1 M. Koskenniemi, ‘The Politics of International Law’, (1990) 1 EJIL 4, at 7–8 (emphasis in original, notes omitted). Note that Koskenniemi does not so much advocate value-subjectivism here, as he attributes that claim to liberals. I address the difficulty created by this strategy at the text following note 21, infra.

2 An analogy with the idea of indeterminacy might also help to make the point: if the value of x is indeterminate, the result of its addition to any determinate quantity (e.g. 3+x) will also be indeterminate.

3 This argument is stated in its general philosophical form by Mackie, John in ‘The Third Theory of Law’, (1977) 7 Philosophy & Public Affairs 3Google Scholar.

4 Such an ‘emotivist’ or ‘projectivist’ view is famously defended by Simon Blackburn; see S. Blackburn, ‘Moral Realism’, in J. Casey (ed.), Morality and Moral Reasoning (1971), at 101 ff.

5 Koskenniemi notes the broader ambition of his thesis right at the start of From Apology to Utopia: ‘This is not only a book in international law. It is also an exercise in social theory and in political philosophy.’ M. Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (2005), at 1.

6 Or by resisting the claim that evaluative judgements figure in the best account of what makes a proposition of law true or false. For such positivist accounts of international law see G. Schwarzenberger, The Inductive Approach to International Law (1976); P. Weil, ‘Towards Relative Normativity in International Law?’, (1983) 77 AJIL 413.

7 The claim that the fact that law requires X is not a sufficient reason to hold that one has a moral duty to do X is central to the work of Herbert Hart, Joseph Raz, and Matthew Kramer; see H. L. A. Hart, The Concept of Law (1994), ch. 9; J. Raz, Ethics in the Public Domain (1994), ch. 9, ‘Authority, Law and Morality’; M. Kramer, In Defense of Legal Positivism (1999), ch. 7. The sharp distinction between legal validity and moral weight also features prominently in the positivism of Hans Kelsen; see text accompanying note 17, infra.

8 Mark Greenberg has offered a powerful general case against this type of theory of law in M. Greenberg, ‘How Facts Make Law’, (2004) 10 Legal Theory 157. Greenberg argues, first, that the determination of legal content is rational and therefore normative in nature and, second, that moral facts are the most plausible candidates as criteria for the rational determination of legal content.

9 J. Finnis, Natural Law and Natural Rights (1980), at 9. The point is, of course, also a staple of critical legal thought.

10 A reviewer suggested that the conception of law as a sort of obligation that is based on sources (typically social practice), rather than on purely deontic propositions, reveals a further and independent sense in which law strives for objectivity – a point partly reflected in the first paragraph of the Koskenniemi quote that opens the article. My own impression is that the reasons why it is important for the rights and duties of international agents to depend on their past practices have less to do with our desire to guarantee that the content of those duties will be as ‘objective’ as possible, and more to do with respect for the autonomy and the reasonable expectations of international agents. Properly elaborated, these reasons could, I think, explain why an international agent's views about the meaning of its past practice are not dispositive of the issue.

11 For a helpful survey of different senses of objectivity and their significance for legal theory, see M. Kramer, Objectivity and the Rule of Law (2007), ch. 1.

12 The idea is prominent in the work of John Rawls; see J. Rawls, Political Liberalism (1996), at 54 ff.

13 B. Stroud, ‘The Study of Human Nature and the Subjectivity of Value’, Tanner Lectures on Human Values, delivered at the University of Buenos Aires, 7 June 1988, at 219 (available at www.tannerlectures.utah.edu). Stroud eventually dismisses value-scepticism on the ground that it draws a metaphysically confused distinction between matters of fact and matters of value (see especially at 220–37).

14 Cf. W. Kymlicka, Contemporary Political Philosophy (2002), at 86. Note also Heidegger's remark that sceptical arguments have ‘something of the attempt to bowl one over’. M. Heidegger, Being and Time (1927), at 229.

15 Cf. B. Williams, Morality (1972), at 14. Williams notes that such a form of subjectivism would entail that ‘there would be no interpersonal moral disagreements; two persons expressing what we would normally take to be conflicting views would not be expressing conflicting views at all, but would be, rather, like two persons on a boat, one of whom says that he feels sick while the other says that he, on the other hand, does not’.

16 This point is emphasized by Tasioulas, John in ‘The Relevance of Ethical Objectivity’, (2002) 47 American Journal of Jurisprudence 254CrossRefGoogle Scholar.

17 Kelsen, H., ‘The Pure Theory of Law and Analytical Jurisprudence’, (1941–42) 55 Harvard Law Review 44, at 45CrossRefGoogle Scholar.

18 Ibid., at 47.

19 Koskenniemi, supra note 5, at 513–14.

20 Ibid., at 412.

21 Koskenniemi, supra note 1, at 7–8.

22 Kennedy, D., ‘A New Stream of International Law Scholarship’, (1988) 7 Wisconsin International Law Journal 1Google Scholar. See also Cass, D., ‘Navigating the Mainstream: Recent Critical Scholarship in International Law’, (1996) 65 Nordic Journal of International Law 341Google Scholar.

23 A. Carty, ‘Critical International Law: Recent Trends in the Theory of International Law’, (1991) 2 EJIL 66; A. Carty, The Decay of International Law? (1986), at 108 ff.

24 For example, in summarizing his argument, Koskenniemi writes that ‘concreteness seemed to require that we exclude not only explicit political opinions from the process of verifying or justifying the law's content but that we also exclude theories of justice. For, it is held, theories of justice are “subjective”, they cannot be verified or justified regardless of the political opinions held by some people’, supra note 5, at 513–14 (emphasis added).

25 I take up this issue very briefly in the conclusion.

26 Koskenniemi apparently holds that value-subjectivism is correct but that this does not preclude the possibility of reaching genuine consensus on matters of value. He writes, ‘To show that the inevitable move to politics in legal argument does not compel apologism requires taking a position against the view that politics (justice, morals) is simply subjective and arbitrary as such. This requires showing that political views can be held without having to believe in their objectivity and that they can be discussed without having to assume that in the end everybody should agree’. Koskenniemi, supra note 5, at 536 (emphasis in original).

27 Ibid., at 18.

28 Koskenniemi, supra note 5, at 531.

29 Ibid., at 218.

30 Carty prefers this formulation; see Carty, ‘Critical International Law’, supra note 23, at 68.

31 Cf. D. Davidson, ‘Objectivity and Practical Reason’, in E. Ullmann-Margalit (ed.), Reasoning Practically (2000), Essay 1, at 3–5.

32 J. Mackie, Ethics: Inventing Right and Wrong (1977), 36–7. Cf. G. Harman, The Nature of Morality: An Introduction to Ethics (1977).

33 Ibid., at 38–9.

34 Cf. J. McDowell, ‘Values and Secondary Qualities’, in T. Honderich (ed.), Morality and Objectivity (1985), 118 ff.; D. Wiggins, ‘Truth, Invention and the Meaning of Life’, (1976) 62 Proceedings of the British Academy 33, at 348–9; Dworkin, R., ‘Objectivity and Truth: You'd Better Believe It’, (1996) 25 Philosophy & Public Affairs 87, at 877 ffCrossRefGoogle Scholar.

35 Dworkin, supra note 34, at 83.

36 Ibid., at 86–7.

37 Mackie, supra note 32, at 4–6; Koskenniemi, supra note 5, at 475–7.

38 McDowell, supra note 34, at 118. Cf. Wiggins, supra note 34, at 349.

39 For a view that defines truth as (simply) justified belief or assertability see M. Dummett, Truth and Other Enigmas (1978), Essay 1, ‘Truth’, at 23–4.

40 This is effectively the position of Hilary Putnam, who defines truth as ‘idealized justified assertability’; see H. Putnam, Realism and Reason (1983), at xviii. For an attempt to apply Putnam's semantics to legal interpretation see N. Stavropoulos, Objectivity in Law (1996).

41 See especially the volume of essays in L. E. Hahn (ed.), The Library of Living Philosophers: The Philosophy of Donald Davidson, Vol. XXVII (1999), with replies by Davidson.

42 One problem with ‘cogito ergo sum’, at least in the context in which Descartes used it, is that it does not establish that I exist, for this is supposed in the first-person use of cogitare. The only non-circular way of reading this statement is ‘I think, therefore there is thought’. This is also Davidson's use.

43 D. Davidson, Problems of Rationality (2004), Essay 1, ‘The Problem of Objectivity’, at 6.

44 On the Cartesian beginnings of Davidson's project see T. Nagel, ‘Davidson's New Cogito’, in Hahn, supra note 41, Essay 7, at 195–8.

45 Davidson, supra note 43, at 8–9.

46 Cf. Davidson, supra note 43, Essay 9, ‘What Thought Requires’, at 135 ff.

47 E.g. consider the range of beliefs that I would need to assume are true to give sense to my belief that ‘my manager's husband put five pounds in my pocket whilst we were watching the solar eclipse’!

48 This idea is developed in the first part of Davidson's famous ‘A Coherence Theory of Truth and Knowledge’; see D. Davidson, Subjective, Intersubjective, Objective (2001), Essay 10, at 138–40. Davidson has taken great pains to emphasize that his theory does not assume that a body of belief is true just by virtue of the fact that it is consistent; see his ‘Afterthoughts’ to ‘A Coherence Theory’, at 154. I deal with this point in the main text.

49 One conspicuous consequence of holism is that my beliefs are not made true because they correspond or refer to particular aspects of the world, for the ideas of correspondence and reference acquire their sense from the false assumption that we experience the world in bits, of which our sentences are supposed to be mental representations. See ibid., Essay 5, ‘Indeterminism and Antirealism’, at 78–9, where Davidson develops this point by talking about the ‘inscrutability of reference’.

50 The oft-quoted sceptical example in this connection is a brain wired so as to produce all the normal sensations but is really placed inside a vat in a scientist's laboratory.

51 Davidson, supra note 48, Essay 14, ‘Three Varieties of Knowledge’, at 204 ff. This essay offers perhaps the easiest route into Davidson's thoughts on the matters of objectivity, truth, and meaning.

52 Koskenniemi, supra note 5, at 318.

53 Davidson, supra note 48, Essay 1, ‘First Person Authority’, at 3.

54 This point is prominent in the early work of Martin Heidegger and the philosophy of Hans-Georg Gadamer and Ludwig Wittgenstein. See Heidegger, supra note 14, at 25–6; H.-G. Gadamer, Truth and Method, trans. J. Weinsheimer and D. Marshall (1989), at 383 ff.; L. Wittgenstein, Philosophical Investigations (1954), at para. 72 ff.

55 Davidson, supra note 43, Essay 9, ‘What Thought Requires’. This does not mean that thoughtless creatures cannot interact with their environment. A worm is perfectly capable of getting by in the world – for example, it can look for and find food, but it does not identify its food as food, nor does it have any belief that it is looking for it.

56 Cf. the useful analogy in Putnam, H., ‘Meaning and Reference’, (1973) 70 Journal of Philosophy 699, at 705CrossRefGoogle Scholar: ‘There are tools like a hammer or a screwdriver which can be used by one person; and there are tools like a steamship which require the cooperative activity of a number of persons to use. Words have been thought of too much on the model of the first sort of tool.’

57 Wittgenstein, supra note 54, at paras. 225–34. For a sceptical reading of Wittgenstein's point see S. Kripke, Wittgenstein on Rules and Private Language (1982).

58 Cf. Davidson, supra note 48, Essay 8, ‘The Second Person’, at 116, where Davidson reads Wittgenstein as saying that ‘without an interpreter, no substance can be given to the claim that the speaker has gone wrong – that he has failed to go on in the same way’.

59 Davidson argues that the only special feature of knowing one's own mind is that one knows what one thinks without the aid of empirical observation, Davidson, supra note 48, Essay 1, ‘First Person Authority’, at 6 ff.

60 Cf. W. V. O. Quine, Word and Object (1960).

61 Some interpreters of Davidson have read the principle of charity as a plea to the interpreter to bring the speaker's beliefs as close as possible to one's own; see I. Hacking, Why Does Language Matter to Philosophy? (1975), at 146–50; C. Taylor, ‘Understanding the Other: A Gadamerian View of Conceptual Schemes’, in J. Malpas, U. Arnswald, and J. Kertscher (eds.), Gadamer's Century (2002), Essay 15, at 291–2. The idea that the interpreter has a choice about how much of his own beliefs to attribute to the speaker in order to understand him is misleading and certainly not Davidson's. As long as one tries to interpret the utterances of a speaker, one is committed to maximizing agreement between the interpreter's and the speaker's beliefs, insofar as there is no cause for thinking that one of the parties labours under a mistake or other misapprehension. So claims such as Taylor's (‘Understanding the Other’, at 292) that ‘Davidson's principle of charity is vulnerable to being abused to ethnocentric ends’ misses the point of that principle, to the extent that it assumes that an interpreter can manipulate it. On this point see D. Hoy, ‘Post-Cartesian Interpretation: Hans-Georg Gadamer and Donald Davidson’, in L. E. Hahn (ed.), The Library of Living Philosophers: The Philosophy of Hans-Georg Gadamer, Vol. XXIV (1997), Essay 3, at 122–5. In Section 5 I argue that charity does have an ethical significance, but of a different kind.

62 Davidson, supra note 51, at 211.

63 Cf. Davidson, ‘On the Very Idea of a Conceptual Scheme’, supra note 48, at 191.

64 Davidson, supra note 48, Essay 8, ‘The Second Person’, at 116 ff.

65 It is possible that the speaker intends to use the word ‘σκύλoς’ to mean table rather than dog, say, in order to make a joke about the fact that both dogs and tables have four legs. But the speaker's intention does not change the meaning of the word he is using. For unless an interpreter had any clue from the speaker that the word is intended to be understood in this extraordinary way (e.g. laughter, a special gesture), he would have no reason to treat its use as anything other than mistaken. On the loose relation between meaning and intention see ibid., at 111–12.

66 Davidson, supra note 48, Essay 6, ‘The Irreducibility of the Concept of the Self’, at 89–90: ‘Mistakes on the part of the interpreter or the speaker are to be expected, but these cannot be the rule, since errors take their content from a background of veridical thought and honest assertion. The crucial difference between the predominant, mostly banal, run-of-the-mill but on target beliefs and assumptions and the occasional deviation is this: errors, confusions, irrationalities have particular explanations; getting things right, aside from hard cases, is to be expected.’

67 Is it not the case that many apparent disagreements about matters of value are exacerbated (if not altogether created) by the fact that the parties offer different accounts of the underlying facts? One need think no further than the relevance of whether Iraq possessed weapons of mass destruction prior to the 2003 invasion on the question of the morality of that invasion.

68 Davidson, supra note 43, Essay 3, ‘The Objectivity of Values’, at 50–1.

69 Cf. C. Korsgaard, The Sources of Normativity (1996), at 139: ‘Philosophers have been concerned for a long time about how we understand the meanings of words, but we have not paid enough attention to the fact that it is so hard not to. It is nearly impossible to hear the words of a language you know as mere noise.’

70 For a similar discussion of the complaint see Tasioulas, supra note 16.

71 Cf. J. Waldron, Law and Disagreement (1999), Essay 8, ‘The Irrelevance of Moral Objectivity’, at 165 ff.; J. Waldron, The Dignity of Legislation (1992), at 61–2. The idea that the concepts of truth and objectivity are largely irrelevant in our everyday practical deliberations, and can therefore be dropped from our vocabulary without loss, has been put in general philosophical form by Richard Rorty; see R. Rorty, ‘Pragmatism, Davidson and Truth’, in E. Lepore (ed.), Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson (1986), at 87.

72 Cf. Dworkin, supra note 34, at 94.

73 Cf. S. Cavell, Conditions Handsome and Unhandsome: The Constitution of Emersonian Perfectionism (1990), at 65, where Cavell attributes to sceptics the tendency to ‘hyperbolize words’.

74 See, e.g., Waldron, supra note 71, at 170: ‘The question of moral objectivity . . . might be an interesting debating topic “for a calm philosophical moment, away from the moral and interpretive wars”’.

75 Cf. R. Rorty, Philosophy and Social Hope (1999), at 35–7.

76 The intersubjective nature of objectivity and truth should be familiar from the discussion of Davidson's work in section 3. My suggestion here is that the idea of truth as an intersubjective concept carries an ethical sense too.

77 My argument here draws on Hans-Georg Gadamer's account of the dialogical structure of interpretation. See Gadamer, supra note 54, at 363 ff. Note that the goal of a genuine and honest debate about the truth of the matter is not to achieve agreement simpliciter, but agreement on the truth. If mere agreement were the goal, we would not be able to explain what moves participants in such debates to eschew a quick mutual accommodation of their views at the first available opportunity.

78 Koskenniemi, supra note 5, at 318.

79 Koskenniemi is not alone in this view. Similar claims are made by R. Unger, Knowledge and Politics (1984), at 66–7; W. Sullivan, Reconstructing Public Philosophy (1982), at 38–40; and M. Kelman, A Guide to Critical Legal Studies (1987), at 64 ff. For a response see W. Kymlicka, Contemporary Political Philosophy (1990), at 98–100.