be justifiable by any theory. Opportunism rejects the idea that
constitutional decisionmaking must be principled, but it accepts some weaker idea—for
example, the idea that that the positive legitimacy of judicial decisionmaking in
constitutional cases requires that some principled explanation for constitutional decisions
be offered (or available).
In sum, all four forms of antitheory reject originalism—which is a normative theory
that provides a single method of constitutional interpretation. Antitheorists can endorse
originalists results reached by particularist, pragmatic, eclectic, or opportunistic methods.
And antitheorists can even endorse reasoning that includes originalist ideas. But no
256
410 U.S. 113 (1973).
See Jack Balkin, The Constitution in 2020, er.... 2016, Bakinization,
http://balkin.blogspot.com/2016/02/the-constitution-in-2020-er-2016.html (February 19, 2016); Eric
Posner, The Coming Flip-Flop in Constitutional Theory, EricPosner.com, http://ericposner.com/thecoming-flip-flop-in-constitutional-theory/ (February 19, 2016).
257
124
The Constraint Principle
antitheorists can endorse originalism (and hence the Constraint Principle) as a general
method for constitutional practice to be applied by all judges in all cases.
b) Pairwise Comparison of Constitutional Antitheory with Originalism and the
Constraint Principle
Pairwise comparison of originalism with constraint and the four forms of
constitutional antitheory suggests favor originalism with respect to both the rule of law
and legitimacy. Each of the antitheoretical positions permits judicial tyranny;
constitutional opportunism embraces rule by decree with relish. Each of the antitheories
undermines the rule of law values of publicity, stability, and certainty. Each of the
antitheories encourages the politicization of constitutional law. And the same is true of
legitimacy. Antitheory authorizes juristocracy and undermines democratic legitimacy.
Some forms of antitheory can be transparent: particularism and pragmatism could be
practiced in a transparent way. Eclecticism seems less transparent, because it offers no
explanation for the choice of one approach over another other than a results orientation
that is unlikely to be disclosed. Opportunism is the opposite of transparency. None of
the antitheories respects limits on the judicial role.
9. Rejectionism: Anticonstitutionalism & Constitutional Replacement
Finally, we have two views that reject a premise shared by the other forms of
nonoriginalism. Most constitutional theorists accept the normative premise that the
communicative content of the constitutional text should make a significant contribution
to the legal content of constitutional doctrine.258 We can use “rejectionism” as an
umbrella term for theories that reject the idea that the communicative content of the
constitutional text should make a significant contribution to the legal content of
constitutional doctrine.
a) Forms of Rejectionism
We will consider two forms of rejectionism: anticonstitutionalism and constitutional
replacement.
(1) Anticonstitutionalism
Anticonstitutionalism is the view that constitutional texts should play no substantial
role in constitutional practice. It is rare to see this view articulated explicitly, but Louis
Michael Seidman has articulated a view like this in his monograph Constitutional
Disobedience.259 Strict anticonstitutionalism rejects the idea of a constitution altogether,
and thus would reject both written and unwritten constitutions—irrespective of their
content. Anticonstitutionalism explicitly rejects the Constraint Principle.
258
The idea of contribution is defined more precisely below in the form of the “Contribution Principle.”
See infra Part II.A, p. 1919.
259
LOUIS MICHAEL SEIDMAN, CONSTITUTIONAL DISOBEDIENCE (2013).
125
Lawrence B. Solum
(2) Constitutional Replacement
Constitutional replacement theories reject the current Constitution of the United
States, but would accept some form of contribution or constraint for a better constitution.
Again, it is rare to see this view articulated explicitly, but my formulation of this
possibility was inspired by Sandy Levinson’s book, Our Undemocratic Constitution:
Where the Constitution Goes Wrong (And How We the People Can Correct It).260
Levinson argues for constitutional replacement, but his view could be extended to include
an explicit rejection of the current constitution pending the adoption of a new one. For
the purposes of this article, we will use the phrase “constitutional replacement” to
designate theories that maintain that contribution and constraint, while appropriate for a
morally acceptable constitution, should be rejected for the current Constitution of the
United States. For example, a constitutional replacement theorist might affirm the
Constraint Principle if the United States Constitution were amended to be more like the
current South African Constitution, but reject the Constraint Principle given the
communicative content of the actual constitutional text.
It goes without saying that both anticonstitutionalism and constitutional replacement
theories reject the United States Constitution in general and the Constraint Principle in
particular.
b) Pairwise Comparison of Constitutional Rejectionism with Originalism and the
Constraint Principle
Finally, we come to rejectionism in both its anticonstitutionalism and constitutional
replacement forms. Both views are radical alternatives to originalism.
Anticonstitutionalism in one form is simply a rejection of the rule of law; that form of
constitutional antitheory raises issues that are far beyond the scope of this paper and far
outside the mainstream of contemporary constitutional thought. This is not to say that
this view does not deserve engagement; rather, it is to say that the terms of engagement
require examination of a set of ideas that are different in kind than those that provide the
organizing focus on this Article.
Constitutional replacement theory certainly rejects the idea that the original meaning
of the current constitutional text should govern constitutional practice, but it is not clear
that replacement theorists reject originalism at a higher level of generality. A new
constitution would require interpretation and construction. Each constitutional
replacement theories must adopt a stance on the question whether the new provisions of
the replacement constitution should constrain judges and other officials. Again, this is a
topic for another day. To the extent that replacement theorists believe that the current
constitution is nonbinding, and therefore that the Supreme Court can decide cases without
reference to the constitutional text, their view converges with the superlegislature theory.
But they might hold some other position, e.g., that the Supreme Court should practice
Thayerian deference until an adequate constitution comes into effect.
260
SANFORD LEVINSON, OUR UNDEMOCRATIC CONSTITUTION: WHERE THE CONSTITUTION GOES
WRONG (AND HOW WE THE PEOPLE CAN CORRECT IT) (2008).
126
The Constraint Principle
***
The complexity of debates about originalism and constraint is in part a function of the
fact that the conceptual space of contemporary constitutional theory is multidimensional.
Many positions in this conceptual space are occupied: witness the nine families of
nonoriginalist theories that reject the Constraint Principle! And there are many possible
variations within each of the nine families: common law constitutionalism, for example,
could have many variations, corresponding to different theories of the common law
combined with various views of constitutional stare decisis.
This complexity gives rise to a practical problem for
constitutional theory. There are too many variations for
each to be given full consideration when advancing the case
for any one theory. The attempt to do so gives rise to what
we might call “Constitutional Theory Whack-a-Mole.”
Every time a particular theory is shown to have a
substantial problem, another theory arises to take its place.
The best that can be done under these circumstances is to identify those families of
theories that seem most salient—either because they have well-developed representatives
or because they occupy important nodes in conceptual space or both. Thus, Dworkin’s
theory is well developed and influential. The “superlegislature” approach does not have
a well-developed representative, but it occupies an important position in the conceptual
space of constitutional theory. Theories can be organized into families around leading
representatives: thus, David Strauss’s theory can serve as a representative of common
law constitutionalism. This seems to be the only feasible way of limiting the list of
alternatives to originalism to a manageable number.
Focusing on families of theories is the only feasible course of action, but it has a
serious disadvantage. Each advocate of a particular theory has a reasonable complaint
that their particular version of nonoriginalism has not been discussed in detail. To the
extent that a particular theory is fully articulated and well defended, the it should be
considered in due course. Nonetheless, it seems reasonable to begin with the theories
that are widely recognized as strong and then proceed to less prominent views.
***
B. The Constitutional Status Quo
At this stage of the argument, some nonoriginalists may object to the method of
pairwise comparison of originalism to its rivals in theoretical theory. Consider the
following statement of the objection:
Originalists are arguing for a change to the constitutional status quo. Pairwise
comparison of originalism to nonoriginalists theories is really beside the point. The
status quo is not theoretical; it is antitheoretical. The relevant comparison should
begin with a recognition that the status quo is Constitutional Eclecticism. Making
out the case for originalism requires a demonstration that adoption of originalism
would be better than what we have now. Once the issue is framed this way, we can
recognize that a move to originalism would be difficult and costly. Pro tanto
127
Lawrence B. Solum
justifications for originalism are all well and good, but they are not sufficient to
justify the many costs that would be imposed by originalism in practice.
Notice that this objection is rooted in a particular conception of constitutional time. The
relevant framework for evaluation of originalism is rooted in the constitutional present,
and the questions that should be asked are about the constitutional future. This
framework entails that constitutional thought experiments about a hypothetical
constitutional past are mostly irrelevant to the question at hand: what should we do now
and in the foreseeable future?
This question is entirely fair, and advocates of the Constraint Principle must provide
an answer. Notice that systematic comparisons between unconstrained eclecticism and
the Constraint Principle are inherently difficulty, because eclecticism can lead to any
approach in any particular case—including originalism (albeit originalism by a particular
constitutional actor or set of actors on a particular occasion). This makes is difficult for a
true constitutional eclectic to object to originalism in general.
This point can be brought out in greater detail by explicit considerations of the
implications of constitutional possibility and constitutional time for comparison for
future-to-future comparisons of unconstrained eclecticism with constrained originalism.
Given the malleability of eclecticism, almost any constitutional future we can imagine is
consistent with eclecticism. As a practical matter, it may well be the case that the effects
of continue eclecticism would depend on who is appointed to the federal bench. And it
seems unlikely that the future of American politics can be predicted with any certainty
once the time frame extends out for more than several months (or a few years at best).
Rule of law arguments apply with special force when comparing originalism with
eclecticism. An eclectic Supreme Court truly does rule by decree creating an especially
vicious version of the judicial tyranny problem. Eclecticism runs into problems with
each of the rule of law values. Indeed, eclecticism introduces a meta-level rule of law
problem that many other versions of nonoriginalism avoids. Eclecticism makes the law
uncertain at the meta-level: judges pick and choose their method of constitutional
construction, making the first-order content of law even more difficult to ascertain in
advance. Eclecticism introduces a particularly vicious form of legal instability at the
meta-level: eclecticism provides no guarantee of continuity in the methods of
constitutional construction, opening a space for radical shifts in constitutional law in
response to whatever motivations cause meta-level shifts in practice at the intra-judicial
and inter-judicial levels. If eclecticism has very serious problems with predictability and
certainty, it is an utter failure when it comes to publicity: indeed, the whole point of
eclecticism is that judicial methods of construction should not be made publicly known at
the meta-level. Finally, when it comes to the risk of politicization, eclecticism leaves the
barn door wide open: eclectic judges can reach almost any result that politics or ideology
would dictate.
Eclecticism fares badly with respect to legitimacy as well. Some forms of
nonoriginalism fare well with respect to democratic legitimacy—the clearest example is
the Thayerian family of nonoriginalist theories. Eclecticism embraces democratically
legitimate forms of nonoriginalism on some occasion, but rejects them on others. The
option to act in a democratically illegitimate fashion is part of what makes eclecticism
truly eclectic. Eclecticism is nontransparent for reasons that have already been discussed
with respect to the rule-of-law value of publicity: the ultimate basis for decision by an
128
The Constraint Principle
eclectic judge is not “out in the open.” But because the judge affirms some set of
principles as the basis for decision in the particular case, the decision has a false
appearance to transparency. The surface level reasons conceal an underlying basis for
choice that is rarely disclosed. With respect to judicial role, eclectic judges sometimes
are “in role” and sometimes “out of role.” When they adopt the constraint principle, as
they may from time to time, their behavior is legitimate. But when the override the text,
as they will on some occasions, their decisions are illegitimate given the appropriate
conception of the proper role of judges.
At this stage, some critics of originalism may complain that eclecticism is not the
status quo, arguing that their preferred version of nonoriginalism is actually the status
quo. This argument has been advanced by advocates of the multiple-modalities view—
particularly by Stephen Griffin in his important essay, Pluralism in Constitutional
Interpretation.261 It is certainly true that judges use multiple modalities of constitutional
argument: no one should dispute that. What is less clear is that they employ the
theoretical framework of the multiple-modalities approach. It is difficult to prove a
negative. Judge Reavely of the Fifth Circuit made explicit reference to the multiplemodalities approach in a 1992 case.262 There is one other occasion upon which a judge
has explicitly relied on the theory. Judge Reinhardt made the following statement in a
denial of a petition for rehearing en banc:
A necessary implication of the constitutional prohibition on congressional
tinkering with the manner of judicial decisionmaking is that Congress may not
eliminate wholesale any tool of judicial reasoning from the judicial toolbox. Courts
in common law systems have developed a panoply of tools to guide them in the
interpretive process, among them inferring rules from text or structure, reasoning
from analogy, and applying rules from precedent. Cf. Philip Bobbitt, Constitutional
Interpretation 12-22 (1991). No one tool takes precedence in any situation, and in
any given case several of the tools may work in tension with one another.263
This passage has never been cited by another judge, and it is the only passage of which I
am aware in which the theoretical structure of the multiple-modalities has been explicitly
endorsed. This one statement does constitute some evidence that the multiple-modalities
view is accepted by the Court, but given the vagaries of opinion writing and the relatively
free hand that some law clerks are given in drafting opinions, this evidence is surely not
decisive. Similar points could be made about common-law constitutionalism—the only
other theory that has a credible claim to being the “status quo.”264
261
See Griffin, supra note 211.
Jones v. Clear Creek Indep. Sch. Dist., 977 F.2d 963, 965 (5th Cir. 1992) (“Of the six forms of
argument recognized in constitutional interpretation, it is the doctrinal arguments that control Establishment
Clause cases.”).
262
263
Crater v. Galaza, 508 F.3d 1261, 1265 (9th Cir. 2007) (Rheinhardt, J., dissenting from denial of
petition for rehearing en banc).
264
So far as I can tell, no court has ever explicitly endorsed common-law constitutionalism.
129
Lawrence B. Solum
***
I read a lot of constitutional cases. If you are reading this article, I bet you do too. My
impression is that eclecticism best describes the constitutional status quo. And I would
guess that some readers will agree. Other readers will not. Some may think that the
multiple-modalities theory is the status quo. If David Strauss reads this article, I am
fairly sure he will think that common-law constitutionalism is the status quo! And some
may think that representation-reinforcement Thayerianism or Dworkin’s theory or the
multiple-meanings view is roughly the status quo—although the emphasis would have to
be on the “roughly.” But they can’t all be right. This disagreement about what
constitutes the status quo is encompassed by eclecticism—it is exactly the situation that
eclecticism would predict. None of the other theories has an easy time explaining the
disagreement among theorists about the nature of the status quo. Eclecticism fits the
primary evidence—what we know about the way opinions are written—and it fits the
secondary evidence—the pattern of disagreement about how to interpret the primary
evidence. No other theory has this virtue.
***
C. Constitutional Compromise
Consider another set of alternatives to the Constraint Principle, which I shall call
“constitutional compromises.” The basic idea of a compromise between originalism and
nonoriginalism is to divide constitutional issues into two sets—one set to be governed by
the Constraint Principle and the other set to be governed by some other principle or
theory. I do not know of any explicit articulation of this kind of theory, but it is “in the
air” so to speak. This idea is best explained by giving examples, three of which are set
out in the next subsection.
1. Three Proposals for Constitutional Compromise
There could be many forms of Compromise with respect to the Constraint Principle. It
simply isn’t practicable to consider them all, but the following three seem to me to be
natural expressions of the idea that are particularly salient given the landscape of
contemporary constitutional theory.
a) Originalism Plus Grandfathered Canonical Cases
Here is one possible compromise. We might accept originalism as the theory that
should guide constitutional practice in general, but make a stipulated exception for the
canonical cases like Brown v. Board of Education. Something like this view was
accepted by Justice Scalia and played a role in his so-called “faint hearted
originalism.”265 One can imagine an originalist like Scalia who advocates a compromise
version of originalism that accepts the Constraint Principle, except for a set of canonical
cases like Brown v. Board of Education.
265
Randy Barnett, Scalia's Infidelity: A Critique of “Faint-Hearted” Originalism, 75 U. CIN. L. REV. 7
(2006).
130
The Constraint Principle
It is conceivable that the compromise might limit the set of canonical cases to a list,
but a more principled form of this position would specify criteria for canonicity. For
example, a canonical case might be one with the following criteria: (1) deep and wide
acceptance by officials and the public at large, (2) the decision is foundational for a
substantial set of legal practice such that reversal of the decision would require wholesale
revision of precedent, statutes, or other legal norms, (3) reversal of the decision would
involve serious moral wrong.
b) Originalism for the Hard-Wired Constitution, Common Law Constitutionalism
for Constitutional Software
Another version of compromise might limit originalism to the so-called “hard-wired
constitution”—those portions of the constitution that specify basic institutional
arrangements, such as bicameralism, the rules for representation in the House and the
Senate, the veto power, and so forth. The original meaning of the hard-wired constitution
would be subject to the Constraint Principle, but common-law constitutionalism would
provide the regime for “constitutional software”—those provisions of the Constitution
that do not provide bright-line unambiguous rules.
c) Short-Term Originalism, Long-Term Common Law Constitutionalism
Another compromise position is represented by David Strauss’s idea that
constitutional amendments should be enforced for some period of time after they are
adopted, but then be supplanted by the common-law method. Strauss is more or less
forced to adopt this position by the reductio of his general theory: it leads to the
conclusion that a Supreme Court decision cannot be overruled by a clear and explicit
constitutional amendment. To that extent, this compromise theory seems more like a
legitimating fig leaf than a serious position. All of the objections to common law
constitutionalism apply to the compromise position, but it the fig leaf is either
disingenuous or an admission that many of the arguments against originalism are
outweighed in the short run. In either case, the position seems unstable.
d) Originalism and Thayerianism
A final compromise solution might combine the Constraint Principle for cases where
the communicative content of the constitutional text is clear (not ambiguous, not vague,
and so forth) with a principle of Thayerian deference for cases where the communicative
content of the text was underdeterminate. In practice, this would mean that the hardwired structural constitution would be governed by originalism, but the vague and opentextured provisions would not be judicially enforceable outside a minimal core of
determinate meaning.
Another version of a Thayerian compromise might involve Representation
Reinforcement Thayerianism. Footnote Four of Carolene products included the rights
enumerated in the Bill of Rights in addition to the protection of democratic procedures of
discreet and insular minorities in its set of exceptions to a general rule of deference
131
Lawrence B. Solum
(implemented by rational basis review).266 One might take an originalist approach to the
first eight amendments, but use Representation Reinforcement Thayerianism for other
constitutional issues. Variations on this approach might extend the domain of originalism
to other constitutional issues, including the hard-wired provisions.
2. The Case Against Constitutional Compromise
The case for the Constraint Principle and hence against the nonoriginalist component
of particular constitutional compromises has already been explicated, but there are some
additional issues raised by compromise positions.
One possible problem the compromise is instability. Constitutional compromise
between originalism and nonoriginalist form of living constitutionalism puts
constitutional actors in the position of employing two incompatible sets of ideas about
constitutional interpretation and construction. This incompatibility results from the fact
that the normative grounds for each half of the compromise: the originalist half is
supported by rule of law and legitimacy concerns, the nonoriginalist half by rejection of
these concerns and by other values. It seems reasonable to believe that such conceptual
incoherence might lead constitutional actors to drift away from comprise and towards a
more consistent position. Drift might occur in either direction, moving the system from
compromise towards originalism and more consistent application of the constraint
principle—or moving in the opposite direction, towards some form of nonoriginalist
living constitutionalism.
3. Constitutional Compromise as an Originalist Second Best
Despite the problems with constitutional compromise, originalists might come to think
of a compromise position as an originalist second best. For example, it might be the case
that the canonical cases are politically entrenched such that their reversal is outside the
feasible choice set. Given the unavailability of the originalist first best—strict adherence
to the Constraint Principle for all constitutional issues, compromise might well represent
the second best alternative.
D. Completing the Project of Pairwise Comparison
The project of pairwise comparison is a large one. Each pairwise comparison would
seem to require a sustained effort—a full length law review article or monograph. But
completing this projects would require many years—decades at the pace at which I work.
This suggests that completing of the project of pairwise comparison will require a
division of labor, with different scholars undertaking different parts of the project at
different times.
VIII. A RESTATEMENT OF THE CASE FOR THE CONSTRAINT PRINCIPLE
At this point, we can summarize the case for the Constraint Principle and against the
rivals of originalism. That summary can begin with the core of the case for constraint.
266
United States v. Carolene Products Co., 304 U.S. 144, 152 n. 4 (1938).
132
The Constraint Principle
A. Summing Up: The Core Case for Constraint
Originalism is a family of constitutional theories that agree on the Fixation Thesis and
the Constraint Principle. The Constraint Principle could take many forms; in this article,
we have examined Constraint as Consistency, the Minimalist Version constraint.
Constraint as Consistency requires that constitutional practice be consistent with the
communicative content of the constitutional text and that all of the communicative
content be reflected in constitutional doctrine. Constraint as consistency serves as the
least common denominator among originalist theories, and hence as the dividing line
between originalism and nonoriginalism.
The Constraint Principle governs constitutional practice, and for that reason it must be
justified by normative argument. One line of argument, not pursued here, could be that
constraint is part of positive law and hence obligates if and only if positive law obligates.
This essay pursues a different line, arguing that the Constraint Principle should be
adopted (even if it is not currently part of the positive law) for reasons of political
morality.
The reasons of political morality that favor the Constraint Principle are organized in
two clusters. The rule of law cluster consists of three arguments. The first argument is
that judicial supremacy without constraint results in rule by decree and hence judicial
tyranny. The second argument is that the constraint principle serves the rule of law
values of publicity, stability, and certainty. The third argument is that violation of the
Constraint Principle risks a downward spiral of politicization that ends in substantial
erosion of the rule of law. The second cluster focuses on legitimacy in three ways. The
first argument focuses on democratic legitimacy, the second on transparency, and the
third on judicial role.
One of the central ideas advanced in the paper is that the case for or against the
Constraint Principle requires an examination of particular forms of nonoriginalism. Each
of the nine varieties of nonoriginalism has its own strengths and weaknesses.
Contemporary constitutional theory in recent years has emphasized three versions of
nonoriginalism, the multiple modalities view, constructive interpretation, and common
law constitutionalism. But there are many other forms of nonoriginalism, some of which
are radically different than the usual suspects. Unconstrained Thayerianism is very
different than common law constitutionalism, and the considerations that bear on a
comparison of those two views with originalism are not at all the same.
One of the most important points that emerges from this discussion may be surprising
to some readers. It is relatively easy to identify pro tanto reasons for favoring the
Constraint Principle, and hence if the Fixation Thesis is true and if the constitutional text
provides at least a moderate degree of determinacy, there are pro tanto reasons favoring
originalism. But we should not be surprised. Once we have identified the pro tanto
reasons for constraint, our attention shifts to the objections to the Constraint Principle.
The objections that seem most powerful are variations on a common theme: originalism
does not lead to all the results that the critic of originalism desires and it may be
inconsistent with some results that most Americans hold dear. One variation on this
common themes is the canonical cases objection. More needs to be said on this topic, but
even at this preliminary point, one thing seems clear: the canonical cases objection has
not been well formulated. It is presented as a conversation stopper, but in fact it merely
133
Lawrence B. Solum
begins a conversation in which the opponents of originalism need to say much more
before their case becomes clear and coherent.
B. Reflective Equilibrium: Considered Judgments and Outcomes
What motivates opposition to the Constraint Principle? Why would anyone who take
the Constitution seriously argue for a power of judicial or legislative override? In the
end, I believe that most opponents of originalism are motivated by their sense that
originalism is inconsistent with outcomes that they strongly believe are correct. This
belief provides the oomph behind the canonical cases objection and the injustice
objection.
At a very superficial level, this objection seems compelling. Who would want to buy
a constitutional theory that does not lead to outcomes they desire? But this superficial
appeal quickly fades. First, there is no plausible constitutional theory that guarantees the
outcome that any particular citizen desires. In the real world, constitutional theories are
implemented by judges selected through a political process. The theories that instruct
judges to achieve good outcomes (most clearly, the Moral Readings approach) will lead
judges to achieve the outcomes that they believe are good—and hence may well lead to
very bad outcomes from the perspective of individual citizens. Second, the idea that
individuals should judge constitutional theories on the basis of narrow reflective
equilibrium does not hold up to sustained examination. Constitutional theories need to be
affirmed by citizens who have different views about the best resolution of controversial
cases. This means that they must be assessed using the method of wide reflective
equilibrium.
These theoretical points can be expressed more directly. If each judge decided cases
on the basis of their own preferences about how the cases should come out, then
outcomes will vary with the composition of the judiciary—especially the Supreme Court.
And if this fact is frankly, acknowledged, there is no reason for the political system to
pick judges on the basis of any criteria other than outcomes. In some political
circumstances, this will lead to a highly politicized court. In other circumstances, it will
lead to gridlock—because the President and the relevant group in the Senate (either a
majority or the 60 Senators required for cloture—depending on the future of filibuster)
are unable to reach agreement. But this outcome is almost certain to have damaging
consequences for the rule of law.
C. Pairwise Comparisons Revisited
Pairwise comparison makes the case for originalism complex, but such complexity is
mostly unavoidable. Nevertheless, it is possible to make some general observations
about the structure of pairwise comparisons by dividing the rivals of originalism into
three broad categories.
First, one set of nonoriginalist theories share the common characteristic that in practice
or in theory that permit judges to rely on their won first-order normative judgments as the
motivating reasons for amending constructions of the Constitution. This property is
shared by the Multiple Modalities, Common Law Constitutionalism, Moral Readings, the
and Supreme Court as Superlegislature. All of these approaches face acute versions of
the rule of law and legitimacy arguments for the Constraint Principle.
134
The Constraint Principle
Second, another set of nonoriginalist theories are constructed so that they are more
democratic than originalism. Popular Constitutionalism and Thayerianism belong in this
group. Moreover, some of these theories do not involve rule by judicial decree and hence
escape the judicial tyranny argument: government by legislative supremacy, the
institutional form of Unconstrained Thayerianism, fits in this category. These theories
have the property that they are substantial departures from the status quo—more radical
than originalism and subject to many of the same objections that are levied at originalism.
They raise an entirely different set of issues than does the first group, and a deep
consideration of these alternatives requires an inquiry into the question whether the
project of American constitutionalism requires radical revision.
Third, a final set of nonoriginalist theories involves an even more radical departure
from the status quo. Antitheoretical views such as eclecticism and opportunism reject the
idea that constitutional practice should answer to a consistent set of principles with
normative grounding. These views suffer from serious problems with respect to the rule
of law and legitimacy, but they are based on premises that call into question the very
ideas of democracy and legitimacy.
The relationship between the core justifications for the Constraint Principle and the
nine pairwise comparisons is summarized in the following table:
Table 2: Summary of Pairwise Comparisons for Pro Tanto Reasons for Constraint
Rule of Law Cluster
Judicial Tyranny
Rule of Law Values
Politicization
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Democratic Legitimacy
Transparency
Judicial Role
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Multiple Modalities
Moral Readings
Common Law Constitutionalism
Popular Constitutionalism
Multiple Meanings
Superlegislature
Thayerianism
Antitheories
Rejectionism
Legitimacy Cluster
Meaning of the Colors
Red = Favors the Constraint Principle
Green = Favors Neither
Purple = Favors Alternative
Blue = Contestable
135
Lawrence B. Solum
D. Does It Matter Whether We Call It Originalism?
At this point, some readers might be willing to accept Constraint as Consistency, but
reject the label “originalist” for their position. They might want to call themselves “new
textualists” or “moderate living constitutionalists” or something else. Could we use some
label other than originalism for the conjunction of the Fixation Thesis and the Constraint
Principle with some plausible theory of original meaning (such as Public Meaning
Originalism)? Does it matter whether it originalism? Yes, and no.
Begin with “No, it doesn’t matter.” As long as we are clear, what matters most is the
substance. If what I am calling “originalism” is accepted but given some other name, my
primary objective as a constitutional theorist will have been accomplished. The
metalinguistic question as to what should be called “originalism” is of secondary
importance.
But then, “Yes, it does matter.” The account of “originalism” on offer in this article
and elsewhere is not arbitrary. The Fixation Thesis and the Constraint Principle provide
the common core for the family of constitutional theories that are self-described as
originalist. It is true that this family of theories is historically connected to the criticisms
of the Warren Court (and the Berger Court and New Deal Court as well). And it is true
that many (but not all) of the self-identified originalists have been associated with
conservatism or libertarianism. It is hardly surprising that a theory that emerged in one
historical context could have implications that its early adherents did not clearly see and
might not fully welcome. And it is no surprise that criticism of originalism has elicited a
response from originalist: that this the way that legal theories evolve and develop. The
opponents of originalism can hardly be blamed for discomfort with the notion that they
might be convinced to accept originalism in the moderate form associated with Constraint
as Consistency, but surely this does not warrant their rejection of the meaning of the word
“originalism” that best fits the theoretical discourse of the last three decades.
CONCLUSION
This Article does not aim to end the debate over originalism and living
constitutionalism. It seems very unlikely that any monograph or article, no matter how
well argued, could produce agreement among legal scholars, lawyers, and judges that
originalism is bunk or that living constitutionalism is junk.
The aim of this article is much more modest but still ambitious. I hope to have made
progress in constitutional theory by clearly identifying the issues that need to be
addressed in order to evaluate the case for and against the constraint principle. And more
than that, I hope that I have given you good reasons to believe that the shape of this
debate is quite different than the conventional accounts would have it. It is in fact false
that there are no good normative arguments for originalism; rather, there is an abundance
of arguments that provide pro tanto reasons for the Constraint Principle. It is in fact false
that originalism is so weak that one does not need to present an alternative; rather, the
case against the constraining force of original meaning can only be made by presenting
and defending a coherent alternative (or set of alternatives) to originalism. It is in fact
false that the canonical cases objection disposes of originalism, so long as one or two of
the canonical cases turn out to be wrongly decided on originalist grounds; rather, the
canonical cases objection is just the first move in a conversation that has barely begun.
136
The Constraint Principle
***
“[I]t is ambition enough to be employed as an under-labourer in clearing the ground a
little, and removing some of the rubbish that lies in the way to knowledge.”
—John Locke267
In a real sense, the most important claim of the article is that this debate has barely
begun. It is for that reason that the first 54 pages engage in elaborate, painstaking, but
conceptually precise “rubbish removal” or “brush clearing.” That effort is motivated by
my belief that the debate over originalism and living constitutionalism is rife with
conceptual confusion and misunderstanding. Before we can resolve the debate over the
constraint principle, we need a clear understanding of the structure of the issues.
Suppose that I am right and that constitutional theorists gradually come to see that the
issues in a new light. Once should then expect the emergence of a new set of arguments,
both for and against the Constraint Principle.
Optimistically, the proponents of originalism might hope that much of the action will
shift. If the Fixation Thesis and either the Constraint Principle or some robust version of
the weaker Contribution Principle came to be accepted as true, then the action might
shift to a third line of defense for living constitutionalism. This new Maginot Line might
be built on a revival and rearticulation of the claim that that the communicative content
of the constitutional text is substantially underdeterminate (or even radically
indeterminate). At the end of the day, that shift would entail a shift of attention from
constitutional theory to more concrete and particularized inquiries into the meaning of
specific constitutional provisions.
This article provides reasons to believe that the Constraint Principle—the single most
important thesis advanced by originalism—is supported by reasons that are powerful and
hence that the case for constraint demands answer. For the debate to move forward from
this point, the opponents of originalism must reply.
267
JOHN LOCKE, AN ESSAY CONCERNING HUMAN UNDERSTANDING ix-x (1836).
137