Every now and again, you hear stories about how Democrats “need” to win Southern white votes or working-class white male votes, or about how Republicans “need” to win Hispanic votes or unmarried female votes. And the rejoinder is, basically, that neither Democrats nor Republicans “need” any particular set of votes – they need a majority of votes. All votes (within any given jurisdiction where an election is being held) are equal.

That’s true from the perspective of each party. But it’s not true from the perspective of the voters.

Voters participate in the general interest. But they also have specific interests – interests that may or may not be aligned with the general interest. The assumption embedded in Federalist #10 is that, in a large Republic, the general interest emerges from a competition between specific interests. Whether or not the design of our Republic facilitates this competition and the emergence of the general interest therefrom may be questioned, but the idea that this is where the general interest comes from marks the big divide between liberal constitutional designs that emerge out of Montesquieu and constitutional designs that emerge out of Rousseau (to pick two French examples; I could have picked examples from British or American intellectual history instead).

From the perspective of the two parties, all votes are equal, and the goal is to assemble a majority – more specifically, to assemble a stable, enduring majority for the lowest cost. A key to lowering cost is to achieve an identification between groups of voters and the party. Once this identification is established, and voters are properly sorted, they can be retained at relatively low cost in terms of policy achievement. Because one party is perceived as their “home” while the other party is perceived as inherently hostile, voters can be mobilized simply to prevent the other “team” from winning, without actually expecting their own “team” to advance the ball significantly.

This process, in the interests of each party, is, logically, against the interest of the voters in question. The very natural process of identifying with a particular coalition, which appears to be advancing that group’s influence, is actually the process that limits that group’s impact. The group is being tricked into entering an arms race with voters of an opposed group. And arms races are, in general, irrational outcomes for both sides.

Effective lobbies, therefore, refuse to be trapped in this fashion. They work both sides of the aisle, and are quite public about the fact that they want their position to become a bi-partisan consensus. From ConAgra to AIPAC to the AARP, the goal of effective lobbies isn’t to become identified exclusively with a single team. It’s to move the entire playing field in their direction as much as possible. And that means engaging with both teams, even if you have a preference between them.

So the assumed formulation I began with isn’t just wrong – it’s backwards. Republicans don’t need non-white votes to win elections – non-white voters need to lobby Republicans to support a policy agenda that advances their interests, if only so that the Democrats (which may remain their party of preference) know they can’t take their votes for granted. Democrats don’t need white evangelical votes to win elections – white evangelicals need to lobby Democrats to support a policy agenda that advances their interests, if only so that the Republicans (which may remain their party of preference) know they can’t take their votes for granted. More generally, groups of voters organized around common interests should strenuously resist the temptation to identify with either political party. The more they play it coy, the more likely they are to advance their interests at lower cost. Moreover, and this is crucial, the more they play it coy, the more the two parties will need to pay attention to their participation in the general interest, because they cannot count on them being “captured” by “team” positions with respect to their specific interest.

The route from here to there, unfortunately, isn’t easy to discern. For one thing, the interests of the leadership of such organizations is not perfectly aligned with the interests of the membership. Indeed, in certain ways their interests are in direct conflict. Identification with a particular “team” provides the leadership of the organization so identified with greater opportunity for political advancement – and this higher political profile provides obvious justification for their organization retaining them in positions of leadership, even if there is little evidence that the ball is actually being advanced faster than it would have been with a different strategy. Moreover, even if this conflict could be resolved, as with any arms race there is the difficulty of persuading anyone to make the first move.

But such moves need to be made. Let me end this by slightly dissenting from one of my favorite commentators, Ross Douthat, in his obit for Charles Colson:

We only have two parties in America, and to be active in politics inevitably requires identifying with one more than the other, and probably becoming directly involved with one or the other as well. But given how unlikely it seems that a political party in a fallen world would have a platform that comports precisely with God’s intentions for human affairs, any serious Christian should assume that there are places where his or her party is getting some important issue wrong, or at the very least giving it insufficient attention. And just recognizing those places or issues is not enough: The Christian Republican or the Christian Democrat has the obligation to focus on them as well, to prioritize them and call attention to them even when it annoys or frustrates their co-partisans, in order to demonstrate where their ultimate loyalties really lie.

I dissent only slightly, because I agree with the first sentence, and I agree with last sentence, without reservation. But I don’t think the middle sentence goes far enough. It is not only unlikely that a political party would comport precisely with God’s intentions – it’s inconceivable. Moreover, it’s extremely unlikely that either party could be identified as precisely contrary to God’s intentions – and this latter assumption is the one that is more likely to motivate extreme team-identification. Few conservative Christians, I suspect, think the Republican party is the party of angels – but too many, I fear, are convinced that the Democratic party is irredeemably damned. A committed Christian, precisely because he or she is guided by what he or she believes to be a truth from beyond this world, should logically be less committed to a partisan identity than someone without such commitments – if for no other reason than that he or she should recognize an obligation not to defeat the opposition, but to convert it.