Hostname: page-component-cd9895bd7-gxg78 Total loading time: 0 Render date: 2024-12-26T04:49:52.623Z Has data issue: false hasContentIssue false
  • English
  • Français

Prichard's Arguments against Ideal Utilitarianism

Published online by Cambridge University Press:  21 March 2017

ROBERT SHAVER
ROBERT SHAVER*
Affiliation:
University of [email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

I give and evaluate Prichard's many (often obscure and idiosyncratic) arguments against ideal utilitarianism.

Type
Research Article
Information
Utilitas , Volume 30 , Issue 1 , March 2018 , pp. 54 - 72
Check for updates
Copyright
Copyright © Cambridge University Press 2017 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Rashdall, Hastings, The Theory of Good and Evil (London, 1924 [1907])Google Scholar, vol. 2, p. 41. See also Rashdall, Theory, vol. 1, pp. 216-17.

2 Rashdall, Hastings, Ethics (London, 1913)Google Scholar, p. 60. Prichard agrees with an even more extreme version of this history, though not with Rashdall's evaluation – hence the view that moral philosophy rests on a mistake, a mistake he thinks Butler and Kant made as well (for Kant, see MO 208, 215, 216, MM 153-9, to Ross 11.12.36, 14.7.38; for Butler, see MO 208, 216, MM 160). (References to Prichard are to Moral Writings, ed. Jim MacAdam (Oxford, 2002) – B = ‘What is the Basis of Moral Obligation?’; DI = ‘Duty and Interest’ [1928]; DIF = ‘Duty and Ignorance of Fact’ [1932]; K = ‘Kant's Fundamental Principles of the Metaphysic of Morals’; M = ‘Does Moral Philosophy Rest on a Mistake?’ [1912]; MM = ‘Manuscript on Morals’ [1928-45]; MO = ‘Moral Obligation’ [1937] – or to the letters in the Bodleian Library, Oxford, MS English Letters c 131, d 116. (Similarly, Carritt claims that: ‘[n]early all moralists since Plato have attempted . . . to prove that certain acts are right . . . generally . . . by deducing the act from the conception of a good or end which it is to achieve’ (The Theory of Morals (London, 1928), pp. 71-2).)

3 See, for example, Darwall, Stephen, ‘Moore to Stevenson’, Ethics in the History of Western Philosophy, ed. Cavalier, Robert J. (New York, 1989), pp. 366–93CrossRefGoogle Scholar, at 378-82.

4 Similarly, Ross objects to ‘ought to be’ that states of affairs do not have obligations ( Ross, W. D., The Right and the Good (Oxford, 1930)Google Scholar, p. 105). Baldwin, Thomas, G. E. Moore (London, 1990)Google Scholar, p. 77 and Hurka, Thomas, ‘Moore in the Middle’, Ethics 113 (2003), pp. 599–628 CrossRefGoogle Scholar, at p. 604 n. 27, agree.

5 See Moore, G. E., Principia Ethica (Cambridge, 1903)Google Scholar, pp. 18, 24-6, 101, 106, 146-8; Ethics (New York: 1965 [1912]), pp. 24-6, 28, 30, 66, 71, 73, 77, 106; also ‘A Reply to My Critics’, The Philosophy of G. E. Moore, ed. P. A. Schilpp (La Salle: 1968 [1942]), pp. 535-677, at 558-9. It is not clear whether, in Ethics, Moore thinks ‘one ought to maximize the good’ is an underived synthetic a priori truth, in which case he could agree with (1.1) (see below), or whether he thinks one can infer directly but non-analytically from ‘x maximizes the good’ to ‘one ought to do x’ (as is perhaps suggested by his claim that actions are right ‘only if and because’ they maximize the good) (Ethics, p. 30).

6 Rashdall, Theory, vol. 1, pp. 135-8; Ethics, pp. 14, 40, 61, 75-6; John Laird, A Study in Moral Theory (London, 1926), pp. 21-2, 25.

7 The counter-intuitive nature of (1.1) is noted by Ewing, A. C., ‘Recent Developments in British Ethical Thought’, British Philosophy in the Mid-Century, ed. Mace, C. A. (London, 1966), pp. 65–95 Google Scholar, at p. 71 and Johnson, Oliver, Rightness and Goodness (The Hague, 1959)Google Scholar, p. 22. As Johnson notes, Ross and Carritt share the unexpected view that production of some goodness is a necessary condition on rightness; see Rightness, pp. 12, 37; Carritt, Theory, pp. 41-2, 54, 86, 139; Ross, Right, pp. 36, 162. Prichard might hold the more plausible position that the badness of pain is not the whole reason why we ought not to inflict pain. In ‘Basis’, he writes that ‘any moral principle . . . includes mention of two things, (a) a good thing which the action will produce, (b) a definite relation in which the agent stands either to another or to himself’ (B 4; also MO 217). The badness of pain is not the whole reason because it does not mention a relation. (For this suggestion, see Hurka, Thomas, ‘Underivative Duty: Prichard on Moral Obligation’, Social Philosophy and Policy 27 (2010), pp. 111–34CrossRefGoogle Scholar, at 117-18. For a different reading, on which the badness is not part of the reason at all, see Dancy, Jonathan, ‘Has Anyone Ever Been a Non-Intuitionist?’, Underivative Duty: British Moral Philosophers from Sidgwick to Ewing, ed. Hurka, Thomas (Oxford, 2011), pp. 87–105 CrossRefGoogle Scholar, at 98-105.) But if this is Prichard's point, he would need to explain why the badness could never be the whole reason, or why it could never be decisive – why (in, say, a case in which the relations were the same) ‘the decision of what we ought to do turns not on the question “Which of the alternative courses of action will originate the greater good?”’ (M 14 n.)

8 Rashdall's quotation marks are around only the ‘ought’, not the ‘ought to be’.

9 Moore, Principia, pp. viii, 17, 119, 121; 118, 128; 120, 64; 99; 121. By writing of the good as what is ‘intrinsically desirable’ or ‘desirable as an end’, he also implies that it is what ought to be desired (pp. 51, 65). Prichard notes some of these passages at MO 224.

10 For debate over Prichard's target, see Dancy, ‘Non-Intuitionist?’, pp. 98-105 and Hurka, ‘Underivative’, pp. 116-17. I thank Hurka for pressing the underived-ought view.

11 For 2.2, see also to Ross 15.4.37, 16.8.47, to Carritt 12.12.36, to Laird 4.1.39, to Broad 4.3.43.

12 See to Ross 11.12.36, 16.12.36, 10.5.40, 8.1.46, to Carritt 12.12.36, to Broad, 4.3.43.

13 Joseph, W. B., Some Problems in Ethics (Oxford, 1931), pp. 61–2Google Scholar.

14 This is another place in which Prichard's target may be the Moore of Principia rather than ideal utilitarianism in general. Moore might think that an action's property of maximizing the good is the same property as my being obligated to do it.

15 To Ross 14.7.38; also to Laird 30.7.38.

16 To Ross 8.1.46.

17 To Ross 15.4.37, 17.1.40, 10.5.40, 8.1.46; to Broad 4.3.43; Carritt to Prichard, 14.12.36; to Laird 4.1.39, 2.3.45; Broad, Critical Notice of Prichard, Moral Obligation, Mind 59 (1950), pp. 555-66, at 564. In a summary of Prichard in Foundations of Ethics (Oxford, 1939), Ross writes that ‘[w]e cannot . . . say “such and such an act is obligatory,” for the act is not there, to be either obligatory or anything else. Nor . . . can we say “such and such an act would be obligatory if it were done;” for clearly its obligatoriness, if it has any, does not depend on its being done’ (p. 56). We can, however, say that the act would be good if it were done, and conclude that, if true, this conditional gives the agent an obligation (that is itself not conditional on the act's being done). This seems to be Ross's own view. Carritt writes that Prichard held the paradoxical view ‘that from the premises: “If I were to do this it would produce good” and “I ought to produce good” we cannot conclude: “I ought to do this.” Both premises can be stated unconditionally’ (‘Fifty Years a Don’ (MS at University College, Oxford), pp. 27-8). The concern with non-existent acts seems to have been common, though the Meinong–Russell literature on the general problem is not cited. Joseph, for example, writes that ‘[p]rovided then that we may speak as if an unperformed action were something to which predicates belong at all’ (Problems, p. 34). For a much later use of the argument, see Ewing, A. C., Second Thoughts in Moral Philosophy (New York, 1959), pp. 54–5Google Scholar.

18 Prichard does note that he rejects the view of ‘Mistake’ and ‘Duty and Interest’ (to Laird 2.3.45, to Ross 11.12.36, 8.1.46). For a time, his positive view was that obligations are based on properties of the circumstances rather than the acts. For example, the obligation to relieve someone's pain arises from the fact that the person is suffering, not from (as Ross and Carritt think) the fact that the act would be a relieving of pain of another; the obligation to keep a promise arises from having made the promise, not from the fact that my action would be a keeping of a promise (to Carritt 15.12.36, to Ross 11.12.36; also to Ross 16.12.36, 28.1.38, 14.7.38). (Sometimes Prichard gives a special role to one property of the circumstances, namely my belief that if I do x, there will be effect y (to Laird 2.3.45, to Ross 8.1.46, DIF 99-100).) But Prichard came to worry that the relevant properties of the circumstances are hypothetical facts (if I do this, that will follow). Since he also held that hypothetical facts cannot ground categorical ones, this led him to doubt the circumstances view, and wonder whether the properties of the act view was not right after all (to Ross 8.1.46, 10.12.46, 16.8.47).

19 Broad, ‘Notice’, p. 564.

20 To Laird 4.1.39, 7.1.39, 10.1.39, 2.3.45; to Ross 10.5.40, 8.1.46, 10.12.46, 16.8.47.

21 This is controversial. It is denied by Sidgwick (The Methods of Ethics (Indianapolis, 1981 [1907]), p. 37) and Broad (‘Notice’, p. 558; ‘Obligations, Ultimate and Derived’, Broad's Critical Essays in Moral Philosophy (London, 1971), pp. 351-68, at 362-3.

22 When he makes the charge just quoted, he thinks the ‘ought to aim’ view has been eliminated and that Sidgwick does not hold the ‘ought to bring about’ view (MM 142, 137).

23 Broad, ‘Notice’, p. 561.

24 For brief discussion of this ‘ought’, see the Appendix.

25 In ‘Manuscript’, he quickly dismisses Moore by noting that it is plain that by ‘I ought to do x’ we do not mean ‘x will cause a good result’ (MM 153). This does not address those like Laird and the later Moore who think the connection between ‘I ought to do x’ and ‘x will cause a good result’ is synthetic a priori.

26 For more on this objection, see Dancy, ‘Non-Intuitionist?’.

27 Oddly, in his survey of arguments for deontology, Gerald Gaus privileges (and endorses) both the objection that utilitarians secure only non-moral ‘oughts’ and (4.1)–(4.6) (‘What is Deontology?’, Part I, Journal of Value Inquiry 35 (2001), pp. 27-42, at 34-5).

28 For an early statement of this tactic, see Rashdall, Theory, vol. 1, pp. 72-3; Ethics, pp. 74, 77.

29 I consider Ross and Carritt in ‘The Birth of Deontology’, in Hurka, Underivative, pp. 126-45. Ross has no objection to adding the good of justice to explain the duty of justice; his objection is to that tactic employed for fidelity, recompense and gratitude.

30 See also to Ross, 20.12.28.

31 Ross, in Right, makes the same exclusion (pp. 36-8, 46-7).

32 Rashdall, Theory, vol. 1, pp. 96-7; Broad, ‘The Doctrine of Consequences in Ethics’, in Essays, pp. 17-42, at 36-8. See also Broad, ‘Certain Features in Moore's Ethical Doctrines’, in Schilpp, Moore, pp. 43-67, at 48-9, and Moore's acceptance of the point at ‘Reply’, pp. 559-60.

33 For more on this strategy, see Hurka, ‘Underivative’, pp. 112-15, and British Ethical Theorists from Sidgwick to Ewing (Oxford, 2014), pp. 125-6, 167.

34 For one recognition of the point, see Ross, Foundations, p. 69.

35 Hurka has suggested to me that Prichard may intend ‘foolish’ to mean ‘causes pain to me’, just as he means by ‘good to me’, ‘causes pleasure in me’. If so, Prichard can consistently say that my inflicting pain on myself is foolish but not something I have reason not to do. This does not, of course, meet the worry that we think there is a reason present (nor does it fit the text so well: it turns ‘we should allow the wanton infliction of pain on ourselves to be foolish’ into a trivial claim).

36 The last is not Prichard's view, since he admits obligations to oneself (M 13). For one ideal utilitarian discussion of the issue, see Ewing, Second, pp. 112-14.

37 Elsewhere Prichard makes the related point that we think of bringing about good for ourself and good for others as different duties. We think the same for bringing about the good of our child, the child of relations, the child of a stranger, or an orphan. The ideal utilitarian is wrong, then, to hold that we think that an act becomes a duty simply by thinking that it brings about something good (MO 217). (This is a rare later appeal to intuitions about cases, though the intuitions concern the classification of duties.) Since this argument concerns only how we think about our duties, the ideal utilitarian could concede that we often do not think of what these duties have in common – that they bring about good to someone. Ideal utilitarianism says only that what makes a duty a duty is that it brings about good to someone, not that we always think this. The argument in ‘Basis’ is stronger because it concludes that the ideal utilitarian is mistaken about our duty, rather than merely mistaken about how we often think about our duty.

38 Thanks to Brad Hooker for noting the different objections here. For the asymmetry, see Slote, Michael, ‘Morality and Self-Other Asymmetry’, Journal of Philosophy 81 (1984), pp. 179–92CrossRefGoogle Scholar. Slote does not give an example of special obligations to oneself, and Prichard does not give the other side of the asymmetry, namely that the man who is not obligated to produce the greater good in himself would be obligated to produce it in others.

39 Oddly, in ‘Mistake’ Prichard suggests this very explanation: it ‘is simply because we can and because others cannot directly modify our disposition that it is our business to improve it, and that it is not theirs, or, at least, not theirs to the same extent’ (M 13). Ross agrees (Right, p. 26).

40 Baldwin, Moore, p. 116. His example concerns promises rather than father-savings.

41 Thomson, Judith Jarvis, The Realm of Rights (Cambridge, MA, 1990), pp. 137–48Google Scholar.

42 For example, it is ‘obvious’ that the ideal utilitarian must say that the good is what ought to be (M 9); ‘it is clear when we reflect’ that the property of an action can give rise only to another property of that action (MO 222); ‘when we consider the matter thoroughly we find ourselves forced to admit’ that when we say ‘X ought to be educating Y’ we attribute a property to X (MO 169); ‘as we recognize when we reflect’ it is ‘obvious’ that actions do not have the property of being obligatory (DIF 99); ‘there is not the slightest plausibility in maintaining that we ought to do a certain action on the ground that it will lead to a certain state of affairs, unless the term “ought” is understood in the non-moral sense of conducive to . . . our actual purpose’ (MM 142); ‘it is really self-evident that an obligation is necessarily an obligation to perform a certain action, and not to have a certain motive’ (MO 199); it is ‘self-evident’ that hypothetical facts cannot ground categorical ones (to Laird 4.1.39).

43 Prichard does say, against Sidgwick, that ‘it is self-evident that there is no such duty’, where the duty in question is a duty to aim at our whole good (MO 206). But it is not clear whether this is a version of the argument noted in VI, against thinking that there is a duty to bring about one's happiness, or whether what Prichard finds self-evident is that there is no duty to aim at one's happiness. (The same goes for his claim that ‘even if we had been purely benevolent beings, there would not have been a duty to be moved by the desire of the good of all others’ (MO 207) Here he might be objecting, in addition, to deriving a duty from a ‘non-moral’ ‘ought’.) MM 135 and 136 seem to reject a duty to bring about (as well as aim at) one's own good, but on MM 137, when Prichard turns to consider that Sidgwick might mean ‘bring about’ rather than ‘aim at’, he writes that ‘it is plausible to say that we think it our duty . . . to make mankind, including ourselves, happy’.

44 Thomson says ‘I will only assume that you believe a certain moral judgment is true if I find it hard to understand how you could possibly believe it false’ (Realm, pp. 32-3). It is easy to understand why one might sacrifice one to save five.

45 Rashdall, Theory, vol. 1, p. 135; also Ethics, pp. 14, 61.

46 Rashdall, Theory, vol. 1, p. 135; Ethics, pp. 14, 72; Ethics, pp. 75; 22; 14; 40, Theory, vol. 1, p. 138, Is Conscience an Emotion? (Boston, 1914), p. 45; Ethics, p. 63 n.2.

47 Prichard, ‘Mistake’, p. 4; Ross, Right, p. 105; Broad, C. D., Five Types of Ethical Theory (London, 1930)Google Scholar, p. 165. Similarly, Laird, who is also targeted by Prichard, is not serious about applying ‘ought’ to states of affairs. ‘When we affirm . . . as we sometimes do, that anything which is good is a thing which ought to be, we do not, I think, intend to make a statement which is true without any qualification . . . When we say that the world contains happiness here, and beauty there, and that beauty and happiness are excellent, we are making statements about values and their incidence, but are saying nothing whatever about what ought to exist. When, inferring from this, we say (if we do) that the world, in these respects, is what it ought to be, we imply that there is a sense in which values may determine existence . . . [T]he “ought” implies agency’ (MO 224; Study, pp. 34-5).

48 For the objection, see Hurka, ‘Moore’, pp. 603-4; British, p. 53. Bertrand Russell raises this as his objection to taking ‘ought’ rather than ‘good’ as primitive (‘The Elements of Ethics’, in Russell, Philosophical Essays (London, 1966 [1910]), pp. 3-51, at 17). For Rashdall on the independence of the good from what we can bring about, see Ethics, p. 12; Theory, vol. 1, p. 136.

49 Sidgwick, Methods, p. 33, quoted with approval by Rashdall at Theory, vol. 1, pp. 103-4.

50 Broad, Types, p. 163.

51 Moore, ‘The Nature of Moral Philosophy’, in Moore, Philosophical Studies (London: 1960 [1922]), pp. 310-39, at 316-23. The same holds for Wilbur Urban, one of Ross's targets. Urban identifies value with what ought to be, but he writes that: ‘[o]f many things we can say that they ought to be, when it would be wholly absurd to think that this notion involved a command to any person or group of persons. Even of things that have already happened . . . we may say that they should have been otherwise. The imperative is but a special case of the category “ought to be” . . .’ (‘Value and Existence’, Journal of Philosophy, Psychology, and Scientific Methods 13 (1916), pp. 449-65, at 462). Unlike Rashdall, Urban would deny that ‘x is good’ could be interchanged with ‘x ought to be’, but only because he thinks what is good is a proper subset of what is valuable (p. 456).

52 Ross, Foundations, p. 45; Hurka, ‘Moore’, p. 604; British, pp. 45, 53.

53 Sidgwick, Methods, p. 33. For more on this reply to the objection, see my ‘The Later British Moralists’, The Routledge Handbook of Philosophy of Well-Being, ed. Guy Fletcher (London, 2016), pp. 95-109, at 99.

54 See Rashdall, Ethics, pp. 14, 76; Theory, vol. 1, p. 138; Conscience, pp. 45-6.

55 Rashdall, Theory, vol. 1, p. 135 n.; also Theory, vol. 1, p. 102; Conscience, p. 46. Similarly, Russell writes that to ‘explain what we mean by Good and Bad, we may say that a thing is good when on its own account it ought to exist, and bad when on its own account it ought not to exist . . . But all such characterizations really presuppose the notions of good and bad, and are therefore useful only as means of calling up the right ideas, not as logical definitions’ (‘Elements’, p. 17). Russell uses ‘ought to exist on its own account’ as a way of stressing that the good he is considering is intrinsic and ‘impersonal’ (p. 18).

56 One might object that Rashdall's claims that there is one notion, or that ‘good’ and ‘right’ are ‘correlative’ notions that cannot be understood apart (Theory, vol. 1, p. 138; Ethics, pp. 14, 76; Conscience, p. 46), make it hard to see how ‘good’ is fundamental. But let the one notion be called ‘ought’. ‘Good’ is what ought to be promoted (in a sense that does not imply ‘can’). ‘Right’ is what ought to be done (in a sense that does imply ‘can’). ‘Good’ is fundamental in that it, combined with claims about what one can do, explains what is right, but not vice versa. One cannot understand the notions apart in that both involve ‘ought’.

57 Rashdall, Theory, vol. 1, p. 135; also Conscience, p. 45.

58 Rashdall, Ethics, p. 14.

59 Rashdall, Ethics, p. 75; also pp. 76-7. When Rashdall writes of an ‘ultimate and unanalysable idea’ elsewhere, he is making the point that there is no reduction of the normative to the non-normative; see Theory, vol. 1, pp. 48, 102.

60 Rashdall, Ethics, pp. 25; also p. 22. In a similar use of ‘ought to be’, James Seth writes that ‘[e]thics is the science of the Good. As distinguished from the natural sciences, or the sciences of the actual, it is a normative or regulative science, a science of the ideal. The question of ethical science is not, What is? but What ought to be?’ (‘Is Pleasure the Summum Bonum?’, International Journal of Ethics 6 (1896), pp. 409-24, at 409).

61 Thanks to audiences at Manitoba and Reading, especially to John Cottingham, Jonathan Dancy, Brian Feltham, Brad Hooker, Jonas Olson, Michael Stack, Philip Stratton-Lake, Chris Tillman and Fiona Woollard; to Ben Caplan, Melissa Hiebert, Joyce Jenkins, and especially Tom Hurka, for discussions of the material; to Tom again, for comments as a referee and for introducing me to Prichard's correspondence (and to Carritt's Fifty Years); and to Dale Miller and an anonymous referee.