1. See Adams, Greg D., “Abortion: Evidence of an Issue Evolution,” American Journal of Political Science 41 (1997): 718–37Google Scholar; Layman, Geoffrey, The Great Divide: Religious and Cultural Conflict in American Party Politics (New York: Columbia University Press, 2001)Google Scholar; Sanbonmatsu, Kira, Democrats, Republicans, and the Politics of Women's Place (Ann Arbor: University of Michigan Press, 2002)Google Scholar; Wolbrecht, Christina, The Politics of Women's Rights: Parties, Positions, and Change (Princeton, NJ: Princeton University Press, 2000)Google Scholar. Also see David Karol, “How and Why Parties Change Positions on Issues: Party Policy Change as Coalition Management in American Politics” (paper presented at the Annual Meeting of the American Political Science Association, San Francisco, CA, August 30–September 2, 2001); and David Karol, “Coalition Management: Explaining Party Position Change in American Politics” (PhD diss., University of California, Los Angeles, 2005) on patterns of partisan change across a range of issues in the second half of the twentieth century.

2. We use “realignment” here to mean a process by which partisan coalitional alignments change. It is worth noting that these theoretical stakes do not depend on the validity of the claim that racial issues drove the changes in mass partisanship in the south from the 1960s to the present. Regardless of whether one agrees with Shafer and Johnston's (2006) thesis that the role of racial attitudes in shaping southern partisanship and election outcomes has been greatly exaggerated, it is nonetheless the case that the shift in the parties' stance on civil rights issues is a crucial development in American politics. In other words, even if the mass electorate did not follow the parties' issue evolution, it is nonetheless critical to understand the parties' changing stances on this central policy battleground. See Shafter, Byron and Johnston, Richard, The End of Southern Exceptionalism: Class, Race, and Partisan Change in the Postwar South (Cambridge, MA: Harvard University Press, 2006)CrossRefGoogle Scholar.

3. While we are the first to use state platforms as a lens to study partisan change on civil rights, Richard Bensel's pioneering study of American industrialization uses an impressive collection of state and national platforms to analyze party coalitional strategies in the late nineteenth century. See Bensel, , The Political Economy of American Industrialization, 1877–1900 (Cambridge, UK: Cambridge University Press, 2001)Google Scholar.

4. Carmines, Edward and Stimson, James, “On the Structure and Sequence of Issue Evolution,” American Political Science Review 80 (1986): 901–20Google Scholar; Carmines, Edward and Stimson, James, Issue Evolution: Race and the Transformation of American Politics (Princeton, NJ: Princeton University Press, 1989)Google Scholar.

5. Party elites are defined as the president, members of Congress, and candidates for high office, while activists include delegates to the national conventions, minor officeholders, and party officials. See Carmines, Edward and Wagner, Michael, “Political Issues and Party Alignments: Assessing the Issue Evolution Perspective,” Annual Review of Political Science 9 (2006): 70Google Scholar; Carmines, Edward and Woods, James, “The Role of Party Activists in the Evolution of the Abortion Issue,” Political Behavior 24 (2002): 363–64Google Scholar; and Carmines and Stimson, Issue Evolution, 162.

6. Carmines and Wagner, “Political Issues and Party Alignments,” 70.

7. Carmines and Stimson, Issue Evolution, 179.

8. Carmines and Stimson, Issue Evolution, 57.

9. Carmines and Stimson, Issue Evolution, 179.

10. Carmines and Stimson, Issue Evolution, 18.

11. Carmines and Stimson, Issue Evolution, 192–93.

12. Eldridge, Niles and Gould, Stephen Jay, “Punctuated Equilibria: An Alternative to Phyletic Gradualism,” in Models in Palaeobiology, ed. Schopf, T.M. (San Francisco: Freeman Cooper, 1972), 82–115Google Scholar; Carmines and Stimson, Issue Evolution, 18.

13. Carmines and Stimson, Issue Evolution, 63.

14. Carmines and Stimson, Issue Evolution, 70. The authors note that, while the difference of means between the two parties' nonsouthern Senate delegations is not statistically significant, its direction is supportive of their claim.

15. Carmines and Stimson, Issue Evolution, 35.

16. Carmines and Stimson, Issue Evolution, 154.

17. We follow Carmines and Stimson in considering convention delegates to be activists rather than party elites. Our evidence suggests that state and local officeholders, allied interest groups, some members of Congress, and party workers participated in writing the platforms, which were then adopted by the convention (see discussion below). While the platform authors included many professional politicians (e.g., state party officials, mayors, etc.), few were national political figures.

18. Published works that consider Issue Evolution to be a significant contribution to the study of partisan change include Pierson, Paul, Politics in Time (Princeton, NJ: Princeton University Press, 2004)Google Scholar; Box-Steffensmeier, Janet and Smith, Renee, “The Dynamics of Aggregate Partisanship,” American Political Science Review 90 (1996): 567–80Google Scholar; Nardulli, Peter, “The Concept of a Critical Realignment, Electoral Behavior, and Political Change,” American Political Science Review 89 (1995): 10–22Google Scholar; and Edsall, Thomas and Edsall, Mary, Chain Reaction: The Impact of Race, Rights, and Taxes on American Politics (New York: W.W. Norton, 1991)Google Scholar.

19. Chen, Anthony, “The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North, 1945-64,” American Journal of Sociology 112 (2007): 1713–74CrossRefGoogle Scholar.

20. Chen, Anthony, “‘The Hitlerian Rule of Quotas’: Racial Conservatism and the Politics of Fair Employment Legislation in New York State, 1941–1945,” Journal of American History 92 (2006): 1238–64Google Scholar. Aronson, Arnold and Spiegler, Samuel, “Does the Republican Party Want the Negro Vote?” The Crisis (December 1949): 364–68, 411–17Google Scholar, comes to a similar conclusion: while Democrats pushed for fair employment practices legislation in many states throughout the 1940s, the GOP was largely indifferent or opposed to these measures.

21. David Karol, “Realignment without Replacement: Issue Evolution and Ideological Change among Members of Congress” (paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago, IL, April 15–18, 1999).

22. Karol, “How and Why Parties Change Positions on Issues”; Karol, “Coalition Management.”

23. Lee, Taeku, Mobilizing Public Opinion (Chicago: University of Chicago Press, 2002)Google Scholar.

24. These actors are positioned at the meso-level in two respects. First, they are neither national elites nor mass publics, instead performing a mid-level function within the party system. Second, while not occupying a national perch, their position within the parties' federal structure is high enough to merit their participation in state party conventions.

25. Freeman, Jo, “Sex, Race, Religion, and Partisan Realignment,” in “We Get What We Vote For … Or Do We?” The Impact of Elections on Governing, ed. Scheele, Paul E. (Westport, CT: Praeger, 1999), 167–90Google Scholar; Wolbrecht, The Politics of Women's Rights. But, see Kira Sanbonmatsu, Democrats, Republicans, and the Politics of Women's Place, for a critique of this view.

26. Foster, James, The Union Politic: The CIO Political Action Committee (Columbia: University of Missouri Press, 1975)Google Scholar. Of course, rank-and-file white union workers were not necessarily supportive of civil rights. See, e.g., Sugrue, Thomas, The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit (Princeton, NJ: Princeton University Press, 1996)Google Scholar. Even though relations between African Americans and unions were by no means smooth, it is still the case that the CIO leadership was a crucial supporter of both civil rights legislation and Democratic candidates.

27. Aronson and Spiegler, “Does the Republican Party Want the Negro Vote?” 411–17; Kesselman, The Social Politics of FEPC; and Anderson, J.W., Eisenhower, Brownell, and the Congress: The Tangled Origins of The Civil Rights Bill of 1956–1957 (Tuscaloosa: University of Alabama Press, 1964)Google Scholar.

28. Gerring, John, Party Ideologies in America, 1828–1996 (Cambridge, UK: Cambridge University Press, 1998)CrossRefGoogle Scholar.

29. See Karol, “Coalition Management”; Skowronek, Stephen, “The Reassociation of Ideas and Purposes: Racism, Liberalism, and the American Political Tradition,” American Political Science Review 100 (2006): 385–401CrossRefGoogle Scholar; and Hans Noel, “The Coalition Merchants: Testing the Power of Ideas in the Civil Rights Realignment” (paper presented at the Annual Meeting of the American Political Science Association, Chicago, IL, September 2007).

30. Brock, Clifton, Americans for Democratic Action: Its Role in National Politics (New York: Public Affairs Press, 1962)Google Scholar.

31. Beginning the analysis in 1920 captures the parties' views in the pre–New Deal period, which enables us to better evaluate the hypothesis that the coalitional partners that became linked with each party during the New Deal pushed the parties in opposite directions on civil rights. The year 1968 is a natural endpoint, both because it is well after the “critical moment” at the core of Carmines and Stimson's theory, and because the splintering of the civil rights movement makes the identification of various positions as pro– or anti–civil rights difficult.

32. Source: ProQuest Historical Newspapers; search terms: platform w/10 (democrat# or republic#) and state, May to Oct 1942/1950. All articles originally appeared in The Chicago Defender, Chicago Tribune, Los Angeles Times, New York Times, or Washington Post.

33. For example, political scientist Stephen Bailey chaired the Connecticut Democrats' committee in 1950. For platform committee members who were listed without an affiliation, we searched politicalgraveyard.com to determine whether they were current or former public officials, or delegates to a national convention.

34. For instance, various coalition groups lobbied the New York State Democrats for the inclusion of pet causes in the party's 1936 platform. These groups include the Association of State Civil Service Employees, which pressed the party to adopt a plank supporting a merit-based system for state employment; a women's organization that urged the party to go on-the-record in supporting gender-neutral labor laws; a teachers group pushing for permanent tenure for schoolteachers; and two representatives of the “state legislative board of the railroad trainmen” lobbying for a platform plank favoring legislation to require “adequate manning” of trains. See “Lehman Predicts a Party Victory,” New York Times, 29 Sept. 1936.

35. A more complete description of the positions in the party structure that platform writers and executive committee members tended to occupy is available in an online appendix, accessible at www.statepartyplatforms.com.

36. Newspaper articles from this period indicate that platform writers took their role as codifiers of party positions seriously. For instance, The Los Angeles Times reports in 1956 that the California Republican State Central Committee was deeply divided over whether to include a statement favoring right-to-work laws in that year's state GOP platform, debating the issue for nearly two hours before rejecting the proposed plank. See “Right-to-Work Plank Rejected by GOP Group,” Los Angeles Times, 15 Apr. 1956.

37. We thank an anonymous reviewer for this formulation.

38. The twenty-two states are: California, Colorado, Connecticut, Hawaii, Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Montana, Nebraska, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, South Dakota, Utah, and Wisconsin. The map in Figure 2 also shows our coverage for two southern states, Texas and North Carolina, which were not included in our analysis.

39. Key, V.O. Jr., Southern Politics in State and Nation (New York: Knopf, 1950), 646Google Scholar.

40. This estimate is based on the conservative assumption that all nonsouthern state parties published platforms biennially in all cases in which we lack evidence to the contrary.

41. Carmines and Stimson, Issue Evolution, 56.

42. Carmines and Stimson also construct a “racial priority index,” which examines the position in each platform in which a paragraph on civil rights first appears, and divides that number by the total number of paragraphs in the platform. Since state party platforms often present issue areas in alphabetical order, a similar analysis would not be useful for our data.

43. A wide range of civil rights issues, including proposed antilynching legislation and integration of the armed forces, among many other subjects, received national attention during the 1920–1968 period. While these and all other civil rights issues are included in our summary score measure, our five issue measures focus on these five specific issue areas because they were the most salient civil rights issues during the late 1950s and 1960s apex of the civil rights movement, each receiving attention in at least one of the major civil rights bills of that period (i.e., the Civil Rights Acts of 1957 and 1964, and the Voting Rights Act of 1965).

44. Although the subjective nature of assigning numerical scores to qualitative accounts makes such work challenging, the fact that our findings were very similar across all three measures gives us confidence that our results are not driven by idiosyncratic judgments.

45. See Budge, Ian, “Validating Party Policy Placements,” British Journal of Political Science 31 (2001): 211CrossRefGoogle Scholar, for a discussion of the advantages of using measures of platform length to gauge party priorities.

46. Carmines and Stimson, Issue Evolution, 116.

47. Carmines and Stimson, Issue Evolution, 56.

48. The finding is substantively identical if one substitutes the median and interquartile range for the mean score plus or minus one standard deviation.

49. Carmines and Stimson, Issue Evolution, 18 and 154.

50. As before, we use 423 sets of paired platforms as our sample here. Those state-years for which we could not obtain both parties' platforms for a given year are excluded from the analysis.

51. Carmines and Stimson, Issue Evolution, 18.

52. The results are substantively identical if one focuses on the average score (on our five-point scale, see Appendix Table A2 for more information) for each party on each issue, as opposed to the simpler ranking of which party took the more liberal position.

53. As previously mentioned, the summary score ranges from −4 to 5. South Dakota's scores range from −1 (signifying that the platform views some government venues as inappropriate for advancing civil rights) to 5 (signifying advocacy of outlawing discrimination broadly in at least two issue areas). Illinois parties take positions ranging from –0.5 (a rarely employed designation referring to opposition to some civil rights measures, but tepid support for others) to 5. Please see Appendix A for a more complete look at how these values were assigned.

54. For comparison, Appendix B also includes graphs for North Carolina and Texas, the two southern states for which we have a sufficient number of platforms to conduct such an analysis.

55. The most striking exception to the Democrats' relative liberalism is New Hampshire. While civil rights were by no means prominent for either party in the 1940s–1950s—perhaps due to the state's lack of racial diversity and its conservatism—Republicans are a bit more likely to take the liberal position until 1960, when New Hampshire Democrats finally embrace racial liberalism. The border state of Missouri is also an interesting partial exception: the state GOP appears as liberal as or more liberal than the Democrats up until 1956, when Democrats embrace an aggressive civil rights platform and the GOP counters by distancing itself from its earlier pro–civil rights pronouncements.

56. A bivariate regression of both same-state parties' summary scores on the proportion of the state's population that is African American shows that African American population is positively correlated with the summary score value. The results of a second regression of the difference between the Democrats and same-state Republicans' summary scores on the state's African American population indicates that the Democrats tended to take the more progressive civil rights position even in states with very small African American populations.

57. We found that the gap between the parties on civil rights was uncorrelated with the “unbossed” and “voters” state party classification variables presented in Marty Cohen, David Karol, Hans Noel, and John Zaller, The Party Decides: Presidential Nominations Before and After Reform (Chicago: University of Chicago Press, forthcoming). In addition, it is not correlated with state party “factional patterns” (i.e., cohesive, bifactional, or multifactional structure) reported in Jewell, Malcolm and Olson, David, American State Political Parties and Elections, (Homewood, IL: The Dorsey Press, 1978)Google Scholar, nor with the strength of local traditional party organizations reported by Mayhew, David, Placing Parties in American Politics (Princeton, NJ: Princeton University Press, 1986)Google Scholar.

58. We plan a more thorough quantitative analysis of the determinants of the state parties' position on civil rights policy. Preliminary analysis indicates that CIO density leads to greater Democratic liberalism on civil rights, while having no effect (or a negative effect) on Republican civil rights liberalism. Jewish population also appears to have a positive impact on state Democrats' civil rights liberalism, though in this case, there is also a small positive effect for the GOP as well. The impact of NAACP membership and African American population is strong at the bivariate level, but our preliminary examination suggests that these effects dissipate when one controls for CIO density. William Collins finds that state-level fair employment laws are more likely in states with strong unions and NAACP organizations, though the size of the state's African American population does not have a significant impact. Collins does not assess whether these variables have differential effects depending on party control of the state government. See Collins, William, “The Political Economy of State-Level Fair-Employment Laws, 1940–1964,” Explorations in Economic History 40 (2003): 24–51CrossRefGoogle Scholar.

59. With the exception of 1948. Carmines and Stimson, Issue Evolution, 56.

60. Carmines and Stimson's discussion of the national parties' civil rights positions acknowledges that some nonsouthern Democrats sought to steer their party in a more liberal direction on civil rights, but they consider nonsouthern Democrats in general to be tepid supporters of civil rights, and suggest that nonsouthern Republicans were typically more supportive during the 1950s (Carmines and Stimson, Issue Evolution, 35–37, 70).

61. Of course, we are by no means the first to make this argument. Indeed, historians' accounts of the period have often depicted northern Democrats as more liberal than Republicans on civil rights. See Zelizer, Julian, The American Congress (New York: Houghton Mifflin, 2004)Google Scholar. But political scientists—relying upon roll call data and national platforms—have generally downplayed northern Democrats' relative liberalism (Carmines and Stimson, Issue Evolution; but see Karol, “Coalition Management;” Chen, “The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North;” and Chen, “‘The Hitlerian Rule of Quotas’” for noteworthy exceptions). See Reiter, Howard, “The Building of a Bifactional Structure: The Democrats in the 1940s,” Political Science Quarterly 116 (2001): 107–29Google Scholar, for evidence that southern and nonsouthern Democrats had increasingly divergent views on a range of issues in the 1940s.

62. See Sitkoff, Harvard, “Harry Truman and the Election of 1948: The Coming of Age of Civil Rights in American Politics,” The Journal of Southern History 37 (1971): 597–616CrossRefGoogle Scholar, on the 1948 campaign.

63. Frederickson, Kari, The Dixiecrat Revolt and the End of the Solid South, 1932-1968 (Chapel Hill: University of North Carolina Press, 2001), 123Google Scholar.

64. Gerring, Party Ideologies in America, 253.

65. Berman, William, The Politics of Civil Rights in the Truman Administration (Columbus: Ohio State University Press, 1970), ch. 5Google Scholar.

66. Sindler, Allen P., “The Unsolid South: A Challenge to the Democratic Party,” in The Uses of Power: Seven Cases in American Politics, ed. Sindler, Allen (New York: Harcourt, Brace, and World, 1962), 233Google Scholar.

67. Heard, Alexander, A Two-Party South? (Chapel Hill: University of North Carolina Press, 1952), 152Google Scholar. Indeed, Truman's decision not to seek the nomination in 1952 was partly rooted in his belief that his candidacy would lead to a southern walkout and permanent split in the Democratic Party (see Berman, The Politics of Civil Rights in the Truman Administration, 196–97).

68. Berman, The Politics of Civil Rights in the Truman Administration, ch. 5; and Frederickson, The Dixiecrat Revolt and the End of the Solid South.

69. Berman, The Politics of Civil Rights in the Truman Administration, ch. 5.

70. J.W. Anderson, Eisenhower, Brownell, and the Congress, ch. 3.

71. Berman, The Politics of Civil Rights in the Truman Administration, 209–10.

72. Delegate numbers were roughly based on the size of a state's congressional delegation, plus between zero and ten at-large delegates.

73. In his speech accepting the Republican gubernatorial nomination for North Carolina, John Parker stated: “I have attended every state convention since 1908 and have never seen a Negro delegate… . The Negro as a class does not desire to enter politics. The Republican Party of North Carolina does not desire him to do so.” (Greensboro Daily News, 19 Apr. 1920, quoted in Watson, Richard Jr., “The Defeat of Judge Parker: A Study in Pressure Groups and Politics,” The Mississippi Valley Historical Review 50 [1963]: 213–34CrossRefGoogle Scholar.)

74. Heard, A Two-Party South?, 166. African American Republicans still controlled the patronage-oriented Mississippi Republican Party; Georgia and South Carolina Republican organizations were mixed.

75. McClosky, Herbert, Hoffman, Paul, and O'Hara, Rosemary, “Issue Conflict and Consensus Among Party Leaders and Followers,” American Political Science Review 54 (1960): 413–16CrossRefGoogle Scholar. More Republicans also favored decreased enforcement than did Democrats (31.7 percent of Republican delegates, as compared to 26.6 percent of Democrats).

76. National Party Conventions, 1831–2004 (Washington, DC: CQ Press, 2005), 228.

77. Farhang, Sean and Katznelson, Ira, “The Southern Imposition: Congress and Labor in the New Deal and Fair Deal,” Studies in American Political Development 19 (2005): 1–30CrossRefGoogle Scholar.

78. Kesselman, Louis, The Social Politics of FEPC: A Study in Reform Pressure Movements (Chapel Hill: University of North Carolina Press, 1948), 151Google Scholar.

79. Anderson, Eisenhower, Brownell, and the Congress, ch. 2.

80. Anderson, Eisenhower, Brownell, and the Congress, ch. 2; Berman, The Politics of Civil Rights in the Truman Administration; Kesselman, The Social Politics of FEPC.

81. Sitkoff, “Harry Truman and the Election of 1948,” 606.

82. Brock, Americans for Democratic Action.

83. Kesselman, The Social Politics of FEPC, 29–31.

84. Caro, Robert, Master of the Senate (New York: Knopf, 2002), 927Google Scholar.

85. Brock, Americans for Democratic Action, 179–80; Martin, John, Civil Rights and the Crisis of Liberalism: The Democratic Party, 1945-1976 (Boulder, CO: Westview Press, 1979), 170Google Scholar. Also see Daniel Schlozman, “The Making of Partisan Majorities: Parties, Anchoring Groups, and Electoral Coalitions,” unpublished manuscript, Department of Government, Harvard University, for an insightful discussion of the role of network ties in cementing partisan coalitions.

86. Gerring, Party Ideologies in America, 253.

87. As noted above, we find that the size of a state's Jewish, Catholic, and CIO member populations (as a percentage of total state population) is positively correlated with the civil rights summary score for the Democratic Party in that state. These correlations between the relative strength of these three demographic groups and the state Democratic Party's civil rights liberalism achieve conventionally accepted levels of substantive and statistical significance.

88. Chen, “The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North,” 1713–74; Chen, “‘The Hitlerian Rule of Quotas,’” 1238–64; Karol, “Coalition Management.”

89. Kesselman, The Social Politics of FEPC.

90. Pittsburgh Courier, 4 Jan. 1947, 4; quoted in Berman, The Politics of Civil Rights in the Truman Administration, 59.

91. See, for example, Griffin, William, “The Political Realignment of Black Voters in Indianapolis, 1924,” Indiana Magazine of History 79 (1983): 133–66Google Scholar.

92. Lawrence, David, The Collapse of the Democratic Presidential Majority: Realignment, Dealignment, and Electoral Change from Franklin Roosevelt to Bill Clinton (Boulder, CO: Westview Press, 1996), 34Google Scholar.

93. Of course, a broad swath of urban, working-class voters (not just African Americans) shifted to the Democratic camp during the New Deal era. Democrats' support for government intervention in the economy to alleviate the Depression struck a chord with an overlapping set of voters: Catholics, Jews, immigrants, the working class, and city-dwellers. See Sundquist, James, Dynamics of the Party System (Washington, DC: Brookings, 1983), 214–23Google Scholar. As previously mentioned, groups tied both to these voters and to the Democratic Party soon became core supporters of the civil rights movement.

94. Lubell, The Future of American Politics, 115.

95. Heard, A Two-Party South?, 233.

96. Myrdal, Gunnar, An American Dilemma (New York: Harper and Brothers Publishers, 1944), 511Google Scholar.

97. Heard, A Two-Party South?, 234. We are planning to explore early public opinion data to allow for a fuller understanding of the timing of African Americans' changing partisan allegiances. For a discussion of the early survey data, see Adam Berinsky and Eric Schickler, “Collaborative Research: The American Mass Public in the 1940s and 1950s,” available at http://web.mit.edu/berinsky/www/nsf.pdf (accessed 17 Aug. 2007).

98. Tate, Katherine, From Protest to Politics: The New Black Voters in American Elections (Cambridge, MA: Harvard University Press, 1994), 70Google Scholar.

99. Holcombe, Arthur, “The Changing Outlook for a Realignment of Parties,” Public Opinion Quarterly 10 (1946-47): 455-69CrossRefGoogle Scholar.

100. Freidel, Frank, F.D.R. and the South (Baton Rouge: Louisiana State University Press, 1965), 91–92Google Scholar.

101. As quoted in Patterson, James T., “The Failure of Party Realignment in the South, 1937-1939,” Journal of Politics 27 (1965): 603CrossRefGoogle Scholar.

102. Patterson, “The Failure of Party Realignment in the South,” 602–17.

103. Heard, A Two-Party South?, 160.

104. James Patterson notes how Republicans, such as Sen. Arthur Vandenberg (R-MI), Kansas governor and 1936 presidential nominee Alf Landon, and former senator George Moses (R-NH), explored the possibilities for a realignment in the late 1930s. Moses wrote Sen. Carter Glass (D-VA) in November 1937: “You and I have often discussed realignment but you have always raised the color question. This condition no longer exists. Jim Farley and the Roosevelt largess have made the colored vote in the North impregnably Democratic. Therefore, with the color line obliterated, why cannot those of us who are free, white, and twenty-one get together and do a job as effective as Mussolini did when he made his march upon Rome?” Similarly, Vandenberg considered forming coalition tickets with southern Democrats for 1940, writing that “a 1940 realignment may not be ‘absolutely essential’ even though my ‘hunch’ runs in that direction.” As quoted in Patterson, “The Failure of Party Realignment in the South,” 605.

105. Frederickson, The Dixiecrat Revolt and the End of the Solid South, 227.

106. Sitkoff, “Harry Truman and the Election of 1948,” 613.

107. Aronson and Spiegler, “Does the Republican Party Want the Negro Vote?”; Berman, Politics of Civil Rights in the Truman Administration.

108. Aronson and Spiegler, “Does the Republican Party Want the Negro Vote?”

109. Lawrence, W. H., “Republicans Woo States Righters,” New York Times, 9 Mar. 1950, p. 23Google Scholar.

110. Savannah Morning News, 3 July 1950; as quoted in Heard, A Two-Party South?

111. Berman, Politics of Civil Rights in the Truman Administration.

112. Sen. Hubert Humphrey (D-MN) seemed to agree, defining a southern Democrat as “a conservative Republican with a southern accent.” Heidepriem, Scott, A Fair Chance for a Free People: A Biography of Karl E. Mundt, United States Senator (Madison, SD: Leader Print Company, 1988), 162Google Scholar.

113. Heidepriem, A Fair Chance for a Free People, 158.

114. Heidepriem, A Fair Chance for a Free People, 163–64. Mundt's voting record, broadly speaking, was not out-of-step with the rest of his party's. His DW-NOMINATE scores place him near the Republican Senate delegation median. See Keith T. Poole and Howard Rosenthal, “Voteview,” available at http://www.voteview.com.

115. Berman, Politics of Civil Rights in the Truman Administration.

116. Black, Earl and Black, Merle, The Vital South: How Presidents are Elected (Cambridge, MA: Harvard University Press, 1992), 188CrossRefGoogle Scholar.

117. Despite Eisenhower's refusal to endorse the Brown decision, his reluctant decision to send troops to Little Rock temporarily set back the GOP cause in the South. But the platform evidence and 1956 convention delegate survey (administered in 1957) indicate that nonsouthern Republicans continued to be more conservative on civil rights than their Democratic counterparts, so that the long-term prospects for GOP gains in the South persisted. McClosky et al., “Issue Conflict and Consensus Among Party Leaders and Followers,” 406–27.

118. Brock, Americans for Democratic Action, 119.

119. Sundquist, Dynamics of a Party System, 264.

120. Richard Bensel emphasizes how the committee system—particularly, the House Rules Committee—helped preserve the Democrats' bipolar coalition in the 1940s and 1950s by blocking civil rights bills from reaching a floor vote. Bensel also finds that sometime between 1937 and 1960, Republicans became more supportive of efforts to block civil rights bills while northern Democrats became more willing to bypass the committee system in order to force action on civil rights. Bensel, Richard, Sectionalism and American Political Development (Madison: University of Wisconsin Press, 1984), 235–41Google Scholar. The evidence that we present in this section suggests that much of this shift had occurred by the time of the FEPC legislative drives of the mid-1940s. We are planning to study signature patterns on discharge petitions throughout this era to tease out the precise timing of these changes.

121. Maslow, Will, “FEPC: A Case History in Parliamentary Maneuver,” University of Chicago Law Review 13 (1946): 419CrossRefGoogle Scholar.

122. Maslow, “FEPC,” 419.

123. For example, House minority leader Martin (R-MA) announced on December 11, 1945, that enough Republicans would sign the petition to get it to the floor, but, as of May 1, 1946, this promise had not been fulfilled. Maslow, “FEPC,” 419. Fifteen years later, in 1960, the fight to discharge a civil rights bill languishing in the Rules Committee played out in a similar fashion. Democratic Study Group members spoke on the floor for a full day, with many of the speakers alleging that House minority leader Charles Halleck (R-IN) pressured Republicans not to sign the discharge petition, in an effort to keep the GOP's conservative alliance with the South intact. When the petition was leaked to the New York Times, it was revealed that, of the 175 signers, 145 were Democrats (See Berman, The Politics of Civil Rights in the Truman Administration, 118). Also see Kathryn Pearson and Eric Schickler, “Discharge Petitions, Agenda Control, and the Congressional Committee System 1929-1976,” (paper presented at the Annual Meeting of the American Political Science Association, Chicago, IL, Aug. 29–Sept. 2, 2007) on discharge petitions more generally.

124. Patterson, “The Failure of Party Realignment in the South,” 304.

125. Individuals who gave multiple civil rights speeches in 1945 are only counted once.

126. Carmines and Stimson, Issue Evolution, ch. 3.

127. Carmines and Stimson, Issue Evolution, 62.

128. These figures were obtained by examining all pages in the 1945 Congressional Record that were mentioned in the index under the headings “colored rights,” “Negro rights,” “Fair Employment Practices Committee,” and “Fair Employment Practices Commission.” (The term “civil rights” does not appear in the index.) We then determined, where applicable, whether each speaker favored or opposed the civil rights issue in question. Items that were inserted in the Congressional Record but were not discussed on the floor are not included.

129. Carmines and Stimson, Issue Evolution, 18.

130. McMahon, Kevin, Reconsidering Roosevelt on Race: How the Presidency Paved the Road to Brown (Chicago: University of Chicago Press, 2004)Google Scholar.

131. Another reasonable counterfactual—though equally at odds with the issue evolution perspective as it is with ours—is that Dewey wins the 1948 election and continues on a national stage with the pro–civil rights policies that he pursued as governor of New York. While it is impossible to determine how a hypothetical Dewey administration would compare to Truman's—particularly since liberal New York afforded Dewey an entirely different and more pro–civil rights political context than Truman's national context—the behavior of the two candidates during the general election campaign provides for a common context to evaluate this counterfactual. During the campaign, Truman became the first U.S. president to speak in Harlem, where he highlighted his pro–civil rights executive orders; establishment of the Civil Rights Committee; the 1948 Democratic platform's landmark civil rights plank; and his Justice Department's role in outlawing restrictive covenants. According to historian Harvard Sitkoff, these highly publicized acts “established Truman more firmly as the leader of the Second Reconstruction.” (Sitkoff, “Harry Truman and the Election of 1948,” 613). While Dewey would have likely pursued pro–civil rights policies in the White House, it is not at all clear that this hypothetical action would have been more vigorous than Truman's or would have drawn a supportive response from rank-and-file Republicans. Notwithstanding his personal views, Dewey would have had to grapple with GOP leader Martin's concerns about offending the business community and Taft's opposition to an FEPC with enforcement powers. It thus seems implausible that the 1948 election was a “critical moment” that would have significantly altered the parties' relative positions for decades.

132. Black and Black, The Vital South, 193.

133. Lowndes, Joseph, The Southern Origins of Modern Conservatism (New Haven, CT: Yale University Press, 2008)Google Scholar.

134. Orren, Karen and Skowronek, Stephen, The Search for American Political Development (Cambridge, UK: Cambridge University Press, 2004)CrossRefGoogle Scholar.

135. Lee, Mobilizing Public Opinion.

136. See Karen Orren and Stephen Skowronek, The Search for American Political Development; also see Thelen, Kathleen, How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan (Cambridge, UK: Cambridge University Press, 2004)CrossRefGoogle Scholar.

137. Riker, William H., The Art of Political Manipulation (New Haven, CT: Yale University Press, 1986), ch. 1Google Scholar.