Skip to main content Accessibility help

Simply Good: A Defence of the Principia



Moore's moral programme is increasingly unpopular. Judith Jarvis Thomson's attack has been especially influential; she says the Moorean project fails because ‘there is no such thing as goodness’. I argue that her objection does not succeed: while Thomson is correct that the kind of generic goodness she targets is incoherent, it is not, I believe, the kind of goodness central to the Principia. Still, Moore's critics will resist. Some reply that we cannot understand Moorean goodness without generic goodness. Others claim that even if Moore does not need Thomson's concept, he still requires the objectionable notion of absolute goodness. I undermine both these replies. I first show that we may dispense with generic goodness without losing Moorean intrinsic goodness. Then, I argue that though intrinsic goodness is indeed a kind of absolute goodness, the objections marshalled against the concept are unsound.



Hide All

1 Moore, G. E., Principia Ethica, ed. Baldwin, T. (New York, 1993), pp. 192, 196–7, 219–20.

2 Thus Philippa Foot writes: ‘it is remarkable how [Moorean] utilitarianism tends to haunt even those of us who will not believe in it. It is as if we for ever feel that it must be right, although we insist that it is wrong’ (‘Utilitarianism and the Virtues’, Mind 94 (1985), pp. 196–209, at 196).

3 Most attention has been paid to Moore's ideal utilitarian account of right action and its allegedly unacceptable consequences. For discussion of perhaps the most powerful objection to the theory and an attractive Moorean reply, see Feldman, Fred, ‘Adjusting Utility for Justice: A Consequentialist Reply to the Objection from Justice’, Philosophy and Phenomenological Research 55 (1995), pp. 567–85.

4 Thomson, Judith Jarvis, ‘The Right and the Good’, Journal of Philosophy 94 (1997), pp. 273–98.

5 ‘The Right and the Good’, p. 274.

6 In using this terminology I follow Zimmerman, Michael, The Nature of Intrinsic Value (Lanham, 2001).

7 This is stated also in her ‘The Right and the Good’, p. 8; Goodness and Advice, ed. A. Gutman (Princeton, 2003), p. 17; and Normativity (Chicago, 2008).

8 She writes in her Goodness and Advice, p. 17: ‘[Goodness] is the property that we would be ascribing to a thing – whether an event or anything else – if we said of it “That's good”; and that is the property such that we are asking whether a thing possesses it when we ask about the thing “Is it good?”’. See also her ‘The Legacy of the Principia’, The Southern Journal of Philosophy 41 (2003), pp. 62–82, at 72, and Normativity, p. 7.

9 Thomson, Normativity, p. 3. This is also suggested in her ‘Goodness and Utilitarianism’, Proceedings and Addresses of the American Philosophical Association 67 (1994), pp. 7–21. It is worth noting that Thomson takes the principle to have broader application than may be apparent here; she assumes that, for Moore, whenever a thing is good in any respect – whether it is a good as a toaster, or good at heating the room, or good for your diet – this is because it possesses the property of generic goodness. See Normativity, pp. 3–6.

10 See also ‘Goodness and Utilitarianism’, p. 8; ‘The Right and the Good’, p. 275; ‘Legacy’, p. 72; and Normativity, pp. 3–7.

11 I suspect Thomson would accept this argument: she reasons similarly in Normativity, p. 10.

12 See Thomson, ‘Goodness and Utilitarianism’, p. 8; ‘The Right and the Good’, p. 273; and Normativity, p. 2. She paraphrases this same passage in Goodness and Advice and ‘Legacy’ but does not provide a citation.

13 Moore, Principia, p. 54.

14 Moore, Principia, p. 3.

15 Moore, Principia, p. 5.

16 Moore, Principia, pp. 21–2.

17 I thus assume with Moore that the intrinsic value of a thing supervenes on its intrinsic features. I recognize this position is disputed; see e.g. Korsgaard, Christine, ‘Two Distinctions in Goodness’, The Philosophical Review 92 (1983), pp. 169–95; Kagan, Shelly, ‘Rethinking Intrinsic Value’, Journal of Ethics 2 (1998), pp. 277–97; and Rabinowicz, Wlodek and Rønnow-Rasmussen, Toni, ‘A Distinction in Value: Intrinsic and for its Own Sake’, Proceedings of the Aristotelian Society 100 (2000), pp. 3351. A defence of supervenience would, I fear, take us too far afield, but for Moorean replies see Zimmerman, Intrinsic Value, pp. 60–4; Bradley, Ben, ‘Is Intrinsic Value Conditional?’, Philosophical Studies 107 (2002), pp. 2344; as well as my own ‘The Pen, the Dress, and the Coat: A Confusion in Goodness’, Philosophical Studies 173 (2016), pp. 1911–22.

18 See also Moore, Principia, p. 68: ‘Every one does in fact understand the question “Is this good?” . . . It has a distinct meaning for him, even though he may not recognise in what respect it is distinct. Whenever he thinks of “intrinsic value,” or “intrinsic worth,” or says that a thing “ought to exist,” he has before his mind the unique object – the unique property of things – which I mean by “good”.’

19 Moore, Principia, p. 78.

20 See Zimmerman, Intrinsic Value, pp. 18–19, for an argument to the same effect.

21 This is, I believe, the orthodox interpretation. See e.g. Fred Feldman, Doing the Best We Can (Dordrecht, 1986), p. 38; Julia Driver, ‘The History of Utilitarianism’, Stanford Encyclopedia of Philosophy, <> (2014); Russ Shafer-Landau, The Fundamentals of Ethics (New York, 2010), pp. 117–21; Shaw, William H., ‘Editor's Introduction’, Ethics, ed. Shaw, W. (New York, 2005), pp. vii–xxxix, at xvi–xvii; and Zimmerman, Michael, Living with Uncertainty: The Moral Significance of Ignorance (New York, 2008), pp. 23 – among many others.

22 Moore, G. E., Ethics, ed. Shaw, W. (New York, 2005), p. 30.

23 Normativity, p. 16. Emphasis mine.

24 Admittedly, Thomson does provide a very brief discussion of this objection in an appendix to her ‘Legacy’. However, it is unclear that she accepts the concern, as she does here.

25 Moore, Ethics, p. 32.

26 See e.g. Zimmerman, Intrinsic Value, p. 19.

27 In his illuminating ‘Hyperventilating about Intrinsic Value’, The Journal of Ethics 2 (1998), pp. 339–54, Fred Feldman makes a similar point with regard to an understanding of these principles that appeals to overall value. And, indeed, my solution here mirrors Feldman's.

28 This principle is perhaps most clearly stated in Moore's ‘The Conception of Intrinsic Value’, Philosophical Studies (London, 1951), pp. 253–75, at 260–1.

29 I say only that if something is intrinsically good then it is necessarily intrinsically good, not that something is intrinsically good just in case it is necessarily intrinsically good (and similarly for the isolation principle). But this omission is only stylistic. The sufficient condition is not informative and may therefore be left unsaid: I believe that it is true, but trivial, that if something is necessarily intrinsically good, then it is intrinsically good. Feldman provides similar modifications for Moore's supervenience principle in ‘Hyperventilating’.

30 In saying these principles are defensible, I do not mean that they are true in this form. In fact, I suspect the best form of the Moorean view will weaken these claims; it will say that these principles specify features of intrinsic value that are individually necessary but only jointly sufficient. (Thus, according to the view I prefer, we say only that if something has intrinsic value, then it has that value intrinsically, necessarily, and in isolation. But this is not so of other kinds of value.) However, regardless, I believe that these principles as stated are close to the truth, consistent with Moore's intent, and provide an excellent place to begin formulating a powerful Moorean view about the concept of intrinsic goodness.

It is important to note that these principles are not supposed to be definitions: for the Moorean, intrinsic value is a conceptual primitive. Rather, they are only supposed to help us grasp the primitive Moore employs, and to show how it differs from other concepts. I explain this further in the next section.

31 Richard Kraut (Against Absolute Goodness (New York, 2011)) and Foot (‘Utilitarianism’) are perhaps most prominent. But see also Nussbaum, Martha, ‘Comment’, Goodness and Advice, ed. Gutman, A. (Princeton, 2003), pp. 97125; Brännmark, Johan, ‘Goodness, Value, Reasons’, Ethical Theory and Moral Practice 12 (2009), pp. 329–43; and Freiman, Christopher, ‘Goodness and Moral Twin Earth’, Erkenntnis 79 (2014), pp. 445–60, among others.

32 Aristotle, Nicomachean Ethics, ed. R. Crisp (New York, 2000), p. 137.

33 Ross, W. D., The Right and the Good, ed. Stratton-Lake, P. (New York, 2002), p. 102.

34 Sidgwick, Henry, The Methods of Ethics (Indianapolis, 1981), p. 382. We might claim that, though these philosophers all use similar language, they are not targeting the same property; see Kraut, Goodness, pp. 10–11, 209–12 for discussion. This is perhaps true – but even if these philosophers have subtly distinct concepts in mind, such differences will not be relevant here.

35 Arneson, Richard, ‘Good, Period’, Analysis 70 (2010), pp. 731–44; Klocksiem, Justin, ‘Perspective Neutral Intrinsic Value’, Pacific Philosophical Quarterly 92 (2011), pp. 323–37; Rowland, Richard, ‘In Defence of Good Simpliciter’, Philosophical Studies 173 (2016), pp. 1371–91.

36 Indeed, some claim further that absolute goodness simply is intrinsic goodness. In an important passage (Goodness, p. 14), Kraut reasons that a thing can be good absolutely only if it is good non-relationally – that is, good in virtue of its non-relational properties. But a thing is good in virtue of its non-relational properties just in case it is good intrinsically. Thus intrinsic and absolute goodness are identical. I reject this argument; I think it relies on a kind of equivocation. To say that a thing is good non-relationally may mean (i) that it is good, but is not merely a good member of a kind, or good for a particular person or purpose (i.e. that it is good, but is not merely good for a K) or (ii) that it is good in virtue of its non-relational (i.e. intrinsic) properties. These interpretations may come apart – to say that something is instrumentally good is not to say that it is good relative to some person or purpose, but instrumental goods do not have their value in virtue of their intrinsic features. Thus I think Kraut's argument unsound. However, little will depend on this. Regardless of whether intrinsic goodness is absolute goodness as Kraut and others such as Arneson maintain, or is simply a kind of absolute goodness, as I believe, it is still true that, if there is reason to reject the concept of absolute goodness, then there is reason to reject the concept of intrinsic goodness.

37 I am thankful to an anonymous referee for raising this concern.

38 Zimmerman, Intrinsic Value, p. 24.

39 For further explication of this kind of reply see Zimmerman, Intrinsic Value, pp. 18–29.

40 I understand an atomic state of affairs to be the instantiation of an n-place universal by n-many particulars (or the instantiation of a second-order n-place universal by n-many first order universals, and so on). Complex states may be formed out of atomic states by conjunction or mereological sum. Conceptually, I take states of affairs to be truth-makers (rather than truth-bearers) as well the fundamental relata of the causation relation. However, for our purposes, most views about the nature of states of affairs will be acceptable. (For a critical discussion of alternatives, see Zimmerman, Intrinsic Value, pp. 46–52.) Further, those who prefer to speak in terms of facts, propositions, events, or tropes, can probably translate. Though these ontological issues will have consequences for our conception of axiology further down the line, such issues will not, I think, be relevant here.

41 Thomson, Normativity, pp. 25–6.

42 Moore (and the tradition he began) understands instrumental value in terms of intrinsic value: the value a thing has as a means is determined by the intrinsic values of what it causes (or prevents). It is thus assumed that, because states of affairs are both the sole bearers of intrinsic value and the fundamental relata of the causation relation, states of affairs are also the sole bearers of instrumental value. For further explication of the concept of instrumental goodness, see Ben Bradley, ‘Extrinsic Value’, Philosophical Studies 91 (1998), pp. 109–26.

43 Still I may be mistaken about this – and I would welcome such a result: I hope only for a defence of the Moorean view. However, I trust that those who believe that intrinsic value is relational will agree that it is worth considering how we might defend the Moorean project if intrinsic value were, as I believe, a kind of absolute goodness. If nothing else, this would be dialectically important: if we can show that the Moorean project could succeed, even if intrinsic value is non-relational, then we may gain a stronger reply to those who reject the Moorean system on such grounds.

44 Geach, P. T., ‘Good and Evil’, Analysis 17 (1956), pp. 3342. There are, of course, other concerns about the notion of absolute goodness – for summary and discussion, see Klocksiem, ‘Perspective’ and Rowland, ‘Defence’. However, the Thomson/Geach objection is, I believe, the most powerful and influential; it will therefore be my primary concern in what follows.

45 See Thomson, Normativity, pp. 14–15.

46 Geach, ‘Good and Evil’, p. 33.

47 In fact, there is a worry, even here. Consider ‘big’. Geach claims that it is logically attributive because ‘this is a big flea’ does not entail ‘this is big’. But if ‘big’ expresses the same property in ‘this is a big flea’ and ‘this is big’, then it seems the entailment holds. Alternatively, if ‘big’ means something different in each occurrence, then entailment fails only because of equivocation. The same seems true of the other examples given – including ‘good’. However, for the sake of argument, I will ignore these concerns.

48 See Geach, ‘Good and Evil’, pp. 33–4.

49 Normativity, pp. 4–6.

50 Geach, ‘Good and Evil’, p. 34 (emphasis mine). Thomson claims that this is slightly too strong; see her ‘The Right and the Good’, pp. 277–8. However, any difference between the conclusion Thomson prefers and what Geach suggests here will not be relevant.

51 Though Geach provides no support for this assumption, a standard rationale is available. The argument is analogical: it first claims that when an adjective is logically predicative (like ‘red’), the truth conditions of its grammatically attributive uses (like ‘A is a red car’) should be understood in predicative terms; thus ‘A is a red car’ is true just in case A is red and A is a car. Similarly, if some adjective is logically attributive, then the truth conditions for its grammatically predicative uses should be understood attributively. Thus ‘B is good’ is true just in case B is good relative to the referent of the contextually supplied substantive. (I am thankful to Bradford Skow for making this clear to me.) However, ultimately, I think this rationale should be rejected. Consider ‘red’: some things are simply red, as Thomson and Geach claim. Others are not (simply) red, but are red-for-hair, red-for-a-face, red-for-an-apple, and so on (see Zimmerman, Intrinsic Value, p. 22). If we accept the rationale given, we must conclude that ‘red’ cannot be used both predicatively and attributively; it thus cannot express both the property of being (simply) red and the properties of being red-for-hair, and so on. If we agree with Thomson and Geach also that a property exists only if it can be expressed, then we must say that either there is no such thing as being (simply) red or no such thing as being red-for-hair, etc. Neither is plausible. The concern is, of course, general: Thomson's argument about ‘famous’ below shows much the same problem.

52 Note that if we say instead that ‘good’ is logically attributive only if every sentence of the form ‘X is a good K’ does not entail that ‘X is good’ and ‘X is a K’, then simply providing a few examples where entailment fails, as Geach and Thomson do, would be insufficient to establish the premise.

53 Thomson, Normativity, p. 14.

54 We might think that Thomson has given up too easily; she should insist that logically attributive predicates can never be used in a way that is genuinely predicative. Thus we say that (e.g.) though Obama is a famous person, he is not famous simpliciter. But this claim is in tension with Thomson's own position, as it is unclear how the kind person in any way determines the relevant standards for being famous. Further, even if this is good enough for ‘famous’, it seems inadequate for other adjectives (see n. 51).

55 This position about ‘good’ is originally advanced in Ross, The Right and the Good, p. 65.

56 Thomson, Normativity, p. 14.

57 Thomson, Normativity, pp. 15–17.

58 Moore makes this plain repeatedly; he writes in the Principia, p. 58, ‘If I am asked “What is good?” my answer is that good is good, and that is the end of the matter. Or if I am asked “How is good to be defined?” my answer is that it cannot be defined, and that is all I have to say about it.’ See also pp. 61, 69, 72, 89 and 111. Of course, some Mooreans do take up Thomson's challenge and provide a definition of intrinsic value; see e.g. Ewing, A. C., The Definition of Good (New York, 2012); Chisholm, Roderick, ‘Defining Intrinsic Value’, Analysis 41 (1981), pp. 99100; Lemos, Noah, Intrinsic Value: Concept and Warrant (New York, 1994); and Zimmerman, Intrinsic Value.

59 Admittedly, we do not use the term ‘intrinsic value’ much outside of axiology. Moore's critics may thus respond that it is a technical term with no pre-theoretical application. But the fact that we do not use the phrase ‘intrinsic value’ does not mean that we do not employ the concept. When we ask (e.g.) whether the world is a good place, we are not wondering whether the world is good for some purpose, or good for ourselves. Rather, I think we are wondering whether it is good in itself. And we may ask similar questions about our lives, and our actions. Thus we do, I think, make use of the concept of intrinsic value outside philosophy – even if we do not employ the term.

60 See Jonathan Schaffer, ‘It is the Business of Laws to Govern’, dialectica 70 (2016), pp. 577–88, at 579.

61 Note, however, that this is a necessary, but not sufficient, reason to reject the concept. Even if Moore's system fails, the concept of intrinsic value may still be significant. Every great treatise in moral philosophy – including the works of Plato, Aristotle and Kant – features discussion of which things are valuable intrinsically, and the notion of intrinsic value is still widely considered fundamental in axiology. Thus more work would need to be done to reject Moore's concept.

62 Thomson sometimes advances a seemingly independent concern: she claims that it is not possible to answer questions about what things are intrinsically good (see e.g. her ‘Legacy’, p. 13). And, admittedly, questions about what things are intrinsically good are more difficult to answer than questions about what things are good toasters, or umbrellas. But most questions in philosophy are difficult to answer; this does not make them bad questions (compare: what is knowledge? Or: what are the fundamental ontological categories?).

63 Many thanks to Eric Wiland, Mark Schroeder, Phillip Bricker, Peter Graham, Bob Gruber, Kim Soland, Jordan Kroll, Luis Oliveira, and an audience at the Central APA for their helpful comments. I am especially grateful to Jean-Paul Vessel, Bradford Skow, Fred Feldman, Lisa Tucker, and two anonymous referees from Utilitas for all their help with this article.


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed