![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
Historians describe the United Daughters of the Confederacy (UDC) as an urban movement, but the reasons for its urban focus have not been adequately explored. This is intriguing because the UDC was most vibrant during the first three decades of the twentieth century, when white and black citizens alike were leaving rural southern communities for urban centers – and as white politicians increasingly enforced Jim Crow restrictions. In this study, we examine the effects of urbanity and race on local UDC chapters and membership across the South. We find that although UDC chapters were not overwhelmingly urban, members largely joined urban chapters – particularly those located in smaller cities. Chapters and members were also more common in cities with larger black populations, suggestive of a racial contact stimulus. These findings lead us to question the UDC’s relationship to the predominately rural South as well as its ability to broadly represent southern white women. They also speak to other research on voluntary associations during this period, underscoring that southern urbanity inspired civil society, albeit based on separation rather than integration.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 If urbanity aided the spread of the AASS, then it also likely helped the AASS transition supporters into Liberty Party identifiers (see Chamberlain Citation2018) and influenced the spread of other voluntary moral reform associations, including those related to temperance and women’s rights.
2 For a more detailed comparison of UDC organizing to that of the Woman’s Christian Temperance Union, the General Federation of Women’s Clubs, and farmers’ organizations, see Appendix I.
3 Many of these groups – including prominent organizations affiliated with national groups such as the National American Woman Suffrage Association – permitted only white members; black citizens, however, created their own vibrant organizations (see Scott Citation1990; Rouse Citation1991; Rabinowitz Citation1992, 151–155; Hunter Citation1997; Gilmore Citation2019).
4 Still, there were efforts to create physical segregation in terms of neighborhood housing, which became especially prominent after New Deal housing legislation (e.g., Connolly Citation2014).
5 We refer to urban communities as “cities” at times, understanding that the latter term varies based on state rules for incorporation.
6 Cities could have more than one UDC chapter.
7 Some readers may question why we did not use only white female population. This is because the 1910 Census does not provide population breakdowns by race and gender at the city level. However, since this information is provided in 1920 and 1930, we checked to see if per capita measures generated with the non-black population and from the native white female population generated significantly different measurements. In 1920, for example, the measures correlated at .98 – essentially, perfect correspondence. Thus, our concerns about using total non-black population for this measure were alleviated.
8 We log population to minimize the effect of the few, very large, outlier cities, such as Atlanta, GA.
9 Although many black citizens moved to the North and West during the Great Migration, which started during World War I, approximately three-quarters of all black Americans still resided in the South in 1930.
10 One point noted earlier was the prevalence of religious organizations in counts from southern cities. Churches were a significant part of these counts, and had significant effects on Southern civic life. Unfortunately, data on church membership in cities is very limited for this era. The Census of Religious Bodies was conducted in 1906, 1916, 1926, and 1936, and the data are only provided for cities of 25,000 or more citizens in 1906 and 1926 (and only for cities of 100,000 or more in 1916). Approximately 5 percent of all southern cities with a UDC chapter were above the 25,000 in population in 1910 (1906 Religious Census) and 1930 (1926 Religious Census). Thus, there are simply too few cases to model. The relationships between white church membership per capita and UDC per capita membership are discussed in Appendix II. There is little connection in 1910 (n = 20) but some positive correlation by 1930 (n = 43).
11 Readers should be reminded that residential segregation is a separate issue that cannot be measured with the data we use. In southern cities, there was greater integration in terms of racial proximity in housing in the era studied. This is a potential avenue for future research that uses local-level chapter membership data (when available) to map out UDC membership and housing segregation.
12 In 1910, UDC data excluded Jackson and Meridian, Mississippi, from the credentials and dues report. As such, these two cities were excluded from that year’s analysis.
13 More complex predicted probabilities, generated by holding other predictors at their mean and/or median values, provide similar results.
14 These illustrations also include median reference lines to show the average marginal effect’s effect at the median value of x (see Pepinsky Citation2018).
15 We note, however, that this does not mean a chapter never existed in these places. It indicates that any chapters that were chartered had little lasting power and thus, diminished ability to shape society in these communities. One example of a city that had UDC activity, but not in 1910, 1917, or 1930, was Jonesboro, Arkansas. Others reported limited activity, such as Monroe, Louisiana, which had membership in 1910 but not in 1917 or 1930.