Self-defeating theories: are the Self-interest Theory and Consequentialism indirectly self-defeating, and what does this show about practical rationality?

Derek Parfit, answered from the texts and cited to the page.

The position in brief

The Self-interest Theory (S) tells me to do whatever is best for me; Consequentialism (C) tells me to do whatever produces the best outcome impersonally.1 Both are indirectly self-defeating in the following sense: if I always try to do exactly what S or C tells me, I will often do worse than if I had had different motives.23 S is therefore committed to recommending dispositions that S itself would not approve — the never-self-denying problem.

C is similarly committed to recommending dispositions that C itself would not approve; if everyone always tried to maximize aggregate good, things would go worse than if some had agent-relative motives. The argument does not refute S or C in the strict sense — it shows that they are correct, by their own lights, only at the cost of being self-undermining.

The structural point matters: a theory that recommends the dispositions that would undermine its own success has a deep philosophical problem even if it is not strictly false. The dualism of practical reason inherited from Sidgwick — between what is rational from the agent's point of view and what is rational from the impartial point of view — is not resolved by treating either side as obviously correct.

The Present-Aim Theory and Critical Present-Aim Theory developed in RP Part II offer alternative frameworks that take the indirect self-defeat of S and C seriously without simply replacing one with the other.45

Sources

  1. Some of these are moral theories; others are theories about rationality. When applied to some of our decisions, different theories give us different answers. We must then try to decide which is the best theory. Arguments about these theories are of many kinds. One argument is that a theory is self-defeating. This argument, uniquely, needs no assumptions.
    Reasons and Persons, p. 20
  2. C is indirectly collectively self-defeating when it is true that, if some or all of us were pure do-gooders, this would make the outcome worse than it would be if we had certain other motives. This would be a bad effect in C's terms. But this bad effect may not occur. There are few people who are pure do-gooders.
    Reasons and Persons, p. 45
  3. But the outcome would be worse than it would be if we had certain other causally possible sets of motives. I asked whether, when C is indirectly self-defeating, it is failing in its own terms. If it would make the outcome worse if we were always disposed to do what would make the outcome best, C tells us that we should not have this disposition. Since C makes this claim, it is not failing in its own terms.
    Reasons and Persons, p. 66
  4. How S Can Be Indirectly Self-Defeating If we call some theory T, call the aims that it gives us our T-given aims. Call T indirectly individually self-defeating when it is true that, if someone tries to achieve his T-given aims, these aims will be, on the whole, worse achieved. On this definition, we do not simply ask whether a theory is self-defeating.
    Reasons and Persons, p. 22
  5. But I often act in one of two ways, believing that neither would be better for me. In these cases I am not trying to do what will be best for me; I am acting on a more particular desire. And this may be true even when I am doing what I know will be best for me. Suppose that I know that, if I help you, this will be best for me. I may help you because I love you, not because I want to do what will be best for me.
    Reasons and Persons, p. 23

Where the argument is made

  • Reasons and Persons Parts I-II, especially Chapters 1-4 on the Self-interest Theory, Chapters 5-9 on rationality and time
  • 'Later Selves and Moral Principles' (1973)

How the argument was received

The structural critique of egoism in Reasons and Persons Parts I-II is a major contribution to action theory and practical reason. The Dancy volume contains six engagements: Dancy, Gauthier, Jackson, Stocker, Pettit-Smith, Brink. The central question — whether the indirect self-defeat amounts to refutation or merely to a structural problem — divides the commentators. Gauthier's Morals by Agreement (1986) is the contemporary analytic response. The Sidgwickian background remains the unresolved 'dualism of practical reason' that both Sidgwick and Parfit acknowledged. The position remains live; the structural critique reshaped how late-twentieth- century moral philosophy thought about the relation between egoism, utilitarianism, and Kantian alternatives.