Sunday, August 23, 2015

Puzzles re: Kant on the Good Will

Two puzzling arguments from Chapter 1 of Kant's Groundwork:

(1) He begins by suggesting that the only thing "good without qualification" is the good will.  Why? Because (i) anything else could turn out to be instrumentally bad, in the hands (or head) of one who lacked good will, and (ii) undeserved happiness lacks positive value.

The first consideration is very puzzling, because a good will could also turn out to be instrumentally bad (e.g. for a person who is anti-reliable at achieving their goals).  So if to be "good without qualification" is to mean that it can't possibly be of instrumental disvalue then a good will is not good without qualification either.  On the other hand, if we are instead merely concerned with whether something is unconditionally pro tanto good, then his primary argument against other putative goods (e.g. intelligence, courage, and wellbeing) collapses.

The second reason is also puzzling, because it merely suggests that we need a more fine-grained description of the relevant good.  Perhaps something like "innocent (non-sadistic) pleasure", or more generally "the wellbeing of innocent (non-evil) beings", is a more plausible candidate for being unconditionally (yet pro tanto) good.  The positive presence of "good will" is not required to make happiness good, for otherwise we could not account for why the joy of playful puppies (and other non-rational beings) is an inherently good thing.

[Incidentally, the idea that a good will is what matters most is just silly.  Suppose an evil demon threatens to torture all sentient beings forever unless you take a "bumbling evildoer" pill that gives you a malicious will (but also magically prevents you from ever acting on it in a way that harms anyone).  Should you take the pill?  It's a no-brainer.  Preventing universal torture is clearly way more important than having a good will, if you are forced to choose between the two.]

(2) The other baffling section of this chapter is when Kant argues that acting from the motive of duty is morally better than acting from sympathy, and indeed that the latter has no moral worth at all:
We have a duty to be charitably helpful where we can, and many people are so sympathetically constituted that without any motive of vanity or selfishness they find an inner satisfaction in spreading joy and take delight in the contentment of others if they have made it possible. But I maintain that such behaviour, done in that spirit, has no true moral worth, however amiable it may be and however much it accords with duty. It should be classed with ·actions done from· other wants, such as the desire for honour. With luck, someone’s desire for honour may lead to conduct that in fact accords with duty and does good to many people; in that case it deserves praise and encouragement; but it doesn’t deserve high esteem, because the maxim ·on which the person is acting· doesn’t have the moral content of an action done not because the person likes acting in that way but from duty.

This is silly. It's not a matter of luck that concern for the wellbeing of others leads to right action (at least on any sensible account of what makes actions right -- Kant's own theory might well be an exception).  On the contrary, the properly sympathetic agent is reliably sensitive to genuine moral considerations -- they simply needn't conceptualize them as such. (They desire what's good, de re rather than de dicto.)

Kant continues:
Now consider a special case: This person has been a friend to mankind, but his mind has become clouded by a sorrow of his own that has extinguished all feeling for how others are faring. He still has the power to benefit others in distress, but their need leaves him untouched because he is too preoccupied with his own. But now he tears himself out of his dead insensibility and acts charitably purely from duty, without feeling any want or liking so to behave.
Now, for the first time, his conduct has genuine moral worth. Having been deprived by nature of a warm-hearted temperament, this man could find in himself a source from which to give himself a far higher worth than he could have got through such a temperament. It is just here that the worth of character is brought out, which is morally the incomparably highest of all: he is beneficent not from preference but from duty.

Kant here seems to imply, bizarrely, that sympathetic motivation is incompatible with acting from the motive of duty.  Surely the more sensible conclusion for him to draw here would be that the depressed agent was acting from the motive of duty (previously together with sympathy) all along.  Post-depression, this may be the first time that we can know that the agent wasn't motivated solely by sympathy, and hence the first time we can know that the agent's conduct has genuine moral worth (by Kantian lights).  But why would anyone think that this must be the first time the agent's conduct has really had genuine moral worth?

* * *

Am I missing something, or are Kant's arguments regarding the "good will" (arguably the foundation of his whole approach to ethics) really this bad?


  1. Preventing universal torture is clearly way more important than having a good will, if you are forced to choose between the two.

    A Kantian would naturally ask what makes it way more important, and would argue that it is because we think preventing universal torture is the sort of thing that we expect to be appropriate to good will. After all, if it is a matter of what is more important to choose, i.e., to will, in what sense is the question, "Which is more important?" not a question about good will?

    On the sympathy/duty point, I don't know how far it addresses your concern, but it might be useful to consider the speculative counterpart to the practical case: drawing conclusions because they feel rewarding (sympathy) vs. drawing conclusions because they are recognized to be the conclusions that actually follow on rational principles (duty). The parallel can often be quite close in Kant's account.

    1. Thanks, Brandon.

      (1) I'd agree that (generally speaking) what's good or important warrants willing, or would be willed by the good-willed person. But insofar as we're interested in the first-order question of which things are so, my point could be restated as that "having/maintaining a good will" should not be the primary concern of the good-willed person. It is not the most important good or value.

      (2) I'm dubious of the analogy between sympathy and wishful thinking. A more accurate practical analogue of wishful thinking would instead seem to be short-sighted hedonism. The truly sympathetic "friend of mankind", at least as I conceive of him, has an intrinsic desire for others' welfare, not just an egoistic desire for "feeling reward" to which helping others is merely instrumental. (He would, for example, choose that others be helped even if he will immediately lose all memory of having made this choice, and even if the alternative would have led to greater personal hedonic reward.) But I guess it helps to make sense of Kant to interpret him as lacking this conception.

    2. (1) But, again, the problem is how one glosses the question in such a way that it does not boil down to the obviously poorly formed question, "Which is more important, willing what's good in this particular case in which the stakes are high, or willing what's good?"

      (2) It's not surprising that wishful thinking and sympathy are not perfectly parallel; it's the larger genus of motivated reasoning, of which wishful thinking is just one kind, that's the parallel, at least when we talk about reasoning that has rational worth (the parallel to moral worth).

      In addition, Kant would say, I think, that you are equivocating. Sympathy in the proper sense is a feeling (fellow-feeling, as we also call it); and such motivations, as feelings, are in themselves not rationally reflective, and therefore cannot concern (directly) something so abstract as "other's welfare", although it may happen that satisfaction of our sympathetic motivation (i.e., getting the felt reward of following through on the felt inclination) contributes to making choices that do happen to contribute to others' welfare -- others' welfare as reason, not feeling, can consider it. The example you give, for instance, involves things that could not possibly be sensed or felt (since it considers an action under an abstract category involving alternative possibilities in hypothetical situations) and thus is not itself a case of sympathy at all: it is an abstract rational conclusion. Kant takes the distinction between sentiment and reason very, very seriously everywhere in his philosophy.

      In any case, the motivational structures are similar: a good first approximation of what Kant would say about any purported ethical motivation can be derived by asking, "If this ethical motivation were carried through consistently so that it governed how we inquire and draw conclusions, what would that involve? What account of rationality would this account of ethics give us?"

  2. I tried to preview a long comment, but when I attempted to preview it, everything got deleted. The gist of what I was going to say is a response to your second set of objections, and it can be found in Barbara Herman's "On the Value of Acting from the Motive of Duty" and Marcia Baron's "On the Alleged Moral Repugnance of Acting from Duty."

    1. Ok, thanks for the pointers! (And apologies for the frustration of dealing with Blogger's bug-infested commenting system...)


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.)