The African American "Great Migration" and Beyond Author(s): Stewart E. Tolnay Source: Annual Review of Sociology, Vol. 29 (2003), pp. 209-232 Published by: Annual Reviews Stable URL: http://www.jstor.org/stable/30036966 Accessed: 17-01-2017 15:54 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted

digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about

JSTOR, please contact [email protected]

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Terms and Conditions of Use

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of Sociology

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

Annu. Rev. Sociol. 2003. 29:209-32 doi: 10.1146/annurev.soc.29.010202.100009

Copyright © 2003 by Annual Reviews. All rights reserved First published online as a Review in Advance on June 4, 2003

THE AFRICAN AMERICAN "GREAT MIGRATION"

AND BEYOND

Stewart E. Tolnay Department of Sociology, University of Washington, Seattle, Washington 98195-3340; email: [email protected] u.washington.edu

Key Words mobility, race, adaptation, South, North

* Abstract During the twentieth century, African Americans participated in one of the most significant demographic events in U.S. history. Their "Great Migration" from the South to the North contributed to profound social, economic, demographic, and cultural changes in northern cities. After the Great Migration, blacks continued to move in search of opportunity as some returned to the South, while others moved to suburbs or better neighborhoods within the North. My review focuses on the Great Migration by discussing research that has examined its causes, the characteristics of the participants, the adaptation of migrants to northern society, and their impact on northern cities. I also briefly review research on return migration to the South and residential mobility by African Americans. Finally, I identify key issues and discuss possible data sources for future research.

INTRODUCTION

Of the traditional trinity of demographic processes (fertility, mortality, and migra- tion), migration is probably the least studied by sociologists. This is somewhat surprising, given that it is the demographic event experienced most frequently throughout an individual American's lifetime. Furthermore, it is more "social" and less "biological" than either fertility or mortality.

During the twentieth century, African Americans participated in two geographic

movements: internal migration and residential mobility.1 These movements, which

are fascinating from a sociological vantage point, had important short- and long- term consequences for individual blacks, the black community, and American society. Extensive social science literature tells the story of African American migration during the past century and investigates its diverse causes and impacts.

1It is conventional to use internal migration to refer to a permanent or semipermanent change

in residence that involves movement within a country's borders but across a meaningful administrative boundary (e.g., between geographic regions or across a county line) and residential mobility to refer to a change of residence within a specified geographic area

(e.g., metropolitan area or city).

0360-0572/03/0811-0209$14.00 209

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

210 TOLNAY

Here, I review that dimension of the African American experience by summarizing what we know and what we have yet to learn.

The "Great Migration" of African Americans out of southern states and into northern cities was one of the most significant demographic events to occur in the United States during the twentieth century. Although numbers alone cannot do justice to the social, economic, political, and cultural importance of the Great Migration, they can at least provide some indication of the magnitude of this phe-

nomenon. A good sense of the demographic impact of the Great Migration on the sending and receiving regions can be gained by considering trends in two measures

(a) the number of southern-born African Americans residing outside of the South

and (b) the size of the black population in nonsouthem states. As the Great Migra-

tion proceeded, the South suffered substantial losses of its native-born black pop- ulation, with over 2.5 million southern-born blacks living outside of the region by 1950 and over 4 million by 1980 (see Figure 1). Equally impressive is the dramatic overall increase in the nonsouthem black population, fueled largely by the southern

migrants and their northern-born offspring. Thus, in purely demographic terms,

the Great Migration produced a dramatic geographic redistribution of the African American population. Furthermore, it had significant consequences for the south- ern region that lost so many of its native-born blacks, for the northern destinations

that absorbed such a large group of newcomers, and for the migrants themselves.

The Great Migration began to wane during the 1960s and was virtually over by the mid-1970s. At that point, the interregional flow of black migrants reversed,

with more moving to the South than were leaving it. During the last quarter of the twentieth century, the stream of migrants entering Dixie comprised a combination

of former southerners returning to their birth region and northern-born "primary

migrants" moving to the South for the first time. Economic decline and restruc- turing in northern cities combined with an expanding economy and improved racial climate in the South to lure African Americans below the Mason-Dixon Line in search of opportunity. However, during this new era of black migration, not all black movers crossed regional boundaries. Some sought safer and more- comfortable housing in the suburbs, or simply a better neighborhood in the central

city. Common to all these mobile African Americans was a desire to achieve a better life in a new place-a new region, a new city, or a new neighborhood-and a willingness to uproot themselves in search of that opportunity.

In the following pages, I explore more thoroughly a variety of issues embed- ded within the preceding thumbnail sketch of internal migration and mobility by African Americans during the twentieth century. I begin with, and focus heav- ily on, the period of the Great Migration. However, to provide a more complete picture of African American migration and mobility, I also devote some attention to the return migration to the South and to residential mobility within regions. The discussion is organized around five general topics that have been the focus of previous research on black migration and mobility. For each topic, I consider the following three organizing questions: What do we know? What critical questions remain? What are the key challenges and opportunities for future research?

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

BLACK MIGRATION AND MOBILITY

Million

Southem-born 12 OTotal in non-South

10

8

6

4

2

0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990

Figure 1 Number of African Americans (total and Southern-born) living in nonsouthern

areas from 1900 to 1990. Data estimated from the Integrated Public Use Microdata Files

available from the Minnesota Population Center (Ruggles & Sobek 2001).

LEAVING DIXIE: WHO WERE THE MIGRANTS AND

WHY DID THEY LEAVE THE SOUTH?

Social scientists have invested much energy in their efforts to establish a profile

of the "typical" participant in the Great Migration. The original and most endur- ing image of the migrants is that of an illiterate sharecropper, displaced from the rural South because of agricultural distress or reorganization (e.g., Chicago Com- mission on Race Relations 1922; Drake & Cayton 1962; Epstein 1918; Frazier 1932, 1939; Mossell 1921; Woofter 1920). This image dominates the many ethno- graphic studies of black migrants living in northern cities during the early stages

211

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

212 TOLNAY

of the Great Migration. Given the general characteristics of the southern African

American population at that time, there can be little doubt that many migrants did

match such a bucolic profile (Grossman 1989). However, researchers are increas- ingly recognizing that the stream of black migrants was probably more diverse than earlier portraits suggested. Marks (1989) has argued that many migrants headed north from southern towns and cities, rather than directly from the rural countryside and, furthermore, that they had more-extensive experience with nona-

gricultural employment than was typically assumed. These are important revisions

to the traditional migrant profile because they have implications for the forces that

were driving blacks from the South as well as for the human capital that they took with them. Marks' inferences about the origins of the migrants, which were

based on rather weak empirical evidence, have received additional support from Alexander's (1998) innovative use of marriage registrations for Pittsburgh and Al- legheny County, Pennsylvania, during the 1930s. Because applicants for marriage

licenses were required to report their birthplace, often including their town of birth, Alexander was able to show that a substantial proportion of black migrants to Allegheny County moved there from towns and cities, rather than directly from

rural areas. Although a very important contribution, the evidence presented by Alexander pertains to only one northern destination, for only a narrow window of

time. Thus, questions remain about whether his conclusions may be generalized to other settings, and other times, during the Great Migration.

The work by Marks and Alexander cautions against oversimplifying the de-

scription of those African Americans who left the South between 1910 and 1970. In all likelihood, it was a heterogeneous group, motivated by a plethora of reasons.

It is also likely that the characteristics of the migrant population varied over time, as

the social and economic forces driving the migration shifted. However, our ability

to obtain a better, statistical sense of the average migrant is handicapped by a lack of information about the migrants before they left the South. Using postmigration

information to describe the characteristics of migrants runs the risk of confounding

cause and effect because the migration experience may have affected postmigra-

tion characteristics. The most useful data source for studying the Great Migration has been a series of Public Use Microdata Samples (PUMS) derived from the decennial U.S. censuses.2 Southern migrants can be identified from these PUMS files by comparing their state of residence with their state of birth. Some PUMS files (e.g., 1940-1970) also include information that gives researchers a better idea of the recency of migration (i.e., state of residence one year or five years before the census). Once identified, the migrants can be described by any characteris- tic that is included in the PUMS file for a particular decade. This approach has yielded valuable information about the postmigration status of migrants and has

enabled important comparisons with nonmigrants in the North (discussed further

2The Integrated Public Use Microdata Series (IPUMS) project at the Minnesota Population Center provides access to all currently available PUMS files (Ruggles & Sobek 2001). The IPUMS archive is available at the following website: http://www.ipums.umn.edu/.

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

BLACK MIGRATION AND MOBILITY

below). However, the PUMS files contain limited information that describes south-

ern migrants before they moved north, for example, their premigration place of

residence, employment status, family characteristics, type of work, etc.3

The PUMS files have proven to be most useful for studying the educational selection of migrants from the South because the education level of an individual changes relatively little after a certain age. That evidence shows that early black southern migrants (in 1910 and 1920) were significantly more likely to be liter-

ate than blacks who remained in the South. In later years (from 1940 to 1970) the migrants had significantly higher levels of educational attainment (years of

schooling) than the sedentary southern black population (Tolnay 1998a; see also, Hamilton 1959, Lieberson 1978b). In contrast, the migrants were less likely to be literate, or had lower levels of educational attainment, than the black population that they joined in the North (Tolnay 1998a).

Researchers have thoroughly mined existing data sources in their efforts to bet-

ter understand which black southerners were more likely to pack up and head north

during the Great Migration. Future progress on this issue will likely come from the innovative use of unanticipated data sources, such as Alexander's (1998) anal- ysis of marriage records for Pittsburgh or Maloney's (2001) use of records from World War I selective service registration. Alternatively, existing census records could be used to create new data sources designed specifically for the study of selective migration. For example, because they are no longer subject to the 72- year confidentiality period, the original census enumerators' manuscripts for 1910,

1920, and 1930 could be used to construct linked census files for 1910-1920 and 1920-1930. That is, southern migrants who moved North between 1910 and 1920, or between 1920 and 1930, could be located in their southern residences at the beginning of the period in which they moved.4 This would provide a richer source of premigration characteristics (e.g., place of residence, occupation, family status) than is available from a single, cross-sectional PUMS file. When combined with a sample of nonmigrants who remained in the South during these decades, the linked

migrant file could advance considerably our understanding of the selection pro- cesses operating during the Great Migration. Most importantly, the premigration

characteristics of migrants could be compared with those of sedentary southerners.

A linked file could also support contextual analyses that would permit researchers to identify the characteristics of local areas (e.g., counties) that were most strongly

related to the out migration of residents.

3The 1940 and 1950 PUMS files include information about farm status and metropolitan area of residence five years and one year before the census, respectively. However, the information is missing for many persons. The 1970 PUMS file includes information about employment status and occupation five years before the census but has no information about

farm or metropolitan residence. 4This tracking can be done using Soundex files that help researchers locate individuals in the original census enumerators' manuscripts by using their names and other limited characteristics.

213

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

214 TOLNAY

There are other types of migration selection that are probably destined to remain

largely the sources of informed speculation. For example, some have suggested that the participants in the Great Migration may have been positively selected from

the larger, black southern population based on their greater ambition, stronger work ethic, and willingness to defer gratification (Lieberson 1978a; Lieberson & Wilkinson 1976; Long 1988, pp. 157-58; Long & Heltman 1975; Rose 1975). These are traits often associated with migrant populations and have been included in explanations for the frequently observed success of migrants in their new places

of residence. However, I am aware of no data source that would allow a sys- tematic comparison of the psychosocial characteristics of southern migrants and

nonmigrants.

Closely related to the question of who the migrants were is the question of what induced them to leave the South. Research into the latter question has been guided strongly by theories of human migration that emphasize a rational decision-

making process in which potential migrants carefully weigh the advantages and

disadvantages of their current residence versus those of potential destinations (e.g., Greenwood 1985, Lee 1966, Ritchey 1976). It is common to refer to those char- acteristics that motivate people to leave a place of origin as push factors, whereas the attractive characteristics of potential destinations are considered pull factors. Economic factors figure prominently in such migration theories and have received

the most emphasis in explanations of the Great Migration. That black southern mi-

grants were strongly motived by economic concerns receives considerable support from both quantitative and qualitative evidence.

The economic deprivations suffered by southern blacks prior to the Great Mi-

gration have been well documented. Plantation agriculture, and the sharecropping system on which it was built, relegated most rural blacks to a landless status, with little opportunity to climb the "agricultural ladder" (Mandle 1978, Ransom & Sutch 1977, Tolnay 1999). Occupational segregation in southern towns and cities concentrated male workers into unskilled jobs and female workers into do- mestic service (Bose 2001, Jones 1992). These conditions, which persisted with little change from emancipation through the early decades of the twentieth cen- tury, represented a powerful incentive for southern blacks to look for opportunity

elsewhere. Indeed, letters that southern blacks wrote to the Chicago Defender and potential northern employers at the outset of the Great Migration offer valuable insights into the strong influence that economic disadvantage and discrimination had on their desire to leave the South (Scott 1919). Many writers mentioned the difficulties they faced in trying to earn a living to support their families in the South and noted their keen interest in relocating to the North to improve their economic fortunes. Oral histories recorded from elderly blacks after the Great Migration reinforce the conclusion that economic motivations were instrumental

in their migration decisions (Faulkner et al. 1982). If the economic conditions for southern blacks were so stultifying for so long,

then why did the Great Migration not begin much earlier? The simple answer to

that question is that southern blacks did not have a feasible alternative. As Collins

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

BLACK MIGRATION AND MOBILITY

(1997) has demonstrated, this situation changed dramatically when World War I and the U.S. adoption of more-restrictive immigration policies forced northern employers to finally consider southern blacks (and whites) as a source of inex-

pensive labor to replace the southern and eastern Europeans. It was this newly created economic opportunity in the North that empowered southern blacks with the ability to make the kind of rational economic decision that is so fundamental to most migration theories. However, although these altered circumstances help to explain the timing of the Great Migration, the pace of exodus from the South varied

cross-sectionally. From perhaps the most extensive aggregate analysis of factors related to migration from southern counties, Fligstein (1981) concluded that those areas affected most strongly by the reorganization of southern agriculture, and by

the increase of farm mechanization, experienced the heaviest out migration. As a result of those changes, large numbers of black tenant farmers were displaced and forced to migrate in search of an alternative economic opportunity–often to southern towns and cities and sometimes outside of the region entirely.

Migration theories also recognize that individuals consider noneconomic, or social, forces in their decisions to stay or move.5 Southern blacks had a number of noneconomic grievances that likely encouraged them to consider migration as a possible remedy. Among the most frequently mentioned are inferior educational opportunities, behavioral restrictions imposed by Jim Crow laws, political disen- franchisement, and racial violence (Henri 1975, Tolnay & Beck 1995, Woofter 1920). These social forces are often included in scholarly treatments of the Great Migration (Ballard 1984, Grossman 1989, Henri 1975, Lemann 1991, Marks 1989, McMillen 1989), and they were mentioned regularly in letters written by poten- tial migrants in the South (Scott 1919), as well as in oral histories of those who migrated (Bunch-Lyons 1997, Faulkner et al. 1982). Compared with economic explanations, however, the social forces related to the Great Migration have less frequently been the focus of quantitative scrutiny. In one exception, Tolnay & Beck

(1992) examined the effect of racial violence on the migration from counties in 10 southern states between 1910-1920 and 1920-1930. Their results showed that, net of economic controls, black out migration was significantly higher in counties that had experienced more black lynchings.

Like the effort to construct a profile of the average migrant, the search for ex-

planations for why the migrants left the South faces significant data challenges. Additionally, this research is conceptually different in that it attempts to understand

the motivations for individual behavior. This task can be approached indirectly by examining aggregate relationships between the rate of out migration and the social

and economic characteristics of local areas. Alternatively, it can be approached more directly by asking the migrants. Comparative approaches have yielded im- portant information about the contextual factors that were associated with higher

50f course, it is sometimes difficult to distinguish economic from noneconomic motives for

migration because these are not necessarily distinct dimensions of the lives of individuals.

215

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

216 TOLNAY

levels of black migration from some southern areas (e.g., Fligstein 1981, Tolnay & Beck 1992). Existing data sources, especially census-based data from southern subregions, offer the potential for further aggregate studies of the environmental

factors promoting migration from the South. The connection of noneconomic areal

characteristics to levels of out migration is especially understudied, but there is considerable potential to examine more thoroughly the role of social forces such as political disenfranchisement and social inequality in the Great Migration. However, such aggregate approaches will face a variety of challenges including (a) adequate measurement of key concepts using existing data (i.e., different noneconomic fac- tors) and (b) the avoidance of the ecological fallacy of drawing conclusions about the motivations for individual-level behaviors from aggregate relationships.

An alternative approach is to gather information about migration decisions di- rectly from the migrants. Although not used widely, oral histories recorded from

southern migrants have provided valuable insights into their motivations and expe-

riences (e.g., Bunch-Lyons 1997, Faulkner et al. 1982, Lemke-Santangelo 1996). Time is quickly running out on our opportunity to gather information from the participants in the Great Migration. Before this valuable repository of information

disappears, researchers should consider the potential of well-designed approaches to record the life histories of migrants. Such qualitative evidence may not be generalizable to the entire population of southern migrants, but this limitation is counterbalanced by the richness of the first-hand information that can be obtained

from the migrants. Whatever methodological approach is used to study the moti-

vations of southern migrants, it is not especially productive to approach the topic as an effort to determine whether economic or noneconomic forces were more

important. Such either/or approaches tend to deflect attention away from the more

interesting, bigger picture that phenomena like the Great Migration are complex processes driven by a large and diverse set of forces.

INTO THE "PROMISED LAND": WHERE DID THE

MIGRANTS GO AND HOW DID THEY FARE?

Southern black migrants settled in virtually all areas of the North and West. How- ever, definite migration streams developed as the Great Migration proceeded, and those streams resulted in a much heavier concentration of migrants in certain places.

Throughout the Great Migration, large metropolitan areas in the Northeast and Midwest were especially popular destinations, with the influx of southern migrants

causing massive growth in the black populations of cities like Chicago, Detroit, New York, and Philadelphia. In contrast, western states did not become a common destination for black migrants until after 1940, when the entry of the United States

into World War II produced a rapid expansion in the defense industry on the West Coast (Johnson & Campbell 1981). Once in the North or West, black southern migrants were overwhelmingly an "urban" population because the industrial em- ployment opportunities that had attracted them were concentrated in larger cities.

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

BLACK MIGRATION AND MOBILITY

An important factor in the development of specific interregional migration

streams was the availability and ease of transportation, especially the routes of

interstate highways and railroad lines connecting southern states to different parts

of the North and West. Many migrants traveled by train, so the existence of con-

venient rail connections influenced their choice of destinations. For example, the Illinois Central Railroad provided potential migrants in Louisiana and Mississippi with relatively direct access to Chicago (Lemann 1991). For those in Georgia and South Carolina, the existing rail and highway connections made Philadelphia, New York, and Boston more-common destinations (Ballard 1994, Kiser 1932). When migration to the West intensified, Highway 66 and the Southern Pacific Railroad were avenues by which migrants from Oklahoma, Texas, Louisiana, and Arkansas

reached California (Gregory 1989). However, the selection of destinations was influenced by more than simple

logistical convenience. In perhaps the most comprehensive effort to incorporate the characteristics of potential destinations into the quantitative study of the Great

Migration, Price-Spratlen (1998; 1999a,b) showed that black migrants were more attracted to areas that offered stronger "ethnogenic" support for the African Amer-

ican community and eased the adjustment for newcomers. As examples of ethno-

gensis, Price-Spratlen (1998; 1999a,b) includes the presence in the community of an NAACP chapter, a mature National Urban League, African American churches, and African American newspapers. He concludes that the impact of ethnogenesis on the selection of destinations weakened over time as migration streams acquired

a momentum of their own. This phenomenon has been observed by many others

(e.g., Ballard 1984, Lemann 1991) who have described the tendency for later Great Migration participants to follow family members and friends who had migrated previously. Once developed, these patterns of "chain migration" funneled migrants from common points of origin to common points of destination and reinforced ex-

isting migration streams. The presence of family or friends in the North improved

the flow of information about specific destinations to potential migrants, especially

the availability of jobs, and eased their transition upon arrival. With few exceptions, such as basic descriptions of the dominant migration

streams (e.g., Florant 1942, Johnson & Campbell 1981, Long 1988) and Price-

Spratlen's work on the linkage between ethnogenesis and the rate of migration to

local areas, the general subject of how black southern migrants selected their desti-

nations has not been explored thoroughly. However, investigators have conducted a number of interesting case studies on the social, economic, and cultural condi-

tions that prevailed in selected northern and western destinations such as Chicago, (Chicago Commission on Race Relations 1922, Drake & Cayton 1962, Duncan & Duncan 1957, Frazier 1932, Freedman 1950, Lemann 1991); Pittsburgh (Bodnar et al. 1982, Gottleib 1987); Evansville, Indiana (Bigham 1987); Philadelphia (Ballard 1984); San Francisco (Broussard 1993); Oakland (Lemke-Santangelo 1996, McBroome 1993); and Milwaukee (Trotter 1985). Although informative, such case studies tell us relatively little about (a) how the characteristics of southern

migrants to different regions/cities within the North and West varied or (b) why the

217

This content downloaded from 148.84.53.55 on Tue, 17 Jan 2017 15:54:27 UTC All use subject to http://about.jstor.org/terms

218 TOLNAY

migrants residing in a specific city moved there rather than elsewhere. There is limited evidence that black southern migrants chose somewhat different desti- nations than their white counterparts. For example, it appears that whites were more likely than blacks to move to the West and to smaller towns or cities within

the Northeast and Midwest (Berry 2000, Tolnay et al. 2002). Furthermore, the selected destinations of both black and white southern migrants determined the opportunity structures available and influenced, for example, the

Related Tags