In the New York Times, David Brooks attributes the trouble to ideological arrogance at both ends of the Bipolarchy. Ideology, he claims, has undone the political modesty that characterized American politics for generations after the Founding.
According to this mentality, a big successful nation exists in a state of equilibrium between its many factions. This equilibrium is fragile because we are flawed and fallen creatures and can’t quite trust ourselves. So all of us, but especially members of the leadership class, should practice self-restraint. Moral anxiety restrained hubris (don’t think your side possesses the whole truth) and self-indulgence (debt corrupts character). This ethos has dissolved, on left and right. The new mentality sees the country not as an equilibrium, but as a battlefield in which the people, who are pure and virtuous, do battle against the interests or the elites, who stand in the way of the people’s happiness. The ideal leader in this mental system is free from moral anxiety but full of passionate intensity. This leader pushes his troops in lock step before the voracious foe. Each party has its own version of whom the evil elites are, but both feel they’ve more to fear from their enemies than from their own sinfulness. Compromise is thus impossible. Money matters should be negotiable, but how can one compromise with opponents who are the source of all corruption?
How did this happen? History isn't Brooks's subject this week; he's just calling things as he sees them. A potential answer might be found in Jonathan Chait's TRB article for the current New Republic. His subject is the "myth of divided government," the ideal of obligatory cooperation now challenged by pessimistic observers. It's a myth that once was fact, Chait suggests, but "the unraveling of a bipartisan consensus" has changed things decisively. In his penultimate paragraph he sketches an important change that took place in the middle of the 20th century, before which, we might infer, bipolarchy was not yet fully in force despite the two-party system.
Electoral politics in a two-party system inherently creates zero-sum competition. This reality was subordinated for a long time, because the peculiar politics of Southern white supremacy created many decades of ideologically amorphous parties with overlapping policy aims. That, in turn, blinded members of Congress to the fact that their success depended upon the other party's failure.
In other words, as reactionary Southerners abandoned their historic alliance with increasingly liberal Northerners inside the Democratic party, and were sought out by ambitious reactionary Republicans as allies in their ultimately victorious struggle with their own party's "Rockefeller" wing, politics became a clear field of polar ideological opposites -- or at least politicians began describing and promoting it as such. The process can be traced back at least as far as 1964, when several southern states voted for Barry Goldwater, or further back to the Dixiecrat revolt against Harry Truman in 1948. It was still in progress in the 1990s, when you could still hear of "the first Republican since Reconstruction" elected to various Southern offices as those elderly Democratic bosses who had not turned Republican finally departed the scene. Depending on your viewpoint, this may have been a good thing because it freed Democrats to become, or at least pretend occasionally to be, a true left-wing party, but it also allowed Republicans to stigmatize them as a left-wing party. Between them, of course, the parties excluded the true left from the national discussion as much as possible. In any event, the Southern turn made it easier for the two parties to sell themselves as irreconcilable ideological opposites who between them (implicitly) had all the possible answers to national problems. It'd be easy to blame it all on racism, if Chait's brief account convinces you, but widespread kratiaphobia (I'm trying out a neologism meaning "fear of political power") outside the South clearly encouraged Northerners otherwise indifferent to race to ally with Southerners battling for local rule against centralized authority.
However it happened, I'd agree with Chait that something important and unhealthy happened in those years. However, Chait doesn't follow up his conclusion on the unleashed pathology of two-party politics with any challenge to the two-party system. If anything, he suggests that pluralist results are more likely under one-party control, noting that the Obama administration enjoyed "support from a broad and diverse coalition" during its first months in power, when the Chamber of Commerce supported the stimulus. But if one-party rule is a desired good, it will force the parties still deeper into "zero-sum competition," as seems to be the case now. Chait should ask whether Americans can afford to wait for another decisive victory for one party or the other, or whether they should decide against the two-party system.