Legaciesof AmericanSlavery:Status
Attainmentamong SouthernBlacks
after Emancipation*
MARTINRUEF,Stanford University
BEN FLETCHER,
Universityof North Carolinaat ChapelHill
Abstract
This study examines the legacy of American slavery at the individual,
levelby analyzinglife-historydatafrom roughly1,400 ex-slaves
intragenerational
andfreeblackscoveringtheantebellumandpostbellumperiods.Wetesta modelof
durableinequalitythat considersthe potentiallyviciouscirclecreatedby status
persistenceacrossinstitutionalregimes.Ourfindingssuggestthat the antebellum
in thelate nineteenthand early
regimeevidencedpartialinstitutionalreproduction
twentiethcenturies,owingto thefact that the antebellumdistinctionoffree blacks
and slaveshad durablestatuseffectslongafteremancipation,but overtime, black
statusattainmentbecamelargelydecoupledfromthe internalhierarchyof slavery.
and
Mediatingeffects,for example,theFreedmenBureau'seducationalinterventions
the blackdiaspora,also servedto curtail the reproductionof antebellumstatus.
Implicationsarepursuedwith respectto bothinstitutionaltheoryand stratification
research.
In December 1865, the American states ratified the ThirteenthAmendment
to the U.S.Constitution,formallyemancipatingover4 million blacksin former
slaveholding states (see Schwartz 1970:25-96 for a legislative history).' The
deinstitutionalizationof slaveryand its effectshavebeen studiedby a numberof
intellectualssince the late nineteenth century,ranging from DuBois's (1935)
* This researchwas
supportedby a Junior FacultyDevelopmentAwardfrom the Universityof
North Carolina at Chapel Hill. We would like to thank Judith Blau, Peter Coclanis, William
Darity, and attendeesat the 2002 SouthernSociologicalSociety meetingsfor helpfulfeedback.
Special thanks also go to Paul Escottfor providing his data set based on a subset of the WPA
interviews.Direct all correspondenceto Martin Ruef, Graduate School of Business, Stanford
University,Stanford,CA 94305-5015. E-mail: [email protected].
Editors'Note: Thisarticlewas evaluatedand acceptedbeforeJudithBlau becamecoeditorof the
journal.
? The Universityof North CarolinaPress
SocialForces,December2003, 82(2):445-480
446 / Social Forces 82:2, December 2003
inquiry into the political potential for greater race and class equality in
southern reconstruction to the large-scale quantitative analyses conducted
under the impetus of the Associationfor the Study of Negro Life and History
(Greene& Woodson 1930; Wesley 1927; Woodson 1918; see also Woodson
1922). Despite this earlyflurryof scholarship,however,more recent efforts in
social historyhave often focused on the institutionof slaveryitself ratherthan
its legacy (Dunaway 2003; Fogel 1989; see also Stinchcombe 1995), while
examinationsof slavery'sconsequenceshave emphasizedaggregate,macrolevel
effects (e.g., Ransom& Sutch 2001). Lackinglongitudinalanalysisat the micro
level, the legacyof Americanslaveryremainspoorly understoodfor individual
blacks emancipatedfrom this "peculiarinstitution"(Stampp 1956).
Sociologicalneglect of the topic is surprising,since instancesof profound
institutionalchange, such as emancipation,offer an unusual opportunityfor
studentsof stratification
to examinethe persistenceof structuredinequalityacross
institutional regimes. After such change, the typically strong persistence of
- may give way to rapid
status- both intergenerationaland intragenerational
or
downward
At
the
same
formal
deinstitutionalization
time,
upward
mobility.
be
remarkable
in
may accompaniedby
stability underlyingmaterialconditions,
norms, and governance structures channeling mobility processes. This
institutionalstabilitywas reflectedin the postbellumSouthby the employment
of ex-slaveson the plantationsof formermasters,in oppressivesharecropping
arrangements,and in other forms of agriculturalpeonage.
Aside from questions of social mobility,institutionalscholarsfind legacies
of formallydismantledinstitutions to be of interest in their own right. Such
legacies hinge on the reproductionof materialand cultural conditions from
outdated institutions (Scott 2001), owing not only to the persistence of
socioeconomicstatusacrossinstitutionalarrangementsbut also to the resulting
ambiguityof identity change for persons embeddedwithin them. Particularly
for individualslocatedin the lowest ranksof socioeconomicstatus,such as exslaves,the subjectivesense that new institutionalarrangementsare merely"old
wine in new bottles" is likely to inhibit productive collective action and
reinforcea perniciouscycleof statuspersistence.Oppositionaltacticsemployed
by southernwhites duringReconstructionand its aftermath- includingwhite
terrorism,the passageof black codes, and the constructionof segregationserved as institutionalsupportsto this cycle.
In this article, we examine the extent to which slavery continued to
influencethe social status of southernblacksafteremancipation.Our specific
emphasisis intragenerational,consideringa sample of 1,471 blackswho were
born before the end of the CivilWar,their statuswithin the plantationsystem
of chattel slavery, and subsequent effects of that system on individual
socioeconomic attainment.To trace these outcomes, we employ life histories
from the WorksProgressAdministration(WPA)FederalWriter'sProject,based
Legaciesof AmericanSlavery/ 447
on interviewsconducted with ex-slaves and free blacks during the late 1920s
and 1930s.These interviewsplace our focus at the "intersection,"
as C. Wright
Mills (1959) termedit, of historyand biography.Applyingquantitativeanalyses
to the life histories, we derive both specific implications for the legacy of
Americanslaveryand more generalinferenceswith respectto theoriesof status
persistenceand institutional change.
Institutional Legacies and Durable Inequality
An institutional legacy refers to the reproduction of material-resourceand
culturalconditionsfrom a social institutiondespitethe fact that the institution
has been formallydismantled(e.g., the ongoing effectsof Americanslaveryafter
the passage of the Thirteenth Amendment). A simple version of such
reproductionoccurs at the intragenerationallevel when individualscontinue
to bear the burden of outdated institutions because socioeconomic status
within present institutional arrangementsis positively correlatedwith status
under former institutions. The resulting status inequalities are especially
durable if they do not necessarily rely on the subjective beliefs of the
individualsinvolvedbut are embodiedin the organizationalformsthat channel
their mobility processes (Tilly 1998).
A concreteexampleof such institutionalreproductioncan be found in the
crop lien and sharecropping systems that emerged during southern
Reconstruction.Prior to emancipation,the majorityof the South's4 million
slaveswere employed as unskilledworkerson cotton, tobacco, sugar,and rice
plantations. Estimates from probate and plantation records place the
percentageof slaves employed in this lowest tier of plantationlabor at nearly
70%(see Olson 1992).Afteremancipation,many of these ex-slavesdriftedback
to unskilled farm work by virtue of necessity but, lacking resourcesof their
own, reliedon creditfrom merchantsto acquiretools, seeds,livestock,and the
like (Foner 1989; Greene& Woodson 1930). Those without land were forced
into sharecroppingtenancy.These arrangementswere generallysecured by a
lien againstthe crops of the ex-slaves,where the conditions of the lien allowed
landlordsand merchantsconsiderableauthorityin dictatingthe type of crop
grown, its quantity,and the method of agriculturalproduction.
Emancipation meant a tremendous increase in the degree to which
freedmen and freedwomencontrolled their privatelives, but, at least for the
majority employed in sharecroppingarrangements,there was not always a
commensurateincreasein economic autonomy.The reasonsweretwofold.First,
the debt load imposed by unscrupulousmerchantsoften forced ex-slavesinto
materialpenury that placed them in a tier of socioeconomic status similarto
that which they had experiencedas field hands (Mandle1992). Eventhose who
448 / SocialForces 82:2, December 2003
were fortunate or talented enough to escape low-status occupations usually
spent a period as unskilledlaborersaccumulatingculturalor economic capital.
Second, the authorityexercisedby merchantsand landlords- and effective
serfdom of many ex-slaves - meant that the cultural distinction between
chattel slaveryand "free"wage labor was likely to seem ambiguous to some.
Clearlythe frequencyof physical coercion was reduced under freedom, but
even this crucialimprovementin the lives of ex-slaveswas far from universal
(Cohen 1976; Daniel 1979).2 In these respects, sharecropping and lien
arrangementscould be seen as reproducingmaterialand culturalconditions
from the antebellumplantationsystem.Otherexamplesof practicesdescendent
from slaveryinstitutions included the enforcement of black codes by police
and the harassmentof blacksby white patrols and vigilantegroups convened
to regulateAfricanAmericanbehavior.3In light of these persistentstructures,
we ask: To what extent did the basic cycle of status persistenceremain after
emancipation?
The riskof institutionalreproductionis presentas long as statusinequalities
under one institutionalarrangementtranslateinto inequalitiesunder another
arrangement.Thispatterncontributesto equilibriumin a systemof stratification,
Marxist,and
despitesuperficialtransformation.Employingstructural-functional,
neoinstitutionaltheories,threecanonicalaccountsmightbe advancedto predict
durableinequality- or lackthereof- acrossinstitutionalregimes(see Figure1).
In CharlesTilly's(1998) structural-functionalexplanation,robustcategorical distinctions among actors are essentialin sustainingstatus inequality.Insofaras new organizationalforms, such as sharecroppingarrangements,match
old status categories(e.g., slavesand owners) with their own categoriesof inequality(tenant farmersand landlords),patternsof inequalityare likely to be
maintained.Thus, accounts by historians often assume that there were constraintson status change after emancipation,noting that postbellum "racism
in slightlynew forms performedthe same function for the social system that
slaveryhad once served,and racialoppressionwas essentialto the ends of the
system- little had changed for the black man, either practicallyor theoretically"(Escott 1979:164).Similarly,sociologicalargumentsreflecton gradations
within the slave hierarchy(Stinchcombe 1995; Tilly 1998) and the tendency
of elites to grant privilegesin freedom along the same distinctions.4
Marxist scholars tend to agree that institutional reproduction rests on
robust categoricaldistinctions,but they draw attention to shifts in the means
of production that may serve to disrupt these categoriesacross institutional
regimes (Wright 1997). In this regard,there were fundamentaldifferencesin
the postbellumsituationof the free blacklaborer,who "owns"him- or herself,
and the antebellumslave(Litwack1979). In particular,two crucialdimensions
distinguish the self-ownershipof free wage laborersfrom the legal status of
Legaciesof AmericanSlavery/ 449
FIGURE 1: Some Canonical Analyses of Institutional Transformation of Social
Status
Statusin
former
Institution
+
Statusin
present
Institution
a. Structural-Functional
Analysis
I- -
Slave :
I
Owner :
'
I
............,
Tenant
Landlord---Landlord!
- ._-- -
Statusin
former
Institution
?
-
Statusin
present
Institution
b. MarxistAnalysis
-?
?...
Slave
-
Changein
+
--
Owner
------
Mode of
Production
.
_
_
+
_- -_
Institution
, Ia-----^
.Non-Persons
L-------------!
-----------Persons I..
----------Act ,j
,
'Free'Labor
--------
Bourgeoisie
-- -- -- -
Statusin
former
?--,?
Status in
present
Institution
Institutional
Fragmentation
'Low-Status
Actors
Actors-: ------High-Status
Actors
- - - - - - --- - - - -
c. Neoinstitutional
Analysis
450/SocialForces82:2,December2003
productive time in pursuits other than work.5Geographicmobility allowed
many freed slaves to sever ties with former owners and abandoninhospitable
areas and conditions in the former slaveholdingstates. Substantialinterstate
mobility, culminating in the GreatMigration of the early twentieth century,
attested to the popularity of this exit strategy for many southern blacks
(Gottlieb 1991). Anotherfundamentaldifferencebetweenthe two institutional
regimes was the possibility for free black laborers to invest labor time in
alternative pursuits, such as education. Despite the opportunity costs and
difficultyof obtainingeducation,it was widely embracedby formerslavesand
their children.6From a Marxistperspective,the gradationalstatus of blacks
within the antebellum regime may not be as important to their postbellum
occupationalattainmentas their relationshipto the new means of production,
mediatedvia worklocaleand educationalinvestment(Figurelb).
Informedby culturalexplanationsof individualismand agency(Meyer1994;
Meyer& Jepperson2000), neoinstitutionalistconcerns with the transition to
freedom are driven by the difficulties inherent in the construction of a new
class of "persons"or "actors."In this regard, freed blacks were not simply
adjustingto a new legal status or means of productionbut were engagedin a
more fundamental process of identity reconstruction, navigating tensions
between formal equality and functional inequality (Meyer 1994). The
mediatingrole of identitychangeimpliesless stabilityin statuspersistencethan
structural-functionalaccounts might suggest. Thus, scholarship on chattel
slaveryas an institution (Stampp 1956;Winsell 1971) suggests that many of
its statusdistinctionsdid not generalizeto the more universalstandardsof free
wage capitalism.Blacks'rejectionof plantationpaternalismand its patternof
personalallegiancescontributedto this decouplingin the postbellumperiod.7
Among other civil liberties,freedmen and freedwomensaw in emancipation
the opportunityto point out the hypocrisyof a systemof paternalismfounded
on both Christianethicalcodes and harshphysicalcoercion (Genovese 1974).
If processesof identity reconstructioncontributedto broaderinstitutional
it is alsounclearthatrelationshipsto the new meansof production
fragmentation,
were as telling as Marxistexplanationsmight propose.To some extent,old and
new institutionalarrangementsexisted side by side in the postbellum South,
leadingto mixedexpectationsforthe freelaborforceengenderedby emancipation.
Given such uncertainty, geographic mobility and control over work locale
oftenbroughtonlylimitedmaterialbenefitsto blacks- at leastwithinthe former
slaveholdingstates.The paradoxof symbolicbenefit in the absenceof material
gain could also be seen in the sharecropping system that evolved in the
postbellumperiod. Sharecroppingwas a victory for ex-slaves,when viewed as
a refusal of supervised labor systems that mirrored the overseer system.8
However,sharecroppingoften transferredwealth systematicallyfrom blacksto
planters- or, as ex-slaveFelixHaywoodexplained,"freedomcould makefolks
proud, but it didn't make 'em rich" (Litwack 1979:449). In the neoinstitutional
Legaciesof AmericanSlavery/ 451
explanation,this materialuncertaintyis reflectedin a loose mappingbetween
postbellum status attainment and legal status in the antebellum period (see
FigureIc).
Historical Background
To providea snapshotof the complexityof slavery'sinstitutionallegacy- and
potentialdisruptionsto that legacy- we reviewa numberof historicalfactors
impingingon the cycleof statuspersistence,beginningwith the statushierarchyof
blacks in the ante- and postbellum South, opportunities for education and
migration,andthe processof identityreconstructionfacedby blacksmovingfrom
slaveryto freedom.In subsequentsectionsof the article,historicalfactorsimpinging
on statuspersistencearemappedto variablescodedfrominterviewswith ex-slaves
and freeblackswho lived in statesand territorieswith substantialslaveholdings
priorto the end of the CivilWar.9
STATUSHIERARCHYUNDER SLAVERY
Statuspersistencebetweeninstitutionalregimesassumes,firstand foremost,that
underformerinstitutionalarrangements.
therewassignificantstatusdifferentiation
Earlyhistoriographicaccountsoften assumed,erroneously,thatthe occupational
structureon southern plantations was relativelyundifferentiated,with most
slavesfallinginto eithera largeclassof unskilledfield laborersor a smallerclass
of household servants.Startingwith Wesley(1927), economic historiansbegan
to discover the large number of skilled black artisansand semiskilled slaves
supportingthe plantation system (see also Moore & Williams 1942). Modern
accounts stress not only that occupational differentiation among slaves on
midsizedand largeplantationswas importantfrom a functionalstandpointbut
that the southern plantersactivelysupportedsuch differentiationin order to
control their slave populations (Fogel 1989;Johnson 1986).
chosen
At the top of the statushierarchywerethe blackoverseers,or "drivers,"
for their loyaltyto the planters,managerialtalents,long plantationtenure,and
"imposingphysicalpresence"(Miller 1979). The driversenjoyed considerable
autonomyin runningthe day-to-dayoperationsof the plantation,leadingto debate
among antebellum planters whether black or white overseers should be
employedin this capacity.Evidencefromprobaterecordssuggeststhat the use of
blackoverseersincreaseduntil the 1830sbut then declineduntil the CivilWar,
possibly due to the Nat Turner uprising (1831) and planter fear of slave
insurrection.
Joining the drivers in the "elite" slave occupations were skilled black
artisans- blacksmiths,masons, mechanics, carpenters,and the like - who
452/SocialForces82:2,December2003
generally enjoyed better living conditions, better vocational education, and
greaterautonomy than other slaves. In urban areas,slave artisanswere often
hired out by their masters, leading to competition with white artisans and
occasionalantagonismfrom the generalpopulace (Greene& Woodson 1930).
Probaterecordssuggestthat around 13%of male slaveswere employedin the
capacityof skilled craftsmen;the number of female slaves so employed was
negligible.
Betweencommon field laborersand the elite slave occupationsof artisans
and overseers, there existed a differentiated status hierarchy of domestic
servants(e.g., waiters,butlers,cooks, barbers),semiskilledworkers(teamsters,
coach drivers,gardeners,clothmakers),and unskillednonagriculturalworkers
(launderers,porters).These stratacomprisedroughly 13%of adult male slaves
and 29%of adultfemaleslaveson midsizedand largeplantations(Olson 1992).
Placementwithin these middle ranks,perhapsto a greaterextent than in the
elite slave occupations,was subjectto ascriptivedecisionsby mastersbased on
the skin complexions, interpersonal relationships, and personalities of the
slaves.10Under slavery,status ranks above field hand status were perpetuated
by marriagepatterns.Skilledmanualworkers,overseers,and to a lesserextent
semiskilled workers usually married domestic servants or other high-status
slaves (Johnson 1986; Schwalm 1997). After emancipation, the institutional
peculiarityof some of these ascriptivedecisionscould renderstatuspersistence
in the middlerangevulnerableto disruption.
Fromthe standpointof a statusattainmentmodel, two other statuseswithin
the slaveryregime complicatethis hierarchy.One is the substantialnumber of
free blacks in the antebellum South, comprising over 5% of the black
population in 1860. Aside from their autonomy and high prestige vis-a-vis
slaves, detailed occupational records from southern urban centers indicate
impressiveoccupationalattainmentamong free blacks.A census of free blacks
from Charlestonrevealeda numberof proprietors(storekeepers,tavernkeepers,
milliners) and a large number of skilled artisans (Wesley 1927); notably,few
free blackswent into domestic servicecomparedto urban slaves.
The other complicationfor a model of statusattainmentinvolvesslaveswho
werechildrenat the time of emancipation.Becauseof the earlyageat whichslaves
wereput to work- someat aroundagethreeor fourandroughlyhalfby ageseven
(Fogel1989)- ageitselfis not an adequateproxyof childhoodunderthisregime.
Rather,our definition of childhood includes those young slaveswho had no
occupationassignedto thembeforeemancipationand thus neverexperiencedthe
Whileaccountsof plantersthemselvessuggest
workingconditionsof slaverydirectly.
that therewerebenefitsto earlyentryinto the slavelaborforce,in termsof skill
formation(seeRoughley1823),the costsof lost childhoodandsocializationwithin
an antiquatedagriculturalgang system are not to be ignored.By avoidingthis
patternof socializationwithin the institutionof slavery,some childrenendured
Legaciesof AmericanSlavery/ 453
its legacyto a lesserextentthan theirpeerswho had alreadyspent time working
as unskilledagriculturallaborersor domesticservants.
ATTAINMENT
AFTERSLAVERY
STATUS
After a brief period of celebration, freedmen and freedwomen faced lives
characterizedby extremelyhard work. The Union army encouragedsouthern
blacks to return to their occupations under slaveryas a practicalsolution to
the need for economic relief and to fear among white elites that the southern
economy was unviable without black labor (Cohen 1991). For agricultural
laborers, returning to preemancipationjobs was probably not good advice.
Many sharecroppersand tenant farmersexisted in abjectpoverty,often living
meal to meal. Yet opportunitiesfor blacks to enter higher-statusoccupations
were few in the Reconstructionera, even in the North. Institutionsand cultural
norms designedto separatethe raceslimited AfricanAmericanstatusmobility
in the late nineteenth and early twentieth centuries. Williamson (1965)
describesa common reactionto segregationistinstitutionsamong blackswho
stayed in South Carolina:"Negroesgenerallysearchedfor perfection within
their half of a dyarchicalsociety" (301).11
WhiteslimitedAfricanAmericans'accessto high-statusworkoutsidethe black
community.Southernwhite eliteshiredaccordingto informallydefinedlimits of
whatconstituted"blackjobs"- typicallyagriculturallabor,semiskilledindustrial
labor,or domesticservicework (Spero& Harris1931).Forthe AfricanAmericans
who continuedto work in agriculture,productivitywas severelyconstrainedby
the relativelackof tools, especiallyfor sharecroppers
who often reliedon planter
loans (Ferleger1993;Wiener 1979).
Structuresconstrainingblackstatusmobilitywerenot universallysuccessful.
The blackcommunityincludedpreachers,skilledcraftworkers,teachers,and a
smallnumberof other professionals.Fiftyyearsafteremancipation,obstaclesto
statusmobilityweresomewhatcounterbalanced
by increaseddemandforindustrial
laborduringWorldWarI. Still,a largeportion of jobs blacksgainedduringthe
war were lost in peace, and prospectsfor occupationalmobilityamong African
Americansdid not dramaticallyimproveagainuntil WorldWarII.
EDUCATION BEFORE AND AFTER EMANCIPATION
The threat posed by education to status persistenceamong blackswas widely
recognizedby plantersin the antebellumSouth. In responseto an increasein
abolitionistactivityin the 1830s,laws againstthe educationof enslavedblacks
were strengthenedin the slaveholdingstates. In some areas,it was a criminal
offense to teach any black, enslavedor free, to read or write. GeorgeAlbright,
a former slave and nineteenth-century politician, explains: "It was only by
trickerythat I learned to read and write - if any slave learned to read and
454/SocialForces822,December2003
write, he was to be punished with 500 lashes on the naked back, and to have
the thumb cut off above the second joint" (Rawick1972-79,6.1:8-19).Official
prohibitionsagainsteducatingblacksand widespreadplanterhostility toward
learningamongslaveslimitedliteracyto a verysmallfractionof the southernblack
populationby the time of emancipation(DuBois 1935).
Northern attempts to develop an educationalinfrastructurein the South
began before the Civil War was over. The Union army trained teachers,
confiscated rebel homes as schools, and sought to instruct ex-slaves in "the
rudiments of civilization and Christianity"(General William Sherman, as
quoted in Blassingame1965).While some of these effortsmay havebeen driven
by moral concern,it must also be rememberedthat the Union'suse of ex-slaves
as soldiers was severely limited by illiteracy and that federal officials saw
education as a means of control as well as enlightenment. Handbooks
distributed by the Freedmen's Bureau were designed to help the recently
emancipated by instilling strong work ethics and ideologies of disciplined
docilityin theirreaders(Hartman1997).
After the Civil War,the Freedmen'sBureauundertook a more systematic
approachto institutionalintervention,whichwas matched,with variousdegrees
of effort, by the developmentof public systems of primaryeducation among
Reconstructiongovernments.Evenat its height,this systemof public education
reachedonly approximately10%of blackchildren(DuBois 1935) and suffered
considerable setbacks with the restoration of southern elites in the 1870s.
Governmentefforts were dwarfedin many areas,moreover,by the grassroots
formation of schools by freed AfricanAmericans,many of whom worked as
builders and teachers without pay (Anderson 1988; Gutman 1987). Maria
Jackson'sstoryrevealsthat,for southernblackfamilies,obtainingeducationwas
a struggle wrought with pragmatic compromise:"[My brothers and sisters]
learnedright well in school. Us other childrenhad to help Daddy in the field"
(Rawick1972-79, 1:267-74).Nevertheless,public education could be counted
as one of the successes of Reconstruction; by the 1890s, black literacy
nationwidehad risen to nearly40% (Greene& Woodson 1930).
The impactof educationon blackstatusattainmentafteremancipationhas
not been studiedsystematicallyat the individuallevel, althoughwell-publicized
debates raged in the late nineteenth century concerning the most effective
forms of education.BookerT.Washington(1901) famouslyadvocateda system
of practical, industrial education, claiming that "the opportunity to earn a
dollarin a factorynow is worth infinitelymore than the opportunityto spend
a dollar in an opera house" (206-26). Fearing that Washington'sapproach
would lead to abjectproleterianization,DuBois ([1903] 1965) counteredwith
proposalsfor collegeeducation,emphasizingthe "talentedtenth"amongblacks.
Eventoday,these debatesare unlikelyto be settledwith the data at hand.What
Legaciesof AmericanSlavery/ 455
can be hypothesized, however, is that education- of any sort - served to
substantiallydisruptthe legacyof the antebellumregime.
MIGRATION
The geographicmobility of blacks following emancipationservedas a second
disruptionto the reproductionof the antebellumregime. In his pathbreaking
analysis,Woodson (1918) tracedthreewavesof southernblackmigrationafter
emancipation:one resultingfrom the immediatedisruptionof the plantation
economy during the Civil War,12a second, westward movement upon the
restorationof reactionarywhite governmentsduringthe late 1870s,and a third,
"greatmigration"to the North, peaking during World War I when wartime
industrieswere understaffedand foreignimmigrationwas reducedto a trickle.
Many southern elites actively tried to stop migration using persuasion,
accommodation,and legalizeddetention (Cohen 1991).At least as late as 1916,
cases exist wherein black agriculturalworkerswere forcefullypreventedfrom
seeking better prospects through migration (Cohen 1976).
DuBois ([1903] 1965) recognized benefits of migration for developing a
pan-African American identity (see also Blau & Brown 2001). The aims of
migrantsthemselves- and of migrantaid societiesin the North and Westwho
assisted them - were far more concrete. John Mathews described to WPA
interviewerswhy he relocatedrepeatedlyacross the cotton belt: "when [the]
end of the year come there was nothing to pay the [farm]hands. I got work at
a saw mill and made enough for us to live on. When the bulldozerstell me to
move, I move" (Rawick 1972-79, 9:1450-60). Still facing profound
discriminationin the postbellumSouth, other blackssaw migrationas a form
of collectiveaction, the only way to "elevate[themselves]to a higher plane of
true citizenship"(U.S. Senate 1880:vol. 7, p. 281; DuBois 1907). Meanwhile,
some black leaders,such as FrederickDouglass (1879), consideredmigration
an abdicationof rightsand a failureto leveragethe sheer number of blacksin
the South to economic and political advantage.
As emphasizedby modernstudentsof collectiveaction,individualadvantage
and collectivedisadvantage
mayexistsideby side.At the individuallevel,interstate
migration after emancipation may have severed economic dependencies on
local white elites, and blacks migrating outside the bounds of the former
Confederacymay have encountered less discrimination.l3At the same time,
other scholars- followingDouglass'swarnings- havesuggestedthat this exit
strategy crippled more fundamental civil rights reform in the South for
decades to come.
456/SocialForces82Z December2003
RECONSTRUCTING
IDENTITIES
For many years, the emphasis of economic historianson materialconditions
blacksslightedthe parallelprocessof culturalreconstruction
amongemancipated
how ex-slavescameto termswith theirchangein identityfromslaveto freedman
or freedwoman.As emphasizedby our theoryof institutionalreproduction,this
dynamicmay be intimatelytied to underlyingstatus persistence.A slave who
had languishedas a field laborer in the plantation economy and was driven
into peonage afteremancipationmay have questionedthe meaning and value
of identity change. In many cases, though, the cycle of culturalreproduction
was brokeninsofaras ex-slaves'valuationsof their identitiesbecamedecoupled
from the material conditions of life, instead emphasizing more abstract
principles of freedom (Litwack1979).
Identityreconstruction
amongex-slavesoccurredin a climateof stigmatization,
as southernwhites questionedtheir social and economic usefulness.Freedom
was an attribute still considered unnaturalfor southern blacks, leading to a
discrepancybetween what Goffman (1963) has referredto as idealized and
actual social identity. Strategies for managing this discrepancy took on a
numberof forms among ex-slaves.Some engagedin wholesalerejectionof the
planter regime, its status hierarchy,and violence. LydiaJefferson,a former
house slave from Louisiana,felt like she came out of "a black hole into [the]
sunlight"after emancipation,stating that the "treatmentwhat some of [the]
slavesgot dat I's see with my own eyes was awful"(Rawick1972-79,6.5:193942). Jefferson expressed this view even though her own existence on the
plantationhad been one of relativeprivilege,at least comparedto field hands.
Other ex-slaves expressed the ambiguity inherent in weighing ante- and
postbellum identities. Talkingto a WPA interviewer,Andy McAdamsnoted,
"well son, I'se expected lots different from freedom than what we got....
news
came one day that we were free and that [same] day they opened the gate and
set the dogs afterus - just like you would a bunch of wild cattlethat you were
going to turn loose in a largepastureto grazeor rustlefor theirliving"(Rawick
1972-79,7.6: 2455). McAdamsrecognizedthat his personalsufferingmight be
accompaniedby collectivebenefitsfor southernblacksas a whole:"usold slaves
has had a hardtime of it but it has been worth all our hardshipscause,look at
the Negro people today.... our people progressedalong to where they don't
have to suffer the hardships we did trying to learn what our white people
wanted us to do" (p. 2456).14
Since individualstories of status persistenceor change differ considerably,
we constructed quantitative measures of ante- and postbellum status from
autobiographicalnarratives.Analysisof these measuresprovides a summary
description of status shifts after emancipationand can be used to assess the
relativevalidity of competing views on status persistenceacross institutional
regimes.
Legaciesof AmericanSlavery/ 457
Data, Measures, and Method
DATA
We trackedchangesin the socialstatusof southernblacksusing interviewsfrom
the WPA FederalWriter'sProject.Followingpilot projectsconducted at Fiske
University,SouthernUniversity,and PrairieView Collegein the late 1920s (see
Cade 1935; Egypt, Masuoka & Johnson 1945), the FederalWriter'sProject
soughtto developa more comprehensiveautobiographicalportraitof ex-slaves.
Between 1936 and 1940, this effort led to the collection of life histories from
over 3,000 formerslavesand free blacksin 25 statesas well as a large number
of secondary materials,such as bills of sale from the antebellum South and
obituariesof ex-slaves (Yetman1984).
The dataused in this articleinclude1,590interviewsin the WPAarchivesthat
elicited informationfrom southernblacks on their socioeconomic attainment
during the postbellumperiod as well as their status under slavery(see Escott
1979; Jacobs 1981; Rawick 1972-79). To be included in further analysis,
respondents had to be born before or during the Civil War and reside in a
slaveholdingstate (or territory)beforeemancipation.Afterremovingcasesthat
failed to meet these criteria, 1,471 remaining interviews were coded for
identifyinginformation,age, gender,familybackground,education,migration,
and occupational attainment of each respondent. Concerns about sample
representativenessand the advancedage of some respondentswere addressed,
as noted below.
Representativeness
In many respects,the WPA interviewsprovide a unique data set for tracking
blackstatusmobilityin the nineteenthcentury.Before1870,U.S.censusrecords
did not identifymost southernblacksby name or occupation;slave schedules
simply enumeratedcharacteristicssuch as number and age of slavesowned by
particularmasters.Other potential sources of data on status mobility- such
as conscriptrecordsfor the Union army- do report former occupationsbut
are obviously conditioned on particular status outcomes for ex-slaves.
Consequently,despitethe shortcomingsnoted below,the WPAarchivesinclude
the most representativedata available on intragenerationalblack mobility
between the antebellumand postbellum regimes.
Although the aim of the Federal Writer's Project was to generate
representativelife histories of former slaves, little systematic sampling was
conducted (Yetman 1984). As a result, the archivescontain narrativesfrom
roughly 2% of the ex-slavesstill living in the 1930s but featuresconsiderable
variationin the number of interviewscollectedfrom state to state.To account
for geographicbias in the sample, we weighted all cases by state of origin to
458/SocialForces 82:2,December2003
correspond to the slave and free black populations at the end of the antebellum
era (see Appendix).
Issues of representativeness also arise from the advanced age of respondents
and the possibility of different mortality rates among respondents from different
status backgrounds under slavery. Demographers have identified two primary
correlatesof mortality for slaves:slave occupation (e.g., domestic versus field hand)
and plantation ecology (type of crop grown) (Fogel 1989). Controlling for crop
type, slaves engaged in domestic and skilled manual labor have been found to have
mortality rates less than half that of field hands of the same age. To account for this
source of differential mortality, we applied a second set of weights based on the
mix of slave occupations in the late antebellum South (using Olson's [1992] sample
of plantation records). We found that unskilled agricultural workers in our data
set tended to be undersampled,whereasdomestics were oversampled(see also Escott
1979). The other characteristic affecting slave mortality - plantation ecology
is highly correlated with state of origin and accommodated by the existing sample
corrections for geographic bias.
RespondentAge
The advanced age of respondents in the WPA archives has been noted as a
shortcoming in using these materials. The median age of respondents in our
subsample is 83, leading to questions concerning accuracy of recall. WPA
interviewers often undertook prior research to establish basic biographical details
on respondents (such as age, family background, and the like). To improve
reliability,many WPA interviewers were also instructed to visit respondents on at
least two occasions, the second visit to "gatherall the worthwhile recollections that
the first talk has aroused" (Alsberg 1937). These multiple interviews reveal
problems of recallrelatedprimarilyto complex chronological sequences (e.g., order
of masters under the antebellum regime) rather than to the more basic
socioeconomic variables employed in our analyses. Moreover, interviews tended
to be conducted around major life-cycle markers (e.g., emancipation, marriage),
an approach that tends to improve the performance of long-term memory (Escott
1979; Hurwicz et al. 1992).
With respect to interview questions about antebellum status, an issue also arises
concerning the amount of time that has passed since slavery.In particular, those
sampled blacks who were very young at the time of emancipation - one-third of
our sample was younger than nine in 1865 - may view the slave regime through
the eyes of childhood (Blassingame 1975; Woodward 1974). To account for such
effects, we control for age in our models and also run separate models for
respondents who were preteens during emancipation.
Legaciesof AmericanSlavery/ 459
STATUSATTAINMENTMEASURES
Our rankingsof status attainmentaim to captureinstitutionalparticularities
of the antebellumand postbellumperiodswhile also permittingcomparability
with contemporaryrankingsof occupationalprestige.Using the class schema
of Erikson and Goldthorpe (1992) as a guide, detailed occupational
descriptionsof respondentswere mapped onto seven statusdistinctionsunder
the antebellum regime and nine status distinctions under the postbellum
regime (see Table 1).15Examining the postbellum rankings,two departures
from contemporarystatushierarchiescan be noted. First,low-skillnonmanual
occupations (classVIII) are considered to be of relativelyhigh status. As in
today's developing countries (see Ganzeboom, de Graaf& Treiman 1992),
clerical and other lower-end, nonmanual occupations in the late nineteenth
century had not yet been deskilledby technology.Thus, white-collarworkers
enjoyed a high degree of autonomy and developed unique skills that were
difficultto replace.Second, there is considerabledifferentiationin the ranking
of agriculturaloccupations,particularlybetweenindependentfarmproprietors
(classV), sharecroppersand rental farmers(placed in classIII), and common
farmlaborers(classI). Thisdifferentiationreflectsthe autonomyofferedby farm
ownershipduring the postbellumera (Foner 1989; Schultz 1998), as opposed
to the exploitive characterof sharecroppingand the similarityof supervised
agriculturallaborto fieldworkunderthe slaveregime.16
Considering postbellum occupational attainment and the longevity of
respondents,an obvious question is whetherthe number of elite occupational
statusesin the sample (especiallyprofessionalsand officials)is representativeof
the blackpopulationas a whole.Becauseof the time periodcovered(1865-1930s),
no strict basis of comparisonis available.However,analysesof census records
around the middle of the period provide indications of whether substantial
bias exists in the sample. In 1890, when the U.S. Bureau of the Census first
distinguishedbetweenwhite and blacklaborers,the top ranksof socioeconomic
status - professionals and officials- comprised merely 1.1% of the black
workingpopulation (U.S. Bureauof the Census 1890). The percentagein our
weightedsampleis actuallyslightlylower (1.0%),with no significantdeparture
from the populationparameterunder randomsampling.Consequently,it does
not appearthat sample selection- in its more extreme forms- occurs on
the dependent variable.
460/SocialForces 82:2,December2003
TABLE1: Occupational Status Attainment of Sampled Blacks, during and after
Slavery
Status
Pre-1865
Unweighted
Weighteda
(Percent)
(Percent)
Slave
I. Unskilledagricultural
II. Unskilled
manual/domestic
III. Semiskilled
agricultural
IV. Semiskilled
manual
V. Skilleddomestic
VI. Skilledmanual(artisan)/driver
22.8
10.9
2.3
2.8
22.9
1.6
36.7
7.4
1.7
3.6
6.7
6.6
Freeorchildofslaves
VII. Freeblack
Childof slaves(no occupation)
2.1
34.6
5.4
31.9
(N= 1,400b)
INDEPENDENT AND CONTROL MEASURES
Education
While a small proportion of our weighted sample (.9%) had formal education
under the antebellum regime, far more received schooling during the
postbellum period - including primary education (39.3%), vocational school
training (.8%), and college education (4.2%). During both periods, some
respondents began formal education but had to stop prematurely because of
personal or social circumstances (8.1%). This was especially true in the
antebellum regime, where masters might vacillate in their opinions of educating
slaves or the seasonal demands of plantation labor could limit schooling to
certain times of the year. Finally, many respondents received informal
education (outside a school environment) from literate blacks or whites who had
no teaching credentials (19.4%).
Migration
Patterns of migration reflect respondents' degree of embeddedness in
communities that formerly supported slavery, as well as ongoing economic
dependencies on white elites in those areas. We code characteristics of
migration for two time periods, analyzing geographic mobility - both
voluntary and coerced - during the antebellum regime and period of
emancipation (through the mid-1860s) and during the postemancipation
regime (after the mid-1860s). Three types of migration are considered:
Legaciesof AmericanSlavery/ 461
TABLE1: Occupational Status Attainment of Sampled Blacks, during and after
Slavery (Continued)
Post-1865
Status
I.
II.
III.
IV.
V.
VI.
Unskilledagricultural
Unskilledmanual/domestic
Semiskilledagricultural
Semiskilledmanual/skilleddomestic
Proprietorfarmer
Skilledmanual/supervisorof unskilled
or semiskilledworkers
VII. Supervisorof skilledmanualworkers/
smallproprietor
VIII.Low-gradeprofessional/nonmanual
worker
IX. Official/high-grade
professional/
proprietor
Unweighted
(Percent)
Weighted
(Percent)
Threshold
12.1
18.5
29.5
21.1
3.1
12.0
16.9
29.4
19.0
3.3
.000
1.141
1.991
2.564
2.701
7.8
9.4
3.184
1.2
1.5
3.290
6.0
7.5
4.446
.7
1.0
n/a
(a)d
(N = 1,392e)
a
Weightsarebasedon the geographicdistribution(1860 census)and occupationaldistribution
(Olson 1992sample) of southernblacks.
b
Seventy-onecases contain incomplete data on statuswithin the antebellumregime (coded as
missing).
c
Percentagesarebasedon a pooled sampleof alljobs held by respondents(2,298 job positions).
d
Figuresare upper-thresholdestimates for an ordered Probit model of highest occupational
attainment;see Table6 (model 4) and equation 2.
e
Eightadditionalcasescontain incomplete dataon statuswithin the postbellumregime (coded
as missing).
interstatemigration;migrationoutside the bounds of the (former)Confederate
states, including movement into slaveholding border states maintaining
neutralityduring the Civil War;and an interactioneffect for the migrationof
blackswho had been free prior to 1865 into non-Confederatestates.The latter
variableis used to explorethe contentionby some earlyscholars(e.g., Douglass
1879) that the lack of a "criticalmass"of freeblacksin statesoutside the South
might have complicatedstatus attainmentand resourcemobilization efforts;
migrationcouldthushaveimposeda relativeliabilityfor thoseblackswho did not
carrythe onus of a formerslavestatus.
462Social Forces 82:2,December2003
Control Variables
Analysesinclude controls for the age and gender of respondents.Exploratory
plots suggest that socioeconomic attainmentcan be characterizedas a linear
function of age, possibly capturing unobserved human capital effects. The
control for genderreflectsdifferencesin nineteenth-centurymobility patterns,
since women were usually favored for domestic work and barred from
manufacturing jobs and other types of strenuous nonagricultural work
(Schwalm1997;Williamson1984).Childbearingand familyresponsibilitiesalso
impededstatusmobilityamongemancipatedblackwomen.
Missing Data
For education,a conditionalmean imputationprocedurewas used to replace
missingvalues and ensurethat a maximumnumberof cases could be retained
(Little1992).Using OLSregression,conditionalmeansarecalculatedas
XT= E(X,)
..., Xp
( 1)
where Xi represents the missing values. Weights are assigned to cases with
imputed education values in order to compensate for increased residual
variance(Little 1992). Caseswith missing values on any of the other variables
were removed by listwise deletion. This reducedthe total number of cases to
1,392. Descriptivestatisticsare reportedin Table2.
STATISTICALMETHODOLOGY
For purposesof analysis,the postbellumstatus attainmentof respondentswas
assessedfortheirinitialoccupationafteremancipationandfortheirhighest-prestige
occupationthereafter.Sincethese dependentvariablesaremeasuredon a ranked
scale, multivariatemethods for ordinal variableswere applied.We estimated
Zavoinaand McElvey's(1975) ordered Probit model based on the following
specification,stated in terms of continuous latent measuresof the dependent
variables(Y*):
<
Y*= Xp + with Y= 0 if Y*?go
1 if Po< Y* <I.1 ..
(2)
J if Y*> P 1
where Y is the observed counterpart to Y* and the p's are free threshold
parametersthat distinguish ordered values. Maximum-likelihood estimates
were derivedusing Greene's(1996) LIMDEPsoftware,with the constraintthat
the first threshold parameter (gto) equal zero.
Legaciesof AmericanSlavery/ 463
TABLE2: Descriptive Statistics for Independent and Control Variables Used in
the Analysis: Weighted Sample
Variable
Demographics
Gender(1 = male)
Age(in 1865)
Migration
Interstate
migration
mid-1860s)
(through
Interstate
migration
(aftermid-1860s)
outsideConfederacy
Migration
(allperiods)
Education
Formaleducation(pre-1865)
Formaleducation(post-1865)
Abbreviated
education(allperiods)
Informaleducation(allperiods)
(N = 1,392afterlistwisedeletion)
Proportion/Mean
.71
13.72
S.D.
9.08
.20
.49
.18
.01
.43
.08
.19
Note:Weightsarebasedon the occupationaland geographicdistributionof southernblacks
(1860).
Results
Table3 providesa descriptivecross-tabulationof antebellumstatusand highest
achievedpostbellumstatusamongthe sampledrespondents.The diagonalof the
mobilitytablesuggestsconsiderablestatuspersistenceamongemancipatedblacks,
with each antebellumstatusgenerallybeing linked to disproportionateodds of
representationwithin a comparablepostbellum status (assuming a model of
statisticalindependence).17
Bythe sametoken,upwardmobilityinto the ranksof
professionalandnonmanualworkerstendsto be relativelyrareforblacksformerly
employedas slavefieldhands,domestics,or semiskilledlaborers,whiledownward
mobility into the ranksof unskilledand semiskilledwage laborersis typically
avoidedby formerslaveartisansand by blackswho were free in the antebellum
South.All of this suggestssome empiricalsupportfor a structuralmodel of status
persistenceacrossinstitutionalregimes.
Preliminaryconclusions in this regardmust be temperedby a number of
caveats.First,the destinationstates consideredin the mobility table combine
occupationsheld immediatelyafterthe CivilWar(often under formermasters)
with lateroccupationaloutcomes.Thistendsto conflateshort-termstatusstability
464/SocialForces 82:2,December2003
TABLE3: Status Mobility of Sampled Blacks between Antebellum and
Postbellum Regimes
Antebellum Weighted
Status
Number
Postbellum
Status(HighestAchieved)
I
II
III
IV
V
VI
52
VII
VIII
IX
Slave
I
509
17a
49
197
101
29
20a
43
b
II
104
1b
18a
38
18
5
6b
2
16
0b
III
22
2
10a
4
3a
2
0
Ob
b
0
IV
50
Ob
7
5b
27a
2
5
0b
4
V
93
1b
13
15b
34a
3b
16
3a
7b
1
VI
92
Ob
0b
19
6
44a
Ob
1ga
3a
78
0b
8
10b
0b
21a
0b
21a
8a
33b
3b
47
6
1b
0
Freeorchild
ofslaves
VII
Childof slaves
(no occupation)
445
11
67
10b
148
111
19
(N = 1,392c)
Note:Foran explanationof occupationalstatusrankings,see Table1.Notations relatedto counts
are limited to cells with an expected count of at least one observation.
a Observedcount exceeds
expectedcount by a ratio of at least 1.4:1.
b
Expectedcount exceeds observedcount by a ratioof at least 1.4:1.
c Row totalsneed not sum to numberof casesbecauseof
roundingassociatedwith caseweights.
with what may be long-term fragmentation of status structures. Second, the
pattern is complicated by the appearance of new classes during the postbellum
period, such as independent farmers and a black petty bourgeoisie. For
instance, low-level supervisors and proprietors (class VII) are drawn
disproportionately from entrepreneurial blacks who were once unskilled field
workers. Third, the simple cross-tabulation does not address other factors
affecting mobility patterns during the postbellum period, including education,
migration, and gender.
Table 4 shows a series of multivariate models predicting the prestige of the
first job held by sampled blacks after emancipation. Status under slavery has
statistically significant effects on status attainment for all of the specifications.
Those blacks who were free in the antebellum South had substantially higher
status outcomes in the postbellum period than those who had been slaves. Since
the magnitude of the ordered Probit estimate does not lend itself to direct
interpretation, we plot the relative probability of different status outcomes,
holding all other explanatory variables at their sampling means (see Figure 2).
The chart shows that former slaves tend to be overrepresented in unskilled
Legaciesof AmericanSlavery/ 465
TABLE4: Ordered Probit Models Predicting Prestige of Initial Occupation
(post-1865) for Blacks Formerly Residing in U.S. Slaveholding
States
Variable
Model1
Model2
Model3
Model4
Intercepta
.135
(.073)
-.038
(.082)
-.075
(.084)
-.099
(.086)
.395***
(.051)
.026***
(.003)
.428***
(.051)
.023***
(.003)
.402***
(.052)
.022***
(.003)
.394***
(.052)
.022***
(.003)
.358***
(.068)
.849***
(.074)
.569***
(.072)
1.104***
(.074)
.581**
(.075)
1.121***
(.079)
.594***
(.075)
1.113***
(.080)
1.058***
1.072***
(.088)
.740***
(.178)
.250*
(.136)
.236*
(.130)
.112*
(.054)
.248***
(.052)
.028
(.186)
1.049***
(.089)
.738***
(.178)
.224
(.137)
.252*
(.131)
.107*
(.054)
.253***
(.053)
.031
(.186)
Demographics
Gender (1 = male)
Age (in 1865)
Statusunderslavery
Childof slaves
Freeblack
Slaveb
Skilledmanual
(.083)
Skilleddomestic
.747***
(.175)
Semiskilledmanual/
agricultural
Unskilled
.215
(.136)
.219*
(.131)
nonagricultural
Communityembeddedness
Migration (out of state)
Migration (to nonConfederatestate)
Freeblacksx migration
from Confederacy
Educationc
Formaleducation
.087
(.512)
.139*
(.075)
Informalor abbreviated
education
-2 log likelihood
Degreesof freedom
4929.62
5
4826.30
9
4811.83
12
4810.34
14
(N= 1,392)
a
Inclusion of interceptrequiresthat t0be constrainedto zero.
b Slaves
workingas fieldhands(unskilledagriculturalworkers)arethe omittedreferencecategory.
c Refersto education receivedbefore
emancipation;"no education"is the omitted reference
category.
* < .05 ** < .01 *** < .001
(one-tailedtests for hypothesizedeffects;two-tailedtests othp
p
p
erwise)
466/SocialForces 82:2,December2003
FIGURE 2: Predicted Probability of Initial Occupational Status (post-1865) for
Free Blacks and Former Slaves
.4
.3
--
.- -_-
0.0
-
|
I
II
III
i
IV
i
V
1
VI
i
VII
Former
Slaves
VIII
IX
InitialOccupationalStatus
Note:Predictionsarebasedon Table4, model 1. Foran explanationof occupationalstatus
seeTable1.
rankings,
occupations and semiskilled agriculture, while blacks who were free in the
antebellum period are overrepresented in all higher-ranking occupations. This
finding reflects several advantages on the part of free blacks. They did not face
the public identity adjustment that former slaves underwent during
emancipation; in other words, free blacks had already established the
autonomous self they presented to employers, authorities, and other members
of society. Moreover, free blacks were much more likely to have accumulated
wealth and belong to communities with relatively powerful social support
infrastructures (Horton & Horton 2001).
Children of slaves, who had not yet been assigned an occupation under the
slave regime, also enjoyed occupational advantages over slaves who were employed on the plantation or in urban slavery. Children were not exempt from
socialization under slavery. They were conditioned to avoid drawing attention
to themselves by both whites and parents concerned with their safety (King
1995). But our results suggest that submissive identities presented to planters
did not have a strong effect on the children's later status attainment, at least
Legaciesof AmericanSlavery/ 467
comparedto the effectsof enculturationon adults.The childhood effect holds
independently of a respondent's chronological age, suggesting that
enculturation in slave labor routines per se had adverse consequences that
could persistwell beyond the antebellumregime.These consequencesinclude
lowered expectationsfor fair labor contractsand submissivework habits (e.g.,
reluctanceto makesuggestionsor negotiatewith powerfulothers).
Status outcomes immediatelyafter emancipationwere also a function of
position within the occupationalhierarchyof slavery(model 2; likelihoodratio
x2 = 103.32,p < .001,vs. model 1). Slaveartisansand craftworkers,in particular,
enjoyed high-status outcomes than the majority of slaves employed as field
laborers.Since slave artisanswere often hired out for wages before 1865, their
more favorablestatusattainmentunder slaveryreflectstheir relativeautonomy
and experience with the marketplace as well as their technical abilities
(Genovese1974). Status-persistenceeffectsbeyond the antebellumregimewere
also evident for household servants and those in other slave occupations,
reflecting an ordered hierarchy that differentiated the statuses of these
occupationsfromthat of unskilledagriculturallabor.
Some forms of black migrationservedto disrupt status-persistenceeffects
(model3). Interstatemigrationthroughthe mid-1860simprovedstatusattainment
outcomesslightly,while exodusfromthe formerConfederatestatesofferedmore
promiseto formerslaves.Both resultscan be attributedto the damagingrole of
communityembeddednessin reproducinginstitutionalarrangementsand their
statushierarchies(e.g.,ongoingeconomicdependenceon formermasters).Notably,
there is no significantdifferencebetween the effects of emigrationon former
slaves and on free blacks.
To ensuretemporalprecedence,we consideredthe effectof educationunder
the antebellumregime on the first job of respondentsafter emancipation.As
shown in model 4, formal and informal education within the planter regime
providedonly limited durablebenefit in terms of status attainmentoutcomes
(likelihoodratio x2= 1.49, nonsignificant,vs. model 3). This could resultfrom
the sporadicnature of such educationor the fact that socializationwithin the
older institutional arrangementdid not offer the adaptabilityneeded after
emancipation.One potential concern with this interpretationof educational
efficacyis that some sampledblackswere relativelyyoung during the antebellum regime. Given that the averagerespondent was less than fourteen years
old in 1865, it is reasonable to expect that a substantial proportion of the
sample would not yet have reached an age at which schooling might be appropriate.To explorethis issue, we split the respondentsinto two subsamples,
one comprisingblackswho were ten years old or younger in 1865 (N= 653)
and one comprisingthose who were older than ten (N = 739). The corresponding models for initial occupationalattainmentduring the postbellum period
are shown in Table5.
468/SocialForces 82:2,December2003
The influence of antebellum status and demographic factors is largely
consistentacrossthe two subsamples.Underthe slaveregime,blackchildrenwere
often expectedto take on adultwork tasksat an earlyage (Fogel 1989), and so
it is unsurprising that status distinctions between young field workers,
domestics, and free blacks would mirror those of their adult counterparts.18
With respectto migrationand education,two notable differencesare evident
between the subsamples. First, interstate migration through the mid-1860s
tendedto havebeneficialstatusimplicationsfor youngerWPArespondentsbut
not for older ones. In particular,this finding appearsto reflect the stronger
embeddedness of adolescent and adult ex-slaves in local communities and
social networks. Second, informal and periodic education during the
antebellumregime proves effectivefor those blackswho were ten yearsold or
youngerin 1865, but it is not effectivefor older blacks.For the lattergroup, it
is possible that the benefits of such education - often offered under
clandestinecircumstances- were counteractedin the late antebellumperiod
by the activehostilityof manyplantersand by effortsat "reeducation."
Examiningthe highest subsequentoccupationalattainmentof respondents
duringthe postbellumperiod(Table6), severalimportantdifferencesfrominitial
occupationalattainmentcan be noted. Whilethe statusdistinctionbetweenfree
blacksand slavespriorto 1865continuesto havetellingeffects,statuspersistence
Formerslave
basedon the occupationalslavehierarchyhasweakenedconsiderably.
artisansand craftworkers,whose skillsappearto generalizeacrossinstitutional
arrangements,haveclearstatusadvantages(p < .001), but otherhigh-statusslave
occupations(e.g.,householdservice)offerlittledurableadvantageovercommon
field labor.Fordomesticand semiskilledworkers,the institutionalreproduction
of the slave regime appearedto be limited by the fact that many of its status
distinctionswere rooted in particularistictraits(skin tone, deferencebehavior)
that became less salient after emancipation (Stampp 1956).
The other notablefactoraffectingupwardmobilityover the life course,but
not necessarilyinitial occupational attainment, is education. As emphasized
by DuBois and Washington,schooling - whether rooted in formal college
education, vocational training, or informal apprenticeship - was the
institutionalinterventionthat could most benefit southernblacksin the long
run (see model 4). To be truly effective, however, such schooling had to be
decoupled from the older educationalmechanisms of the antebellum South
(cf. Table4, model 4), which were ad hoc at best and often rooted in the
patriarchal ideology of the planters. Although blacks faced barriers to
professional employment throughout the period of the study, education
broughta host of cumulativeadvantagesover the life course,includinggreatly
increasedaccessto academicand vocationalknowledgeas well as a much lower
likelihood of victimization in labor contracts.Surprisingly,both formal and
informal education affect status attainmentwith about the same intensity in
model 4. A plausibleexplanationfor this equivalenceis that formal education
Legaciesof AmericanSlavery/ 469
TABLE5: Models Predicting Prestige of Initial Occupation (post-1865) for
Separate Subsamples of Younger and Older Southern Blacks
10YearsOld or
Youngerin 1865
Intercepta
Demographics
Gender(1 = male)
Age (in 1865)
Statusunderslavery
Childof slaves
Freeblack
-.300
(.157)
-.163
(.147)
.488***
(.078)
.035**
(.013)
.392***
(.075)
.025***
(.005)
.731***
(.114)
1.328***
(.160)
.571**
(.193)
1.005***
(.121)
1.076***
(.302)
.652
(.548)
.285
(.197)
.868***
(.096)
.661**
(.230)
.179
(.170)
.313
(.191)
.318**
(.104)
.365***
(.097)
-.120
(.218)
-.022
(.073)
.176*
(.078)
.203
(.617)
-.026
(.476)
.446**
(.151)
.233
(3.819)
.029
(.101)
Slaveb
Skilledmanual
Skilleddomestic
Semiskilledmanual/
agricultural
Unskillednonagricultural
Communityembeddedness
Migration(out-of-state)
Migration(to non-Confederate
state)
Freeblacksx migrationfrom
Confederacy
Educationc
Formaleducation
Informalor abbreviated
education
OlderThan
10in 1865
2186.92
2558.51
-2 Loglikelihood
14
13
Degreesof freedom
739
653
N
a Inclusionof
interceptrequiresthat mo be constrainedto zero.
b Slaves
working as field hands (unskilledagriculturalworkers)are the omitted referencecategory.
c Refersto
highestlevel of education:"no education"is the omitted referencecategory.
*p .05 **p< .01 ***p<.001
470/SocialForces 82:2,December2003
TABLE6: Ordered Probit Models Predicting Highest Prestige of Subsequent
Occupations (post-1865) for Blacks Formerly Residing in U.S.
Slaveholding States
Variable
Intercepta
Demographics
Gender(1 = male)
Age (in 1865)
Statusunderslavery
Childofslaves
Freeblack
Slaveb
Skilledmanual
Model 1
Model 2
Model 3
Model 4
1.341
(.092)
1.322
(.100)
1.289
(.104)
.928
(.117)
.465***
(.055)
.017***
(.003)
.446***
(.056)
.014***
(.003)
.428***
(.057)
.013***
(.004)
.398***
(.057)
.018***
(.004)
.111*
(.064)
.827***
(.077)
.182**
(.070)
.940***
(.080)
.191**
(.070)
.986***
(.083)
.188**
(.071)
.903***
(.084)
(.085)
.161
(.155)
-.075
(.107)
.153
(.112)
.782***
(.088)
.155
(.155)
-.040
(.108)
.166
(.111)
.648***
(.090)
.164
(.154)
-.005
(.102)
.180
(.115)
.029
(.047)
.173**
(.057)
-.195
(.559)
.042
(.046)
.114*
(.057)
-.224
(.564)
-.781***
Skilleddomestic
-
Semiskilledmanual/
agricultural
Unskilled
nonagricultural
Communityembeddedness
Migration(out-of-state)
Migration(to nonConfederatestate)
Freeblacksx migration
from Confederacy
Educationc
Formaleducation
-.473***
(.086)
Informalor abbreviated
education
-2 Loglikelihood
Degreesof freedom
-.502***
(.087)
5045.58
5
4999.02
9
4992.98
12
4970.92
14
(N= 1,392)
a
Inclusion of interceptrequiresthat P0be constrainedto zero.
b Slaves
workingas fieldhands(unskilledagriculturalworkers)arethe omittedreferencecategory.
c Refersto
highestlevel of education;"noeducation"is the omitted referencecategory.
*
p < .05
erwise)
**
p < .01
***
p < .001(one-tailedtestsfor hypothesizedeffects;two-tailedtestsoth-
Legaciesof AmericanSlavery/ 471
was of highly variable quality during the period of our study (Anderson 1988).
Black public schools ranged from Hampton-model institutions - designed to
produce more productive agricultural laborers to black-run schools that were
oriented toward academic achievement but often suffered severe resource
shortages.
Discussion
For the last generation of blacks born under American slavery, we have shown
that the legacy of this institution evidenced both persistence and disintegration
in the decades after emancipation. Status distinctions between free blacks and
slaves under the antebellum regime continued to influence occupational
attainment in the postbellum era. Slaves who were emancipated after reaching
adulthood evidenced socioeconomic scars for years to come in contrast with
children who only glimpsed the conditions of the slavery regime. But other
features of the slave regime disintegrated over time. Except for those who
learned skills in manual trades, occupational status within the system of slavery
carried few durable occupational advantages or disadvantages in the late
nineteenth century.
Mediating factors, such as migration and education, also served to disrupt
institutional reproduction after 1865. Emigration was consistently defamed by
southern elites as a move away from home to dangerous territory, outside the
"protective"arms of paternalism (Horton & Horton 2001). But former slaves who
confronted disenfranchisement and economic exploitation in the South often
found new opportunities outside the former Confederacy. At the same time,
there was often an absence of benefits for those migrating within the
Confederacy, particularly for black adults. This could be accounted for by a
number of contextual factors. By the early 1880s, blacks encountered
antivagrancy laws throughout much of the South. New arrivals to a formerly
Confederate town could expect to be jailed or forced into penitentiary labor
if they failed to immediately find work (Cohen 1991; Williamson 1984).
Moreover, blacks who migrated within the South were often those who were
most harshly persecuted; they often had to leave home in a hurry, regardless
of employment prospects at their destinations (Gutman 1976; Litwack 1979).
Educational opportunities during Reconstruction helped to substantially
offset the forced illiteracy of the slavery regime. The success of such education
in opening new economic opportunities for freed blacks was all the more
impressive considering that formal schooling in the antebellum period had
been largely ineffective. Our analysis thus suggests that substantial barriers to
status mobility among blacks were produced by the legacies of slavery, but it
does not support the contention that mobility processes were equivalent across
the two regimes. Rather, our findings imply that mobility among freedmen and
472/SocialForces82:2,December2003
freedwomendepended on many of the same factorsdeterminingmobility in
contemporaryAmerica.Freedmenand freedwomenneeded opportunitiesfor
education,the resourcesto move to areasdemandingskilledlabor,information
about job opportunitiesthey could reasonablytrust, a childhood free of hard
laborand physicalabuse,and, ideally,work experiencein skilledcrafts.
Conclusion
With respectto our canonicalexplanationsof institutionaltransformationand
social status (see Figure1), our results suggest qualifiedsupport for all three.
In the immediate aftermathof the antebellum regime, status structuresevidenced recalcitranceand developed consistent mappings between categories
of inequalityunder slaveryand freedom.This substantiatesan account of durableinequalityacrossthese regimes (Tilly 1998). Mediatingcharacteristicssuch as labormobilityand investmentin human capital- also exercisedsome
influence on initial status attainmentamong the emergingpopulation of free
blacks.Consistentwith Marxistaccounts, these featuresof the means of production within free wage capitalismbecame centralinfluences on the mobility regime over time. Simultaneously,institutionalfragmentationcontributed
to loose couplingof socialstatusacrossthe regimesin the long run.While some
detailsof this processare particularto the deinstitutionalizationof slavery,the
generalpattern is one familiarto neoinstitutionalscholars.When regulatory
changeslead to formaldismantlingof one institutionalregime,the beliefs and
norms that supportedthat regime often linger,contributingto conflictingexpectationsand fragmentationof authoritysystems(Meyer& Scott 1983). Such
patterns of "partial"reproductionwere reflected in the Reconstructionand
Restoration eras during the latter half of the nineteenth century (DuBois
1935).
Naturally,understandingthe institutionallegacy of the antebellumsystem,
as well as the legacy of American slavery for individual blacks, requires
consideration of additional levels of analysis. Organizationaland political
dynamics,in particular,merit consideration.Whatcooperativeorganizations
schools, churches,banks,newspapers,and the like- were createdin the black
community to break patterns of status persistence (DuBois 1907)? To what
extent were these organizations supported by federal interventions during
Whenwereorganizedoppositionaltacticsemployedby southern
Reconstruction?
white elites to counter these efforts?What specific effects did Ku Klux Klan
oppression have on black status attainment in the postbellum period
(Hadden 2001)? These questions of institutional reproductionat the macro
level require careful attention to the ecology of social movements and
organizationsthat filledthe niche vacatedby the dismantlingof the antebellum
regime.
Legaciesof AmericanSlavery/ 473
Even at the individuallevel, our findingsmust be regardedas preliminary.
We have consideredintragenerationalpatternsof status persistence,although
many of the more telling legacies of slaverywere played out in the first generation of southern blacks born without the onus of forced plantation labor
(see Darity, Dietrich & Guilkey 2001 for one long-term study of
intergenerationalpersistence).Our analysesmay also benefit from an enriched
conception of black identity during the postbellum era, which has been limited to considerationsof status change. Lengthiernarrativespublishedby exslavesand their kin in the late nineteenthand earlytwentiethcenturiespresent
more nuancedviews on slavery,freedom,and racerelations(Blassingame1975).
Despite these caveats,our study offers severalgeneralinsights for students
of stratificationand social institutions. One is that the tools of stratification
researchcan - and should - be applied to even the most extreme cases in
the scale of socioeconomic status. All too often, researchers ignore those
individualslocated in the bottom rungs of socioeconomicattainment,because
they seem difficultto classifyin termsof standardoccupationalscales.However,
models of mobilitymust take these individualsinto account,particularlywhen
they constitutea populationas formidableas the 4 million AfricanAmericans
who were emancipated in 1865. Our finding that intragenerationalstatus
mobility was limited in the South afterthe Civil Waralso has implicationsfor
intergenerationalmobility. Leavingthe South altogetherwas an easier choice
for the first generationborn in freedom than for the generationof freedmen
and freedwomen,in partbecausehopes of improvingracerelationsin the South
had been destroyedtoo often duringthe Jim Crowperiod (Litwack1998). The
portrayalof freedpersonstatus mobility in this article suggests that the first
generation born free witnessed few opportunity structures shaping their
parents'careers.19
Furthermore,our studyshowsthat mobilitymodelsshould incorporatelegal
statuses, such as slave status, since these qualities have long-term effects on
occupationalattainment.Individualswho were freeat the time of emancipation
fared much better throughouttheir lifetime. It remainsunclearwhether such
differencesweregeneratedby the improvedcapacityof freeblacksto accumulate
resourcesin the antebellumperiod,the difficultyencounteredby formerslavesin
constructing a new identity, unobserved variation in skin tone (i.e., color
stratification),or the greaterlikelihoodthat free blackshad a white or mulatto
parent.20
Revisingconventionalmodelsof mobilityin suchcasesrequiresattention
to historicaldetail and opens a dialoguebetweenstudies of social stratification
and institutions.
Institutionalanalysisalsobenefitsfromsuchdialogue,sinceoppressedgroups
can serveas a crucialimpetusto systemicreform,as Marxpointed out long ago.
Whenstatuspersistencedisplaysa qualitativebreakbetweeninstitutionalregimes,
it can lay a foundation for further institutional change. Conversely,status
persistence across regimes - accompanied by status-consistent identity
474/SocialForces 82:2,December2003
valuation - is indicative of cosmetic institutional change, rather than
fundamental reconstruction. In this regard, the issue of institutional persistence
is not merely of historical interest, since the efficacy of more recent civil rights
movements may be viewed in a similar light. As Tilly (1998) emphasizes, mere
attitudinal changes toward discrimination are typically insufficient to alter
structured patterns of inequality. Our analysis of postbellum status attainment
suggests that investments in human capital (education, moving persons to
match work contexts) and fundamental reorganization of workplaces
(particularly their categorical distinctions) may generate more effective
interventions.
Notes
1. Lincoln'searlier,and more famous, EmancipationProclamationof 1863 was issued
under his war powers and has been questionedfrom the standpointof constitutional
validity.
2. The distinctionwas especiallyambiguousin those states- Alabama,Florida,Georgia,
Mississippi, North Carolina, and South Carolina- that enacted laws supporting
peonage, in which ex-slaveswhere legallybound to serve a creditoruntil a debt was
paid.Althoughsuchlawswereunconstitutionalfromthe standpointof federalauthorities,
they remaineda fact of life for many formerslavesin the South.
3. Wedo not meanto implythatantebellumslavepatrolsand postbellumwhitepatrollers
wereidentical.Slavepatrolswereinstrumentsof the lawbefore1865,whereasnightriders
and other race terroristsoperatingafter 1865 were criminals.Rather,we suggestthat
racistterroristgroupsandvigilantegroupsusedtacticsdevelopedby and drewinspiration
from the slave patrols(see Hadden2001 for a reviewof this continuity).
we do not intend to equatethese perspectives
4. In using the term structural-functional,
on institutional reproduction with the familiar Davis and Moore (1945) theory of
inequality.Rather,the accounts are structural-functionalinsofar as they assume that
organizationsadoptingcategoricalinequalitiesfrom formerinstitutionalarrangements
continue to be favoredin new institutionalcontexts (Wright1998).
5. Thesecharacteristics
can be describedmore abstractlyas controloverworklocaleand
control over work time. From a Marxistperspective,a definingelementof precapitalist
means of productionis the absenceof one (or both) of these elementsamong exploited
classes. Thus, slaves lack either form of control in the work process, while serfs lack
control over work locale (at least, when seen as ideal types). Nominally free wage
laborers,by contrast,exercisesomecontrol on both dimensions.
6. The Freedmen'sBureau,establishedto easethe transitionof blacksafteremancipation,
reportedthe developmentof 740 black schools in the formerslave statesin 1866, with
90,589 students; by 1870, those numbers had grown to 4,239 schools under the
supervisionof the bureau,educatingsome 247,333 students (DuBois 1935).
Legaciesof AmericanSlavery/ 475
7. The degreeof paternalism's
degenerationis a topicof debate.Ochiltree(1998)questions
the outright demise of paternalismbut agrees with the prevailingconclusion in the
literaturethat paternalismdeclineddramaticallyafter 1865.
8. Cohen (1991) arguesthat sharecroppingwas forced on plantersby a freed working
class that refused to participate in fieldwork that reminded them of slavery. The
sharecropping system was an improvement over slavery in that it gave workers
substantiallygreaterfreedom over day-to-daywork in the field and did not share (at
least in theory) slavery'spatternsof regulationthroughphysicalpunishment.
9. Our criterion for identifying states with substantial slaveholding activity is the
enumerationof at least 2,000 slavesin the 1860U.S. census.Statesmeetingthis criterion
are Alabama,Arkansas,Florida,Georgia,Kentucky,Louisiana,Maryland,Mississippi,
Missouri,North Carolina,South Carolina,Tennessee,Texas,and Virginia,as well as the
Indian Territory(Oklahoma).Excludedfrom analysisare New Jersey(which reported
18 "indenturedservants"in the 1860 census), Delaware,Kansas,and Nebraska.
10. Physiologicalfactors,thought by plantersto correlatewith slaves'physicalability,
figuredheavilyin fieldlaborand skilledlaborassignments.An elaboratesystemof"hand"
ratings among planters viewed height, strength, age, and gender as primary
considerations(Fogel 1989).
11. Therewere other reactions,of course, includingprotestand migration.
12. Migrationimmediatelyafteremancipationcan largelybe attributedto attemptsto
reunify separatedfamilies, searchesfor economic opportunity,and desires to create
distancefrom formerowners (Davis 1993;Foner 1989).
13. The effectsof migrationon statusattainmentamongblackswho had been freewithin
the antebellumregimearemore complex.Theseindividualshad moreextensiveresources
and supportnetworksin their southerncommunitiesthan ex-slavesdid and may well
have sufferedstatuslosses as a resultof migrationto the North or elsewhere(Woodson
1918).
14. Note that all of these presentationspertainto socialidentity- identityas perceived
by strangers- ratherthan personalidentity- as perceivedby friends and intimates
(Goffman 1963). Some ex-slavesmay have shown equivocalattitudestowardfreedom
in public(especiallywhen interactingwith whites),whileimbuingtheirnew identitywith
positivevaluationin private.
15. Even for postbellumstatus,the constructionof prestigerankingsis complicatedby
several factors. One typical approach to describing occupational desirability socioeconomicstatus (SES)scaleconstruction(Blau& Duncan 1967;Wegener1992)requiresinformationon educationand incomethatis generallyunavailablefor the period
consideredhere. Althoughwages were recordedfor industrialworkersin turn-of-thecentury census records, agriculturaland professional earnings cannot be recovered.
Educational experience is recorded in the 1900 census, but the corresponding
"occupational"classificationsin census data are industry-based.Moreover,education
and income may be less robust determinants of occupational prestige during
Reconstructionand the late postbellum period (Katz 1972) than in modern society.
476/SocialForces 82:2,December2003
Despitethesedifferences,correlationsbetweennineteenth-centuryprestigerankingsand
twentieth-centurySESscalestend to be high (Hauser1982).
16. Independentfarmersstill rankbelow skilledmanuallaborersalong dimensionsof
income and advancementpotential.In 1900,the averageblack-ownedfarmearnedless
than half the annual salary of a single carpenter(U.S. Bureauof the Census 1900).
Nonagriculturalworkersalso enjoyed chancesfor promotion to supervisorystatus.
17. Note that the comparablepostbellum status rank for antebellumclassV (skilled
domestics) is classIV,not V (which refersto independentfarmers).
18.The designation"childof slaves"in the tableis formallyindependentof chronological
age (referringto anyrespondentwho hadyet to be assignedworkdutiesin the antebellum
period) and thereforeappliesto both subsamples.
19. Note that the influence of pull factors in the North also played a critical role in
stimulatingmigration- the differencein northernand southernopportunitystructures
createda decisivecontrast(Tolnay1998).
20. The limited evidence available in the WPA archives suggests that one obvious
explanationof advantageamong free blacks- viewing it as a form of social capital
handeddown by those who had white ancestry- is not especiallyviable.In particular,
we find no significantassociationbetween freedomduringthe antebellumperiod and
white parentage.
References
to the StateDirectorsof the FederalWriter'sProject."
Alsberg,Henry.1937."Memorandum
WorkProjectsAdministration,
D.C.,July30.
Washington,
of North
Anderson,JamesD. 1988.TheEducation
of Blacksin theSouth,1860-1935.
University
CarolinaPress.
InstitutionforNegroes,1862John.1965."TheUnionArmyAs an Educational
Blassingame,
1863."Journalof Negro Education34:152-59.
. 1975."UsingtheTestimony
of Ex-Slaves:
andProblems."
Journal
Approaches
ofSouthern
History41:473-92.
Blau,Judith,andEricBrown.2001."DuBoisandDiasporicIdentity:TheVeilandthe Unveiling
Project."
Sociological
Theory19:219-33.
Structure.
Blau,PeterM.,andOtisD. Duncan.1967.TheAmerican
Occupational
Wiley.
Cade,John.1935."Outof the Mouthsof Ex-Slaves."
JournalofNegroHistory20:294-337.
Servitudein the South,1865-1940:A Preliminary
Cohen,William.1976."NegroInvoluntary
Analysis."Journalof SouthernHistory42:31-60.
. 1991.At Freedom'sEdge:BlackMobilityand theSouthernWhiteQuestforRacialControl,
1861-1915.LouisianaStateUniversityPress.
of Slavery,1865-1900."
Journalof AmericanHistory
Daniel,Pete.1979."TheMetamorphosis
66:88-99.
Legaciesof AmericanSlavery/ 477
APPENDIX: Distribution of Sample across Former Slaveholding States and
1860 Census Weights
State
Cases
Case-to-PopulationRatio
CaseWeight
SlavePopulation
Alabama
Arkansas
Florida
Georgia
Kentucky
Louisiana
Maryland
Missouri
Mississippi
North Carolina
Oklahoma(Indian Territory)
South Carolina
Tennessee
Texas
Virginia
102
131
12
159
42
73
5
45
196
93
8
143
96
196
69
1:4,265
1:848
1:5,145
1:2,907
1:5,369
1:4,544
1:17,438
1:2,554
1:2,228
1:3,560
1:313
1:2,814
1:2,872
1:931
1:7,114
1.479
.294
1.784
1.008
1.861
1.576
6.046
.886
.772
1.234
.109
.976
.996
.323
2.467
FreeBlackPopulation
All states (pooled)
30
1:7,699
2.577
Note: Figuresare reportedfor sample of blackswith known social statuswithin the antebellum
regime.
Darity,William,JasonDietrich,and David Guilkey.2001."PersistentAdvantageor Disadvantage?
Evidence in Support of the Intergenerational Drag Hypothesis."American Journal of
Economicsand Sociology60:435-70.
Davis, JackE. 1993. "ChangingPlaces:Slave Movement in the South."Historian 55:657-76.
AmericanSociological
Davis,Kinsley,andWilbertMoore. 1945."SomePrinciplesof Stratification."
Review10:242-49.
Douglass, Frederick.1879. "NegroExodus."AmericanJournalof SocialScience11:1-21.
DuBois, W.E.Burghardt.[1903] 1965. The Souls of BlackFolk.Penguin.
. 1907. "Economic Co-operation among Negro Americans."Proceedingsof the Twelfth
Conferencefor the Studyof Negro Problems.Atlanta University Press.
.1935. BlackReconstructionin America.Russell& Russell.
Dunaway,Wilma. 2003. Slaveryin theAmericanMountain South.CambridgeUniversityPress.
Egypt, Ophelia, J. Masuoka, and Charles Johnson. 1945. Unwritten History of Slavery:
AutobiographicalAccountsof Negro Ex-Slaves.Fisk University Social Science Institute.
Erikson,Robert, and John H. Goldthorpe. 1992. The ConstantFlux:A Studyof ClassMobility
in IndustrialSocieties.Oxford University Press.
478/SocialForces 822, December2003
Escott, Paul. 1979. Slavery Remembered:A Recordof Twentieth-CenturySlave Narratives.
Universityof North CarolinaPress.
Ferleger,Louis A. 1993. "SharecroppingContracts and Mechanizationin the LateNineteenth
CenturySouth."AgriculturalHistory67:31-46.
Fogel,Robert.1989. WithoutConsentor Contract:TheRiseand Fallof AmericanSlavery.Norton.
Foner,Eric. 1989.Reconstruction.Harper& Row.
Frazier,E. Franklin.1939. The NegroFamily in the UnitedStates.Universityof Chicago Press.
Ganzeboom,Harry,Paulde Graaf,and Donald Treiman.1992."AStandardInternationalSocioEconomic Index of Occupational Status."Social ScienceResearch21:1-56.
Genovese, Eugene. 1974. Roll JordanRoll: The Worldthe SlavesMade. Pantheon.
Goffman, Erving. 1963. Stigma:Notes on the Managementof SpoiledIdentity.Prentice-Hall.
Gottlieb, Peter. 1991. "Rethinkingthe Great Migration."Pp. 68-82 in The GreatMigrationin
HistoricalPerspective,edited by J.Trotter.Indiana UniversityPress.
Greene, Lorenzo,and CarterWoodson. 1930. The Negro WageEarner.AMS Press.
Greene,William. 1996. LIMDEPVersion7.0. Bellport, N.Y.:Econometric Software.
Gutman, Herbert. 1976. TheBlackFamilyin Slaveryand Freedom,1750-1925.Random House.
. 1987. "Schools for Freedom: The Post-Emancipation Origins of Afro-American
Education."Pp. 260-97 in Powerand Culture:Essayson theAmericanWorkingClass,edited
by Ira Berlin. Pantheon Books.
Hadden, Sally E. 2001. SlavePatrols.HarvardUniversityPress.
Hartman, Saidiya. 1997. Scenesof Subjection:Terror,Slavery,and Self-Makingin NineteenthCenturyAmerica.OxfordUniversityPress.
Hauser,Robert M. 1982. "OccupationalStatus in the Nineteenth and Twentieth Centuries."
HistoricalMethods 15:111-26.
Horton, JamesOliver,and Lois E. Horton. 2001. Hard Road to Freedom:The Storyof African
America.RutgersUniversityPress.
Hurwicz, M., C. Durham, S. Boyd-Davis,M. Gatz, and V. Bengston. 1992."SalientLife Events
in Three-GenerationFamilies."Journalof Gerontology47:11-13.
Jacobs,Donald (ed.). 1981. Index to the AmericanSlave.GreenwoodPress.
Johnson,Michael P. 1986."Work,Culture,and the SlaveCommunity:SlaveOccupationsin the
Cotton Belt in 1860."LaborHistory27:325-55.
Katz,MichaelB. 1972."OccupationalClassificationin History."
Journalof Interdisciplinary
History
3:63-88.
King, Wilma. 1995. Stolen Childhood:Slave Youthin Nineteenth-CenturyAmerica. Indiana
UniversityPress.
SocialScience
Lantz,Herman. 1980."Familyand Kin as Revealedin the Narrativesof Ex-Slaves."
Quarterly60:667-75.
Little,Roderick.1992."Regressionwith MissingX's:A Review."Journalof theAmericanStatistical
Association87:1227-37.
Litwack,Leon. 1979. Been in the Storm So Long:The Aftermathof Slavery.Knopf.
.1998. Troublein Mind: Black Southernersin the Age of Jim Crow.Knopf.
Legaciesof AmericanSlavery/ 479
Mandle, JayR. 1992. "BlackEconomic Entrapmentafter Emancipationin the United States."
Pp. 69-84 in TheMeaningof Freedom:Economics,Politics,and CultureafterSlavery,edited
by FrankMcGlynn and Seymour Drescher.Universityof PittsburghPress.
Meyer, John. 1994. "The Evolution of Modern Stratification Systems."Pp. 730-37 in Social
Stratificationin SociologicalPerspective,edited by D. Grusky.WestviewPress.
Meyer, John, and Ronald Jepperson. 2000. "The 'Actors'of Modern Society: The Cultural
Construction of Social Agency."SociologicalTheory18:100-120.
Meyer,John,and W. RichardScott. 1983."Centralizationand the LegitimacyProblemsof Local
Government."Pp. 179-98 in OrganizationalEnvironments:Ritual and Rationality,edited
by John Meyer and W. RichardScott. Sage Publications.
Miller, Randall. 1979. "The Man in the Middle: The Black Slave Driver."AmericanHeritage
30:40-49.
Mills, C. Wright. 1959. The SociologicalImagination.Oxford UniversityPress.
Moore,Wilbert,and RobinWilliams. 1942."Stratificationin the Ante-BellumSouth."American
SociologicalReview7:343-51.
Ochiltree,Ian D. 1998."'AJustand Self-RespectingSystem'?BlackIndependence,Sharecropping,
and PaternalisticRelationsin the American South and South Africa."AgriculturalHistory
72:352-80.
Olson, John. 1992. "The Occupational Structure of Southern Plantations during the Late
AntebellumEra."Pp. 137-69 in WithoutConsentor Contract:TheRiseand Fall ofAmerican
Slavery(Technical Papers,vol. 1), edited by RobertW. Fogel and StanleyL. Engerman.
Norton.
Ransom,Roger,and RichardSutch.2001. One Kind of Freedom:TheEconomicConsequencesof
Emancipation.2d ed. CambridgeUniversityPress.
Rawick,George (ed.). 1972-79. The AmericanSlave:A CompositeAutobiography(Series 1-2,
SupplementalSeries 1-2). GreenwoodPress.
Roughley,T. 1823. TheJamaicaPlanter'sGuide.Longman, Hurst, Rees, Orme, and Brown.
Schultz,Mark.1998."TheDreamRealized?AfricanAmericanLandownershipin CentralGeorgia
between Reconstructionand WorldWarII."AgriculturalHistory72:298-313.
Schwalm,LeslieA. 1997. A Hard Fightfor We:Women'sTransitionfrom Slaveryto Freedomin
South Carolina.Universityof Illinois Press.
Schwartz,Bernard.1970. StatutoryHistoryof the UnitedStates- CivilRights.ChelseaHouse.
Scott, W. Richard.2001. Institutionsand Organizations.2d ed. Sage Publications.
Spero, Sterling D., and Abram L. Harris. 1931. The Black Worker:The Negro and the Labor
Movement.Columbia UniversityPress.
Stampp, Kenneth. 1956. The PeculiarInstitution:Slaveryin the AntebellumSouth. Knopf.
Stinchcombe, Arthur. 1995. Sugar Island Slavery in the Age of Enlightenment:The Political
Economyof the CaribbeanWorld.Princeton UniversityPress.
Tilly, Charles. 1998. DurableInequality.Universityof CaliforniaPress.
Tolnay,Stewart E. 1998. "EducationalSelection in the Migration of Southern Blacks, 18801990."Social Forces77:487-514.
480/SocialForces 82:2,December2003
U.S. Bureau of the Census. 1890. "SpecialReport on the Statistics of Occupations in 1890."
Department of Commerce,Washington,D.C.
. 1900. TwelfthCensusof the United States. Bureau of the Census.
U.S. Senate.1880."NegroExodusfrom the SouthernStates."Testimonybefore the SenateSelect
Committee, Washington,D.C.
Washington,Booker T. 1901. Upfrom Slavery:An Autobiography.A.L. Burt.
Wegener,Bernd. 1992. "Conceptsand Measurementof Prestige."Annual Review of Sociology
18:253-80.
Wesley, Charles. 1927. Negro Labor in the United States, 1850-1925: A Study in American
EconomicHistory.VanguardPress.
Wiener,JonathanM. 1979. "ClassStructureand Economic Development in the South, 18651955."AmericanHistoricalReview 84:970-92.
Williamson, Joel. 1965.AfterSlavery:TheNegroin South CarolinaduringReconstruction,18611877. University of North CarolinaPress.
.1984. The Crucibleof Race:Black/WhiteRelations.Oxford University Press.
Winsell, Keith. 1971."BlackIdentity:The Southern Negro, 1830-1895."Ph.D. diss., University
of California-LosAngeles.
Woodson, Carter.1918.A Centuryof NegroMigration.Association for the Study of Negro Life
and History.
.1922. TheNegro in Our History.AssociatedPublishers.
Woodward,C. Vann. 1974. "Historyfrom Slave Sources."AmericanHistoricalReview 79:47081.
Wright, Erik Olin. 1997. Class Counts: ComparativeStudies in Class Analysis. Cambridge
UniversityPress.
.1998. "Metatheoretical
Foundationsof CharlesTilly'sDurableInequality."
Paperpresented
at the Social Science History Conference, Chicago.
Yetman,Norman. 1984. "Ex-SlaveInterviews and the Historiography of Slavery."American
Quarterly36:181-210.
Zavoina,Richard,and William McElvey.1975."AStatisticalModel for the Analysisof Ordinal
LevelDependent Variables."
Journalof MathematicalSociology4:103-20.