The “Great Man”/Inorganic Theory Of Re-alignment
It is now time to discuss our initial set of maps.
Map A is the map of McCain counties and Huckabee counties from the 2008 primaries. McCain’s counties are in blue, Huckabee’s counties are in red. Map B is the DeMint-Tenenbaum 2004 Senate race. Map C is the second Democratic gubernatorial primary between Burnet Maybank and Wyndham Manning. Counties in red went to Manning, counties in blue went to Maybank. Map D shows the results of a referendum on liberalizing liquor sales. Red counties voted “no,” blue counties voted yes.
In all four maps we see the same basic divisions between piedmont and the coastal plains. All four maps certainly look similar.
But here is the rub. The elections represented in Map A and Map B occurred in 2008 and 2004, respectively.
The Burnet Maybank/Wyndham Manning primary took place in 1938. The liquor referendum took place in 1940 and represented South Carolina’s decision on ending prohibition in that state.
V.O. Key, writing in his seminal work “Southern Politics” in 1948, described the 1938 race thusly:
Regardless of the disposition to explain a sectional vote in terms of patronage, deals, arrangements, alliances, the chances are that the sectional groupings manifest in the 1938 Maybank vote represents a deeper sectional unity that provided a framework within which alliances could be made. . . . [T]he uplanders perhaps still feel an antipathy toward aristocratic Charleston, as it was pictured in the agrarian crusade, an antipathy kept alive by persistent evangelical condemnation of that city as the symbol of all sin. Then, too, Charleston has always insisted on home rule, and its neighboring counties, with a lesser approach to unanimity, oppose prohibition.”
Read that carefully. In 1938, the division in “one-party” South Carolina was between upcountry evangelical populists proselytizing against sin, while the coastal plains voters favored a more liberal, aristocratic approach to governing. The division goes back further; after all, upcountry voters represented the home base of populist “Pitchfork Ben” Tillman, and later of New Dealer Olin Johnston.
The amazing thing is that, seventy years later, the exact same division played itself out in the Republican primary between evangelical, populist Mike Huckabee and the establishment’s candidate, John McCain. Even in the 2004 Senate race, the socially conservative DeMint’s base is shared with Huckabee and, earlier, Col. Manning.
A few more examples are worth mentioning. Examine these maps:
Here, the underlying maps are all the same (I lightened the blue so you could see the crosshatches better). The red counties are counties carried by Bob Corker in his 2006 Senate race. The blue counties are those carried by Harold Ford, Jr.
So what are the cross-hatches? In Map A, the crosshatches represent counties that gave Thomas Dewey a majority of the vote in 1944 against FDR. Note the strong correlation here. In 2006, Bob Corker carried every county that Thomas Dewey carried almost sixty years ago. To be sure, his victory depended on his expanding his coalition, and he did. But it is expanded from the same base Dewey had in the state.
Map B takes us back farther. The crosshatches here represent the counties that Charles Evans Hughes carried against Woodrow Wilson. Again, we see the same pattern. Nearly 100 years later, Corker carried every county that Hughes carried except for two.
Let’s go back even farther. Map C represents counties that voted against secession in 1861. Same pattern.
Part of this, to be sure, is the impact of race. But not all of it. The counties in the West that voted for Corker, Dewey, Hughes, and against secession – those that abut the so-called Western Highland Rim — were heavy slave counties, that nonetheless stuck with the Union, then the Republican party, and do so today. One more example:
In Maps A and B here, red counties are those that Richard Burr carried with a majority of the vote in his 2004 Senate race; blue counties were carried by the Democrat. But the crosshatches in Map A represent counties carried by Warren Gamaliel Harding in 1920. In 1940, the Democrats fared about 20 points better than in 1920, so the crosshatches represent the counties where Wendell Wilkie received 30% of the vote or better. Again, we see the same pattern. In 2004, Republican strength had seeped outwards from the base that the party had in the 1920s and 40s, but it was the same basic base. Burr carried every county that Harding had carried but two and every county where Wilkie received more than 30% of the vote (he got 25% statewide).
This shows the underlying stability in American politics. It should not be surprising that the aftershocks of the Civil War still reverberate today. But that the base of Republican support in Tennessee today is largely the same as it was 150 years ago is shocking to me. More surprising is that cleavages exposed in South Carolina drinking regulation referenda and Democratic primaries 60 years ago still manifest today. Keep in mind that in the 1940 Presidential election, fewer than 100,000 people voted in South Carolina. In 2004, it was over 1.5 million. Yet despite that massive increase in the electorate, and the emergence of true two party politics that are slowly fading back into one-party politics, the map remains basically the same.
This is why I am skeptical of the “great man” theory of re-alignments. Voting patterns change very, very slowly. Though this is something of heresy for a Republican such as myself to say – and to be clear, I consider Reagan a great President, the second greatest of the twentieth century – but he did not make the new alignment. Nor did he re-make the American electorate. Take a look at the following table, which shows the percentage of the electorate that considered itself conservative, moderate, and liberal, taken from exit poll data such as it is available on Wikipedia. As you see, from 1976 to the present, the electorate has been remarkably stable. Reagan’s great success came from taking an electorate where conservatives outnumbered liberals 3:2, recognizing it, and running with it. In other words, he took a conservative America, and made the policies of the Republican party match up nicely with the electorate
Year Lib% Mod% Cons% 1976 18 51 31 1980 18 51 31 1992 19 48 34 1996 19 48 34 2004 21 45 34
Most stories of the modern alignment start around 1964, with the passage of the Civil Rights Act. These accounts would have you believe that the Southern Republican party was suddenly made of whole cloth, and that the most solidly Democratic region in the nation then shifted almost entirely on the basis of race alone to the Republicans, largely as a result of an insidious “Southern Strategy” practiced by Nixon in 1968, and by Republicans ever since.
There’s something to this, but it is by no means the entire story.
It is also important to note that the ideological underpinnings of the “Reagan re-alignment” were set in motion much, much earlier (and as we’ll discuss tomorrow, the political manifestations of this occurred earlier than commonly assumed as well). In short, we are still living in the aftershocks of the re-alignment set in motion with the Great Depression and FDR’s election in 1932.
In Southern Politics, V.O. Key argues that the “solid south” was a gross misnomer, and that most Southern states had vibrant factions, growing out of the debate between populist Democrats and pro-business, conservative Democrats around the turn of the century. Key even intuited, though he dare not argue it as early as 1948, that but for the strange career of Jim Crow, there might be two party politics in the South. The section that my South Carolina maps were drawn from was even entitled “Latent Bipartisanism Smothered By Racism?”
We can see this play out in the following maps. This first map contains Poole-Rosenthal DW-NOMINATE scores for all Congressional districts for 1902 (on all of the following maps, if you want more detail, you can click on them for a hi-res version).
Basically, DW-NOMINATE is a way of ranking members of Congress by looking at all of their votes. If you really want, I can explain in the comments, but the bottom line is that a dark blue district is represented by a very liberal member, and a dark red district is represented by a very conservative member. Also, for reasons that I can go into in the comments, DW-NOMINATE filters out votes on civil rights issues, so what follows is almost irrespective of what was occurring with the shift of the national Democrats to the left on Civil Rights.
Clearly, the South, with its Jeffersonian distrust of large corporations and support of assistance for hard-pressed agrarian constituents, is the most liberal area of the country at this time period, with the North being a block of solid red. There are some pockets of blue in the North, but these are limited to enclaves of white ethnics in places like Boston, Chicago, and Manhattan.
Next stop: 1922:
Here, we can see the South purpling somewhat, but it is still pretty blue. This is in part because the Populist movement was flaming out, and the red scare was discrediting government intervention in the minds of many Southerners. But also note that the upper Midwest and the West coast are beginning to “purple.” This is related to the rise of progressive Republicanism in those areas as a distinct contrast from conservative Republicanism.
By the time of the 72nd Congress (1930-1932), the Great Depression was in full swing.
We see further “purpling” in the upper Midwest and somewhat in the West Coast, while the South is still fairly blue, though it is noticeably more conservative than it was in the 1900s and 1920s.
Enter FDR. FDR’s political strategy was forged in three Presidential elections: 1920, 1924 and 1928. In 1920, FDR as the Dems’ Vice Presidential candidate suffered what is still worst popular vote loss in U.S. History, losing 34%-60% to Warren G. Harding. But that was the low point for the Democrats. In 1924, FDR noticed that the Republicans were held to 54% of the vote, with 28.8% for the Democrats and 16.6% for the Progressives. Combined, the Progressives and Democrats added up to 45% of the vote. In 1928, although the Democrats lost badly, the nomination of Catholic Al Smith (while depressing Democratic turnout in the South) had turned out white ethnic voters in droves, and had flipped MA and RI to the Democratic column. While this does not seem significant today, it was extraordinary without the benefit of hindsight. After all, except for the badly divided election of 1912, no Democrat had carried either of these states since 1836 (no Democrat had carried MA in a contested election (save 1912) since 1804).
So what if Roosevelt could take the rump Democratic party of Southern Jeffersonians, keep the newly-engorged big city machines, and add to it Progressive Republicans who were disenchanted with their party’s pro-business stance? And with the collapse in the farm economy, maybe add some of midwestern farmers as well?
It worked better than hoped. Roosevelt ran as a national healer, without a clear platform. He was slippery, and sometimes contradictory in what he aimed to do. Even in office, he veered back-and-forth between conservative and liberal tendencies, offering the repeal of Prohibition for Catholics, strengthening unions for the city machines and progressives, but at the same time resisting the more left-leaning policies of Huey Long. Heck, even Ayn Rand supported him in 1932. And he won a landslide.
But when he finally planted himself on the left in 1934, he set forces in motion that reverberate to this day. By the end of the year, all three of the last Democratic Presidential nominees had publicly repudiated him. His moves to the left were vehemently opposed by Southern Democrats, who increasingly abandoned the President. As he moved into his second term, notwithstanding his massive majorities, he lost control of the Congress.
In 1937, after the Fed raised reserve requirements to ghastly levels to combat phantom inflation, the economy went into another spiral. Democrats lost badly in 1938, and New Dealers within the party fared even worse. Out of the ashes of the 1938 election rose a new governing force in Congress: The Conservative Coalition.
Southern Democrat Richard Russell paired with Republican Robert Taft to exercise virtual veto power over liberal legislation in the Senate. Similar coalitions formed in the House. New Dealers fared well in the North, but in the South they became virtually non-existent. And the liberal domestic agenda was finished for thirty years. By 1942, the ideological map of Congress looked like this:
Look at how much the South has changed! By now it is almost completely purple, and even has some areas in Texas, Louisiana and Mississippi that are more red than blue. This is because these Congressmen were voting more with Republicans than with Democrats. This is how much of the New Deal came to be dismantled in the 40s, and how Roosevelt reached the point that a veto of a bill lowering taxes was overridden in a Congress he controlled. In other words, while the Democrats generally controlled the House in the 40s, it was not a liberal House by any means. Truman’s Fair Deal went nowhere because of this coalition. In fact, before the 1944 election, Roosevelt mused that the future of the Democratic party might lay in a combination of machine Democrats, liberal Democrats, and progressive Republicans, while conservative Democrats and Republicans merged together to form their own parties. It took six decades, but he was correct.
Note also that the change isn’t limited to the South. The Northern industrial areas like Pittsburgh, Chicago, and now the outer boroughs of New York are blue. What Republicans there are are distinctively purple. And even though they were still largely Republican, the West coast and upper Midwest are ideologically not that far from Democrats. Fast forward to 1962.
By now, of course, civil rights are a major issue, but of course as I noted above, those votes are basically filtered out. There are also now twelve Republicans from the South, two in Texas, two in Virginia, four in Tennessee, two in North Carolina, and two in Florida.
But that doesn’t explain the “redness” of many of the Southern Districts. At this point, a near-majority of Southern Democrats have voting records that place them on the right side of the ideological spectrum, even excluding civil rights votes.
In the present Congress, a DW-NOMINATE score of -.2 (on a scale of roughly -1 to 1) puts you around Gene Taylor, Melissa Bean, or Henry Cuellar – pretty darned conservative Democrats. Of the 96 Southern Democrats in the 88th Congress, only 18 are to the left of that. In the 110th, only a single Democrat has a DW-NOMINATE score to the right of -.1 (Barrow). In the 88th, 61 of 96 – almost 2/3 of the Southern Democrats – are to the right of -.1.
The continued purpling of the North – especially the Northeast – is also evident – unlike the South, those 50/50 districts are held by Republicans, not Democrats. The other districts tend to be very blue. And by 1972, the map really begins to flip:
And this is how re-alignment really occurred in the South. It was (and still is) a process lasting nearly a century. Yes, the “Southern Strategy” was an important part of it, but it was not the only part, and arguably was not the major part. The re-alignment was organic, and had to do with the South’s agrarian, rural culture finding itself pitted against a party that increasingly found itself rooted in industrial, modern cities. As the South became increasingly detached from a Democratic party that continued moving Leftward, it was bound to find a home in a different party sooner rather than later.
In many Southern states, there were longstanding divides between conservative Democrats and populist Democrats, dating back to the turn of the century. Once Jim Crow was weakened in the 50s, and then removed in the 60s, the states with such well-developed factions quickly developed a two-party system. Virginia was the first state to move into the Republican fold, where a Republican base in the Southwest and Northeast was eventually padded by a Democratic party that was split between conservative Byrd Democrats and progressive Democrats in some of the cities. Indeed, from 1948-1964, Republicans regularly won the tidewater-based 1st district and the Northern-Virginia-based Tenth, and occasionally won the 6th district in the Shenandoah.
States such as Texas, North Carolina and Tennessee all had rump Republican parties and Democratic parties riddled by de facto two-party politics in the 1930s and 40s, which evolved into true two-party politics in the 1950s. Tennessee had a strong Republican base in the East (little known fact – the Second Congressional district has not been represented by a Democrat since 1854, the longest continuous representation by a single party in the country). Texas had a growing Republican base in the cities and elected a pair of Republican Congressman in the 1950s and 60s, and a Republican to the seat held by Lyndon Johnson well before the Civil Rights Act of 1964 was passed. North Carolina likewise elected its first Republican Congressman since the Depression in 1952, and elected a pair in 1962.
Southern states such as Alabama, Mississippi, Florida, and Georgia, had no residual Republican base, and Alabama and Florida lacked strong geographical divides. They re-aligned fully at the federal level at later times (Florida is something of a special case, given that it was rapidly filling up with wealthy Northerners).
Arkansas and Louisiana were states without strong regional divides, and without a rump Republican party with which the more conservative faction could join in the wake of Jim Crow’s demise (and it should be noted that conservative does not mean racially intolerant – some of the most repugnant racists were from the populist wing of the party which mostly kept its Democratic affiliation). They were the last states to re-align. Arkansas has actually not yet re-aligned at the state level in any meaningful fashion, and is only marginally Republican at the federal level (there was some historic Republican activity in the Northwest portion of the state, but that’s about it). Key referred to Arkansas as “pure one-party politics” in his book; that truly monolithic Democratic party has been the most successful Democratic party in the South to date.
So the South did not realign to the Republicans all at once, and it certainly did not do so under Reagan or as an immediate effect of the civil rights movement. Reagan simply continued a progression that had begun in the 1950s and 1960s. By the 1940s, the South was primed for a two-party system, and by the early 1960s, it was ideologically estranged from the Democrats. And for all the talk of the New Deal majority from 1932-1968, it really only commanded a majority for six of those years (1932-1936 and 1964-1966).
Nor do these maps display any type of “pendulum” – probably the most inappropriate analogy in politics. They display stability. This country has generally had a center-right governing majority for most of the past century, interrupted by some “hiccups” to the left, notably from 1932-1937 and from 1964-66. Power has moved back-and-forth between Republicans and Democrats with some regularity, but in terms of “real change” in the governing philosophy of America, the swings have been overstated.
In other words, Obama is unlikely to manufacture a new progressive majority. Reagan didn’t, and even FDR didn’t. What those Presidents did was recognize an extant majority coalition for them, and give it life.
If the progressive majority is already out there (a claim we shall turn to later), perhaps Obama will be able to give it new life. But all the eloquence in the world will not manufacture for him a majority that does not already exist.