With the highly partisan votes on tax reform in the House and the Senate (no Democratic legislator voted in favor of either bill; only one GOP Senator and 13 GOP House members voted against the respective bills), partisan polarization has again hit the headlines. A Pew Research Center report in 2014 showed public opinion becoming increasingly polarized in the last 20 years. The same is true for Congress. The National Journal reported recently that, in 1982, 344 House members were ideologically positioned in this middle ground of the House. The National Journal defined the middle as Representative who were ideologically positioned between the most-liberal Republican House member and the most-conservative Democratic House member. Thirty years later, in 2012, almost 95 percent of the middle had evaporated. Only eleven House members, out of 435, remained between the most-liberal Republican and the most-conservative Democrat.
The puzzle is why. Why have the public and legislators become more ideologically polarized? But “political polarization” is simply another way to ask about the ideological coherence of parties and the ideological overlap between the parties. Albeit, “political polarization” sounds like a negative thing while “ideological coherent” parties sounds like a good thing, or at least “ideologically coherent” sounds like a thing that good parties should be. In any event, why are legislators and the public becoming more ideologically coherent; why is the ideological overlap between the parties disappearing?
My own hunch is all the talk of polarization stems from the country returning to what would otherwise be normally expected patterns of partisan identification. It’s just that the normally expected pattern was disrupted. The thing is, the cause of the disruption in normally expected patterns occurred so early in the nation’s history, and had such a long half-life, modern Americans largely assumed that the exception was normal. Thus, we take the return to normality to be the puzzle rather than understanding the puzzle to be that for such a long time the U.S. had political parties with such a large ideological overlap.
The great disruptor was the Civil War. Because of the bitterness of that conflict, for well over a century the U.S. had sectionally-defined parties rather than ideologically-defined parties. Actually, that’s too strong. Here’s a more careful hypothesis: For over a century U.S. political parties have had significant sectional components that, for many partisans, although not all of them, overshadowed intra-party ideological disagreements. This created significant ideological overlap between the two parties.
For over a century, many southerners would never vote Republican because it was the party of Lincoln. It was the “solid South” of “yellow-dog Democrats,” i.e., voters who would rather vote for a yellow dog than for a Republican. In Texas as late as the 2000s, one could see campaign signs marketing so-and-so as a “Conservative Democrat” for political office.
Republican partisanship in the north was never quite as monolithic, partly because of a large Democratic presence in large northern cities. Nonetheless, plenty of northerners in rural areas and small towns would never vote Democratic, the party of “rum, Romanism, and rebellion.”
The bitter antipathies created by the war lasted almost a century. Indeed, the “solid south” began breaking up, and then gradually, with the Goldwater campaign in the 1964. Trying to nip the bud, a group of liberal Republicans founded the Ripon Society in the early 1960s to promote the presidential candidacy of Nelson Rockefeller. The Rockefeller wing of the Republican Party pretty much died about the same time he did.
A century or more of political parties organized with significant sectional components meant plenty of conservative Democrats and plenty of liberal Republicans during that time. As a result, the parties had a lot of ideological overlap. In today’s parlance, that meant there was little partisan polarization.
After a century of partisan identities tied to Civil War divisions, memories began to fade. At least memories linking party labels to the conflict. The children of once-Democratic Southern conservatives increasingly sorted into the Republican Party; the children of once-Republican Northern liberals increasingly sorted into the Democratic Party. With diminishing sectional components, both parties lurched toward greater ideological coherence. But greater ideological coherence meant less ideological overlap between voters and office holders who identify with the respective parties. Partisan polarization.
The irony is that Americans lived so long with political parties that had strong sectional components, sectional components that blurred ideological distinctions, Americans got used to the overlap. We thought it the natural state of affairs. Indeed, it was the normal empirical state of affairs. But once the partisan remnants of America’s bitter sectional divide began to dissipate, American political parties increasingly became simply coalitions of more or less like-minded individuals. To be sure, parties are extremely broad coalitions. But the measure is relative to the other party. So while there is ideological distance between the Sanders wing of the Democratic Party and the Clinton wing of the Party, there is nonetheless less distance between either of those wings and the Republicans. So, too, for now at least, the distance between the Trump wing of the Republicans, and the libertarian, Main Street, and social conservative wings of the Republicans is less than the distance between any of those groups and the Democrats.
The Civil War hypothesis doesn’t explain everything about polarization. It’s not intended to. There remains the issue of “asymmetric polarization,” that is, why Republicans seem to be moving to the right faster than the Democrats are moving to the left, and the issue of intra-party polarization, which we see in both parties. Nonetheless, writ large, my hunch is that increasing polarization reflects the displacement of sectional commitments in the parties, making them more cohesively ideological.