Sunday, September 16, 2007

Reflecting on Irrational Desires

In Chapter 6 of Reasons and Persons, Parfit contrasts three desire-based views of what we have reason to do. The Instrumental Theory (IP) claims that we have most reason to do whatever would best fulfill our present desires. The Deliberative Theory (DP) instead appeals to the desires we would have on ideal reflection. Finally, the Critical Present-aim theory (CP) claims that some desires may be "intrinsically irrational", and others may be rationally required, and we have reason to fulfill only the latter. But it isn't clear to me how DP and CP differ.

After suggesting that we may consider apocalyptic desires (etc.) to be irrational and to provide no reason for acting, Parfit writes (p.119):
[The Deliberative theorist] might insist that his theory is adequate, since those who were thinking clearly and knew the facts would not have such desires.

Whether this is true is hard to predict. And even if it were true, our objection would not be fully met. If certain kinds of desire are intrinsically irrational, any complete theory about rationality ought to claim this. We should not ignore the question of whether there are such desires simply because we hope that, if we are thinking clearly, we shall never have them. If we believe that there can be such desires, we should move from the Deliberative to the Critical version of the Present-aim Theory.

Bizarre. It may be hard to predict whether a desire would be rejected on ideal deliberation, but that is just to say that it is hard to tell whether the desire is truly irrational. For surely persistence on ideal reflection provides the very criterion of what it is for a desire to be rational. (DP then tells us how to derive reasons from rationality. Parfit seems to be doing the opposite: taking reasons as given, and defining 'irrationality' as the objective absence of these.)

Parfit is thus using 'intrinsic rationality' as a term of dogmatic endorsement. He demands that a "complete theory of rationality" be more like the Ten Commandments: an explicit list of which desires are to be deemed 'good' or 'bad', without any need for actual reasoning. But that would be a theory of revelation, not rationality. What we want from our complete theory of rationality is a process of inquiry by which we can discover which desires are ir/rational. We shouldn't expect to be given the answers right at the outset, at least not explicitly. Instead, the theory has implications for what conclusions are or are not rational. This is the only sense in which it "ought to claim" such things.

It's especially strange when Parfit talks about how "[w]e should not ignore the question of whether there are [irrational] desires..." Of course we shouldn't! (Does he really take the Deliberative Theorist to disagree with this?) We should inquire as best we can - undergo a process of rational deliberation, do philosophy - and see where we end up.


  1. This is how I understood Parfit:

    I imagined the deliberative theory to be one like Railton's: He (roughly) claims that desires are not worth fulfilling if I wouldn't continue to hold them after gaining full factual (non-evaluative) information. Thus desires are subject to deliberation and rational criticism, but no desires are ruled out merely on the grounds that people ought not to hold them.

    In contrast, the CP-theorist thinks that some desires are, in themselves, irrational. They might deliberate towards this conclusion, but they can do so by appealing to evaluative facts as well as positive ones.

    It's not that the CP-theorist does not need to deliberate, but rather their conception of deliberation allows greater scope of criticism of desires. Arguably, given this, the deliberative theory is poorly named.

  2. That's an interesting suggestion, though it doesn't seem to fit with what Parfit says in the section I quote above. Note that DP explicitly includes the requirement that one be "thinking clearly" in addition to being in possession of all the relevant empirical facts. And surely for any sense of "rational" whatsoever (however broad), there must be some sense in which having irrational views entails a flaw in one's process of thought, i.e. one is thinking insufficiently clearly.

    Yet Parfit says it is "hard to predict" whether "those who were thinking clearly and knew the facts would not have such desires." What grounds are there for calling something 'irrational' if not that it would be rejected by this rational process?

    So it is this that makes me think that Parfit is mistakenly using the term 'intrinsically rational' when really he means 'intrinsically good', or something along those lines.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)