Saturday, September 10, 2011

Voting as Sufficient, Con't

I just read a healthy-sized essay by John Rawls on the question of what creates an obligation to obey the law. But here's my problem: he's stipulating that we're talking about constitutional democracy, and indeed, in most of the scenarios he sketches out the person who has to decide whether or not they are obliged to follow a law they disagree with has in fact voted. A typical example: suppose you have a referendum or direct-democratic process for choosing between two income tax schemes, Person X votes for Law A, but the majority votes for Law B and therefore Law B is enacted. But it seems to me that, in such a situation, there is nothing more to say about whether Person X has an obligation to obey the law. Presumably constitutional provisions specify that, as a result of this referendum, the tax code that receives majority support will be enacted. If you vote in that referendum, you know that, or at least you ought to know that. Therefore, by participating in the process, you accept that. Honestly, what's so hard about this? I'm open to arguments that the obligation of someone who has never voted requires a more complex theory, but it simply does not need explaining why someone much obey a law that passes over their vote to the contrary. If "social contract" means nothing else, it means this.


UPDATE: The next essay I'm reading begins with a takedown of Robert Nozick's critique of this theory of Rawls'. Nozick gives several hypothetical examples that he thinks are fair schemes of cooperation but that standard moral intuition would not tell us generate obligation. The problem is that in all of his examples the person he wants us to think Rawls thinks is obliged but he thinks is not obliged is an outsider to the system, someone who had no part in creating the system, has no opportunity to try to alter the system, and has essentially had the benefits of such a system showered upon them from without. John Simmons, the guy writing this essay (which is arguing against Rawls overall), says that the fair-play principle Rawls is arguing for only includes those who are a participant in any given fair scheme of cooperation. But, uhh...

If you're a participant, you're a participant!!!

Do we really need a theory for why participants in a scheme of cooperation have to cooperate? If anything, we would need to come up with some kind of special pleading to give us any opportunity to participate in a scheme of cooperation without doing our fair-share part. Is it really this hard?

No comments:

Post a Comment