...will have an ongoing optics problem. All the proposals of the Partisan Party will be bipartisan. That is, a few members of the other party will, predictably, peel off and cross the aisle to stands with the Partisans. None of the proposals of the Bipartisan Party, on the other hand, will ever be bipartisan...Result: the Partisan party, thanks to its unremitting opposition to bipartisanship, will be able to present itself as the party of bipartisanship, and be able to critique the Bipartisan Party, with considerable force and conviction, as the hypocritically hyperpartisan party of pure partisanship.Again, I'd say -- true enough. I'd add yet another problem for the less-unified party; they will constantly be facing stories about how disorganized they are. But both parties will have to fight hard to keep their marginal voters in line (it's just a different set of marginals; the unified party will be dealing with the last one or two potential dissenters, while the less-unified party will be dealing with).
But wait! Did you notice that little "all things equal" I stuck in above? All things, it will readily be apparent, are not equal. Parties cannot be simply "unified" in the abstract. Parties are going to behave more unified, over the long run, if they actually are more unified -- that is, if they fundamentally agree on things. In the short run, of course, a particularly effective leader might make a bit of difference, but over time, a more diverse party is going to act as if it was more diverse. But that doesn't mean it is destined to lose. In fact, the unified party is, and again all things equal, destined to be the smaller of the two parties. After all, the diverse party can fight for practically any constituency, while the unified party -- in order to stay unified -- must surrender any constituencies who oppose the principles or issues around which it unites.
You might ask, then: can't that yield a death spiral in which the unified party constantly narrows itself, purging those who were formerly considered loyal but now are on the fringes of the (ever-shrinking) party mainstream? Ah, here we're on ground that I've covered before, but it's worth going over again because it's potentially very important. It shouldn't, because of the electoral incentive. Normal parties want to win. In fact, one way of looking at democracy is that its an ingenious system for coordinating incentives of self-interested individuals for the public good through the electoral incentive. See, in a normal situation, everyone within a party wants to win elections. Politicians and those who want to serve on their staffs want to win because their careers depend on it directly: they need to win to be employed. Electioneering professionals want to win because it helps their reputations, which means more and more lucrative future clients. Party-associated interest groups want to win in order for their policy demands to be satisfied. Even "purist" activists, who may choose pure stances on issues of public policy over victory, still prefer winning to losing, even if it might not be their highest priority. As long as winning is more important than party unity, then, there's no real danger of a death spiral. A party at risk of losing elections will give up unity as a strategy if it is costing them seats.
The problem for the GOP right now, as I and others have said, is that it's not clear that the electoral incentive applies. Important portions of the Republican network appear to have an incentive to be in the minority, because it's good for book sales, TV and radio ratings, and page views. Others may find that its just as lucrative, if not more so, to organize Tea Party protesters than it is to organize winning electoral campaigns. Even candidates may not be fully dedicated to winning if they know there's a very soft landing available to them as lobbyists, or as Fox News contributors.
OK, back to the effects of unity, and moving into more speculative material. Interest groups, one would think, still have an incentive for their party to win. That's why I've been so interested in the defection of doctors and groups during the health care fight. If Republicans would prefer symbolic victories over actually affecting policy, it seems to me that their associated interest groups may well defect to the Democrats; better to fight for policy as a minority faction in the majority party than to control the issue positions of the minority party if it is truly a minority party, not just the (temporary) out-party. But at the same time, the "unity" party may well gravitate anyway to purely symbolic issues over substantive issues, since its easier to unite against flag burning or for a balanced budget (in the abstract, as in a balanced budget amendment) than it is to unite over complex policy, which tends to have winners and losers. It certainly is my impression that Republicans are relatively more interested in symbolic issues than are the Democrats, but that's really just a guess -- one could, however, go through party platforms or some such exercise to check on it, but I'm not going to, at least this week.
Last thing: what all this suggests is that "unity" might well be best seen in the abstract not as a potentially good strategy, but as an effect of a party that is shrinking, especially a party that is shrinking because it has become dangerously divorced from normal electoral incentives. Is that what's actually happening to the Republicans right now? I don't know! I do think, however, that it's rapidly becoming probably the biggest current question worth exploring in the empirical or theoretical study of American political parties. I am sure, however, that against that possibility, the "optics" (man I hate that word) of partisanship and bipartisanship is not at all important.
(Update: small edit for a collapsed sentence. Gotta stop posting while sleepy!).