The fundamental principle of practical reasoning

International Journal of Philosophical Studies

Volumn 6, Issue 2, 1998, Pages 189-209

  Wedgwood, Ralph ^a  

^a MASSACHUSETTS INSTITUTE OF TECHNOLOGY   (United States)

Author keywords

Good; Practical reasoning; Rational choice; Rationality

Indexed keywords

EID: 60949446672     PISSN: 09672559     EISSN: None     Source Type: Journal    
DOI: 10.1080/096725598342109     Document Type: Article

References (19)

1. The genera] idea of this principle seems to be tacitly assumed in Donald Davidson, 'How is Weakness of the Will Possible?', in his Essays on Actions and Events (Oxford: Clarendon Press, 1980)
2. compare also Philip Pettit and Michael Smith, 'Practical Unreason', Mind, 102 (1993), pp. 53-79
3. and David Wiggins, Needs, Values, Truth, 2nd edn (Oxford: Blackwell, 1989), Essays VI and VII
4. Compare Christopher Peacocke, Thoughts: An Essay on Content (Oxford: Blackwell, 1986), p. 46
5. The same point has been made by many writers in many ways. For example, Wiggins presents what is essentially the same point as the first 'mark' of truth, when he says, 'Truth is a primary dimension of assessment for beliefs', Needs, Values, Truth, p. 147
6. Compare Bernard Williams' account of the practical ought in 'Ought and Moral Obligation' and 'Practical Necessity', in his Moral Luck (Cambridge: Cambridge University Press, 1981), especially pp. 118-20 and 124-5
7. For a rational agent, an option typically consists of a general type of behaviour, and some particular stretch of time during which the agent might behave in that way, such that the agent believes herself able to behave in that way at that time. Sometimes, however, an option is a general policy, rather than a particular course of action; in this case the option consists of a general type of behaviour and a type of circumstance in which one might have a policy of behaving in that way. Compare Michael Bratman, Plans, Intentions and Practical Reasoning (Cambridge, Mass.: Harvard University Press, 1989), pp. 56-7
8. This account of what counts as 'enough' options is closer to the optimizing approach than to the satisficing approach. I think, however, that I can still accommodate the cases in which satisficing seems rational: this is mostly because I am concerned with the evaluation of options, not the evaluation of states of affairs. I cannot go deeply into these questions here. (Arguments for satisficing appear in Michael Slote, Beyond Optimizing (Cambridge, Mass.: Harvard University Press, 1989)
9. Martin Hollis, The Cunning of Reason (Cambridge: Cambridge University Press, 1987), chap. 8
10. Jon Elster, Sour Grapes (Cambridge: Cambridge University Press, 1983), pp. 16-18.) Cases where the set of options is infinite and contains no optimal members are irrelevant, since it is a necessary truth that a finite rational being can only actually consider finitely many options. Note that my account of what counts as 'enough' options is not an optimizing account in the sense of implying that one rationally should at all times be doing the best possible thing: the best possible thing would presumably not involve deliberating at all! (An omniscient and perfectly rational being would not need to deliberate in order to determine what the best thing to do was. He or she would just do it, without deliberating.) It is also important to note that the question of whether one has deliberated 'enough' depends on the correct evaluation of higher-order options (viz., continuing to deliberate versus making one's decision straight away). The relative value of these higher-order options is not determined solely by the value of the lower-order options being considered, but by other factors as well
11. It would probably be preferable to replace this second occurrence of 'believe' by 'accept', contrasting 'belief and 'acceptance' in something like the way suggested by Michael Bratman, 'Practical Reasoning and Acceptance in a Context', Mind, 101 (1992), pp. 1-15. If one accepts that something is the case, then one bases one's practical reasoning on the assumption that it is the case. Acceptance differs from belief in being context-sensitive in a way in which belief is not: it may be entirely rational to accept that something is the case, for the purposes of making one decision, and at the same time not to accept it for the purposes of making another decision; it may also be entirely rational for what one accepts to be influenced by practical pressures, as well as by evidence for the truth of what one accepts
12. I should note that this need only be a minimalist form of expected utility theory. It need not be committed to the view of many decision theorists that a person's "utility function' is entirely determined by the person's actual preferences. (For a clear presentation of such a 'minimalist' form of expected utility theory, see John Broome, Weighing Goods (Oxford: Blackwell, 1993), chap. 4.)
13. Compare Kant, Critique of Practical Reason, chap. 2
14. and Christine M. Korsgaard, Creating the Kingdom of Ends (Cambridge: Cambridge University Press, 1996), especially p. 122: 'what makes the object of your rational choice good is that it is the object of a rational choice'
15. More generally, my proposal is also incompatible with the idea that this concept of 'a good thing to do' has no content except the bare idea of 'whatever the rational deliberator aims at'. If that were the case, then the idea of the good could not offer any guidance in deliberation; there would have to be some other principles of practical reasoning that would determine whether a course of action really was of the kind that the rational deliberator aims at. (J. David Velleman develops just this objection to attempts (like mine) to base accounts of practical reason on the good, in 'The Possibility of Practical Reason', Ethics, 106 (1996), pp. 694-726.) I shall try to indicate in the third section how the concept of 'a good thing to do' does indeed have more content than this, by exploring its relations to information about our plans and projects, our interests, and moral considerations
16. See Hobbes, Leviathan, I, chap. 6, p. 28
17. and Elements of Laws I (Human Nature), chap. 12, §§1-2
18. compare also Terence Irwin, A ristotle's First Principles (Oxford: Clarendon Press, 1988), chap. 15, §§173-80
19. To record my own opinion: I hold that the mental state standardly expressed by such an evaluative statement is a belief in the fullest sense. Unfortunately, I cannot defend this point here. What I say here is meant to be compatible with non-cognitivist analyses of evaluative attitudes. (For an argument against such non-cognitivist analyses, see my 'Non-Cognitivism, Truth and Logic', Philosophical Studies, 86 (1997), pp. 73-91.)