I am grateful to Neil Walker and Bernard Jackson for comments on and discussion of previous drafts of this paper. I am also grateful to an anonymous CJLJ reviewer’s comments and suggestions regarding the treatment of Reid’s philosophy.

1. A particularly heated debate on this issue was conducted between Searle and Derrida. See: Derrida, J., “Signature, Event, Context” (1977) 1 Glyph 172 Google Scholar; Searle, J., “Re-iterating the Differences” (1977) 1 Glyph 198 Google Scholar; Derrida, J., “Limited Inc. abc” (1977) 2 Glyph 162 Google Scholar; Searle, J., “The World Turned Upside Down” New York Review of Books 30.16 (27 October 1987) 74 Google Scholar; Derrida, J., “Afterword: Toward an Ethic of Discussion” in his Limited Inc. (Baltimore, MD: Johns Hopkins University Press, 1977) 111 Google Scholar; Searle, J., “Literary Theory and its Discontents” (1994) 25(3) New Lit. Hist. 637.CrossRefGoogle Scholar

2. On the role of philosophy in the university curriculum during the Scottish Enlightenment see, Davie, G., The Democratic Intellect: Scotland and her Universities in the Nineteenth Century (Edinburgh: Edinburgh University Press, 1961) at 3–25 Google Scholar.

3. The view that Austin subscribes to the volitional theory of meaning may appear at first sight incongruous, in particular in the light of Austin’s commitment to the Wittgensteinian notion that meaning is determined according to use. His desire however to distinguish between properly executed and failed speech acts leads to a re-introduction of intention into his theoretical scheme. See Culler, J., “Convention and Meaning: Derrida and Austin” (1981) 13 New Lit. Hist. 15.CrossRefGoogle Scholar

4. See Hume, D., A Treatise of Human Nature, ed. by Selby-Bigge, (Oxford: Clarendon Press, 1978) at 7 Google Scholar (§1). Hume also refers to his position as a “first principle”.

5. Reproduced in Philosophical Works I/II (Hildersheim: Olms, 1983) at 117–18 Google Scholar. Wittgenstein appears to reach a similar conclusion: Wittgenstein, L., Philosophical Investigations (Oxford: Blackwell Publishing, 1976) at §86.Google Scholar

6. Reid, supra note 5 at 230.

7. Ibid. at 441 et seq.; see also at 230 et seq.

8. See ibid. at 439: “In this kind of proof [reductio ad absurdum] … we suppose the contradictory proposition to be true. We trace the consequences of that supposition in a train of reasoning; and, if we find any of its necessary consequences to be manifestly absurd, we conclude the supposition from which it followed to be false; and therefore its contradictory to be true.”

9. Ibid. at 442.

10. Ibid. at 445. “If any many should think fit to demand a proof that the thoughts he is successively conscious of, belong to one and the same thinking principle—if he should demand a proof that he is the same person to-day as he was yesterday, or a year ago—I know of no proof that can be given him: he must be left to himself, either as a man that is a lunatic, or as one who denies first principles, and is not to be reasoned with” (232).

11. Ibid. at 232.

12. This distinction appears to be justified on the basis of the different temporal coordinates of the mental operation (which is by definition present) and its object (which may be past, continuous or future).

13. Supra note 5 at 450, 233 Google Scholar. There is a possible parallel here with the core concept in Habermas’ discourse ethics: “The condition for the truth of statements is the potential agreement of everyone else.” Habermas, J., Vorstudien und Ergänzungen zur Theorie des Kommunikativen Handelns (Frankfurt: Suhrkamp, 1984) at 137 Google Scholar.

14. Supra note 5 at 449 Google Scholar. This includes the innate capacity to link certain signifiers and signifieds.

15. Ibid. at 233: “Although some writers on this subject have disputed the authority of the senses, of memory and of every human faculty, yet we find that such persons, in the conduct of life, in pursuing their ends [such as arguing in favour of scepticism (T.R.)], or in avoiding dangers, pay the same regard to the authority of their sense and other faculties, as the rest of mankind. By this they give us just ground to doubt of their candour in their professions of scepticism.”

16. Ibid. at 234.

17. Much of the following analysis of speech acts in Reid’s writings is inspired by Schumann, K. & Smith, B., “Elements of Speech Act Theory in the Work of Thomas Reid” (1990) 7 Google Scholar Hist. Phil. Q. 47.

18. Letter to Dr Gregory, James, 26th August 1787, supra note 5 at 71 Google Scholar.

19. Reid also offers the following argument: “That the parts of speech should be conceived before speech was in use, and that speech should at first be formed by putting together parts of speech, which before had got names, seems to me altogether incredible; no less incredible than if it should be said that before men had the conception of a body, they first formed the conception of matter, then the conception of form, and, putting these two together, they got the conception of body, which is made up of matter and form.” Ibid. at 71. The following thought experiment also supports Reid’s view: image that we find ourselves before a person with whom we share no language to whom we wish to communicate the fact that we are hungry. It would be absurd first to endeavour to establish a concept of selfhood, followed by a concept of a transient state of being, together with the concept of a need for nutrition, and then finally to combine the three elements into a meaningful whole. We should rather try to negotiate a common signal ex Pressing the concept “I am hungry”.

20. Ibid. at 72. Reid also argues that certain features of the structure of language are universal (e.g., nouns, adjectives, verb voices, verb tenses, adverbs, prepositions and conjunctions: supra note 5 at 233), which is arguably a precursor of Chomsky’s generative grammar. This does not however necessarily contradict his position that the phrase is the first manifestation of language in human society; it is plausible that a deep underlying generative grammar (if it does indeed exist) only becomes apparent in the form of analytical units within a sentence once a language has reached a certain degree of sophistication.

21. Smith, B., “Towards a History of Speech Act Theory” in Buckhardt, A., ed., Speech Acts, Meanings and Intentions: Critical Approaches to the Philosophy of John R. Searle (Berlin: W. de Gruyter, 1990) 29 Google Scholar at 29.

22. Reid’s short treatise on Aristotle’s logic (supra note 5 at 681-714, originally published in 1774 as part of Lord Kames’ Sketches on the History of Man) ex Presses dissatisfaction with the restriction to propositions. Considering other types of speech, such as questions, commands, promises and contracts, Reid states: “yet I apprehend that an analysis of such speeches, and of the operations of mind which they ex Press, would be of real use, and perhaps would discover how imperfect an enumeration the logicians have given of the powers of human understanding, when they reduce them to Simple Apprehension, Judgement, and Reasoning” (Section V).

23. Reid, , supra note 5 at 509 Google Scholar et seq.

24. Ibid. at 663-64.

25. Also discussed ibid. at 245: “The ex Pression of a question, of a command, or of a promise, is as capable of being analysed as a proposition is; but we do not find that this has been attempted.” See infra note 30.

26. Ibid. at 664.

27. Ibid.

28. See Reid, supra note 5 at §4.

29. Ibid. at 238. Some of these operations are repeated elsewhere, e.g., at 244, 664.

30. Cited by Calabrese, O., Breve Storia della Semiotica: dai Presocratici a Hegel (Milano: Feltrinelli, 2001) at 27–28 Google Scholar.

31. Austin, J., How to do Things with Words (Oxford: Oxford University Press, 1976) at 153–63 Google Scholar.

32. Supra note 5 at 664.

33. Ibid. at 449.

34. Ibid. at 233.

35. Ibid. at 232.

36. An utterance is a meaningful speech act where it is made in appropriate circumstances with the requisite intention and by a person in possession of the appropriate authority.

37. Supra note 5 at 532 (discussing commands) and at 667 (discussing contracts).

38. Ibid. at 664. Habermas, to cite one contemporary example, goes against Reid, claiming that: “The felicity of the illocutionary act is understood as the disclosure from the manifest meaning of what is said of the communicative intention of the speaker and the intended illocutionary purpose.” (Theorie des kommunikativen Handelns Bd. 1 (Frankfurt: Suhrkamp, 1981) at 389).

39. See Austin, supra note 31 at 109, “a locutionary act … is roughly equivalent to uttering a certain sentence with a certain sense and reference.” J. Searle recognises that some “very simple sorts of illocutionary acts” can be performed “while standing outside language”, but “this should not obscure the fact that in general illocutionary acts are performed within language in virtue of certain rules, and indeed could not be performed unless language allowed the possibility of their performance.” (Speech Acts: an essay in the philosophy of language (London: Cambridge University Press, 1969) at 38.) Austin too concedes the possibility of non-verbal speech acts, but he too gives complete priority to language—supra note 31 at 121.

40. Supra note 5 at 667.

41. Supra note 5 at 664. One should not read too much into Reid’s choice of terminology. Nevertheless, his reference to ‘acting’ would in present times be regarded as a reference to the structural linguistics of the Prague School. See Greimas, A.-J., Sémantique Structurale (Paris: Larousse, 1966) at 180 Google Scholar et seq., developing the theory initially presented in Propp, V.J., Morfologia della Fiaba(Roma: Newton, 2003)Google Scholar, especially chapter two on methodology. An interesting and compelling application of Greimasian semiotics to legal theory—analysing the determination of fact and of law within the courtroom—is made in Jackson, B.S., Law, Fact and Narrative Coherence (Roby, Merseyside: Deborah Charles, 1988)Google Scholar.

42. Austin defines infelicities in terms of “not having the requisite feelings … thoughts … intentions”—supra note 31 at 40; thus, to use Reid’s terminology, Austin holds the felicity of a speech act to be entirely dependent on solitary operations. Searle’s view is somewhat more nuanced: “Illocutionary acts consist simply in uttering words in sentences in certain contexts, under certain conditions and with certain intentions.” Supra note 39 at 24-25.

43. Supra note 31 at 14–15 Google Scholar, 76.

44. Ibid. at 52.

45. Sbisà, M. & Fabbri, P., “Models (?) for a Pragmatic Analysis” (1981) 4 Google Scholar J. Pragmatics 301.

46. Ibid. at 309.

47. Ibid. at 313.

48. Ibid. at 313.

49. Ibid. at 304. See also supra note 39.

50. Supra note 31 at 18. The notion that incorrect uses could be made of language acts did not begin with Austin. A possible precursor is for example Hobbes, who speaks of “abuses of speech” ( Hobbes, Thomas, Leviathan, ed. by MacPherson, C.B. (London: Penguin Books, 1985), part 1 Google Scholar, ch. IV—cf. Austin who speaks of: “misfires” and “abuses”); the first occurs “when men register their thoughts wrong, by the inconsistency of the signification of their words” (cf. misinvocation), the second “when they use words … in another sense than that they are ordained for” (cf. mis-execution), the third “when by words they declare that to be their will, which is not” (cf. insincerity) and the fourth “when they use [words] to grieve one another.” That Hobbes reaches such conclusions is not surprising, given his understanding of meaning in language; “The generall use of Speech, is to transferre our Mentall Discourse, into Verbal; or the Trayne of our Thoughts, into a Trayne of Words” (part 1, ch. IV, ‘The use of Speech’).

51. E.g., an attempted divorce in a country which did not allow it.

52. E.g., an attempted marriage by unilateral declaration.

53. E.g., a marriage scripted as part of a play performed on stage between actors.

54. On this point it is interesting to note the criticism of Reid brought by Schumann and Smith (supra note 17) in §4 that Reid does not pay sufficient attention to infelicities. The remark is however only made in passing and not within the context of a full consideration of the place of infelicities within Reid’s theory. Such an approach reveals the pitfalls inherent in any consideration of a theory (such as here speech act theory) on that theory’s own terms and which does not challenge the assumptions underlying the theory.

55. Supra note 31 at 155-56. The list at p.156 includes ‘order’ and ‘command’.

56. As in Kurzon, D., It is Hereby Performed (Amsterdam: Benjamins, 1986)CrossRefGoogle Scholar.

57. Jackson, B., Semiotics and Legal Theory (London: Routledge, 1985) at 233.Google Scholar Kelsen states that “the meaning or the significance of the act or acts that constitute the legislative process, is a norm.” (Pure Theory of Law (Berkeley: University of California Press, 1967) at 7.)

58. Hart acknowledges that certain assumptions about the social circumstances of statements underlie his theory. See Hart, H.L.A., Essays in Jurisprudence and Philosophy (Oxford: Oxford University Press, 1983) at 27 CrossRefGoogle Scholar: “It is obvious that the use of sentences [ex Pressing legal obligations] silently assumes a special and very complicated setting, namely the existence of a legal system.”

59. Hart recognises a terminological difficulty in English: “One of the difficulties to be faced in the analysis of the general notion of ‘imperative’ is that no word exists for what is common to orders, commands, requests and many other varieties, i.e., the ex Pression of intention that another should or should not do some action.” (The Concept of Law (Oxford: Clarendon Press, 1961) at 235, note to page 19.)

60. MacCormick, N., H.L.A. Hart (London: Arnold, 1981) at 15 Google Scholar.

61. An example is the acceptance of the relevance of the ‘family resemblance’ notion for law and rules: supra note 59 at 234: “Wittgenstein’s advice … is peculiarly relevant to the analysis of legal and political terms.”

62. In the essay “Definition and Theory in Jurisprudence”, Hart endorses a Wittgensteinian focus on use, citing Bentham as authority: “we must never take [legal] words alone, but consider whole sentences in which they play their characteristic role. We must not take the word ‘right’ but the sentence ‘You have a right’.” Supra note 58 at 26; see also at 33. Hart also states in the introduction to this book that the linguistic philosophy of Wittgenstein and Austin contained a “conviction that longstanding philosophical perplexities could often be resolved not by the deployment of some general theory but by sensitive and piecemeal discrimination and characterisation of the different ways … in which human language is used.” Ibid. at 2. See also B. Jackson, supra note 57 at 148.

63. MacCormick, supra note 60 at 13.

64. Hart, supra note 59 at 17.

65. Ibid. at 15.

66. Ibid. at 14.

67. Ibid. at 54-55.

68. Ibid. at 114.

69. Ibid. at 89.

70. Ibid. at 19.

71. Ibid. at 89.

72. Paraphrasing Hart, supra note 58 at 28.

73. Hart, supra note 59 at 18.

74. Ibid. at 44.

75. Ibid. at 48.

76. Ibid. at 235 (note to page 18).

77. This is a specific application of Hart’s general theory of meaning. The theory is discussed by Jackson who notes “a refusal to detach the meaning of the words from the act (and intention) of the person uttering those words.” Supra note 57 at 150. An obvious practical difficulty in founding meaning upon intention is how one actually identifies the intention. Hart recognises the “dif-ficult[y] of proof” of legally relevant mental states so that “the law … may often adopt what are called external or objective standards, which treat certain forms of outward behaviour as conclusive evidence of the existence of mental states or impute to an individual the mental state that the average man behaving in a given way would have had” (“Problems in the Philosophy of Law” in Hart, supra note 58 at 96.) Despite Hart’s claim that “there are divergencies between the legal and non-legal use” of the notions of ‘will’, ‘intention’ and ‘motive’, it is difficult to see how similar problems of proof of the intention behind imperatives cannot be resolved with reference to external rules imputing mental states. Reaching this conclusion would however directly challenge Hart’s position that the production of meaning of legal rules is structurally different from the production of meaning of imperatives.

78. Hart is acquitted of the charge of advocating a will theory of law by Olivecrona: Olivecrona, Karl, Law as Fact (London: Stevens, 1972) at 83 Google Scholar. For Hart’s position on the gunman’s imperatives, see supra note 59 at 22, where it is stated that he “ex Presses his wish that the clerk should do something.”

79. See Hart, supra note 59 at 78-79, where he appears to describe primary rules as imperatives: “Under rules of the one type, which may well be considered the basic or primary type, human beings are required to do or abstain from doing certain actions, whether they wish to or not.” Law is described as the “interplay” between these rules and secondary rules (at 79). It is difficult to see any fundamental difference between the imperatives of the gunman and Hart’s ‘primary rules of obligation’; on imperatives: “the statement that a person was obliged to obey someone is, in the main, a psychological one referring to the beliefs and motives with which an action was done” (81); on primary rules of obligation: “Rules are conceived and spoken of as imposing obligations when the general demand for conformity is insistent and the social Pressure brought to bear upon those who deviate or threaten to deviate is great” (84). If obligations are accounted for in terms of the ‘ Pressure for conformity’, then obedience of such rules must too be described in terms of ‘beliefs and motives’.

80. Cf. Olivecrona’s ideatum and imperantum, supra note 78 at ch. V.

81. “The acts whose meaning is a norm are acts of will.… ‘Norm’ is the meaning of an act by which certain behavior is commanded, permitted, or authorized.” Supra note 57 at 5.

82. “The norm functions as a scheme of interpretation. To put it differently: The judgment that an act of human behavior … is ‘legal’ … is the result of a specific, namely normative, interpretation.” Ibid. at 4.

83. Jackson, B., Making Sense in Jurisprudence (Liverpool, UK: Deborah Charles, 1996) at 101 Google Scholar.

84. Supra note 57 at 7. Kelsen goes on to argue that “the command of the gangster to hand over to him a certain amount of money has the same subjective meaning as the command of an income-tax official, namely that the individual at whom the command is directed ought to pay something” (8).

85. Ibid. at 9.

86. “The custom becomes the ex Pression of a collective will whose subjective meaning is an ought.” Ibid.

87. This formulation is perhaps over-simplistic. It is certainly true that the nature of the social fact of the legal norm is not determined with reference to the intention, but rather to an external feature such as another norm providing a scheme for interpretation. However, this does not fundamentally affect the argument above. Kelsen’s legal norm can be equated with a felicitous speech act (which derives its efficacy from performance in appropriate circumstances and according to appropriate procedures) which is a successful realisation within a given environment of the intention inherent in the speech act (or in Kelsen’s terms, the successful confirmation of the subjective meaning of an act of will as the objective meaning of an act of will). Meaning is an all cases ultimately derived from the intention of the speaker as manifested in the linguistic articulation which is indicative of the rule.

88. Kelsen, supra note 57 at 71. For a concrete application, see ibid. at 260: “The contract as a norm-creating fact, and the norm created by this fact, need to be clearly distinguished.”

89. Supra note 57 at 234.

90. Kelsen notes that, other than the objective meaning derived form another norm, there is no difference in the commands of a robber and law, as they are both “the ex Pression of a will directed toward the behavior of another individual.” Supra note 57 at 44.

91. Ibid. at 3.

92. Ibid. at 4.

93. Kelsen refers to the Grundnorm using different terminology, describing it as the “meaning of an act of thinking”, though this is a rather loose translation of the original “gedachte Norm“ (lit. “thought norm”). Since however the Grundnorm is by definition incapable of being objectively validated by a legal norm, it is closer to the subjective meaning of an act of will. The rule conferring validity on the rules of the robber gang for instance remains the subjective meaning of an act of will unless and until the gang overthrows the institutional powers of state, at which point this rule is assumed as a new Grundnorm.

94. Supra note 57 at 45-46. He later argues that the necessity of the existence of the Grundnorm can be established transcendental argument. Ibid. at 202. See further infra note 101.

95. Ibid. at 47: the robber gang cannot presuppose a basic norm because their coercive order has no lasting effectiveness.

96. Reid’s derivation of first principles from Common Sense bears striking similarities with Kant’s use of transcendental arguments. “It is absurd to conceive that there can be any opposition between reason and common sense …We ascribe to reason two offices …. The first is to judge of things self-evident; the second to draw conclusions that are not self-evident from those that are.” Supra note 5 at 425. See also Grave, S.A. in The Scottish Philosophy of Common Sense (Oxford: Clarendon Press, 1960)Google Scholar: “Indeed all the Common Sense philosophers … are sooner or later willing or anxious to use on occasions ‘common sense’ and ‘reason’ interchangeably” (115).

97. It is perhaps on this issue that the conclusions of the pure theory are most tenuous. Kelsen considers whether the commands of robbers could be analysed as legal rules (i.e., being derived form a presupposed Grundnorm), rejecting this possibility by relying on the sociological argument that such rule systems have no “lasting effectiveness”. Supra note 95. It is however not clear why effectiveness should be “lasting” in order for rules to be understood as being legal; at a certain level, no system of rules is lasting because of the possibility of revolution, annexation by another state, etc. Furthermore, the issue over whether a normative system will still be effective in future does not of itself alter its current efficacy. Kelsen however must make this distinction, as it is central to his separation between legal and non-legal rules.

98. Supra note 59 at 105.

99. “Without their voluntary co-operation, thus creating authority, the coercive power of law and government cannot be established.” Ibid. note 59 at 196. We might speculate over three possible manifestations of this cooperation:

• –

  – Is Hart endorsing a theory of social contract under which the voluntary aspect is merely a hypothetical historical element? If so, the pertinent historical facts are assumed. Furthermore, the authority according to which this social contract would be binding is assumed, as is the authority upon which this authority is founded, and so on. Attempts to found the legitimacy of law on a social contract fall prey to an infinite regress. This criticism is not new and was made around the turn of the century by Petrazycki: see Rudzinski, A.W., “Petrazycki’s Importance for Contemporary Legal and Moral Theory” (1976) 21 Am. J. Juris. 107 at 115.CrossRefGoogle Scholar

• –

  – Is the argument based on the normative claim that human society necessarily needs authority in order to survive? If so, the relevant observations on human nature and communal life are assumed. Furthermore even if the necessary empirical basis could be provided, this argument would still not be able to account for any qualitative difference between the dictatorial authority of the gunman and the institutional authority of a legal system; some authority may be required, but it would not necessary be legal or governmental.

• –

  – Is this requirement of a voluntary aspect a normative conclusion of Hart’s, under which there exists an extra-legal duty to accept the ‘secondary rules’ of the legal system? If so, the mechanism by which the extra-legal duty is binding on the subjects (such as natural law) is assumed. Moreover, this argument too would fall prey to a malign infinite regress, because the extralegal duty would subsist by virtue of an extra-legal authority. Thus the problem of the nature of authority would be pushed back onto a higher level of (transcendent) normativity.

Hart’s attempt to establish a qualitative basis to authority and avoid the need to assume it, itself necessarily appears to invoke some sort of assumption, which either falls into an infinite regress or blurs the distinction between legal and plain rules. For the purposes of this paper a presumption is of a metaphysical nature where its acceptance is mandated as a matter of faith (as opposed to a transcendental presumption, which is necessary in order for an activity to be possible or meaningful).

100. It may be argued that Hart justified authority on the grounds of the sociological effectiveness of the secondary rules, in the sense that an internal aspect exists in respect of the system’s officials (supra note 59, see ch. VI, in particular at 113). This criterion can however only be regarded as an acceptable foundation for authority if an additional rule is presupposed according to which any rule of recognition which has sufficient sociological support is presumed to be valid. This point is accepted by Hart when arguing against the use of sociological criteria to determine the validity of primary rules:

Were this not to be presumed then it would be necessary to search for the authority according to which this higher rule conferring validity on the rule of recognition were itself valid. This would however bring the argument full circle by reposing the initial problem of establishing the ultimate ground for a system of rules.

101. It would perhaps be more accurate to term Kelsen’s argument as “contingently transcendental”. This is because it is founded on a notion of rule which can be opened to debate, as it is not necessarily so that we conceptualise rules as intentions. Kant argues inter alia that time and space are a priori necessary because it would be impossible for us to make sense of our perceptions without them. Kelsen argues that the Grundnorm is a priori necessary because we could not make sense of intentions as having objective validity as legal rules without it. The difference between the two arguments lies in the fact that we have no choice as to whether to make sense of our perceptions, whereas Kelsen has made an active choice in favour of conceptualising rules as intentions. This means that the ground can be pulled from under the argument by challenging the volitional view of rules upon which it is premised. To the extent therefore that Kelsen enjoys a choice over whether to sign up to the volitional theory of meaning, his argument is contingently transcendental which is of course an oxymoron.

102. Supra note 31 at 28. Searle offers a more complex analysis, though the conclusion is essentially identical to Austin’s; authority is regarded as a “preparatory condition” of an order (see supra note 39 at 64). Preparatory conditions appear to be those circumstantial elements which must accompany the enunciation of a speech act in order for it to be meaningful.

103. Some correction of this statement is offered several lines later: “the person, to be the object of the verb ‘I order to …’ must, by some previous procedure, tacit or verbal, have first constituted the person who is to do the ordering an authority.” This claim is however open to the same objections which have been brought above against the similar claim made by Hart.

104. Speech act theory is subjected to similar criticism by Sbisà and Fabbri who propose that any theory of speech acts must take account of the intersubjective construction of meaning, together with certain instabilities inherent in this process. The authors argue that the participants in the act must ‘negotiate’ (this terminology was introduced by Jackson, supra note 83 at 123) the meaning of the speech act, and “build up their context.” Supra note 45 at 316.

105. Cf. Kelsen, supra note 57 at 2.

106. In Petrazycki’s terms, positivist legal theory is “lame” along the following lines:

Similarly, the positivist account of law (whether as the “union of primary and secondary rules” or as a behavioural standard derived from a Grundnorm) can be applied to all normative statements, irrespective of whether they are legal or legitimate or not.

107. The structural distinctions should be familiar: Hart argues that law consists of the union of primary and secondary rules, whereas non-legal commands are simply primary rules; Kelsen argues that the meanings of acts of will have legal status when the meaning is conferred by a superior authorising norm, whereas non-legal commands derive their meaning simply from the subjective content of the act of will. Kelsen recognises this potential weak point, in seeing that his structural scheme could arguably also be applied to gangsters’ commands (see supra note 78), though he ultimately seeks refuge in the sociological distinction of effectiveness. Hart does not consider the argument that primary rules must, in order to be recognised as such, contain a secondary element of recognition; for example, the physical gun acts as the means for endowing the enunciations of the gunman with normative status, since any person who is not familiar with guns and does not recognise them as lethal weapons will not understand the primary enunciation as being normatively applicable to their own behaviour.

108. A strong sociological approach can be traced back at least to Petrazycki, whose theory of law (or more generally of “normative experiences”) was based on the notion of a command as an “impulsion”: Petrazycki, L., Law and Morality (Cambridge, MA: Harvard University Press, 1955)CrossRefGoogle Scholar at 31 et seq. Petrazycki’s advocacy of a move in conceptual focus from that of norm-sender to norm-receiver is founded on a critique of the volitional view, which he argued was based on a misunderstanding of the word ‘will’:

The foundation of theory on colloquial language is to be avoided. As soon as the will is set aside as the origin of normative meaning, factors lying outwith the subject must be considered, and the nature of the analysis becomes by definition sociological.

Nelken has argued against a sociological approach, claiming that “both law and sociology necessarily operate with their own ‘internal’ distinctions—in terms shaped by their own disci-plines—between what they each deem to be external and internal to law.” Thus sociologists argue in favour of a sociological approach to law because their world view is naturally framed in sociological terms. ( Nelken, D., “Can There Be a Sociology of Legal Meaning” in Nelken, D., ed., Law as Communication(Aldershot, UK: Dartmouth, 1996) 107 Google Scholar at 112). Whilst it is a truism to say that the sociological approach to law involves a reduction of law to sociological terms, it is submitted here that this reduction is not imposed by any choice of the sociologist (as was done in the theories which Nelken criticises), but rather that positivist legal theory of necessity reduces itself into sociology. It has been argued above that positivist theory is premised on presuppositions relating to context and authority, both of which can only be analysed in sociological terms. If legal theory attempts to retain its specificity as a separate discipline, it can only do so by obscuring such foundational presuppositions. This is because any acceptance that rules in general can only be properly understood where the positivist presuppositions of context and authority are fully accounted for necessarily results in any investigation into law in particular being prefaced by a sociological analysis of these presuppositions. Where therefore such an analysis is not undertaken (and legal theory is not therefore ‘contaminated’ by sociology), the meaning of ‘context’ and ‘authority’ remains unclarified and these concepts acquire the status of metaphysical foun-dational notions upon which the whole edifice of legal positivism as an autonomous discipline is constructed.

109. Cotterrell, Roger, “Why Must Legal Ideas Be Interpreted Sociologically?” (1998) 25 J. L. & Soc’y. 171 at 181.Google Scholar

110. The colonialist metaphor is taken from Cotterrell, supra note 109 at 191, and its use by him appears to be pejorative in the sense that it accuses certain sociologists of law of unjustifiably sup Pressing the distinctive and autonomous character of law in favour of a way of thinking which is alien to that of lawyers. Cotterrell justifies the adoption of a weak sociological approach on the basis of a distinction between ‘normative’ and ‘empirical’ legal theory, which in turn is founded on a par-ticipant/observer dichotomy ( Cotterrell, R., Law’s Community: Legal Theory in Sociological Perspective (Oxford: Clarendon Press, 1995) at 24–25 Google Scholar). Therefore, “the numerous approaches to legal analysis that can be categorised as sociological in the broadest sense are unified only by their deliberate self-distancing from the professional viewpoint of the lawyer” (25). Normative legal theory on the other hand deals with “the construction of a professionally plausible and logically coherent concept of law as doctrine” (25). He goes on to make the somewhat enigmatic assertion that: “This is notwithstanding the fact that the techniques of normative legal theory can be and are applied to analysis of normative systems other than that suggested by lawyers’ pro-fessional doctrine” (25). This paper argues that the necessary consequence of the possibility of applying these techniques to any normative system is the colonisation of legal theory by sociology.

111. Ehrlich argues that “the inner order of associations” provides normative standards for people’s conduct, and that the law is simple a manifestation of this inner order. Ehrlich, E., Fundamental Principles of the Sociology of Law (Cambridge, MA: Harvard University Press, 1936) at 37.Google Scholar Ehrlich argues that ‘law in books’ is an unnecessary abstraction from the more fundamental living law (i.e., ‘law in action’):

112. Harris, J. W., Legal Philosophies (London: Butterworths, 1997) at 257.Google Scholar

113. This is based on the factual claim that “the statement that legal institutions are based exclusively on legal norms is not true”: supra note 111 at 55. Indeed, Ehrlich takes comfort in the fact that this is not the case, since “life would become a hell if it were regulated by law alone” (58). There are many types of social behaviour which are subject to legal regulation but which are in fact regulated by non-legal social practices and where it may be improper for a person to insist on a legal right where social practice would class a different course of action as normal. This most commonly occurs where the social relationship between two people is destined to outlive the particular legal relationship at issue. The law treats social situations in isolation, whilst social practices take into account the fact that those involved may have to co-exist (and even cooperate with one another) in future. Examples can be found in: trade practices which de facto modify the provisions of contract law; the cooperative reality of family life, even in all but the most dysfunctional families, as opposed to the rights and obligations stipulated by family law. It is clear that family life would descend into hell were it to be regulated by the law alone; in fact it could be argued that the family as such would cease to exist were its internal dynamics determined entirely by the relevant provisions of family and contract law.

114. Ibid. at 39.

115. Cf. Ehrlich, ibid.:

116. The use of the colonialist metaphor is intended as an explicit reference to Cotterrell’s pejorative dismissal of certain strands of sociology of law. Supra note 104. The same terminology is used because the process being described is essentially the same, namely that sociological concepts be grafted onto legal theory in preference to an analysis exclusively framed in terms of blackletter law. The precise meaning of the terms does however differ between the two contexts in that Cotterrell argues that this sociological imperialism results from a conscious choice of the sociologist (disguised as a legal theorist), which is in fact both unnecessary and unjustified, whereas this paper has argued that the legal theorist must incorporate a sociological frame of reference in order to be able to make sense of the phenomenon of ‘rule’ in general, and of ‘law’ in particular. In both cases the effect of the adoption of the sociological frame is to undermine law’s status as an autonomous discipline. However, whereas Cotterrell’s imperialism is a hostile venture forced upon legal theory from without, the imperialism advocated in this paper is a cooperative venture instigated from within by legal theory itself which is unable adequately to account for the notion of rule solely on its own terms, and hence invites colonisation in order to save itself from its own shortcomings. Although this latter usage of the concept of colonialism may appear somewhat flippant, it should be noted that there exists a similar debate at the political level on the merits or demerits of colonisation, i.e., whether the “white man’s burden” was just a façade for shameless greed, or whether the purpose of European colonisation was to civilise areas which were regarded as more backward.