Disclaimer. Don't rely on these old notes in lieu of reading the literature, but they can jog your memory. As a grad student long ago, my peers and I collaborated to write and exchange summaries of political science research. I posted them to a wiki-style website. "Wikisum" is now dead but archived here. I cannot vouch for these notes' accuracy, nor can I say who wrote them.
Key. 1949. Southern politics in state and nation. Knoxville: University of Tennesee Press (new edition, 1984).
X: The proportion of blacks in an area (size of "black belts")
Y: Lots of things, but especially the dominance of the Democratic party
In their efforts to maintain white rule in the black belts, whites in the black belts have largely made southern politics what they are. In the Civil War, it was these whites (plantation owners, mostly) who won the vote for secession, even though many yeoman whites preferred not to go to war (recall the West Virginia actually left Virginia over this). In the agrarian Populist uprisings of the 1890s, similar cleavages formed; once again, the black-belt whites won. The black-belt whites maintain one-party rule to (1) prevent partisan competition for the black vote and (2) present a unified opposition to any Federal attempts to interfere in racial policy.
Four competing explanations: (1) Demographic (urban-rural); (2) Institutional (county-unit system); (3) Historical (neglect of farmers--top of 118); (4) Personalistic (Talmadge, see 108).
Sure, Talmadge and son built up a major political faction. They would place candidates in other races and, as shown in figure 109, there was sometimes a high correlation between the vote that Talmadge (or son) got and the vote that a crony in another race got. But there was never a legislative-executive coalition among those elected from pro-Talmadge districts that was strong enough to be a proper faction. And although Talmadgism helped unify the opposition, the opposition never became a solid political faction, either.
Much of Talmadge's success stemmed from his ability to win the rural vote (see table on pg 119 for why the rural vote mattered disproportionately) while acting in concert with bankers and merchants against rural interests.
Point: Talmadge may have been a sufficient condition for GA's shift toward two-factionalism. But without the urban-rural split (allowed a consistent base), the county-unit rule (allowed his base to win), and populist history, Talmadge might not have been what he was. He became one of the few demagogues capable of transfering his support to others (109).
Since Jeff Davis, agrarianism died out without leaving much of a residue (184). If a split had remained, some political entrepreneur would have emerged to exploit it. As a result: the best explanation of how much electoral support you get is friends in neighbors: where you're from. They'll vote for you nearby, or they'll vote for you if you pay them off (or persuade the county machine that you're the best).
More than in other states, Arkansas had a true one-party state: there weren't persistent factions within the party or other persistent voting coalitions. Generally, candidates for statewide office succeeded by winning over local politicians (some of whom were bought, some of whom looked for the "best qualified" candidates). There weren't consistent factions because there was general political conformity. The biggest movement that can be called such was one by returning WWII GIs for honest elections. They managed to win local races against corrupt officials, and McMath even won the governorship. Nevertheless, the local GI "revolts" were independent and never became a cohesive movement.
NC has been different since the War. It joined the rebellion only hesitantly, after VA and SC made it an island of southern loyalism. Slavery was less important, and the black belts were smaller, so the minority white areas had less influence. NC ended up being much better for blacks than most of the south; when NC had an educational revival around 1900 (establishment of public universities etc), blacks were also educated. NC's main cleavage has been east-west. NC's cleavage is so sharply factional that it comes as close to being a two-party state as any in the south, though the upper-classes tend to hold the political majority.
X: Small black belt. Y: More liberal politics.
X: Occasional Republican victories (led by Western voters). Y: keep the Democrats organized. (like "conditional party government": if your party is closer to losing power, it will be more cohesive).
Y: Durable competition, group discipline, collective concern for the faction's welfare, and redistribution, which happens more in NC than elsewhere.
(1): One-party states vary greatly (in their degree of factionalism). On one extreme, TN and NC come close to having two parties. The opposition's strength unites the majority. On the other hand, AK and SC are almost pure one-party states. In these states and others, multifactionalism is the norm.
X: Two factions. Y: Cohesive factions
X: Multiple factions. Y: Loose alliances, sometimes grouping around localism, economic and social divisions or the old Populist battle (between poor white farmers and plantation regions), especially during times of crisis.
(2): Limitations of factional leadership: Things that parties do but factions don't.
These effects are strongest in places where there are more factions.
Implied hypothesis: Greater organization leads to greater redistribution (necessary condition) (307): the "haves" can obstruct taxation/redistribution without great organization. But you need organization for the have-nots to get past this obstructionism. In modern "veto points" language: preferences alone (by the rich) can obstruct redistribution; but it takes organization to overcome this veto. In BDM's terms (selectorate theory): do you only need to please the elites, or do you need to please an organized electorate? Organization helps that have-nots at the expense of the haves. (THIS COULD BE TESTED COMPARATIVELY ACROSS STATES MORE THAN IT HAS BEEN).
(3): Isolation from national politics
The South's factional system "fails to provide the political leadership necessary to cope reasonably well with the governmental problems of the South" (310). Usually, the two-party debate over national issues seeps into local campaigns, but in the south it often does not.
"If state politics must be organized fundamentally along the same lines of division as national politics, the maintenance of a disorganized state politics depends fundamentally on a continuation of those conditions that induce southern unity in national politics" (311). In other words: Southern politics would reflect national politics (i.e. two-parties) if it weren't for the fact that certain issues force Southern unity: race politics and the legacy of the War.
In one-party states, the primary is the election. Thus, all southern states have adopted primaries (not conventions), and most have adopted a two-stage primary (a runoff).
When donations come to candidates/factions rather than to a party, the relationship between "donor and donee" is different. This chapter has three sections:
Research by the same authors
Research on similar subjects