The political geography and electoral consequences of the Slavery and Civil Rights Eras in American history.

摘要

Scholars charge the study of political realignments, or critical elections, as non-falsifiable events so rare that no two realignments share the same catalyst. Despite those noteworthy criticisms, non-direct ways to look for evidence of realignment are valuable and there are two historic periods, the Slavery and the Civil Rights Eras, which share shifts in racial policy as the catalyst. Armed with caches of data and novel spatial statistics, political geographers are moving this research program forward. They confront troublesome global analyses with a recognition and demonstration of the more refined, local nature of electoral geography. Following them, I thread racial issues through time periods separated by more than 100 years using the theoretical structure of issue evolution proffered by Carmines and Stimson (1989) and the power theory prevalent in political science. The Moran's I cluster statistic and geographically weighted regressions (GWR), a local modeling technique, drive the analyses. Results do not succinctly follow the patterns predicted in the issue evolution framework however through the Moran's I, there is solid evidence that the Civil Rights Era was a political realignment. The Slavery Era dynamics are tested and the innovative application of the Moran's I is advanced as a viable statistic for measuring realignment. The power theory and subsequent intergroup conflict/threat hypothesis is applied to both eras. The Slavery Era GWR analyses, negating some historic narratives, demonstrate the 1860 election as a slavery-driven ideological vote. Other historic discrepancies are examined. The Civil Rights Era GWR analyses shows strong evidence of a threat response and supports the contention that the election was a realigning one. Overall, analyses strengthen and elucidate the argument that global analyses obscure important patterns and details and support the importance of investigating electoral behavior at a localized scale.