Bullshit non-ideal theory and non-bullshit non-ideal theory

This might make no sense to anyone other than political theorists/philosophers, but, well, that’s never stopped half of my specialized audience posts. The following is from an e-mail I recently sent to a colleague, in the context of a discussion about a paper that we’ve both read.

Compare two kinds of non-ideal argument. 1) “In ideal conditions, X would have a duty to do P, but Y has created non-ideal conditions which makes P-doing bad or futile, so X does not have a duty to do P/X has a duty to refrain from P.” 2) “In ideal conditions, X would have a duty to do P, but X has created non-ideal conditions which makes P-doing bad or futile, so X does not have a duty to do P/X has a duty to refrain from P.” It seems intuitively true that 1) is legitimate, while 2) is not. 2 is not legitimate because the non-ideal conditions are the fault of the person who is now asking for an exemption from a moral duty because of them. The more natural response is just to say “X’s duty is to P, and to stop creating the non-ideal creations that make P’s effect go wrong, damnit.” Arguments of 2) form seem like a kind of bad faith (in Sartre’s sense, even) — a denial of one’s own responsibility for the conditions “preventing” one from doing one’s duty. Or, perhaps, it’s better characterized as an illegitimate form of an “ought implies can” move (I am channeling Gerry Cohen today, aren’t I?). Anyway, it seems like total crap.

The argument here — like, as far as I can tell, the rest of the [literature omitted to protect the guilty] is of sort 2) — that is, the same people who are responsible for their [bad voting and economic behavior, leading to non-ideal situation] are also those who are responsible for [the policy that is supposed to be obligatory in ideal conditions and counterproductive in non-ideal conditions], and thus are the object of duties in both cases — that is, those people are the members of the public. Some of this is concealed by the standard [omitted literature] reification of the state, as if [aforementioned] policy is the duty of the state and the political and economic behavior making the situation non-ideal are attributable to a bunch of people, and those two entities for some reason don’t ultimately come down to the same people, viz., a bunch of democratic citizens. But since that reification is wrong, it doesn’t help…

So there, nameless political philosopher who has annoyed me.


9 Responses to “Bullshit non-ideal theory and non-bullshit non-ideal theory”

  1. Matt Says:

    a bit too abstractly put for me to know how to respond. For one, I’d like to know a lot more about how X has “brought about” the situation. Sometimes this is straight-forward, but often it’s not. My worry is that some would extend this approach to the point where they deny the truth that considerations of justice can’t require you to be a sucker.

  2. Richard Says:

    Ordinary, I should give to charity. But I’ve set things up (irreversibly) so that a nuclear bomb will detonate and kill millions if I ever give to charity. So: should I give to charity, in these circumstances? Obviously not.

    My prior act was wrong, of course. But that’s no reason to compound the wrong by making things even worse. So I’m in full agreement with your unnamed philosopher: Bad Voters Should Opt Out.

  3. Paul Gowder Says:

    Matt, yeah, sorry about the abstraction, but I’m trying to avoid slamming a particular quite influential literature and person on the blog in less than a full argument. (Partly, this is a civility/respect issue, and partly this is a “I don’t really want to piss on important people in my own field in public without an extended presentation because I have some minimal sense of self-preservation” issue.) Broadly, the idea is that the state is obligated to do P iff the state is already doing Q, and obligated to not do P if the state is not doing Q, and a world where the state does P and Q together is much more just than the world where P and Q are not done. And the argument that’s making me so steamed is “because voters aren’t going to go for Q, the state ought not to do P.”

  4. Paul Gowder Says:

    Richard, I think there’s a difference between reversible and irreversible non-idealness. In your example, sure, you shouldn’t give to charity, although you’re blameworthy (obviously) for setting up the bomb. But suppose you can defuse the bomb? It’s hard to resist the claim that your obligation is to defuse the bomb and give to charity.

    In your voting post:

    Now, you might claim that people are obligated to vote well. But as a matter of non-ideal theory, it doesn’t follow that they should vote (simpliciter). Whether they should vote depends on whether they actually would vote well or badly.Clearly the ordering of the possible actions, from best to worst, is:
    (1) Vote well
    (2) Don’t vote
    (3) Vote badly
    Merely saying ‘vote’, simpliciter, does not provide sufficient information to determine whether it’s advisable or not. That’s true even if you think option 1 is obligatory, because committing an egregious wrong [3] rather than a minor wrong [2] is all the worse — not something to be advised.

    I’m not sure that’s right. Suppose someone is a pyromaniac and a kleptomaniac, and can satisfy the various antisocial urges he has either way. We’d say, I presume, that the similar moral ordering is:
    1) Don’t set fires or steal,
    2) Steal something small,
    3) Set a small fire.

    In such a case, it seems to follow from the fact that he ought not to set fires or steal that he ought not to steal, even though there’s a “non-ideal” sense in which, conditional on him not doing 1) he’d be doing best to do 2). But the “conditional on him not doing 1)” is in his control, so he’s still blameworthy in such a situation… and, more importantly, I think the discussion just stops at the end of 1. We wouldn’t advise our antisocial friend to steal, because we would advise him to resist the damn urges, period.

    Note that Brennan’s paper relies crucially on the high cost of voting well to undermine the notion that it’s obligatory — I think this is absolutely necessary, because it suggests that doing 1) isn’t genuinely in the voters’ control.

  5. Richard Says:

    Here are a bunch of claims (let me know which you disagree with). One should pick the best option available. But non-ideal interference (whether by oneself or others) can change which option would be best. Even so, it may be best to undo the interference, if it is reversible. But that may be so even if the interference was originally caused by someone else. So your “two kinds of non-ideal argument” don’t seem to be tracking the relevant difference here.

    re: my voting post, I’m not sure what you disagree with. I agree that we would recommend the ideal combined act “P and Q”, if that is still an available option. But it doesn’t follow that we should advise P alone. If we know the person is very unlikely to do Q, then we might (i) yes, blame him for this, and (ii) note that this suggests he should no longer do P (unless he manages to do Q after all).

    Again, advising against P doesn’t let him off the hook for failing to meet his obligation to perform the conjunctive act (P and Q). But if he’s not going to do Q, then nothing is gained by trying to pressure him into doing P regardless. P is (ex hypothesi) only worthwhile if done in combination with Q. So to advocate P blindly, regardless of whether in the actual situation it is likely to make things better or worse, would be sheer foolishness.

  6. Paul Gowder Says:

    Richard, I suppose I’m disagreeing with: But non-ideal interference (whether by oneself or others) can change which option would be best. It seems like when the interference is by oneself, one is asking the wrong question — the notion of conditional moral obligations, where the antecedent of the conditional is something that is in one’s own control, misses the point of a moral obligation, which is not to select between blameworthy actions but to command actions that are not blameworthy when those are available to the agent.

    Put a different way, moral obligations ought not to depend on an agent’s character. We ought to insist that an agent with a bad character act entirely like an agent with a good character by doing both Q and P, not accept the agent’s bad character (which, ex hypothesi, makes him less willing to do Q) and advise him to act like a less-bad character by not doing either.

  7. Paul Gowder Says:

    Richard, it occurs to me that this might just come down to a difference between consequentialists and deontologists. I can see how a consequentialist would admit of the notion of a moral ordering where 2) is worse than 1) but less bad than 3), but that doesn’t seem coherent to this deontologist — both 2) and 3) violate a duty…

  8. Richard Says:

    I’ve continued this in a new post

  9. Uncommon Priors » When does non-ideal political theory really exist? How moral and political theory come apart. OR: Why Gerry Cohen is Right About Everything, Part. 9823948790. Says:

    [...] have objected to this style of non-ideal theory before. Richard disagrees. I take Richard’s point, but I think it has a limited sphere of [...]

Leave a Comment