System and Repetition in Legal Discourse: A Critical Account of Discourse Analysis of the Law

Transcription

1 System and Repetition in Legal Discourse: A Critical Account of Discourse Analysis of the Law Nick Hartland For those who observe courts, the most obvious fact about their operation is whatever gets done, gets done in and through language. Indeed, most lawyers who practice in courts are bush linguists (mirroring linguists concern with laws). While the majority of analyses of legal language are un-technical or concerned with specific legal competencies, there is now a substantial literature applying discourse analysis to courts. This paper reviews this literature and its directions. There is something beguilingly obvious about connecting discourse analysis and the law, both for those who study courts and those who specialise in discourse analysis. The ability of discourse analysis to make immediate sense of what happens in courts hides some important theoretical problems. In this paper I argue the connection of the repetitions which can be seen in discourse, to systems of discourse with defined social functions, is one of the prime reasons for the attraction of discourse analysis. Put simply, this connection allows discourse analysts to claim their readings of discourse identify wider social effects and relationships than those which can be derived from discourse alone. This connection is rarely made out in analytic terms. In relation to analyses of legal discourse, this means readings of specific legal texts are not rigorously connected to analyses of the role of law in society. The paper has three further sections. The next section provides a brief critique of the central themes of discourse analysis; introducing the notions of repetition and system in discourse in more detail. The third and 89

2 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 fourth sections read two analyses of discourse in courts on the basis of this analysis, before conclusions are proffered in the final section. Discourse Analysis Pursuing the literature claiming the title discourse analysis, or has the title placed on it, reveals a heterogeneous group of studies from linguistics, sociolinguistics, semiotics, symbolic interactionism, phenomenology, marxist accounts of ideology, speech act theory, and ethnomethodology.1 They share very little. There is a number of explicit attempts to define a discipline of "discourse analysis".2 These works aim to produce an autonomous domain of investigation, with its own distinct concepts, methodologies and objects of analysis. Each of them only applies an existing technique of analysis to discourse, with the possible exception of Foucault.3 Stubbs work derives from linguistics; Coulthard s is based in speech act theory and sociolinguistics; and Pecheux uses Althusser s4 interpretation of ideology, as does Macdonell. At the moment there is no one discourse analysis, merely a number of competing theories that valorise one approach at the expense of others. A corollary of this observation is it is possible to present discourse analysis in a number of ways. What I focus on in this article is the status of systems of discourse in these analyses. Discourse analysis often attempts to establish the existence of organisations of discourse that occur somewhere above the level of a single word, phrase, sentence or conversation, and somewhere 90 T van Dijk, "Introduction: Discourse Analysis as a New Cross-Discipline", in T van Dijk (ed), Handbook of Discourse Analysis: Volume 1, Disciplines of Discourse, Academic Press, London, 1985, p 1; and W Corsaro, "Sociological Approaches to Discourse Analysis", in T van Dijk (ed). Handbook of Discourse Analysis: Volume 1, Disciplines of Discourse, Academic Press, London, 1985, p 167. M Coulthard, An Introduction to Discourse Analysis, Longman, London, 1985; M Foucault, The Archaeology of Knowledge, Tavistock, London, 1972; "Politics and the Study of Discourse" (1972b) 3 / & C 7; "The Order of Discourse", in R Young (ed), Untying the Text: a Post-Structuralist Reader, Routledge and Kegan Paul, London, 1981, p 48; D Macdonell, Theories of Discourse: an Introduction, Basil Blackwell, Oxford, 1986; M Pecheux, Language, Semantics and Ideology: Stating the Obvious, Macmillan, London, 1982; M Stubbs, Discourse Analysis: the Sociolinguistic Analysis of Natural Language, Basil Blackwell, Oxford, Foucault s "The Archaeology of Knowledge" is probably the most rigorous attempt to theorise discourse analysis as an autonomous discipline. "The Archaeology of Knowledge" is not a theoretical rendering of Foucault s empirical studies, but an immensely compelling analysis of a number of abstract theoretical and methodological problems. L Althusser, "Ideology and Ideological State Apparatuses", in Lenin and Philosophy, Basic Books, New York, 1971, p 127.

3 SYSTEM AND REPETITION IN LEGAL DISCOURSE below the level of a language. Such systems are usually, though not necessarily, located by reference to social institutions; so it is possible to speak of the discursive system of medicine, the discursive system of accounting, and the discursive system of the law. As importantly, these studies attempt to use systems of discourse to explain the repetitions of discourse. Notions of "discursive systems", or "discursive formations" in Foucault s terms5, are intimately connected to the perception of repetition within discourse. Obviously there are many types of repetition in discourse: the same words can mean different things in different circumstances; different words can make up the same statement6 or produce similar effects; and there are standardised uses of words in specific situations. These repetitions - of words whose meaning differs, of statements whose words differ and of stock statements - do not exhaust the repetitions of discourse. They do already imply that the meaning of a statement is not given by the content of words or by the structure of language, and some organisation operates at a different level to that of language. This is precisely the place discursive formations occupy, they are meant to explain the socially available repetitions of discourse. Discursive formations try to make sense of the paradoxical qualities of statements, whereby the same word can make different statements and the same statements can be made by different words. For example, the idea there is a system of legal language seems to account for the repetitions of talk in courts, and the ways in which talk in courts differs from everyday language. In this way discourse analysis is intuitively applicable to legal language. This is not without problems. In working out how systems of discourse relate to repetitions in discourse, a tension arises for discourse analyses. The concept "discursive formation" must do work of two types, and these do not necessarily mix well. To tease out this tension I will now briefly examine two definitions of "discursive formations". Pecheux, in an argument that pits Althusser s theory of Foucault, above, n 2, p 38. I will use the terms discursive formation and discursive systems interchangeably in this paper. There is an extensive literature ext the boundaries of statements; see Foucault, above, n 2; E Goffman, Forms of Talk, Basil Blackwell, Oxford, 1981; E Schegloff, "Goffman and the Analysis of Conversation", in P Drew and A Wootton (eds), Erving Goffman: Exploring the Interaction Order, Polity Press, Cambridge, 1988, p 89. Goffman and Schegloff do not, in the end, find the concept "statement" to be a useful one. It should be noted that their criticisms are not directed at Foucault, but at each other. 91

4 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 ideology against traditional semantics, defines discursive formations in this way:7... the meaning of a word, expression, proposition, etc., does not exist "in itself' (i.e., in its transparent relation to the literal character of the signifier), but is determined by the ideological positions brought into play in the socio-historical processes in which words, expressions and propositions are produced (i.e. reproduced). This thesis could be summed up in the statement: words, expressions, propositions, etc., change their meaning according to the positions held by those who use them, which signifies that they find their meaning by reference to those positions, i.e. by reference to the ideological formations... in which those propositions are inscribed. Henceforth I shall call a discursive formation that which in a given ideological formation, i.e., from a given conjuncture determined by the state of the class struggle, determines "what can and should be said (articulated in the form of speech, a sermon, a pamphlet, a report, a programme, etc.)"... This formulation differs from Foucault, who argues:8 Whenever one can describe, between a number of statements,... a system of dispersion, whenever between objects, types of statement, concepts or thematic choices, one can define a regularity (an order, correlations, positions and functionings, transformations), we shall say, for the sake of convenience, that we are dealing with a discursive formation. At one level, each definition simply has a slightly different emphasis.9 Clearly Foucault gives far less importance to notions of class struggle and ideology, and more importance to features to be found within discourse, though discursive practices are not absent from Pecheux. Clearly also 92 Pecheux, above, n 2, p 111 (original emphasis). Foucault, above, n 2, p 38 (original emphasis). On this choice, see M Cousins and A Hussain, Michel Foucault, Macmillan, London, 1984; "The Question of Ideology: Althusser, Pecheux and Foucault", in J Law (ed), Power Action and Belief: a New Sociology of Knowledge? Sociological Review Monograph 32, Routledge and Kegan Paul, London, 1986 p 158; and Macdonell, above, n 2.

5 SYSTEM AND REPETITION IN LEGAL DISCOURSE Pecheux s definition gives a greater role to social structures and processes, though this is not to say Foucault ignores these questions. At another level, the differences between these two definitions are symptomatic of deeper problems in discourse analysis. Discursive formations are defined by two, qualitatively different, sets of social determinations. On the one hand, discursive formations are defined by social institutions and social structures. The existence of a discursive formation is thus established by examining its external aspect; the way it fits into other social institutions and processes. On the other hand, discursive formations are defined by the regularities of discourse itself. The existence of a discursive formation is thus established by examining its internal aspect; the repetitive features of statements within its domain. In abstract analyses, like that of Foucault and Pecheux, it is possible to incorporate both of these aspects. Foucault and Pecheux take account of internal and external features of discursive formations, and of course attempt to some extent to dissolve the distinction. In empirical studies, one of these aspects of discursive formations comes to dominate the other. In effect discursive formations are defined by external processes. In turn, this leads to a gap between the analysis of discursive formations and the socially available repetitions of discourse. The disjuncture between its account of repetition and its account of system, destroys the original attractiveness of discourse analysis. Before proceeding, it must be noted the pair "intemal/extemal" are analytic constructions. They attempt to capture the way conceptions of "discursive fonnations" try to balance two sets of determinations. The distinction between the internal and external aspects of a discursive formation does not correspond to the more common dichotomies in the analysis of discourse. It does not coincide with a distinction between theory and data, or ideal types and the empirical, because both aspects of discursive formations are made available by theory. Nor does the distinction parallel the distinction between langue and parole. The distinction between langue and parole is part of one system of determinations, the structure of a language compared to empirical instances of language use, while the internal and external aspects of a discursive formation concern two different systems of determinations. Though the distinction between the syntagmatic and paradigmatic aspects of language concerns more than one plane of determinations, it does not map onto the intemal/extemal pair. Neither does my distinction coincide with sociology s distinction between micro and macro phenomena. Internal and external phenomena can both involve micro and macro processes and phenomena, and of course the distinction between micro and macro is an 93

6 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 unstable one anyway. Finally the different levels here do not necessarily correspond to the distinction between the discursive and the non-discursive. The external aspect of discursive formations may concern its relationship to other discursive formations. The task of this paper is to read a selection of discourse analyses of the law to see how they balance the internal and external features of discursive systems. Following the general development of discourse analysis, the study of discourse in legal settings has blossomed over the last ten years. Indeed, in discourse analysis courtrooms are a favourite ground for illustrating the connection of language to social factors like power and inequality. Pecheux uses legal utterances to fill out more abstract points on a number of occasions.10 Two approaches to the discourse analysis of the law will be distinguished and examined: sociolinguisticsu, and the analysis of rhetoric in the law12. These two approaches are analysed not because they constitute the sum total of discourse analysis of the law, or even because they are particularly representative of discourse analysis. Rather they are examined because the balance between the internal and external aspects of discursive formations is an especially pertinent problem for them. Their resolution of this problem leads to a gap between their analyses of discursive formations, and their analyses of repetition in discourse. Other sociological techniques like ethnomethodology,13 phenomenology,14 94 Pecheux, above, n 2, p 110. B Danet, "Language in the Legal Process" (1980) 4 Law and Society Review 445; "Legal Discourse", in T van Dijk (ed). Handbook of Discourse Analysis: Volume 1, Disciplines of Discourse, Academic Press, London, 1985, p 273; B Danet, and B Bogoch, "Fixed Fight or Free-for-all? An Empirical Study of Combativeness in the Adversary System of Justice" (1980) 7 British Journal of Law and Society 36; S Harris, "Questions as a Mode of Control in Magistrates Courts" (1984) 49 International Journal of Social Language 5; R Nofsinger, "Tactical Coherence in Courtroom Conversation", in R Craig and K Tracy (eds), Conversational Coherence: Form, Structure and Strategy, Sage, Beverly Hills, 1983, p 243; W O Barr, Linguistic Evidence: Language, Power and Strategy in the Courtroom, Academic Press, New York, 1982; S Philips, "Strategies of Clarification in Judges Use of Language: From the Written to the Spoken" (1985) 8 Discourse Processes 421; R Wodak, "Discourse Analysis and Courtroom Interaction" (1980) 3 Discourse Processes 369. P Goodrich, ' The Role of Linguistics in Legal Analysis" (1984) 47 The Modern Law Review 523; Reading the Law: a Critical Introduction to Legal Method and Techniques, Basil Blackwell, Oxford, 1986; Legal Discourse: Studies in Linguistics, Rhetoric and Legal Analysis, Macmillan, London, J Atkinson, "Sequencing and Shared Attentiveness to Court Proceedings", in G Psathas (ed), Everyday Language: Studies in Ethnomethodology, Irvington, New York, 1979, p 257; "Ethnomethodological Approaches to Socio-legal Studies", in A Podgorecki and J Whelan (eds), Sociological Approaches to Law, Croom Helm, London, 1981, p 201; "Understanding

7 SYSTEM AND REPETITION IN LEGAL DISCOURSE symbolic interactionism,15 the analysis of the law s narrative grammar,16 and the idiosyncratic but highly interesting work of McBamet,17 will be left to one side in this presentation. They do not look for distinct formations in the discourse of the law. For instance in Jackson s case:18... legal discourse is only one particular semantic investment of discourse in general, and [...] the same narrative grammar applies to it. Formality: the Categorisation and Production of "Formal" Interaction" (1982) 33 The Britisk Journal of Sociology 86; J Atkinson and P Drew, Order in Court: the Organisation of Verbal Interaction in Judicial Settings, Macmillan, London, 1979; W Beach, "Temporal Density in Courtroom Interaction: Constraints on the Recovery of Past Events in Legal Discourse" (1985) 53 Communication Monographs 1; W Bennett and M Feldman, Reconstructing Reality in the Courtroom: Justice and Judgement in American Culture, Rutgers University Press, New Brunswick, 1981; E Bittner, "The Police on Skid-Row: a Study of Peace Keeping" (1967) 32 American Sociological Review 699; A Cicourel, The Social Organization of Juvenile Justice, John Wiley and Sons, New York, 1968; P Drew, "Accusations: the Occasioned Use of Members Knowledge of Religious Geography in Describing Events" (1978) 12 Sociology 1; R Dunstan, "Contexts for Coercion: Analysing Properties of Courtroom Questions" (1980) 7 British Journal of Law and Society 61; H Garfinkel, Studies in Ethnomethodology, Prentice- Hall, Englewood Cliffs, 1967; M Lynch, "Closure and Disclosure in Pre-Trial Argument" (1982) 5 Human Studies 285; D Maynard, Inside Plea Bargaining: the Language of Negotiation, Plenum, New York, 1984; M Pollner, "Mundane Reasoning" (1974) 4 Philosophy of the Social Sciences 35; "Explicative Transactions: Making and Managing Meaning in Traffic Court", in G Psathas (ed), Everyday Language: Studies in Ethnomethodology, Irvington, New York, 1979, p 227; A Pomerantz and J Atkinson, "Ethnomethodology, Conversation Analysis, and the Study of Courtroom Interaction", in D Muller, D Blackman and A Chapman (eds), Psychology and Law, John Wiley and Sons, New York, 1984, p 283; D Sudnow, "Normal Crimes: Sociological Features of the Penal Code in a Public Defender Office" (1965) 12 Social Problems 255. M Los, "Law from a Phenomenological Perspective", in A Podgorecki and C Whelan (eds), Sociological Approaches to Law, Croom Helm, London, 1981, p 187; and H Taylor-Backner, "Transformations of Reality in the Legal Process", in T Luckmann (ed), Phenomenology and Sociology, Penguin, Harmondsworth, 1978, p 311. A Brittan, "The Symbolic Dimension of Law and Social Control", in A Podgorecki and C Whelan (eds). Sociological Approaches to Law, Croom Helm, London, 1981, p 167; and P Carlen, Magistrates Justice, Martin Robertson, London, B Jackson, Semiotics and Legal Theory, Routledge and Kegan Paul, London, D McBamet, Conviction: Law, the State and the Construction of Justice, Macmillan, London, Jackson, above, n 16, p

8 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 Such an approach excludes consideration of systems at any level other than that of society as a whole. For a review of the analysis of courtroom interaction see Hutton.19 Sociolinguistics Sociolinguistics is not a homogeneous discipline. There is very little agreement over appropriate methods, variables and concepts in the literature, and this is reflected in the work on courts. However at a very abstract level, analyses which declare themselves to be sociolinguistic generally share two related themes. First, a starting point for the sociolinguistic analysis of courtroom talk is the observation language in courts is a specific type of language, language in courts is part of a discursive formation. O Barr in fact defines sociolinguistics as the study of "... patterns of variation in language...".20 Second, sociolinguistics is concerned with the relationship between language use and non-linguistic social factors.21 Sociolinguistics emerged in the 1960 s and 1970 s and is the study of the interrelations between patterns of language use, social characteristics of speakers, and features of social systems. Virtually all sociolinguists attempt to conceptualise this interrelationship as a two way exchange:22 The general sociolinguistic thesis attempts to explain how symbolic systems are both realizations and regulations of the structure of social relationships. These two themes of sociolinguistics broadly correspond to the internal and external aspects of discursive formations distinguished above. In the case of sociolinguistics the distinction between internal and external aspects of discursive formations does map onto a distinction between the discursive and the non-discursive. Here internal aspects of discursive formations solely 96 N Hutton, "The Sociological Analysis of Courtroom Interaction: a Review Essay" (1987) 20 Australian and New Zealand Journal of Criminology 110. O Barr, above, n 11, p 8. Danet, above, n 11, p 453. B Bernstein, "Social Class, Language and Sociolinguistics", in P Giglioli (ed), Language and Social Context: Selected Readings, Penguin, Harmondsworth, 1972, p 158.

9 SYSTEM AND REPETITION IN LEGAL DISCOURSE conc ern linguistic phenomena, while the external determinations of discursive formations only concern non-discursive structures and processes. This section focuses on the approach of Danet. This is not a full exposition of the work of this author, as I am primarily interested in how Danet constructs discursive formations and the regularities of legal discourse. My argument is in Danet s work there is a gap between the analysis of discursive formations and of repetition. Danet uses the tools of linguistics to describe what she calls "Legal English". Danet s analysis identifies a number of regularly recurring features, or variables, of Legal English, features also seeming to characterise legal discourse in other languages, though no thorough study exists here.23 These distinct features occur on four levels; the lexical, roughly the level of vocabulary,24 syntactic, the mles of grammar,25 prosodic, the "word music", or the way the language sounds,26 and what Danet calls the discourse level, for Danet this refers to the organisation of the unit of interaction as a whole, say the entire legal document under consideration.27 Danet attempts to use all of these features to establish Legal English is a specific variety of English and I will return to the relationship between the linguistic features and Legal English below. As Danet recognises28 the distinctive nature of Legal English can be construed in a number of ways. Danet prefers to think of it as a register of English, rather than a dialect of English. Dialects characterise groups of speakers, all people from X speak dialect A, while those from Y speak dialect B, whereas registers "... are a function of situation or use".29 With a register the user of a language has available two, or more varieties, and employs them according to their appropriateness in a situation.30 Legal English is construed as a specific type of register by Danet; a diglossia31. A diglossia is a register of a language acquired through formal education, which enjoys a high prestige, Danet, above, n 11 at and pp Danet, above, n 11 at and pp 279, 281. Danet, above, n 11 at and pp Danet, above, n 11 at 482. Danet, above, n 11 pp ; Coulthard, above, n 2, p 9. Danet, above, n 11 at and p 275. Danet, above, n 11 at 471. Coulthard, above, n 2, p 39. Danet, above, n 11 at 473 and p

10 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 and has a specific function, among other traits.32 The concept "diglossia" basically fills the same analytic space as the concept "discursive formation", because it is a distinct system of discourse which occurs between the level of individual instances of language use and the language as a whole, and which governs repetition in language use. It is important to note Danet s empirical analysis concerns the variables of legal documents. Danet recognises a number of distinctions: legal discourse covers both spoken and written legal language; legal discourse occurs in a variety of situations, legal documents, lawyers briefs, examination, testimony, arguments on law, lawyer-client discussion, and informal lawyerlawyer discussion are all examples of legal discourse; and these "modes of language use" vary in their formality.33 Danet examines the most formal types of legal writing, legal documents, because the features of the legal register are most prominent in them.34 The purest distillation of the legal register is no doubt found in written documents... This statement is important because it indicates the status of the legal register in relation to the regularities of legal talking and writing. The regularities of legal discourse are conceived of in terms of variables. These variables, in the dimensions outlined above (lexical, syntactic, prosodic and so on), are the analytically available repetitions of legal discourse. In Danet s analysis registers do not exist in and through the variables of language use. Rather registers exist outside of these variables; registers are idealised systems of variables which exist beyond any empirical instance of language use. However, registers cannot be thought of as ideal types, because the distinction between register and variables is not a distinction between theory and data, but a distinction between two levels of theory and two corresponding levels of social organisation. To the degree a piece of legal talk or a legal document contains these variables, it is correspondingly closer to or further from the register. But no matter how many variables a piece of talk or a document contains, it is never the same thing as a register. An instance of legal language can only ever be a more or less pure distillation of the diglossia of Legal English. Therefore the relationship of instances of 34 Danet, above, n 11 at 473. Danet, above, n 11 at and pp Danet, above, n 11 p 277; quotation from Danet, above, n 11 at 472; my emphasis. 98

11 SYSTEM AND REPETITION IN LEGAL DISCOURSE legal discourse and their identifiable variables to the diglossia of Legal English is always open to question. If it is possible to say diglossia roughly correspond to discursive formations, or diglossia are particular ways of construing discursive formations, then clearly what is at issue is the way the internal regularities of discourse (the variables) define the existence of discursive formations. It is interesting this is the weakest link in Danet s analysis. Despite claims it is possible to describe Legal English by reference to the variables, Danet fudges the relationship between internal repetitions and discursive formations. It is unclear how many variables a statement actually has to contain to be in a register. Are all statements by legal personnel in the legal register? Nor is it clear if the existence of a variable is sufficient to put a statement in the legal register. For instance, would a statement that contained semantically similar terms to legal documents, but spoken by a judge at a dog show, be an example of Legal English? Danet does not adequately theorise the relationship of repetition in discourse to discursive formations. In fact the notion of a diglossia of Legal English is not really used to understand the repetitions of specific documents or discourses. Rather the concept "Legal English" enables Danet to make an argument about the legal system s place in society. The notion of the diglossia of Legal English is used to make a general conclusion about the role of law in society. In examining the effect of legal discourse on social relationships, Danet is not concerned with the effect particular documents might have on what happens outside of courts, but with the diglossia of Legal English, and how it affects access to the law. Danet claims the existence of the diglossia of Legal English reproduces inequality in access to the law. Thus the analysis can suggest Legal English as a whole should be made more accessible, though Danet eventually prevaricates on the question of reform. The existence of Legal English as a distinct register, or what Danet elsewhere refers to as the "thickening" of English in legal discourse, may serve "socially important symbolic functions", such as the need for the certainty this language conveys.35 Thus Danet s analysis of the existence of the diglossia of Legal English only makes reference to its external functions. It is the position of legal discourse in society which defines the discursive formation of the law. This elides the relationship of the repetitions of legal statements to the diglossia of Legal Danet, above, n 11 at , 540 and and p

12 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 English. It means there is a gap between the analysis of the internal regularities of Legal English (the analysis of the variables) and the analysis of the diglossia of Legal English. Rhetoric The term "rhetoric" is even more shaky than categories like sociolinguistics. Rhetoric can be crudely defined - at least in its classical sense - as the study of persuasion, or of the effects of language on an audience.36 Rhetoric emphasises the relationship of language to the exercise of power, or the political effects of speech, rather than the relationship of speech to truth or facts. Goodrich s aim is to produce a sociologically and linguistically sophisticated rhetoric, and through this, to read the discourse of law as a historical and political discourse.37 It is moreover the case that a critical linguistic orientation also suggests that the separation or isolation of questions of law and legality from wider considerations of the organisation of discursive processes as a modality of power and of intergroup relations is hollowly ideological. Even at this stage it is possible to see the analysis will embody two ends: Goodrich wants to produce an understanding of power and class relations, and a reading of the law and of legal discourse. Needless to say, these two ends correspond to the distinction between external and internal aspects of discursive formations. The question is; how do these two aims come together? Goodrich uses the term rhetoric to flag a number of theoretical issues. The approach attempts to dissolve any distinction between a legal text or statement, and a supposedly deeper level of organisation - an immutable structure. Such a distinction is fundamental to an approach like Jackson s analysis of the narrative grammar of the law.38 Goodrich argues referring solely to a deep structure excludes adequate consideration of the history of legal discourse. A good deal of "Legal Discourse" is devoted to distinguishing this rhetorical analysis from related approaches:39 Goodrich, above, n 12, pp Goodrich, above, n 12, p 20; quotation from Goodrich, above, n 12, pp Jackson, above, n 16, pp Goodrich, above, n 12, p

13 SYSTEM AND REPETITION IN LEGAL DISCOURSE... it has been argued that both in the case of linguistics and in that of comparable developments in the philosophy of law, the constitution of a scientific project took place by means of excluding history. In linguistics this initially meant an emphasis on the form of language and the rules which govern it to the exclusion of semantics and speech... The outcome has been that problems of history and semantics have been, and continue to be, treated within the oppositional framework of the necessary and the contingent, the objective and the subjective, with the analysis of the context and usage of language in terms of their eventual subjective unknowability being really little more than a palliative. Goodrich s aim is to undo this structure of understanding which linguistics and jurisprudence have created. By interpreting linguistic practices in terms of the oppositions above, linguistics and jurisprudence split the structure of language, the necessary and the objective, and practices of legal discourse, the contingency of actual language use and its so-called subjectivity. These analyses become directed towards a "deep structure" at the expense of actual legal discourses. Against both linguistics and the philosophy of law Goodrich disposes of such oppositional frameworks and analyses legal discourse as a "social practice".40 This move also opens up the possibility of analysing law as a discursive formation. By examining law as a social practice it is clear Goodrich is concerned with both the internal and external aspects of discursive formations. In fact Goodrich makes extensive use of Pecheux s definition of discursive formations to emphasise both components have to be taken into account. The analysis examines the implication of legal discourse in power relations and class relations, and the institutional supports of law which operate as external determinations of legal discourse.41 Goodrich also analyses the more narrowly discursive organisation of the law, which concerns how the law positions itself in relation to other discourses; interdiscourse, and how the law is ordered internally; intradiscourse.42 In Goodrich s schema, interdiscourse concerns the external aspects of discursive formations, and Goodrich, above, n 12, p 2. Goodrich, above, n 12, p 77 and p Goodrich, above, n 12, pp

14 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 intradiscourse the internal aspects. Thus in Goodrich s analysis, internal and external do not correspond to discursive and non-discursive. Goodrich is able to reach some interesting conclusions about the discursive organisation of the law:43 Law, as a linguistic register or as a literary genre, can be described linguistically or, more importantly, discursively, in terms of its systematic appropriation and privileging of legally recognised meanings, accents and connotations (modes of inclusion), and its simultaneous rejection of alternative and competing meanings and accents, forms of utterance and discourse generally, as extrinsic, unauthorised or threatening (modes of exclusion). To understand the coherence of this process of linguistic and semantic inclusion and exclusion is to introduce the problem of the relationship of law to power, and to some extent to explain the characteristic modes of legal utterance as social discourse - as a hierarchical (stratified), authoritarian (distanced), monologic (uniaccented) and alien (reified) use of language. These modes of inclusion and exclusion, and the hierarchical, authoritarian, monologic and alien characteristics of legal utterances concern the internal regularities of legal discourse, and some aspects of the external functioning of the discursive formation of the law (the way it positions itself in relation to other discursive formations). These discursive aspects of legal discourse are not enough to define the existence of the discursive formation of the law. In order to account for the reproduction of these features of legal language, Goodrich in effect gives a primacy to the ideological and class determinations of discursive formations, just like Pecheux.44 Indeed this is marked in the long quotation immediately above by the introduction of the notion of power to explain the coherence of the process of linguistic and semantic inclusion and exclusion. The ideological and class determinations which define the discursive formation of the law solely concern the external features of the discursive formation of the law. Since Goodrich draws on Althussarian analyses of the structure of society, through Pecheux, the fact ideological and class 44 Goodrich, above, n 12, p 3. Goodrich, above n 12, pp 77, 144,

15 SYSTEM AND REPETITION IN LEGAL DISCOURSE determinations define the existence of the discursive formation of the law means the primary role of this discursive formation is to reproduce society. External factors thus come to define the existence of the legal discursive formation in Goodrich s analysis. This opens up a gap between Goodrich s construction of the discursive formation of the law, and the empirical details of actually occurring legal talk and legal texts. In other words, the analysis of the discursive formation of the law, and the analysis of repetitions of legal discourse do not connect. Goodrich notes the most pressing problems facing research into legal discourse are those of:45... attempting to formulate and substantiate the complex relationship of the structural features, or regularities, of systems of communication as discursive formations, to their agency or manifestation in the specific, empirical practice. In terms of the state of the art, I shall indicate that the most pressing need facing the theories of discourse is precisely that of locating the specific linguistic and semantic tools which will enable the generation of as yet absent empirical and particular analyses of text and reception, of utterance and audience. Like Danet s distillation of Legal English in legal documents, this quotation is interesting because it indicates the status of specific statements, empirical practices, compared to discursive formations. In this passage Goodrich argues specific statements are the empirical manifestation of discursive formations. The statements of actors are thus animated by the agency of a discursive formation which exists beyond them. Even if Goodrich were to produce the presently absent "linguistic and semantic tools" to generate a rigorous empirical analysis of discourse, this would not necessarily close the gap between the repetitions of legal discourse and the discursive formation of the law. So long as specific statements are conceived of as the manifestation of formations of discourse existingyond these statements, there is a necessary gap between the discursive formation and the statements. The origin of this gap is not hard to find. While reproduction is the dominant function of the discursive formation of the law, the external features of the discursive formation will in effect define its existence. Because the concept "discursive formation" is defined by external processes, there will always be Goodrich, above, n 12, pp

16 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993) 9 a gap between the analysis of the existence of this discursive formation and the analysis of specific actually occurring legal practices. Conclusion The analyses of Danet and Goodrich are not the same. They conceive of the analytically available features of society and discourse in very different terms. In moving between Danet and Goodrich I move from an analysis of linguistic variables to an analysis of rhetorical figures, and from an analysis of inequality as the uneven distribution of social goods, to an analysis of inequality as asymmetrical power relationships. All of these terms are incommensurable. Danet and Goodrich do share a gap between their analyses of the repetitions in discourse and discursive formations. In both cases, this gap is brought about by the dominance of the external aspects of discursive formations over the internal aspects of discursive formations. Effectively, the analyses establish the existence of the discursive formations of the law only by examining the external determinations of these formations. Given the analyses aim to make sense of the internal repetitions of legal discourse, this dominance is problematic. The gap between repetition in discourse and discursive formation annuls the immediate attraction of the concept "discursive formation". I argued above the ability of discursive formations to explain repetition in discourse was an obvious attraction of the concept. The origin of this problem is also similar in the two analyses examined here. Danet and Goodrich s concern with reproduction leads them to look almost entirely to external determinations when they are establishing the existence of discursive formations. Obviously again, what is being reproduced and how it is being reproduced are very different in the two analyses. It should also be noted the status of claims that society is reproduced, have been questioned from a number of perspectives.46 Nevertheless the attraction of reproduction is fairly clear. If it can be claimed discursive formations help reproduce significant aspects of society, then analyses of these formations are also socially significant. I do not deny legal talk contributes to the reproduction of social relationships outside of the court, though clearly the exact construction of the relationship of discursive formations to external social processes makes a great deal of difference. Attention needs to be given first to the status of discursive 46 B Hindess and P Hirst, Mode of Production and Social Formation, Macmillan, London, 1977; P Hirst, On Law and Ideology, Macmillan, London,

17 SYSTEM AND REPETITION IN LEGAL DISCOURSE formations and the analytically available repetitions of discourse. The prima facie attraction of discursive formations derives from the fact repetition and system are intuitively related. A rigorous account of the relationship between statements and discursive formations is needed. Moreover, this account must be able to explain the repetitions of statements by reference to the nature of the discursive formation. This trick, if it can be accomplished, will provide a basis for an analysis of the role of discursive formation in society. 105

18 106 AUSTRALIAN JOURNAL OF LAW AND SOCIETY (1993)9