Matthew Yglesias correctly dings Richard Cohen for this passage:
Congressional Republicans have made a stand on the stimulus package, just as they did on the original bank bailout when they refused to accommodate a president of their own party, George W. Bush. These Republicans are as wrong as wrong can be, and history, I am sure, will mock them, but they were not elected by history, and they are impervious to mockery from the likes of me. They come from conservative districts, and they are voting as their people want them to. That’s partisanship. It is also democracy…
Reality is real. No amount of lofty rhetoric is going to change the way members of Congress are elected. Most of them come from exquisitely gerrymandered districts created by computers that could, if good taste allowed, part the marital bed, separating husband from wife if they were of different political parties. This system created districts that are frequently reliably liberal or conservative. The computer has deleted the middle.
As Yglesias notes, “There’s something to this, but it can’t explain the fact that all the House Republicans voted ‘no.’ Nor can it explain the Senate, where many Senators representing states Obama won voted ‘no.’ There’s more to legislative partisanship than manipulation of district boundaries.”
In fact, the evidence suggests that gerrymandering has played a relatively small role in the increase in polarization. As UT-Austin’s Sean Theriault emphasizes in his book 
Both pundits and scholars have blamed increasing levels of partisan conflict and
polarization in Congress on the effects of partisan gerrymandering. We assess whether there is a
strong causal relationship between congressional districting and polarization. We find very little
evidence for such a link. First, we show that congressional polarization is primarily a function of
the differences in how Democrats and Republicans represent the same districts rather than a
function of which districts each party represents or the distribution of constituency preferences.
Second, we conduct simulations to gauge the level of polarization under various “neutral”
districting procedures. We find that the actual levels of polarization are not much higher than
those produced by the simulations. We do find that gerrymandering has increased the Republican
seat share in the House; however, this increase is not an important source of polarization.
I still support redistricting reform to maximize the number of competitive House elections, but as McCarty et al. note, its effects on polarization are likely to be limited given the way the party system works and the apparent growth in geographic sorting.
PS I love the line “Reality is real.” Maybe Cohen has been taking logic lessons from Mike Huckabee or MIMS…
Update 2/18 9:53 AM: See also Emory’s Alan Abramowitz et al., who conclude that “redistricting has not made [House]
elections less competitive,” and Duke’s David Rohde (one of my advisers) and coauthors, who find that redistricting has increased polarization but “the effect is relatively modest.”
Update 2/20 9:46 PM: In this week’s issue, the New Yorker’s Hendrik Hertzberg gets the causes of the death of bipartisanship right but the effects of gerrymandering wrong:
[W]hat might be called America’s parliamentary parties have come to resemble their disciplined European counterparts. As recently as the nineteen-sixties, for reasons of history and origins, the Democrats were a stapled-together collection of Southern reactionaries, big-city hacks, and urban and agrarian liberals; the Republicans were a jumble of troglodyte conservatives, Yankee moderates, and the odd progressive. Ideological incoherence made bipartisanship feasible. The post-civil-rights, post-Vietnam realignment, along with the gerrymandered creation of safe districts, has given us—on Capitol Hill, at least—an almost uniformly rightist G.O.P. and a somewhat less uniformly progressive array of Democrats.