by Jonathan Bernstein

Yesterday's big electoral news was a rumor that Charlie Crist might really bolt and run a third-party campaign for the Senate.  I posted yesterday about the implications for voting in the Senate during the remainder of the current Congress, but I agree with those who believe that these sorts of purist primary challenges are important to the future course of the Republican party.  Here's Ezra Klein:

[I]t would be bad for Democrats -- and I'd say for the country -- if Crist simply loses to Marco Rubio in the Republican primary. The better conservatives get at mounting effective primary challenges against moderate Republicans, the more impossible it is for moderate Republicans serving in Congress to act like anything but hardline conservatives...The only way to stop that trend is to convince Republicans it's bad for them: New York's 23rd already went for a Democrat, and now if Specter leaves and wins, and Crist leaves and wins, that'll really discredit the effectiveness of the primary approach at electing conservative alternatives.

Two things.  First, one of the possible outcomes here is for the Republicans to marginalize themselves so much that they become a clear minority party for a longish stretch of time.  That's not bad for the Democrats! 

Second, I think he's right that the main way to reverse this is to convince Republicans that it's electoral poison to nominate extremists (defined, by the way, just as candidates who support policies that are very popular among conservative Republicans but not popular among the rest of the electorate).  For better or worse, however, I think it's implausible to believe that a Crist third-party win will convince anyone of anything.  Republicans are extremely likely to wake up on November 3 this year with an election they perceive as a landslide; (even if they only win 2-4 seats in the Senate and 20 in the House they're going to think of it as a big win), and consequently they're likely to interpret everything that happened since the 2008 election as helpful.  In fact, Republicans are likely to do well in the 2010 cycle just because the Democrats did so well the last two times (and therefore control lots of swing seats, while Republicans only have to defend seats that survived two good Democratic years), and because the out-party tends to do well in mid-term elections.  

Let's say that Republicans pick up 35 House seats and six Senate seats, but that Crist does bolt, wins as an independent, and caucuses with the Democrats and votes like Joe Lieberman; let's also say that Republicans lose another Senate seat and five or so House seats that they could have won by nominating more moderate candidates.  Conservative activists aren't going to look at those results and conclude that if only they had backed off on various challenges that Republicans might have done even better; they're going to look at those results and declare it a huge victory for conservatives, and reinforce their belief that conservative actions since 2008 -- a rejectionist strategy in Congress, extreme rhetoric against the president on the talk shows and the blogs, and strong conservative candidates across the board -- were electorally smart.

The larger lesson is that politicians, and political actors, tend to interpret political events based on their own biases and interests.  Polling on health care right now is ambiguous, but conservatives are absolutely convinced that the Democratic health care reform bill is massively unpopular, while liberals are equally convinced that it's quite popular.  So conservatives interpreted the 2008 election as a repudiation not of conservatism, but as a repudiation of Congressional and presidential deviations from conservatism (in fact, the 2008 election was more of a reaction to a deep recession than anything else -- of course, that just moves the argument to whether liberal or conservative parties were responsible for the recession).  It's not impossible for pols and activists to learn useful lessons, but the evidence is that they learn slowly and inefficiently.

Bottom line: yes, it is electorally bad for parties to nominate unelectable ideological candidates.  No, Republicans aren't going to stop doing it even if it costs them seats in the 2010 election cycle.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.