Jackson at AOTP pointed me to this post from "The Great" Tyler Cowen regarding Cowen's theory of political coalitions. Cowen notes the ways in which certain elements of each "side" (most obviously, politicians) changes its arguments tend to result in extremely inconsistent arguments even within the same issue (a critical element of what I would call "pu-pu platter partisanship."). He suggests that this indicates that the two "sides" are less concerned with specific principles than they are about preventing certain (non-coalition member) groups from rising in relative status:
Take the so-called "right wing." I believe that some people on the right do not like those they perceive as "whiners." They do not want these whiners to rise in relative status. That means they must argue against the whining and also they must argue against the presuppositions behind the whining.
Take the so-called "left wing." Some of these people favor a kind of meritocracy. They feel it is unfair that money so determines access in capitalist society and they do not want the monied class to rise in relative status, certainly not above the status of the smart people and the virtuous people. It is important to fight for the principle that the desires of this monied class have a relatively low priority in the social ranking. Egalitarianism is the rhetoric of the day, and readjusting the status of other Americans to the status of this monied class often receives more attention than elevating the very poorest in the world to a higher absolute level.
This excerpt doesn't really do the whole post justice, but I don't want to quote the entire thing. I think to a large extent Cowen is on to something here. Certainly, his theory goes a long way to explaining "pu-pu platter partisanship" (i.e., the holding of beliefs that apparently lack any discernible philosophical principle).
Where I think he goes (slightly) wrong is in articulating the ultimate goal of these group members as being to prevent certain groups from rising in status. Instead, I think the crux of the matter is simply that there is nothing that can hold a coalition together more than a perception of an imminent threat. An atmosphere in which the values or way of life of particular coalition members are "under attack" is an atmosphere in which those coalition members have great cause to fight back against those threats, especially by voting against the coalition that contains the source of those threats.
Importantly, if you have a perception that your way of life is under threat from the members of one coalition ("Party A"), then suddenly your primary goal in political life is to prevent "Party A" from gaining or holding on to power. The result of this, importantly, is that the voter/coalition member will become a dedicated voter of "Party B," as long as "Party B" is the coalition most capable of stopping "Party A." What matters, then, is not so much whether you agree with "Party B," but rather whether "Party B" is big enough to stop "Party A." It is a magnificent deterrent to voting for third parties; it is also a magnificent way of insulating the leadership of "Party B" from criticism by coalition members - it ceases to be particularly important that the leadership of "Party B" agree with its constituents, but is instead only important that the leadership of "Party B" is not "Party A."
In other words, if you feel threatened by "Group X" within "Party A," then you will support "Party B" so long as "Party B" claims to be against "Group X" and to want to restrict "Group X." You suddenly care less about keeping "Party B" honest and holding them to the fire when they deviate from principle than you care about supporting "Party B" because of your fear of what will happen if "Party A" maintains or obtains power. So you don't recognize or, at least, care enough to leave "Party B" when "Party B" does something that either actively hurts you or at least actively prevents you from achieving your policy goals.
The theory Cowen advances strikes me as being one in which politics for certain people is primarily a function of disdain for particular groups or a belief in which those groups are essentially beneath the coalition members of "Party B." I would instead characterize the issue as being one in which politics for these individuals is primarily a function of fear of, rather than disdain for, particular groups. This may only be a difference in semantics, as the effects of both characterizations are similar. But I think it's a noteworthy difference nonetheless.