I would like to express my gratitude to my colleagues Louis Romero, Wanda Wiegers and Ken Norman for their helpfulness and generosity in reading and commenting upon earlier drafts of this essay.

1. I rely in this paper upon Wittgenstein, L. Philosophical Investigations, trans. Anscombe, G.E.M. (Oxford: Basil Blackwell, 1958).Google Scholar For a brief account of the distinction between the linguistic theory of Wittgenstein’s early period and that of his later period which I shall appeal to in this paper and discuss in some detail below, see Patterson, D. “Wittgenstein and the Code: a Theory of Good Faith Performance and Enforcement Under Article Nine” (1988), 137 U. Penn. L. Rev. 335 at 356–69.Google Scholar

2. See Langille, B. “Revolution Without Foundation: The Grammar of Scepticism and Law” (1983), 33 McGill L.J. 451 at 454–65.Google Scholar Langille has described the extent to which “strong interdeterminacy critics” (citing, among others, J. Boyle, Mark Tushnet, Sanford Levinson and Stanley Fish) rely upon Wittgensteinian arguments, with or without citing Wittgenstein, in order to justify their views.

3. See, e.g., Langille, supra, n. 2 and Patterson, supra, n. 1. These writers also appeal to Wittgenstein’s theory of language to support particular interpretive approaches to legal questions. Langille appeals to Wittgenstein to support an interpretative approach to the Canadian Constitution. Patterson offers a Wittgensteinian interpretation of the “good faith” provisions of Article Nine of the Uniform Commercial Code.

4. Langille, id., at 455, citing also Strawson, P.F. “Review of the Philosophical Investigations” (1954), 63 Mind 70 and Hacker, Insight and Illusion, revd. ed. (New York: Clarendon Press, 1986), preface.Google Scholar

5. Langille, supra, n. 2, al 486–7.

6. Hutchinson, A.C. Dwelling on the Threshold, Critical Essays in Modern Legal Thought, (Toronto: The Carswell Company Limited, and London: Sweet and Maxwell Limited, 1988).Google Scholar See infra, text accompanying notes 7–25.

7. Id.

8. See supra, n. 2. Although the central arguments of this essay are directed against the general sceptical position I shall here describe, I will also more specifically address arguments of two particular theorists in addition to Hutchinson. These are Mark Tushnet, infra, text accompanying n. 45–51, and Saul Kripke, infra, text accompanying n. 63–74. Tushnet and Kripke are particularly important because they both attribute radically sceptical conclusions to Wittgenstein. I shall try to show why their conclusions are mistaken.

9. It is important to acknowledge that it is not Wittgenstein’s writings which have inspired Hutchinson’s scepticism. His mentors are primarily the continental philosophers, Jacques Derrida and Michel Foucault. I intend here to set out only the barest outline of Hutchinson’s argument in order to indicate the sceptical presumptions from which his argument proceeds. It is impossible in the context of this essay to make a full argument for the accuracy or even the reasonable fairness of my interpretation of Hutchinson. Certainly, he often denies that he is committed to radical scepticism. See Hutchinson, supra, n. 6 at 43, 46–9. While I believe that these denials are inconsistent with Hutchinson’s general arguments, it is not essential to my argument in this essay to pursue that point, for I believe that in any case the arguments which I here outline, in varying forms and degrees, are reflected widely in the literature of scepticism. Accordingly, while I appeal to Hutchinson as both inspiration and illustration for the sceptical arguments here presented, I shall attribute the sceptical conclusions only to the hypothetical theorist, “The Sceptic”.

10. Id., at 25.

11. id., at 25.

12. Id., at 27.

13. Id., at 37.

14. Id., at 27.

15. Maclntyre, A. After Virtue, 2d ed. (Notre Dame: University of Notre Dame Press, 1984).Google Scholar

16. Hutchinson, supra, n. 6 at 27.

17. Id., at 28.

18. Id., at 29.

19. “The history of philosophy is philosophy tout court. To understand a philosophical proposal and its career, it is important to appreciate its historical context; it can only begin to make sense in terms of an agenda of pressing issues, the methodologies on offer, the bounds of political propriety and the like. To put it crudely, philosophers begin and end with their own prejudices as understood within the larger social context of available values”.

“The critical upshot of this argument (i.e., the inextricable relation of philosophy and history) is that modes of thinking and discourse are themselves revealed as contingent. The great systems of philosophy and styles of theorizing about the human condition are not fixed nor immutable. Id., at 32.

20. Id., at 33.

21. Hutchinson’s seems to be an overly simplistic view of Derrida’s theory which more fundamentally also attacks the idea that meaning can be situated contextually, since context is itself boundless. See Patterson, supra, n. 1 at n. 58 for a summary of Derrida’s claim and critiques of his theory. Patterson concludes: “Derrida’s claims for the boundless nature of context have no purchase on non-totalizing theories of social knowledge of the type advocated by the later Wittgenstein nor, for that matter, speech act theories such as those of Austin, Searle and Grice.”

22. Hutchinson, supra, n, 6, at 36.

23. Id., at 39.

24. Id., at 288.

25. Id., at 297.

26. Wittgenstein, supra, n. 1.

27. See subsection A, infra.

28. See subsection B, infra.

29. See subsection C, infra.

30. Langille, supra, n. 2 at 489–498.

31. Wittgenstein, supra, n. 1 at para. 201.

32. Langille’s primary example is Mark, Tushnet “Following the Rules Laid Down: A Critique of Interpretivism and Neutral Principles” (1982), 96 Harv. L. Rev. 781.Google Scholar

33. Langille, supra, n. 2 at 465.

34. See supra, text accompanying n. 7–25.

35. Wittgenstein, supra, n. 1 at para. 201.

36. To express the issues in terms of my schematic account of sceptical arguments, we might say that Langille is primarily concerned to show that Wittgenstein was not himself a ‘P-sceptic’ and that he offered a theory of language which was not ‘P-sceptical’, for he did not doubt the effectiveness or stability of language or the possibility of following a rule. I agree with Langille here and his argument is important to my general thesis. However, I want to focus, perhaps more than Langille does, upon what I perceive to be Wittgenstein’s ‘X-scepticism’, for I believe that it is equally important, and that it has its own implications for legal theory. Indeed, I am perhaps more ready than Langille to accept that some of the conclusions about the nature of legal reasoning drawn from Wittgenstein’s insights by critical legal scholars are correct.

37. See supra, text accompanying n. 10–11.

38. See Wittgenstein, supra, n. 1, at para. 66: “…don’t think, but look!”

39. In Philosophical Investigations Wittgenstein presented much of the argument in the form of a dialogue with an imaginary interlocutor whose comments are usually distinguished in the text by quotation marks.

40. Wittgenstein, supra, n. 1 at paras. 208–210.

41. 41. See id., paras. 78–80 for passages in which Wittgenstein attempts to show us the way in which our endorsement of or commitment to an abstract definition may fluctuate depending upon our experiential judgments.

42. Id., at para. 79.

43. This is the point about the difference between knowing and saying in paragraph 78:

78. Compare knowing and saying:

how many feet high Mont Blanc is—

how the word “game” is used—

how a clarinet sounds.

If you are surprised that one can know something and not be able to say it, you are perhaps thinking of a case like the first. Certainly not of one like the third.

There are many other passages of Philosophical Investigations, id., also relevant here. Consider, for example:

150. The grammar of the word “knows” is evidently closely related to that of “can”, “is able to”. But also closely related to that of “understands”. (“Mastery” of a technique.)

44. Also relevant here are paragraphs 84, 85, and, best of all, 87. These are quoted below. See text accompanying n. 55.

45. Tushnet, supra, n. 32 at 822. Tushnet indicates that this example is suggested by Winch, Peter, The Idea of a Social Science and lis Relation To Philosophy (London: Routledge and Paul, 1958) at 29–32,Google Scholar who in turn draws on Wittgenstein. Langille quotes more extensively from this passage from Tushnet’s article in Langille, supra, n. 2 at 464.

46. The latter may seem less obvious, but Kripke’s ingenious account of the “paradox” shows that this is so. See, infra, text accompanying n. 63–74 and in particular n. 64.

47. Tushnet, supra, n. 32 at 822–823.

48. See text accompanying n. 45.

49. Tushnet, supra, n. 32 at 822.

50. Although the main force of Tushnet’s argument is simply that context is essential to rule derivation, at some points in the argument he seems to come close to drawing the further odd and unjustifiable conclusion that this means that judges can “do whatever they want”. See Tushnet, id., at 824. But this, of course, would require that the judge disassociate herself not only from a particular context, but from any context at all. Yet the point of the argument is to show that this is not merely impossible, as a practical matter, but is conceptually incoherent. Rule following loses its meaning when abstracted in this way.

51. It is interesting to note that Hutchinson has himself fallen into this very trap in discussing Tushnet’s puzzle in a recently published reply to Langille’s article. After describing the theoretical ambiguity Tushnet identified in determining which two numbers come next in the series which begins “ 1,3,5,7...”, considered abstractly, Hutchinson makes these comments:

Where does all (his leave the student of mathematics or law? The only Langillean answer must be..—who knows? If “precedent” and “intention” are no guide, the hapless student is on her or his own. In such circumstances, it makes no sense to require a following of existing practice when the character of such practice is the very matter in issue. Even to advise a choice for elegance or symmetry seems self–defeating and question–begging. Whatever the case in mathematics, the legal situation is doubly or even trebly problematic. (Hutchinson, “That’s Just the Way It Is: Langille on Law” (1989), 34 McGill L.J. 145 at 157.)

The question which one must ask Hutchinson, however, is this: whatever the case in law, what difficulty does Hutchinson think that this puzzle creates for the student of mathematics! Is it not obvious that it creates no difficulty whatsoever for the student of mathematics? Moreover, if there is a difficulty for law, it is not one of theoretical ambiguity. It can only be one of actual ambiguity. There will, of course, be many problems of actual ambiguity in law, but it is important to recognize that Tushnet’s example does not speak to them. It does not illustrate the universality or the intransigence of actual ambiguity. Nor, for that matter, does the puzzle say anything at all about the possibility of challenging particular presumptions which may operate in any particular context. The induction puzzle does not show this to be problematic. It merely shows that the challenge will itself, of necessity, emanate from some context, and cannot be merely abstract and acontextual if it is to be coherent.

52. See supra, text accompanying n. 31.

53. Wittgenstein, supra, n. 1 at para. 86.

54. This is closely related to the point made in paragraph 152 where Wittgenstein is investigating (id., at paras. 143–155) what it is to “know how to go on” i.e., to give the next number in a series:

152...“B understands the principle of the series” surely doesn’t mean simply: the formula “a(n) =...” occurs to B. For it is perfectly imaginable that the formula should occur to him and that he should nevertheless not understand. “He understands” must have more in it than: the formula occurs to him. And equally, more than any of those more or less characteristic accompaniments or manifestations of understanding.

55. Id., at paras. 84–87.

56. Id., at para. 133.

57. Id., at paras. 201–02.

58. Id., at paras. 217,219.

59. As Saul Kripke, for example, seems to interpret these passages. See Kripke, Saul A., Wittgenstein on Rules and Private Language (Cambridge: Harvard University Press, 1982) at 10, 17, 55.Google Scholar

60. See supra, n. 55.

61. A good example of this mistake is illustrated by the following footnote in Tushnet’s article, where he is discussing Dworkin’s interpretative theory of adjudication:

....Dworkin’s analysis might alternatively be read as a call for a metatheoretical defense of interpretivism—that is, for a demonstration that a suitably flexible interpretivism is the mode of judicial decision making that best accords with our broader vision of a just society. This alternative project recalls an anecdote about Indian cosmography. A traveler, it is said, asked a wise man to describe the earth’s place in the universe. The wise man replied that the stars were on a bowl above the earth, that the earth rode on the back of an elephant, and that the elephant stood on the shell of a huge turtle. But what, asked the traveler, does the turtle rest upon? ”Ah, after that it is turtles all the way down,” was the reply.

Tushnet seems to be suggesting that no “interpretation” could satisfy us, for any “interpretation” can only create a need for a new “interpretation”, “all the way down.” See Tushnet, supra, n. 32 at 791–92 n. 32.

62. Wittgenstein, supra, n. 1 at paras. 198, 199. These passages are also quoted by Langille, supra, n. 2 at 489–90.

63. Kripke, supra, n. 59 at 7. Kripke describes this paradox as “a new form of scepticism....the most radical and original sceptical problem that philosophy has seen to date, one that only a highly unusual cast of mind could have produced.” (At 60.) Kripke’s description of this aspect of Wittgenstein’s argument as a “paradox” is significant, for it reveals Kripke himself as one who appreciates better than many the power of Wittgenstein’s arguments in support of ’X–scepticism’, and yet one who is ultimately unable to escape’the power of the theoretical picture which Wittgenstein’s arguments were meant to dispel. Wittgenstein’s successful challenge to the picture results in “paradox” only if one believes that the picture must, in some sense, be accurate. Although Kripke himself acknowledges that Wittgenstein offered a “solution” to this “paradox”, he also thinks that the solution is itself a “sceptical” one. It is “sceptical”, as far as Kripke is concerned, because it leaves untouched the conclusions of ’X–scepti-cism’, with which it begins. A “straight” solution, for Kripke, would show that ’X–scepticism’ is itself a mistake. (At 66–67.)

Many have written critically of Kripke’s argument, including Langille, supra, note 2. The most complete argument is that of Baker, CP. and Hacker, P.M.S., Scepticism Rules and Language (Oxford: Basil Blackwell, 1984).Google Scholar My own critique is, I believe, consistent with but somewhat distinct from both of these. For a more sympathetic account of Kripke’s argument see Yablon, Charles M., “Law and Metaphysics” (1987), 96 Yale L.J. 613.Google Scholar

64. Id., at 8–9.

65. In thinking of this, I was reminded of the card game “Blackjack” or “21” in which the winning hand is the one in which the face value of the cards comes closest to a total of 21 without exceeding that sum. Any hand which exceeds that sum automatically loses.

66. If it did not, then our past practice might itself have been actually ambiguous. E.g., we might have been in the process of learning the game, not yet fully understanding how this function would finally fit in.

67. For example, Kripke considers a reply to the sceptic that in the past I did not merely rely upon a finite number of examples of addition, but, rather, that I “learned and internalized instructions for—a rule which determines how addition is to be continued....I proceeded according to an algorithm for addition that I previously learned.” (Kripke, supra, n. 59 at 15–16.) This cannot extricate us from the sceptic’s problem, he argues, because if we seek to expound such a rule in terms of the concept of counting, for example, the sceptic’s puzzle simply recurs at another level.

68. This conclusion is, of course, attributed to Wittgenstein. See these passages in Kripke, id.,

We can put the problem this way: When asked for the answer to ’68+57’, 1 unhesitatingly and automatically produced ’125’, but it would seem that if previously 1 never performed this computation explicitly 1 might just as well have answered ’5’. Nothing justifies a brute inclination to answer one way rather than another. (At 15.)

The sceptical argument, then, remains unanswered. There can be no such thing as meaning anything by any word. Each new application we make is a leap in the dark; any present intention could be interpreted so as to accord with anything we may choose to do. So there can be neither accord, nor conflict. This is what Wittgenstein said in paragraph 202. (At 55.)

...Wittgenstein’s main problem is that it appears that he has shown all language, all concept formation, to be impossible, indeed unintelligible. (At 62.)

69. Langille, supra, n. 2 at 465, n. 58.

70. See Langille, id., at 491–497.

71. This view of the role of community, in Wittgenstein’s writing, in determining the meaning of a rule, is the subject of some controversy. See Norman Malcolm, “Wittgenstein on Language and Rules” (1989), 64 Philosophy 5. Malcolm disputes the position taken by CP. Baker and P.M.S. Hacker that the relationship between a rule and what it requires is essentially an “internal” relationship. Malcolm believes that it is a question of communal consensus. Baker and Hacker argue that communal agreement has only a limited significance in that shared language games assume “a framework of agreement in behaviour.” ( Baker, G.P. and Hacker, P.M.S., Wiitgenstein: Rules, Grammar and Necessity (Oxford: Basil Blackwell, 1986) at 249, quoted in Malcolm, at 16.)Google Scholar In general, however, a practice itself, they argue, must be sharable but need not be shared. Thus they believe that for Wittgenstein a community, per se, is not a necessary condition for a language. Malcolm argues that meaning must as a matter of logic be independent of what an individual speaker believes and that “this independence condition can be satisfied only if there is a community of speakers who use the sign in a customary way.” (At 28.) I should not like to dispute with Malcolm, Baker or Hacker about what Wittgenstein himself actually thought about this. However, it is my view that the analysis I have presented here supports only the double point (consistent with the Baker/Hacker position) first, that meaning is inherently sharable (and therefore not “private” in a sense that would render this impossible) and second, that most language games of any degree of complexity do, as a contingent matter, depend upon a community of agreed practice for their logic or grammar.

72. See supra, text accompanying n. 58.

73. Wittgenstein, supra, n. 1 at para. 78.

74. Id., at paragraphs 29–30. See also paragraph 31 :

....the words “This is the king”....are a definition only if the learner already ’knows what a piece in a game is’. That is, if he has already played other games, or has watched other people playing ’and understood;—and similar things....Vie may say: only someone who already knows how to do something with it can significantly ask a name.

75. Id., at para. 21, and see also, para. 23.

76. In the sense discussed supra, n. 71.

77. Wittgenstein, supra, n. 1, at para. 23.

78. Id., n. al bottom of English page 18.

79. A similar point can be made about creative use of language, which is common in poetry but may occur in even the most prosaic contexts. The idea of “community practice” should not be understood as an entrenchment of the “status quo”. Both challenge and creativity are possible, and, indeed, may themselves be established practices! Nonetheless, creative use is parasitic upon established use. It is established use which makes it possible and gives it a point.

80. Wittgenstein, supra, n. 1 at para. 33. And see also para. 31: “We may say: only someone who already knows how to do something with it can significantly ask a name.”

81. Id., at paras. 66 and 67. There is another point in this passage which is worth mentioning. That is Wittgenstein’s insistence, “Don’t say: There must be....Look and see!” The passage quoted, of course, illustrates the wisdom of that imperative. Wittgenstein shared The Sceptic’s scepticism oí abstract rationalism. But this did not, of course, render him philosophically mute. It suggested a better way of doing philosophy.

82. Wittgenstein, supra, n. 1 at para. 70.