Why Philippa Foot Changed Her Mind (And I Haven't)

This is a followup to my post on Philippa Foot's paper "Morality As a System of Hypothetical Imperatives."¹ I already knew Foot later came to repudiate the view expressed in that paper — a view that resonates with me — but I didn't know why she changed her mind or what became her new position. So I set out on a little investigation, half-expecting she would lead me to new insight.

If anyone wants to follow Foot's trail without being biased by my take on things, you'll find a thorough explanation in her book Natural Goodness, especially Chapter 4. It will also be very helpful to read Warren Quinn's paper "Rationality and the Human Good," included in the collection Morality and Action which Foot put together after his death. Natural Goodness is also dedicated to Quinn's memory.

[Added] Neil S. points out that Foot wrote a one page "Recantation" in Moral Discourse and Practice, edited by Darwall, Gibbard, and Railton.

Acting Morally and Acting Rationally

In "Morality As a System of Hypothetical Imperatives," Foot wrote:

"Irrational actions are those in which a man in some way defeats his own purposes, doing what is calculated to be disadvantageous or to frustrate his ends. Immorality does not necessarily involve any such thing."

It's this definition of what it means for an action to be rational that Foot later questioned and rejected. In other words, she rejected the idea that a person could conceivably act rationally yet immorally by pursuing his own desires and interests. In Natural Goodness, Foot looks back to her earlier paper:

"I now wonder why, given the obvious indigestibility of the idea that morality is indeed a system of hypothetical imperatives, I should have accepted it even for a short while. What seemed to force it on me was the sheer difficulty of showing a practical rationality that was independent of desire or interest."²

So she implies she has found a better definition of what it means for an action to be rational, one which goes beyond acting toward one's own desires and interests. In fact, one which includes moral considerations.

"Can it be the case that someone who does what is wrong thereby acts in a way that is contrary to reason? May we add considerations that are about right and wrong to the list of rationalizing considerations given above?"³

Foot's inspiration for answering these questions with a 'yes' comes from Warren Quinn's article "Rationality and the Human Good."

The Virtue of Rationality

Quinn tries to show that if we accept a definition of rationality which is indifferent to moral concerns — what he calls "Neo-Humean rationality" — we cannot at the same time regard rationality as "our chief excellence as agents."⁴ He argues that if rationality allows (or even recommends) a contemptible action, then such rationality itself must share in that contemptible quality. He goes on to say, "[S]urely it is doubtful whether either the excellent or the worthwhile can be contemptible. The ideas do not seem to cohabit comfortably in the same logical space."⁵

Back to Foot:

"And Quinn asked, in the crucial sentence of the article, what then would be so important about practical rationality? In effect he is pointing to our taken-for-granted, barely noticed assumption that practical rationality has the status of a kind of master virtue, in order to show that we cannot in consistency with ourselves think the that Humean account of it is true. Seeing this as a move of great originality and extreme importance, I asked myself why [...] it should be supposed that there is [a morally independent concept of rationality] with which the requirements of moral goodness must somehow be shown to be consonant."⁶

Both Foot and Quinn are viewing morally indifferent theories of rational action — like the one Foot used to hold — as setting up rationality as an authority over morality. Not only that, but a supposedly virtuous authority which turns out to not be so virtuous. The fix, then, is to stop putting the cart before the horse (Foot's actual analogy!) and make goodness the "master virtue" which restricts what it means for an action to be rational.

"This now seems to me to be the correct way of meeting the challenge that I myself issued in 'Morality as a System of Hypothetical Imperatives' and that at the time despaired of meeting: namely, to show the rationality of acting, even against desire and self-interest, on a demand of morality. The argument depends on the change of direction that Quinn suggested: seeing goodness as setting a necessary condition of practical rationality and therefore as at least a part-determinant of the thing itself."⁷

That's...quite a change. From what I can gather, a breakdown of Foot's new definition of rational action would be:

1. An action is rational if demanded by morality, or if it is morally permissible and promotes one's own desires/interests.
2. An action is irrational if forbidden by morality, or if it is morally permissible but works against one's own desires/interests.

An Example

To lend intuitive support to rejecting a "present desire" account of rational action, Foot points out that we have no qualms about saying a person is acting irrationally if he indulges in a trivial pleasure which risks his future health.

But do we necessarily consider such behavior irrational? Driving to the movie theater is a risk to my future health. Is skydiving irrational? What about people who willfully sacrifice their lives for a cause they presently support? If one's present desire for something risky outweighs present desire for future security, I don't think we can so quickly dismiss the risky option as 'irrational.'

Even if we do find "present desire" theories of rationality insufficient, it would be a small adjustment to consider theories which take into account both present and future desire. We would still be squarely in a desire/interest based understanding of rational action. Foot only threw in the 'present' qualifier for her supposed counter-example.

But, again, even if we do find desire/interest theories insufficient overall, this doesn't automatically make the case that practical rationality is mostly about taking moral actions and avoiding immoral actions. Why doesn't Foot give readers an example of acting against personal desire/interest to perform a morally required action which even 'Neo-Humeans' would have a hard time calling irrational? That would be an intuition lending positive support to her new position.

The Root of the Problem

My investigation has been informative, but not enlightening. I still maintain that the rational action and the moral action may be at odds, if a person truly has no desires or interests which bring the two together. Suppose I have the option of gaining a million dollars, but I know it would require the death of ten healthy people I've never met. If I take the million dollars, I haven't necessarily demonstrated faulty reasoning; what I have shown is the moral failing of preferring a million dollars over the continuance of ten lives.

Why would Quinn and Foot feel so uncomfortable judging rationality and morality in separate terms? I suspect it boils down to a presupposition Quinn implies in his paper:

"I spoke earlier of rationality as authoritative. By the authority of one excellence over another I mean the ability of the former to prevail over the latter in determining what the agent should, in some unqualified and unrelativized sense, do."⁸

I deny that any "unqualified and unrelativized" sense of 'should' is valid. Or, positively, I affirm that normative 'should,' 'ought,' 'must not,' etc. uniformly imply a related end. See my post on Stephen Finlay's paper "Oughts and Ends."⁹ I'm comfortable with the possibility that what a person should_rationally do is different from what a person should_morally do, but Quinn would insist on a further answer about what a person simply should do. As far as I can tell, this is like insisting on being told which number is produced when dividing an integer by zero; if terms like 'should' are relative to an end, demanding a non-relative 'should' is futile.

Stepping Back

How does all this apply to the question of whether moral imperatives are hypothetical or categorical? If you recall, the essential difference between the two is that hypothetical imperatives are necessary to fulfill a desire/end (actual or potential) while categorical imperatives don't refer to anything beyond themselves.

Foot spent time in her earlier paper showing how moral virtues can be properly thought of as having ends, which people may or may not desire. If this counted in favor of characterizing morality as a system of hypothetical imperatives back then, I don't see how Foot's new position about rationality would change that. It would merely be irrational to flout hypothetical imperatives, even when one's own interests or desires alone encourage doing so.

Overall, I get the feeling much of the conversation is a fight over terminology. Does the word 'rational' include moral considerations? We could define it either way. Foot was lambasted for suggesting morality is "a system of hypothetical imperatives," even though I doubt most of her critics would have as much of a problem with the idea that moral virtues have ends; they were probably upset because they understood "hypothetical imperative" to mean something else. Foot's repudiation of her earlier paper did, in fact, focus on something else. I'm much more interested in sorting out ideas than getting lost in the words.

At least I did find the answers I set out to find; not a bad result!


1. http://wordsideasandthings.blogspot.com/2011/01/on-morality-as-system-of-hypothetical_11.html
2. Foot, Philippa (2001). Natural goodness. Oxford: Oxford University Press. p 61.
3. ibid. p 62.
4. Quinn, Warren (1993) Rationality and the Human Good. In P. Foot (Ed.), Morality and action. New York: Cambridge University Press. p 221.
5. ibid. p 226.
6. Foot (2001). p 62. 
7. ibid. p 63.
8. Quinn (1993). p 212.
9. http://wordsideasandthings.blogspot.com/2010/12/on-oughts-and-ends.html