Iberica 13 Ibérica 21 (2011): 93-116 ISSN 1139-7241 Abstract The last 20 years have seen increasing interest in the way in which meaning is made in different professional and academic disciplines. Central to this issue is the notion of disciplinary values, that is, qualities which define what is prized or stigmatised by different professional communities. In the present paper, the notion of disciplinary values is used to examine the way legal writers communicate meaning in different genres. To this end, six adjective/adverb sets which have a prominent place in legal discourse (“clear/ly”, “important/ly”, “reasonable/ly”, “appropriate/ly”, “correct/ly” and “proper/ly”) are identified. Their collocates and semantic preferences are studied in four 500,000-word corpora consisting of texts from the area of commercial law: academic journal articles, case law, legislation, and legal documents. Although the frequency and use of “clear/ly” and “important/ly” appear not to differ greatly from those found in other corpora of written and academic written texts such as the British National Corpus (BNC) and the British Academic Written English corpus (BAWE), “reasonable/ly”, “appropriate/ly”, “correct/ly” and “proper/ly” were found to be salient in some or all of the subcorpora. The reasons for this are then analysed within the framework of disciplinary values. These words appear to convey attributes that have particular importance in the legal profession, reflecting disciplinary values that cross the boundaries between various written genres. Keywords: disciplinary values, genre, legal discourse, discourse analysis, corpus linguistics. Resumen Valores disciplinares característicos del discurso legal: un estudio de corpus En los últimos 20 años ha surgido un interés creciente en la generación del significado en las distintas disciplinas académicas y profesionales. Un concepto Disciplinary values in legal discourse: a corpus study Ruth Breeze Universidad de Navarra (Spain) rbreeze@unav.es 93 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 93 Ibérica 21 (2011): 93-116 RUTH BREEzE clave es la noción de que existen “valores disciplinares”, es decir, cualidades que definen lo que se valora o se censura en las distintas comunidades profesionales. El propósito de este artículo es profundizar en el concepto de los valores disciplinares en el discurso de los profesionales del derecho mediante el estudio de un corpus de distintos géneros textuales del área jurídica. Se identifican seis adjetivos/adverbios que destacan en el lenguaje del derecho (clear/ly, important/ly, reasonable/ly, appropriate/ly, correct/ly y proper/ly). Las palabras contiguas y las preferencias semánticas se estudian en cuatro corpus especializados de 500.000 palabras cada uno, del área del derecho comercial y corporativo: artículos de revistas, sentencias, leyes y documentos legales. Aunque la frecuencia y el uso de clear/ly e important/ly no se distinguen de manera significativa de los que se encuentran en otros corpus de textos escritos y científicos, como el British National Corpus (BNC) y el British Academic Written English corpus (BAWE), reasonable/ly, appropriate/ly, correct/ly y proper/ly son especialmente frecuentes en los corpus de lenguaje jurídico. Tras el análisis efectuado en el corpus sobre su potencial significado en el mundo jurídico, se concluye que dichas palabras comunican cualidades que se consideran especialmente importantes en dicho ámbito y reflejan valores disciplinares que cruzan las fronteras entre los distintos géneros textuales. Palabras clave: valores disciplinares, género textual, discurso legal, análisis del discurso, lingüística de corpus. Introduction Over the past 20 years, growing interest has centred on the way that knowledge and ideas are presented in a wide range of academic and professional genres (Knorr Cetina, 1999; Hyland, 1999; Biber, 2006). Phenomena such as evaluative or relational language, modality and hedging have received considerable research attention, and major differences have come to light concerning discursive practices in disciplinary communities. Analysis of the lexical and syntactic choices made in different professional and academic texts may shed light on the specific assumptions and practices that affect the way people handle and represent experience within particular communities. The degree of personality or impersonality in the text, the use of hedging, emphatics or attitude markers, and various aspects of lexical choice, all project the epistemological premises of the discipline and the value system that operates within it. Central to this issue is the notion of disciplinary values, that is, qualities which define what is prized or stigmatised by different professional communities. For example, in a cross- 94 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 94 disciplinary study of academic publications, giannoni (2009) found differences in the metaphorical language used to express approval and disapproval in applied linguistics, law, medicine and economics. In the present paper, the notion of disciplinary values, developed by Hyland (1999) and further explored by giannoni (2009), is used to examine the way legal writers communicate meaning in different genres. To this end, six salient adjective/adverb sets are examined, as they appear in four 500,000-word corpora of text from the area of commercial law: academic journal articles, case law, legislation, and legal documents. Corpus linguistics and disciplinary values When writers produce texts, they are engaging in a social process, and the choices they make are shaped by their knowledge of genre conventions, their previous experience of professional discourses, and the professional and personal ends that they hope to achieve. As Hyland states (1999: 100), “Textual meanings (…) are socially mediated, influenced by the communities to which writers and readers belong”. The texts themselves thus serve as evidence of the way a particular professional community generates and manages knowledge. The language choices made by individual writers of professional texts are rarely arbitrary, and they are not generally creative or original, as they might be in the case of a novelist or poet. The word often used to describe lexical choices in professional and academic contexts is “appropriate”, and whether or not a particular word or phrase is appropriate depends on the tacit consensus within the disciplinary community, as well as on broader considerations such as register, the parameters of which are set by broader social consensus. In the last instance, a word is generally considered appropriate because it is the one that is most commonly used in a particular context, and therefore appropriacy is closely related to frequency. To access the discourses of professional groups, corpus methodology provides a useful tool which generally proves more reliable and objective than introspection or observation. Although some authors have postulated a “cultural divide” between corpus linguistics and discourse analysis (Leech, 2000), on the grounds that discourse analysis relies on the integrity of text whereas corpus analysis tends to work on decontextualised fragments, there is now a growing consensus that the two areas are mutually complementary DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 95 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 95 and may even overlap (Stubbs, 2007). The evolving understanding of corpus linguistics is that although patterns in language can be studied through concordancing and other tools, the analysis of meaning should always have a central role (Sinclair, 2004). In linguistic terms, investigation of samples of texts affords direct access to parole, or what the professionals in question actually write or say on particular occasions, in the form of horizontal concordance lines. However, when concordance lines are viewed vertically, more enduring patterns come to light that can provide considerable information about the discourses that operate among particular groups. Although the term “discourse” is not always used in corpus research, this article will use the framework set out by Stubbs (2007) based on Tuldava (1998), in which text is a understood as a static product of discourse, which is a dynamic process. Discourse is meaningful social action which is not reducible to text, but may be accessed through it. Corpus methodology enables us to perceive the “pervasive routine of most language use” (Stubbs, 2007: 146), and allows us to document in detail the patterns and repetitions that are essential to the way language functions in real social contexts (Adolphs & Durow, 2004). Although this approach may be more problematic if the object of study is the language system itself, when the focus of interest is a clearly delimited area such as the discourse of a particular community of practice, it would seem clear that analysis of patterns that recur over large samples of text can provide important insights. In this context, the particular frequency of specific lexical items that convey attitude and epistemic stance in texts produced within different disciplines has led some linguists to propose the notion of disciplinary values, as a way of characterizing the distinctive set of qualities or attributes to which worth is ascribed within a particular community of practice. Such values may not be immediately obvious, not least because professional and academic writers tend to conceal their identity behind a cloak of objectivity and rarely make explicit personal declarations concerning professional values. However, small clues are available, for example, in the form of evaluative or epistemic adjectives conveying value judgements, which can help outsiders to form an impression of the qualities that are considered important within the particular discipline (Tutin, 2008; Dahl, 2009; giannoni, 2009). The presence or absence of value-laden lexis may in itself shed light on the nature of the claims being made by the writer, and thus on the epistemological framework within which the discipline operates (Hyland, 1999). RUTH BREEzE Ibérica 21 (2011): 93-11696 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 96 In the case of academic writing, one area that has received scholarly attention is the presence of certain key adjectives, which may reflect the writer’s underlying values and influence the communication of facts, ideas and interpretations within a community of practice. Swales and Burke (2003) investigated the use of evaluative adjectives in spoken academic language, and found there to be more evidence of intensification and polarization in academic speech than in written genres. Pérez-Llantada (2008) analysed stance adjectives in research articles from applied linguistics and information technology, and established that both disciplines tended to rely heavily on a small set of evaluative descriptors, such as “important”, “good”, “difficult” and “clear”, and that both used a slightly higher proportion of relevance adjectives (“important”) than assessment adjectives (“good”). Taking this type of analysis a stage further, giannoni (2009) examines the use of the adjectives “broad”, “clear” and “strong”, all of which have metaphorical connotations, and other members of the corresponding word families, in academic articles from four disciplines, and identifies considerable variation between areas in the extent to which each one is used. He attributes these results to differences in the values held by members of the respective discourse communities. In his view, these evaluative terms, which might conventionally be regarded as “dead metaphors”, that is, metaphors which have become so conventionalised that the links to their original meaning have been severed (Pawelec, 2006), actually influence and are influenced by the values that people hold in particular communities, much in the way that a conceptual metaphor in an idiomatic expression reflects and perpetuates a particular way of understanding the world (Lakoff, 1987). By way of example, in giannoni’s (2009) study, the writers of linguistics articles were found to use more metaphors related to clarity, whereas academic lawyers favoured words that connoted breadth or inclusiveness: he surmises that this result may be coloured by the linguists’ struggle to extract knowledge from text, on the one hand, and the lawyers’ quest for general, widely-applicable norms, on the other. To research these concepts, one promising method is that proposed by Stubbs (2007), based on the analysis of extended units of meaning explained in Sinclair (1998). This involves the use of corpus techniques to obtain detailed information about the way that words are used, focusing particularly on collocation (the relation between the node word and other word-forms that co-occur frequently with it); colligation (the relation between the node word and particular grammatical structures with which it is found); semantic DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 97 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 97 preference (the relation between the node word and lexical sets of semantically-related lemmas) and semantic prosody, sometimes called discourse prosody (the pragmatic or attitudinal function of the unit in the text). Of these, the first and third attribute appear to be of particular relevance to the purpose of this study: the words and semantic sets with which a particular adjective frequently co-occurs would be likely to provide rich information about the sense in which that adjective is used, and its particular resonance within the discipline or the genre. Identification of evaluative adjectives and adverbs Against this background, the present study sets out to obtain deeper knowledge of the values embodied in legal discourse by identifying and studying high-frequency lexical items. Rather than comparing academic law articles with publications from other spheres of academia, I aim to investigate disciplinary values in law by taking a broad overview of four important areas of legal discourse: academic articles, case law (law reports and opinions), legislation and legal documents. The first two fields (academic law and case law) belong firmly within the scope of the same parent discourse community, since there is real overlap between academics and leading legal practitioners, and both sets of professionals read each other’s texts (vass, 2004). Moreover, both lawyers and academics deal with legislation on a daily basis, even though they are less likely to be involved in its drafting. Finally, practising lawyers are likely to be thoroughly familiar with the language of legal documents within the realm of private law. The corpus which forms the basis for this study consists of 2 million words from the area of commercial law in English, divided into four sub-corpora of approximately 500,000 words each from: academic law articles on commercial and corporate law, case law (law reports and court opinions), legislation (Companies Acts) and legal documents (contracts, merger agreements and so on). These four subcorpora can be seen to serve as a sample of four interlinked legal discourses. Corpus methodology is used, first to identify possible value-laden lexical items, and then to bring out the similarities and differences in the way that these are used in the different types of legal text. In order to identify words with possible epistemic and evaluative connotations, the following procedure was followed. First, WordSmith tools RUTH BREEzE Ibérica 21 (2011): 93-11698 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 98 (Scott, 2007) were used to generate wordlists ordered in terms of frequency. These lists were then scanned in order to identify frequent adjectives. The 20 most frequent evaluative adjectives from each subcorpus were identified. Since the focus of the study was on evaluative language that might relate to disciplinary values, items that seemed to refer to frequency, possibility, generality/specificity or necessity were not included in the list. The final lists were then compared, and all the adjectives/adverbs that appeared within the top twenty on two or more lists were assessed for inclusion in the study. The following epistemic or evaluative adjectives/adverbs appeared in two or more lists: “appropriate”, “fully”, “greater”, “proper”, “properly”, “reasonable”, “reasonably”, “sufficient”, “approximately”, “clear”, “close”, “complete”, “correct”, “direct”, “directly”, “fair”, “free”, “full”, “fundamental”, “important”, “ordinary”, “particularly”, “relevant”, “significant”, “simply”, “unfair”. Of these, “reasonable”, “appropriate” and “proper” appeared in all the lists, being close to the top in all except that for academic articles, while “clear”, “important” and “correct” were among the first ten words on at least two of the lists. Once the six words had been selected, frequencies were obtained for all the adjectives and their corresponding adverbs. The adjective forms were more frequent than the adverbs in all cases except “clearly” (legislation subcorpus) and “properly” (academic articles, case law and legal documents subcorpora). Overview of adjective/adverb frequency The frequency of the six adjectives in the four legal subcorpora is shown in table 1, and that of the six corresponding adverbs in table 2. The data are normalized to frequency per 10,000 words. In order to provide a point of comparison, the frequencies of the respective items in the British National Corpus (BCN) and the British Academic Written English corpus (BAWE) are also displayed in both tables. DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 99 AREA reasonable clear important correct proper appropriate Academic articles 2.2 3.1 4.8 1.8 0.9 1.7 Case law 4.6 4 2.1 1.7 1.3 2.8 Legislation 4.3 0 0.1 0.4 1.5 4 Legal documents 9.8 0.7 0.4 1.7 0.9 0.9 BNC 0.6 2.5 3.9 0.7 0.6 1.1 BAWE 0.7 2.6 6.7 1.2 0.3 1.7 Table 1. Frequency of adjectives per 10,000 words. 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 99 For the sake of simplicity, a frequency twice the frequency of the BNC or BAWE (taking the highest of the two) was taken to indicate salience of the item concerned in the subcorpus in question. For example, in the case of “appropriate”, the highest incidence is that of the BAWE (1.7), and so for this item to be salient, the frequency in the subcorpus would have to be greater than 3.4, which only holds for the legislation subcorpus. The salience of “reasonable” and “reasonably” in all four corpora, and its particular prominence in legal documents, are especially noticeable. As far as the other items are concerned, the pattern is more complex. “Clear”, “clearly”, “important” and “importantly” are frequent in academic articles and case law, while “correct” is frequent in all the subcorpora except legislation. “Proper” is salient in case law and legislation, whereas “properly” is frequent in all the subcorpora but salient only in case law and legal documents. Finally, “appropriate” is salient in legislation, but no instances of the corresponding adverb were encountered in this subcorpus. Once the initial frequencies had been established, the collocates and semantic preferences of the chosen words were investigated in depth. In what follows, the full research process is explained in the case of “reasonable/reasonably”, while in the case of the other words, only the most important findings are described. Case study: “reasonable/reasonably” The construct of the “reasonable person” is familiar to anyone who has had contact with legal discourse, since it is a legal fiction, that is, a construct that is widely used as a standard against which the understanding or behaviour of individual people can be measured (green, 1968). What constitutes “reasonableness” has, of course, been a central philosophical question in western thought, particularly in the last two centuries (Nubiola, 2009). In the area of law, the notion of “reasonableness” seems to have been originally RUTH BREEzE Ibérica 21 (2011): 93-116100 AREA reasonably clearly importantly correctly properly appropriately Academic art 1 1.6 0.6 0.6 0.9 0.3 Case law 1.7 2.2 0.5 0.8 1.7 0.1 Legislation 2.6 0.4 0 0.1 0.9 0 Legal docs 7.7 0.3 0 0 1.7 0.1 BNC 0.3 1.5 0.1 0.2 0.6 0.1 BAWE 0.3 1.8 0.3 0.3 0.3 0.1 Table 2. Frequency of adverbs per 10,000 words. 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 100 applied to the behaviour of individuals, particularly regarding their civil liability in cases of negligence, and is thought to date back to the leading tort case of vaughan v. Menlove (132 ER 490, Common Pleas 1837) (see Schwartz, 1989, for discussion of the reasonable person in negligence cases). It is now also frequently applied in an epistemological sense to the issue of how legal documents (contract, treaties, codes and so on) should be interpreted (Knapp, Crystal & Prince, 2003; Orts, 2005), and is therefore particularly common wherever the meaning of written material is in dispute. The jurist Oliver Wendell Holmes Jr. explained the theory behind the reasonable person standard as stemming from the impossibility of “measuring a man’s powers and limitations” (Wendell Homes, 1881: 108) and the need to presume that a kind of average conduct exists above and beyond individual peculiarities and quirks. The notion of the “reasonable person” has been the object of considerable discussion in jurisprudence, touching on both its philosophical basis and its practical application (for accounts of the debate in legal theory, see Di Matteo, 1996, and gilles, 2001). This construct has come in for harsh criticism, since it seems to presuppose the existence of “an independent, mystical interpreter” (Di Matteo, 1996: 301) and entails an element of subjectivity, but although the notion of the “reasonable person” may be perceived as being less than robust, no satisfactory alternative has been devised. In fact, there is a need for such terms, which are useful precisely because they entail an element of openness and adaptability, and can be applied across a wide range of differing cases. There is a sense in which such words function as “wild cards”, whose precise application is left to the discretion of the court or relevant authority in each particular instance (Orts, 2005). This would seem to be an instance of the strategic role of vagueness and semantic indeterminacy in legal language, analysed in depth by Endicott (2000), which is counterbalanced in practice by the presence of authoritative interpreters of the law, in the form of judges, who can apply the general but unspecific concept to a concrete, particular case (Engberg & Heller, 2008). In order to investigate the use of “reasonable” and “reasonably” in legal discourse, the WordSmith concordance tool (Scott, 2007) was used to detect the n-grams and patterns in which these terms appeared, and to research any related semantic preferences. DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 101 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 101 It is noteworthy that 16 of the 114 instances of “reasonable” in academic articles were in inverted commas (either the word itself, or as part of a longer quotation). Although it is hard to know the force of quotation marks in a concordance line, the examples seemed to indicate that the writer was problematizing the concept of reasonableness, or at least throwing doubt on a previous instance in which it had been used. RUTH BREEzE Ibérica 21 (2011): 93-116102 L2 L1 R1 R2 and with prudence reasonable faith and attorneys and have reliance Time price not not reasonable Price time Table 3. Patterns in academic articles (node = “reasonable”). L2 L1 R1 R2 had and inference that support all opportunity interpretation the with probability efforts fair only reasonable interpretation and Table 4. Patterns in case law (node = “reasonable”). L2 L1 R1 R2 took all grounds expenses within the lengths fees take and diligence for fair with reasonable Fees and Table 5. Patterns in legislation (node = “reasonable”). L2 L1 R1 R2 use commercially efforts and shall use satisfaction fees and its detail the its all reasonable discretion including Table 6. Patterns in legal documents (node = “reasonable”). 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 102 DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 103 L2 L1 R1 R2 and with prudence reasonable faith and attorneys and have reliance time price not not reasonably price time Table 8. Patterns in academic articles (node = “reasonably”). L2 L1 R1 R2 could could have have were can infer infer the not practicable and would and reasonably well expect Table 9. Patterns in case law (node = “reasonably”). T Subcorpus (per 10,000) Academic art. (2.2) Case law (4.6) Legislation (4.3) Legal docs (9.8) R prudence 17 R reliance 7 R inference* 37 R interpretation 11 R probability 10 Fair and R 7 R attorneys’ fees 7 Support a R inference that 8 R probability of a business opportunity 6 Had a R probability of 5 R steps 39 He took all R steps 19 R expenses 13 R time 12 R grounds 9 R expenses incurred 9 R period 8 R cause 8 R care 6 Failed to take all R steps 6 R efforts 66 commercially R efforts 39 R commercial efforts 23 R detail 17 R fees 17 R attorneys 17 R access 15 R notice 12 R period 11 R discretion 8 R time* 6 R opportunity 6 R satisfaction 6 (clusters) reasonable (semantic preference) attorney’s fees/costs, time, price, cost, basis, construction, damages, formula, conclusion*, risk taking, parties, relationship, view, period, doubt, sum, mind*, prediction, way, businessman, substitute, hourly rate, connection, compensation, foreseeability, effort*, R to rely on, intelligence, expectation, chance of success, R to assume, take a R view certainty, effort*, decision*, view, objection*, steps, prospect*, fees, diligence, doubt, likelihood, sense, person*, grounds, alternative, time, cause, belief, arrangement*, proposal*, bases, expectancy, attitude, observer, care, opportunity enquiries, belief, likelihood, basis, expenses, time, accuracy, period, fees, progress, judgement, costs, expectation, satisfaction, intervals, enquiry, measures, conduct, prospect, diligence, purpose, person, compensation, witness, amount, precautions, opportunity, excuse, notice, means, accuracy, evidence approval, request*, assistance, basis, cooperation, manner, business, investigation, expenses Table 7. Clusters and semantic preferences of “reasonable” (R) in the four subcorpora. 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 103 From the foregoing tables, it is clear that although the semantic preferences of the high-frequency words “reasonable” and “reasonably” are similar across the four legal corpora, they are not interchangeable. It can be noted that, although both words were most frequently found in legal documents, RUTH BREEzE Ibérica 21 (2011): 93-116104 L2 L1 R1 R2 the may practicable believes soon director believed regarded state not regarded after will and reasonably likely expected Table 10. Patterns in legislation (node = “reasonably”). L2 L1 R1 R2 could not advisable request that may request required would could satisfactory expect and would reasonably acceptable Table 11. Patterns in legal documents (node = “reasonably”). Subcorpus (per 10,000) Academic art. (1) Case law (1.7) Legislation (2.6) Legal docs (7.7) R expect 7 R equivalent 6 could R 9 R be 8 may R 19 R practicable 16 R require 15 R be 12 R believes 10 Be R 10 R be expected 90 may R request 27 expect, estimate, induce, afford, construe, read, ask, decide, believe, act, view, rely, reject, determine, adjudge, project rely, expect, draw, consider, incur, relate, come to conclusion, reflect, infer, interpret, propose, differ, want, describe, include, think, disclose, approve, anticipate, act believe, regard, consider, be aware, require, understand, anticipate, exercise, incur, request, resolve, estimate, inform expect, determine, incur, require, believe, cooperate, request, act (clusters) reasonably (semantic preference:verbs) (semantic preference: adjectives) equivalent, foreseeable, short, confident, certain prompt, susceptible, available, necessary, circumspect, practicable, satisfied, confident, well informed, likely, obtainable, capable practicable, likely, necessary, open, sufficient, attentive, sufficient, able, diligent necessary, likely, satisfactory, practicable, acceptable, advisable, approximate, possible Table 12. Clusters and semantic preferences of “reasonably” (R) in the four subcorpora. 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 104 the patterns in which they occurred were more restricted than in the other subcorpora, being more formulaic, with a larger number of n-grams. Their semantic preferences were limited to a few nouns, verbs and adjectives. On the other hand, the greatest number of possible combinations using “reasonable” was found in academic articles, while the greatest for “reasonably” was in the case law corpus. Closer examination of the items that collocated with “reasonable” revealed that there was a semantic preference for words related to money and time across all four subcorpora, while epistemic judgements (“doubt”, “certainty”) were found in the academic, case law and legislation subcorpora. In the legislation subcorpus, “reasonable” also combined with a wide selection of actions such as “steps” and “measures”, and reasons such as “cause” and “grounds”. The legal documents subcorpus also combined “reasonable” with actions (“efforts”, “approval”, “request”), though in a more limited range, and was the only subcorpus to include the category of help, embodied in “assistance” and “cooperation”. In the case of “reasonably”, the semantic preference was clearly for verbs of thinking and anticipation. “Expect” was one of the commonest collocates in all the subcorpora except legislation, where “believe” headed the list of collocating verbs. “Request” and “require” were common collocates of “reasonably” in legislation and legal documents, but not in the other two subcorpora. In the academic articles and case law, however, “reasonably” was found in combination with a wider range of epistemic verbs (“reflect”, “interpret”, “construe”, “think”, “view”, “infer”) and verbs reporting speech acts (“describe”, “propose”, “disclose”). In general, as might be expected, there was somewhat greater overlap between legislation and legal documents, and a rather greater divide between these and the other two subcorpora. Not only are legislation and legal documents more repetitive and formulaic, but they are also more likely to focus on procedural aspects, on actions, requests and requirements, than are the areas of case law and academic writing. The greater degree of commonality between academic articles and case law may also be attributed to rhetorical aspects, such as the need to argue a case, which would tend to necessitate a greater range of epistemic verbs. It is hardly surprising that the collocations of “reasonable” and “reasonably” in these genres should differ sharply from those found in the prescriptive language of laws and legal documents, which exist in order to set out clear rules for people to follow. DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 105 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 105 Case study: “clear/clearly” As Table 1 shows, “clear” was most common in academic articles and case law, with higher frequency than in the BNC and BAWE, but was infrequent in legislation and legal documents. It seems logical that this should be the case, since “clear” is an epistemic adjective which is frequently used in the discourse of argument. This use would seem to be confirmed by the frequency of “be* clear” and “make* clear” in these subcorpora (73 and 23 occurrences in academic articles and 89 and 46 occurrences in case law, respectively). The semantic preference of clear in academic articles was for “whether”, and for “failure”, “example”, “framework”, “evidence”, “implication”, “manifestation”, “words”, “instance”, “line”, “provisions”, “leader”, “rule”, “violation”, “directive”, “agreement”, “reference”, “separation”, “consent”, “contractual language”, “exception”, “distinction”, “custom”; in other words, for a range of lexical items connected with concrete evidence or intellectual constructs. In case law, in addition to “whether”, the pattern was similar: “acknowledgement”, “view”, “breach”, “instance”, “error”, “analysis”, “invitation”, “message”, “terms”, “notice”, “answer”, “case”, “distinction”, “shortcomings”, “example”, “evidence”, “provision”, “choice”, “intent”, “meaning”, “conflict”, “language”. “Clear” only occurred twice in the entire legislation subcorpus, while in the legal documents subcorpus it appeared 20 times, in the combination “free and clear (of all/any liens)”. It was notable that the academic articles subcorpus was the only one in which “unclear” also occurred (frequency 0.4/10,000), which points to a difference between academic and judicial discourse: academic writers take an objective stance to complex issues, and may find it expedient to identify areas of law or enquiry that are unclear; the judge’s function is to rule on particular cases, which means that his/her discourse is likely to emphasise clarity rather than its absence. “Clearly” was present in academic articles at around the same frequency as in the BNC and BAWE (1.6 as compared to 1.5 and 1.8), was markedly more frequent in the case law subcorpus (frequency 2.2), and was present to a negligible extent in the other two subcorpora. In academic articles it was found with the verbs “limit”, “allow”, “exceed”, “raise”, “intend”, “protect”, “preclude”, “suggest”, “reflect”, “incorporate”, “lay down”, “anticipate”, “exist”, “demarcate”, “meet”, “feel”, “draft”, “express”, “cover”, “separate”, “imply”, “promote”, “consider”, and with the adjective “unfair”. In case law, it co-occurred with “state”, “acknowledge”, “estimate”, “exclude”, “define”, “express”, “consider”, “warrant”, “constitute”, “refer”, “intend”, RUTH BREEzE Ibérica 21 (2011): 93-116106 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 106 “foreshadow”, “prefer”, “communicate”, “presuppose”, “cover”, “contemplate”, “believe”, “relate”, “fail”, “comprehend”, “regard”, “foresee”, and with the adjectives “identifiable”, “honest”, “correct”, “contrary”, “erroneous”. In the foregoing, it is noticeable that epistemic verbs are present to a considerable degree among the collocates in case law, whereas academic texts include a greater preponderance of verbs with a concrete meaning. This probably reflects the special preoccupation with intentions and interpretations in case law. In case law, “clearly” also often occurred adverbially, separated from the rest of the sentence by commas (“Clearly, the defendant was not aware …”). Adverbs are generally more often found in spoken than written registers (Biber, 2006), and so the greater presence of “clearly” here may be due to the fact that law reports and court opinions are heavily influenced by the spoken language of the courtroom. Case study: “important/importantly” As Tables 1 and 2 show, “important” was frequent in academic articles and fairly frequent in case law, while “importantly” was salient in academic articles and frequent in case law. Both “important” and “importantly” were largely absent from the other two subcorpora. In academic articles, important was frequently used in comparative or superlative forms (25 instances of “most important” and 21 of “more important”). “Important” appeared with the following nouns: “role”, “implications”, “policy”, “factor”, “aspect”, “case”, “point”, “question”, “characteristic”, “element”, “concern”. This item thus seemed to be used in discursive functions, chiefly for discriminating between different parts of a whole. In case law, “important” was followed by “issue”, “interest”, “provision”, “question”, “company”, “element”, “witness”, “role”, “point”, “matter”, “part”, “link”, “similarity”, “element”, “term”, “question”, “insight”, “role”, “omission”, “fact”. The wider semantic range of these nouns points to “important” being used across a broader range of uses, including the evaluation of players in the case at hand (“company”, “witness”). “Importantly” was mainly used as an adverbial discourse marker in both academic articles (30 out of 31 instances) and case law (25 out of 26 instances). In both cases, the comparative and superlative forms were more frequent than the unmodified adverb. The high frequency of this adverb in comparison to BNC data seems to point to its being a regular feature of argumentative legal discourse. DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 107 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 107 Case study: “correct/correctly” “Correct” appeared in academic articles, case law and legal documents much more frequently than in the BNC (1.8, 1.7 and 1.7, respectively, as compared to 0.7 in the BNC). However, it was infrequent in the legislation subcorpus (0.4). In academic articles, the most frequent combinations were “correct answer” (36 instances) and “correct interpretation” (9 instances), as well as “be* correct” (23 instances), often used impersonally. In case law, “be* correct” was also common (48 instances), and “correct” was also found qualifying “evaluation”, “assertion”, “amount”, “statement” and “approach”. In other words, “correct” appears to be an adjective that indicates the writer’s approval of a particular answer, approach or interpretation, and is thus basically epistemic, indicative of certainty. In legal documents, however, the use of “correct” was quite different, since almost all the instances refer to “accounts”, “copies” and “lists”, that is, to the accuracy of concrete documentation. In the legal documents subcorpus, the most frequent clusters are “true and correct” (31 instances), “complete and correct” (14 instances) and “correct and complete” (14 instances). “Correctly” is salient in academic articles and case law (frequencies of 0.6 and 0.8 respectively, as compared with 0.2 in the BNC and 0.3 in BAWE), and is found with verbs of noticing, thinking and deciding such as “hold”, “decide”, “interpret”, “note”, “identify”, “see”, “anticipate”, “demonstrate”, “consider”, “allow” and “distinguish” in articles, and with “point out”, “argue”, “note”, “observe”, “presume”, “identify”, “describe”, “conclude” and “hold” in case law. Case study: “proper/properly” “Proper” was salient in case law and legislation, perhaps because there is a particular focus on procedure in both of these areas. “Properly”, by contrast, was salient in case law and legal documents. In case law, proper was found with the following range of nouns: “interpretation”, “notice”, “decisions”, “means”, “grounds”, “opportunity”, “place”, “purposes”, “consideration”, “measure”, “course”, “disclosure”, “standard”, “construction”, “analysis”, “mechanism”, “negotiation”, “scope”, “time”, “consequences”, “treatment”. These would seem to group into four categories: procedures (“steps”, “measures”), interpretations (“analysis”, “construction”), reasons (“grounds”, “purposes”) and conditions (“times”, “places” and so on). The semantic preferences of RUTH BREEzE Ibérica 21 (2011): 93-116108 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 108 “proper” in legislation centred mainly on conditions: “person”, “delivery”, “purpose”, “information”, “party”, “person”, “accounts”, “instruments”, “proceedings”, “address”, “form”, “official”, “claimant”, “business”, “officer”. In case law, “properly” was found in combination with “act”, “allege”, “assert”, “belong”, “view”, “understand”, “consider”, “describe”, “consult”, “describe”, “pay”, “establish”, “make”, “put forward”, “give”, “find”, “accept”, “incur”, “conclude”, “apply”, “represent”, “refuse”, “relate”, “present”, “frame”, “bring a claim”, “serve”, “distribute”, “reflect”, “hold”, “grant”. In legal documents, it co-occurred with “bring”, “make a complaint”, “sign”, “address”, “submit”, “accompany”, “incur”, “endorse”, “execute”, “maintain”, “reflect”, “require”, “file”, “explain”, “administer”, “convene”, “deal with”, “exercise2. In this subcorpus, it was particularly common in the n-grams “properly completed” (16 instances) and “properly completed and signed” (8 instances). In both cases, “properly” seemed to be used across a range of functions to indicate that something is done in the legally prescribed manner. Case study: “appropriate/appropriately” The occurrences of “appropriate” were mainly in case law and legislation. In case law (frequency 2.8), “appropriate” combined with “meaning”, “approach”, “forum”, “order”, “value”, “share”, “recommendation”, “date”, “advice”, “party”, “measure”, “proportion”, “method”, “principle”, “case”, “response”, “defence”, “fee”, “claim” and “form of relief ”. In legislation, where “appropriate” was extremely frequent, accounting for 4 out of every 10,000 words, its most frequent collocates were “condition”, “term”, “arrangement”, “authority”, “rate” and “qualification”, and other collocates were: “modification”, “official”, “standard”, “person”, “enquiry”, “number”, “approval”. “Appropriate” was also found in the phrases consider* appropriate (21 instances) and “deem* appropriate” (11 instances). The adverb “appropriately” was only more frequent in academic articles than in the BNC, and was found with the verbs “measure”, “treat”, “protect”, “resolve” and “define”. As in the case of “proper”, the basic meaning of “appropriate” seems to imply a quality relating to actions that have the stamp of legal approval. The difference would appear to be that “appropriate” implies something that is approved because it is proportionate (“share”, “value”) or fitting in a context DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 109 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 109 (“defence”, “advice”, “recommendation”), whereas proper may also be applied to ratiocinatory phenomena such as “interpretation” and “analysis”. In other words, “proper” means “right”, but “appropriate” only means “right in context”, and does not convey a strong enough sense of security for describing activities of epistemological inquiry. On the other hand, both “proper” and “appropriate” seem to be applicable to material conditions, such as “date” or “fee”. A similar contrast was apparent in the case of the corresponding adverbs, so that “properly” co-occurred with a wide range of epistemic and action verbs, whereas “appropriately” was mainly found with certain specific action verbs. The main pattern emerging from the case studies of “appropriate” and “proper” is that in case law, “proper”, with its implicit meaning of procedural or intellectual rightness, is both stronger and more widely applicable than “appropriate”, which is understood as a relative matter, indicating rightness in a particular context. Thus important intellectual activities to do with the reasoning of legal cases, such as interpreting, construing or analysing, can be done “properly”, whereas “appropriate/appropriately” seemed to be more generally used to refer to actions and their suitability in a particular context. In legislation, however, “proper” and “appropriate” both referred mainly to conditions and procedures, as was also the case for “properly” in legal documents. Discussion It is known that the epistemic and evaluative lexis in academic and professional writing varies from one discipline to another, and that the observable differences may stem from different underlying epistemological frameworks in different areas of academia (Meyer, 1997; Hiltunen, 2006; giannoni, 2009). Against this background, this paper explored the use of six salient stance-denoting adjective/adverb pairs across four subcorpora representing different areas of legal discourse. Their relative frequency in the different areas of legal writing was considered, and the n-grams and semantic preferences associated with these words in each text type were examined. In very general terms, “reasonable/reasonably”, “correct/correctly”, “proper/properly” and “appropriate” emerged as being particularly prominent in particular areas of legal discourse, and as being likely to embody community-specific values that shape the disciplinary culture of law. RUTH BREEzE Ibérica 21 (2011): 93-116110 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 110 Looking first at the comparisons of word frequency that were made between subcorpora, complex patterns emerged, but in general it can be stated that there was a certain degree of common ground as far as the frequency and use of the lexical items considered here in academic articles and case law (law reports and court opinions) were concerned. This similarity could be explained by the similarity in rhetorical purpose (to argue and persuade, as opposed to prescribing or proscribing), or by the fact that these genres are freer and tend to be characterised by a more personal voice, whereas legislation and legal documents tend to be more formulaic and more repetitive. However, further research would be needed to establish whether the similarities and differences between genres that were detected actually extend to other aspects of stance, such as hedging or verb choice. Regarding the actual adjective/adverbs themselves, the results of this study provide an interesting overview of certain salient values in legal discourse, and the role they play in text. While “clear/clearly” and “important/importantly” appear to have a persuasive function, being used to project epistemic certainty and order points in the discourse, the other four pairs (“reasonable/reasonably”, “correct/correctly”, “proper/properly” and “appropriate/appropriately”) appear to convey disciplinary value- judgements, and despite their apparent similarity, seem not to be interchangeable. Specifically, the finding of specific patterns as regards n- grams and semantic preferences tends to support the notion that these words have different discipline-specific roles which are hard for non- specialists to define precisely. In general, they appear to belong to the class of lexical items which has been described as “sub-technical vocabulary” (Dudley-Evans & St. John, 1998: 80; Flowerdew, 2001: 78) or “crypto- technical vocabulary” (Howard, 1991: 15); that is, words which are neither truly common-core vocabulary (“good”, “useful”) nor genuine technical terms (“tort”, “duress”), but which can be described as “re-designated general language items” (Sager, Dungworth & McDonald, 1980: 24). However, other authors have distinguished between common words which occur regularly in a particular discipline and have acquired a specialised meaning (e.g. in medical English, words such as “complications” or “examination”), which in their view can be defined as “sub-technical vocabulary” in a strict sense, and common-core words that are simply more frequent in a specialised field, but which maintain their usual meaning, and are an essential component in the language of the discipline (Alcaraz, 2000; Pérez-Paredes & Sánchez, 2005). In the present study, even though the DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 111 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 111 words are all in general usage, and belong to the first thousand word families of English, their meaning within the specialized discourses of law has been found to be both more specific and more circumscribed than might be expected. It appears that these words can be situated in the second of these categories, that is, as common words that are markedly frequent in legal language, although one of the words (“reasonable/ly”) appears to cross the borderline between frequent words and sub-technical words as defined above. Of all the items studied here, “reasonable/reasonably” appears to be the most specific instance of a word with a special resonance in the field of law. Its extremely high frequency across all four subcorpora, and the large number of high-frequency n-grams associated with it, point to its having a particular, legally-specific significance. However, even though the construct of the “reasonable person” is familiar as part of legal discourse, the vast majority of occurrences in these corpora did not make a direct association between “reasonable” and “person”. Instead, most of the instances documented here would appear to extend that construct further, by some process of transfer, so that the commonest combinations were with personal characteristics (“prudence”), actions (“efforts”, “steps”), presumably such as would be expected of the “reasonable person”, and then with epistemological aspects such as “interpreting”, “inferring”, “concluding”, and so on, which presumably project the positive quality of reasonableness to the writer him/herself. It is thus viable to propose that reasonableness is truly a characteristic disciplinary value that crosses the borders of different legal genres, and is applied both in a classically technical sense, as a construct needed to measure human behaviour, and in an extended sense, as a value that legal writers wish to appropriate for themselves. Regarding “correct”, “proper” and “appropriate”, the boundaries dividing their semantic fields are harder to demarcate. Corpus studies in other areas have done much to elucidate the semantic preferences of apparent synonyms (cf. Partington, 1998, 39-47 on the preferences of “sheer”). In academic articles and case law, “correct/correctly” are generally found with verbs or nouns of thinking, saying and deciding. This is not wholly surprising in case law, because the discourses of case law are inevitably concerned with weighing up and determining the right course of action. It is more interesting that the same terminology is used almost to the same degree in academic writing, where its connotations of normativity and prescriptiveness might make it seem rather incongruous. Notably, “correct” RUTH BREEzE Ibérica 21 (2011): 93-116112 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 112 was found to be much more frequent in the academic law articles in the present study than in the BAWE corpus of cross-disciplinary academic written English. This evidence would seem, again, to indicate that the particular value of correctness, which may have originated in the context of the courtroom, has become enshrined in legal discourse across a range of genres. It was also noteworthy that “correct” also appeared frequently in the legal documents subcorpus, but that its use there was quite different, since here it referred almost exclusively to the accuracy of specific documents. As far as “proper” and “appropriate” and their respective adverbs are concerned, “proper” was found to combine with a relatively wide range of items, including epistemic verbs/nouns, in case law, and with a narrower range centring on correctly executed actions and procedures, in legislation and legal documents. “Appropriate” had a narrower range of application, and seemed mainly to denote a quality of proportionality or suitability to context. “Proper” was three times as common in academic law as in BAWE, as was “properly”, while “appropriate” was found to have the same frequency in academic law as in BAWE. Extrapolating from this, we can perceive that properness is a more pervasive legal value, which can apply to analyses and interpretations as well as to correctly executed procedures, and which appears to indicate that something is done as it should be, with a stamp of legal approval. Moreover, its use extends to the academic sphere, where it would not be expected. In this, the meaning and semantic preferences of “proper” are close to those of “correct”, although their fields of use differed: “proper” was particularly common in case law and legislation (areas where procedures are foregrounded), whereas “correct” was frequent in all the subcorpora except legislation. “Appropriate”, on the other hand, could be seen to denote suitability in context, rather than an abstract quality of rightness, and was not found with epistemic verbs except in the set phrases “consider appropriate” and “deem appropriate”, which again emphasize suitability in context rather than an abstract value. In summary, this paper has addressed the question of disciplinary values in law by detailed study of the behaviour of six adjective/adverb pairs in four subcorpora of legal texts. It is clear that academic authors and judges who deliver rulings and opinions construct an authoritative voice by signalling the value (or non-value) of their own contributions and of other knowledge claims that are under discussion, and that they rely on a specific disciplinary value framework to underpin their discourse. We have seen here that certain key lexical items of an evaluative nature (“reasonable”, “correct”, “proper”, DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 113 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 113 “appropriate”) appear to play an essential role in their value system. Moreover, it has also been shown that there is considerable overlap between the salience and use of these items in academic and case law, and in other areas of legal discourse, since the same words also play a significant role in the language of legislation and legal documents. Further research is needed to establish whether the discourses of other disciplines, such as economics or medicine, also make use of specific sets of prominent value-laden lexical items across a variety of different academic and professional genres. [Paper received March 2010] [Revised paper accepted August 2010] References RUTH BREEzE Ibérica 21 (2011): 93-116114 Adolphs, S. & V. Durow (2004). “Socio-cultural integration and the development of formulaic sequences” in N. Schmitt (ed.), Formulaic Sequences, 107-126. Amsterdam: John Benjamins. Alcaraz, E. (2000). El inglés profesional y académico. Madrid: Alianza Editorial. Biber, D. (2006). University Language: A Corpus- Based Study of Spoken and Written Registers. Amsterdam: John Benjamins. Dahl, T. (2009). “The linguistic representation of rhetorical function: a study of how economists present their knowledge claims”. Written Communication 26: 370-391. Di Matteo, L. (1996). “The counterpoise of contracts: the reasonable person standard and the subjectivity of judgment”. South Carolina Law Review 48: 293-236. Dudley Evans, T. & M. St John (1998). Developments in English for Specific Purposes. Cambridge: Cambridge University Press. Endicott, T. (2000). Vagueness in Law. Oxford: Oxford University Press. Engberg, J. & D. Heller (2008). “Vagueness and indeterminacy in law” in V. Bhatia, C. Candlin & J. Engberg (eds.), Legal Discourses across Cultures and Systems, 145-168. Hong Kong: Hong Kong University Press. Flowerdew, J. (2001). “Concordancing as a tool in course design” in M. Ghadessey, A. Henry & R. Roseberry (eds.), Small Corpus Studies and ELT: Theory and Practice, 71-92. Amsterdam: John Benjamins. Giannoni, D. (2009). “Disciplinary values in English academic metaphors”. Linguistica e filologia 28: 173-191. Gilles, S. (2001). “On determining negligence: hand formula balancing, the reasonable person standard, and the jury”. Vanderbilt Law Review 54: 814-861. Green, E. (1968). “The reasonable man: legal fiction or psychosocial reality?” Law and Society Review 2: 241-257. Hiltunen, T. (2006). “Coming-to-know verbs in research articles in three academic disciplines” in K.Neumann, R. Plo Alastrué & C. Pérez-Llantada Auría (eds.), Proceedings of the Fifth International AELFE Conference, 246-251. Zaragoza: Prensas Universitarias de Zaragoza. Howard, R. (1991). “Teaching medical English systematically”. English for Medical Purposes Newsletter 8: 15-21. Hyland, K. (1999). “Disciplinary discourses: writer stance in research articles” in C. Candlin, & K. Hyland (eds.), Writing: Texts, Processes and Practices, 99-121. London: Longman. Knapp, C.L., N.M. Crystal & H.G. Prince (2003). Problems in Contract Law. Cases and Materials. New York: Aspen. Knorr Cetina, K. (1999). Epistemic Cultures: How the Sciences make Knowledge. Cambridge MA: Harvard University Press. Lakoff, G. (1987). Women, Fire and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Illinois Press. Leech, G. (2000). “Grammar of spoken English: 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 114 Ruth Breeze has a Ph.D. in second language writing, and teaches legal English and media English at the University of Navarra, Spain. Her research interests include legal English, discourse analysis and media language, particularly focusing on persuasive discourses. DISCIPLINARy vALUES IN LEgAL DISCOURSE Ibérica 21 (2011): 93-116 115 new outcomes in corpus-oriented research”. Language Learning 50: 675-724. Meyer, P.G. (1997). Coming to Know. Studies in the Lexical Semantics and Pragmatics of Academic English. Tübingen: Narr. Nubiola, J. (2009). “What reasonableness really is”. Transactions of the Charles S. Peirce Society: A Quarterly Journal in American Philosophy 45: 125-134. Orts Llopis, M.A. (2005). “Semejanzas y diferencias léxicas en el lenguaje contractual en inglés y en castellano”. Ibérica 10: 23-40. Partington, A. (1998). Patterns and Meanings. Amsterdam: John Benjamins. Pawelec, A. (2006). “The death of metaphor”. Studia Linguistica Universitatis Iagellonicae Cracoviensis 123: 117-121. Pérez-Llantada Auría, C. (2008). “Stance and academic promotionalism: a cross-disciplinary comparison in the soft sciences”. Atlantis 30: 129- 145. Pérez-Paredes, P. & P. Sánchez Hernández (2005). “Examining English for academic purposes students’ vocabulary output: corpus-aided analysis and learner corpora”. Revista Española de Lingüística Aplicada 1: 201-212. Sager, J.C., D. Dungworth & P.F. McDonald (1980). English Special Languages: Principles and Practice in Science and Technology. Wiesbaden: Oscar Brandstetter. Schwartz, W. (1989). “Objective and subjective standards of negligence: defining the reasonable person to induce optimal care and optimal populations of injurers and victims”. Georgetown Law Journal 78: 241-280. Scott, M. (2007). Oxford WordSmith Tools 4.0. Oxford: Oxford University Press. Sinclair, J. (1998). “The lexical item” in E. Weigand (ed.), Contrastive Lexical Semantics, 1-24. Amsterdam: John Benjamins. Sinclair, J. (2004). Trust the Text. Language, Corpus and Discourse. London: Routledge. Stubbs, M. (2007). “On texts, corpora and models of language” in M. Hoey, M. Mahlberg, M. Stubbs & W. Teubert (eds.), Text, Discourse and Corpora: Theory and Analysis, 127-161. London: Continuum. Swales, J. & A. Burke (2003). “It’s really fascinating work: differences in evaluative adjectives across academic registers” in P. Leistyna & C. Meyer (eds.), Corpus Analysis. Language Structure and Language Use, 1-18. Amsterdam/New York: Rodopi. Tuldava, J. (1998). Probleme und Methoden der quantitativ-systemischen Lexikologie. Trier: Wissenschaftlicher Verlag. Tutin, A. (2008). “Evaluative adjectives in academic writing in the humanities and social sciences”. Paper presented at InterLAE conference: Interpersonality in Written Academic Discourse. Jaca, University of Zaragoza, Spain. Vass, H. (2004). “Socio-cognitive aspects of hedging in two legal discourse genres”. Ibérica 7: 125-141. Wendell Holmes, O. (1881). The Common Law. Boston: Little, Brown and Co. 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 115 05 IBERICA 21.qxp:Iberica 13 10/03/11 17:21 Página 116