Calpundit has an in-depth post about classifying political ideologies, spinning off from a Matthew Yglesias post on the same topic. They're both concerned about the limitations of the two salient ways of classifying politics these days -- either the popular (especially in the media) single liberal/left-conservative axis, or the two-axis scheme proposed most forcefully by the libertarians.

A brief summary of the libertarian two-axis scheme: people are divided into four groups based on how much freedom from the government people should have in social or economic issues. So you get conservatives advocating government involvement in social but not economic issues, liberals/leftists advocating government involvement in economic but not social issues, authoritarians advocating government control of everything, and libertarians wanting no government involvement. It's become the standard criticism of this model that it's designed to make libertarianism attractive to fence-sitters by using loaded terms like "freedom" and "authoritarian." But I think this scheme is a good way of understanding the political ideology of libertarians, because it reflects how the world looks from their standpoint (I used the diagram for just that purpose in my feminist geography class). Dividing people according to their views on government control reflects libertarians' preoccupation with the issue of freedom from government coercion. To others of us (the reason you don't see liberals arguing that bigger government is necessarily better is not because they're trying to hide their authoritarian side, it's because size of government isn't in an of itself important), however, the libertarian scheme creates some absurdities. For example, it puts advocates of politically correct speech codes in the same basket as people who want to legislate conservative Christian morality. To me, a person's vision for how society ought to be constructed is at least as important as their view on whether the government is an appropriate tool to get there.

In general I tend to be skeptical about neat master-classifications of ideologies (which extends beyond politics to things like the hierarchy-fatalism-egalitarianism-individualism scheme for attitudes toward the environment, or various ways of classifying schools of thought in geography). They can be useful devices for capturing where people stand with reference to certain salient issues, but when looking for a sort of general-purpose classification they run up against the fact that the real world is not so orderly. The terms in which a faction or party defines itself are rarely the same as those on which other factions define themselves -- or the terms the second faction uses to define the first. This point has been impressed upon me just in reading a general history of the United States. The political ideologies and preoccupations of the various pre-Civil War parties and factions defy categorization into any simple scheme I could devise.

So to some degree I think Matthew Yglesias is on the right track by rejecting a deductive approach in favor of an inductive one, and simply asking which party (and I'd add "which faction within the party") a person supports in practice. The parties that exist today are a disorderly reflection of the major pools of political thought. And Matt's scheme is a good way to understand how people's political ideologies will have practical consequences in the world. But at the same time, the party power structure distorts people's political views. I don't mean to suggest that people's ideologies are somehow fully independent of the political power context they operate in. But it seems that, for most purposes that you'd be interested in defining a person's political ideology, there would be a meaningful difference between, say, someone who agrees with all of the Green party's policies but votes Democratic because the Greens don't have a chance, and someone who genuinely agrees with the Democrats' policies. The first person's willingness to vote strategically (or "sell out," from a hardcore Green perspective) in the current environment is important as well, though. It's a difficult situation all around.

I also think Calpundit is doing something worthwhile in challenging readers to come up with their own political classifications. But I think the key is not to treat the exercise as making up an objective, universally applicable model of politics. Rather, it's an issue of defining what the political field looks like from the modeler's perspective. What issues define your position? Where do you see other major factions falling in relation to those issues? I may return and give this problem a shot next year (though this may turn out to be a Josh Marshall promise).