• A. Introduction

  I. Background

  The present article investigates the influence of Plain Language (PL) in legal academic research, and more specifically, in legal research articles (LARs), a side-genre to more traditional legal realizations such as normative-statutory texts (i.e. statutes, laws, contracts, etc.) and forensic ones (i.e. monologues at trials, cross-examinations, etc.). The Plain Language Movement (PLM), which originated and quickly developed especially in Anglophone cultures and Common Law systems, considerably affected the way legal experts and practitioners working within these systems use the language in professional contexts, both in writing (i.e. when drafting regulations and norms) and in oral situations (i.e. when defending a cause or discussing cases in tribunals).¹xM. Adler, ‘The Plain Language Movement’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012. The assumption at the basis of the present analysis is that the exposure to and experience with such discursive practices both in legislative and forensic environments is likely to have influenced also the way experts write in research-related and academic settings.
  Studies in the legal domain have revealed “a number of regularities in the ways that legal language operates, suggesting that legal orders around the world attempt to harness linguistic regularities in an effort to make language both clear and definitive”.²xL. Solan & P. Tiersma, ‘Introduction’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 4. Such regularities concern both micro- and macro-linguistic features, ranging from terminology and syntax to text typology and genre. One of the main reasons of the existence of such templates, on the one hand, and their consistent use in legal contexts, on the other, is that conventionalized and ‘frozen’ forms are thought to better capture the possible complexity of the content, conferring authority – or ‘legal-ness’³xG. Cornu, Vocabulaire Juridique, Paris, PUF 2007; H. Mattila, ‘Legal Vocabulary’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012. – to its wording.⁴xF. Bennion, ‘Confusion over Plain Language Law’, The Commonwealth Lawyer, Vol. 16, No. 2, 2007. As a consequence, recognizability of such forms on the part of the recipient is believed to facilitate interpretation, hence boosting textual effectiveness. The main criticism to this view is that traditional and crystallized language resources run the risk of making meanings impenetrable (if not for the competent expert), since “complexity in subject matter does not call for complicated, convoluted language.”⁵xR. Eagleson, ‘Untitled Article’, Sydney Morning Herald, 20 January 1985, p. 8 (reprinted in Plain English and the Law. Report No. 9 by the Victorian Law Reform Commission, 1987). PLM was therefore intended to simplify the language, so that “people who are the audience of that communication can quickly and easily find what they need, understand what they find, act appropriately on that understanding.”⁶xN. James, ‘Defining the Profession: Placing Plain Language in the Field of Communication’, Clarity, No. 61, 2009, p. 35. In other words: “‘Plain language’ means language and design that presents information to its intended readers in a way that allows them, with as little effort as the complexity of the subject permits, to understand the writer’s meaning and to use the document”.⁷xM. Cutts, The Plain English Guide, Oxford, Oxford University Press 1996, p. 3.

  II. Legal Language and Plain Language

  Traditional legal language – which is still widely employed in Civil Law systems and is usually derogatorily referred to as ‘legalese’ – is characterized by such features as a highly impersonal style, the consistent use of passive constructions, archaisms and Latinisms, of long and complex sentences, and the lack of emotive connotations.⁸xM. Gotti, Specialized Discourse. Linguistic Features and Changing Conventions, Bern, Peter Lang 2003, pp. 35-36; P. Tiersma & L. Solan (Eds), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012. Among emphasis-containing resources, the handling of the impersonal tone and the control of emotive meanings are probably the most strategic ones.⁹xD. Mellinkoff, The Language of the Law, Boston, Little, Brown & Co 1963; C. Williams, Tradition and Change in Legal English. Verbal Constructions in Prescriptive Texts, Bern, Peter Lang 2007. Impersonal structures have often been recognized as being typical of legal language,¹⁰xIbid., pp. 36-38. often coupled with the use of passives, in order to conceal or contain individual – hence possibly subjective – agency on the part of the legislator (for instance by using terms like the Judge, the Jury, etc. instead of the first person pronouns) or for semantic transparency (through labels such as the Landlord, the Tenant, the Grantor, etc. rather than vaguely using the second and third person pronouns to refer to the same actors), besides emphasizing the result of a given action rather the role of the agent (in sentences like the request cannot be accepted, etc.¹¹xGotti 2003; M. Gotti, Maurizio & P. Anesa, ‘Professional Identities in Italian Arbitral Awards: The Spread of Lawyers’ Language’, in V. Bhatia & P. Evangelisti Allori (Eds.), Discourse and Identity in the Professions, Bern, Peter Lang 2011, p. 100. This way of using the language is so central to legal contexts that very often it is replicated in such domains as International Commercial Arbitration so as to make its pronouncements ‘sound’ and ‘qualify’ as fully fledged normative documents, that is, having the same performative character that is almost automatically associated to legal texts codified in such a language.¹²xGotti et al., 2011. Strictly connected to depersonalization, that is, the concealing of the legislator in text, is the tendency to avoid emphatic language: in order to enhance denotative precision and circumscribe the possible suspicion of bias or any ideological inclination, legal texts are characterized by the lack of emotive connotations. The tone of such texts is usually neutral and their illocutionary force hinges on the way concepts are organized and acts proscribed rather than on the use of emphatic formulations,¹³xGotti 2003, pp. 35-36. since such documents do not presuppose emotional involvement of the receiver.¹⁴xIbid., p. 141. These emphasis-controlling resources are very strategic in order to boost text effectiveness and corroborate its deontic quality.
  In systemic-functional terms, perlocutionary effectiveness depends on the way ideational material, that is, the content, is presented (i.e. the clarity, transparency, monoreferentiality, and conciseness of the expressions used to codify meanings) as much as on the use of interpersonal resources (i.e. those meant to express attitude and either enhance or downtone certain meanings according to the expectations and stance of the recipient). In traditional legal discourse, specifically in normative texts, ideational aspects (i.e. lexis, syntax, logical connectedness, and associative meanings) represent the main concern and focus. Language is primarily used as a function of meaning (with respect to the referent) rather than a function of communication (as a form of persuasion),¹⁵xA. Trosborg, Rhetorical Strategies in Legal Language: Discourse Analysis of Statutes and Contracts, Tübingen, Narr 1997, p. 25. since legal texts are to be interpreted on the basis of the “proper meaning of the words”,¹⁶xL. Solan, ‘Linguistic Issues in Statutory Interpretation’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 88. and of its conventionally accepted reference, in accordance with their purpose and their social context.¹⁷xEven the documented phenomenon of semantic vagueness, which appears to be at odds with this focus on clarity and comprehensibility, very often stems from the concern for semantic transparency, exhaustiveness, all-inclusiveness, and, from the standpoint of perlocution, in view of the wider applicability of a given norm to an extended range of unspecified or underspecified contexts (V. Bhatia, J. Engberg, M. Gotti & D. Heller (Eds.), Vagueness in Normative Texts, Bern, Peter Lang 2005; R. Poscher, ‘Ambiguity and Vagueness in Legal Interpretation’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012).
  Things seem to be slightly different in PL contexts. As a matter of fact, despite the natural prominence conferred to the representation of ideational material, ‘plainers’ (practitioners and campaigners) are engaged in limiting as much as possible the use of crystallized, stereotyped, and normalized formulations – although conventionally accepted and recognizable (proscribing, for instance, the use of euphemisms, archaisms, convoluted forms, long sentences, clichés, vague formulations, etc.) – whereas they attribute a considerable emphasis to interpersonal resources of the language. In fact, when synthesizing the main features of PL, Adler (2012) opens the list of requirements with the following suggestion: “adopt an informal a tone as is appropriate in the circumstances (which include the reader’s and the writer’s preferences and the relationship between them)”.¹⁸xAdler 2012, p. 77. According to studies and surveys,¹⁹xM. Adler, ‘British Lawyers’ Attitude to Plain English’, Clarity, No. 28, 1993; M. Adler, Clarity for Lawyers, London, The Law Society 2007; J. Kimble, Lifting the Fog of Legalese, Durham, Carolina Academy Press 2005. this tendency towards inclusion and reader-friendliness seems to be the preferred way of writing legal texts in Anglophone countries.

  III. Legal Research Articles

  Legal language, maybe more than other specialized languages, is far from being a monolithic unit, especially as regards its communicative functions in different settings. Legal discourse may in fact have a marked regulative and deontic function in legislative texts (i.e. laws, courtroom pronouncements, contracts, etc.), an argumentative function in forensic contexts (i.e. council/witness examinations, courtroom monologues, etc.), and an informative function in pedagogical texts (i.e. legal textbooks).²⁰xTrosborg 1997, pp. 20-22; M. Gotti, ‘Text and Genre’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press, 2012a. In this article, we will consider the case of LRAs, which belong to a very specific genre combining both an informative purpose and an argumentative one, being in fact meant to both discuss abstract principles or practical cases and, at the same time, persuade experts of the domain about the relevance and importance of the writer’s point in such discussions. Since this argumentative function is typical of all LRAs, they may be legitimately expected to be all written and organized in the same or in very similar ways. In reality, although they share some generic and linguistic features, LRAs vary consistently according to a variety of parameters such as the writers’ stance, their status, their education, their ideological orientation, the juridical system of reference, their mother tongue, etc.²¹xTrosborg 1997; P. Tiersma, Legal Language, Chicago, University of Chicago Press 1999; S. Gozdz-Roszkowski, Patterns of Linguistic Variation in American Legal English. A Corpus-Based Study, Bern, Peter Lang 2011; M. Sala, ‘Legal Expertise as a Cultural Identity Trait’, in V. Bhatia & P. Evangelisti Allori (Eds.), Discourse and Identity in the Professions, Bern, Peter Lang 2011; Tiersma & Solan 2012. The present article will investigate one of these factors of variation, namely, the influence of PL ideologies in academic writing. More specifically, the hypothesis at the basis of this analysis is that, irrespective of their generic realizations (informative-argumentative vs. constitutive genres), texts generated in Anglophone and PL-receptive contexts are likely to originate from and reflect the same concerns (i.e. effectiveness, transparency, reader-friendliness, and inclusion) which are distinctive of PL ideologies. Based on a comparison between a subcorpus of 40 LRAs written by English, American, and Australian authors and 40 LRAs authored by experts working in other linguistic and juridical contexts – thus, much less affected (or at least not so distinctively influenced) by PLM ideology – this investigation seeks to establish the main differences in the two subcorpora focussing especially on the interpersonal level of discourse and, more precisely, on the use of metadiscursive interactional strategies such as epistemic modality markers (i.e. certainty and uncertainty markers, intended to facilitate interpretation by either emphasising or mitigating assertiveness), personalization (used to lexically represent the author within his/her own text as the one who is responsible for it), and interactive markers such as code glosses (meant to paraphrase or reformulate meaning to both simplify and control interpretive processes).

• B. Material and Methodology

  I. Common Law and Civil Law Rhetoric

  One important aspect to assess at this point regards the criteria to evaluate the influence of PLM on legal academic writing and, more precisely, to establish how it is possible to identify texts produced in PL contexts. According to Adler,²²xAdler 2012. PL ideology first concerned (starting in the 1970s) and still distinctively affects Anglophone countries – namely, USA, Britain, Australia, New Zealand, and Anglophone Canada – whereas its bias is less marked and noticeable in Continental and Eastern Europe, East and South East Asia, parts of the Middle East, much of South America, Francophone Canada, and Africa. This distinction reflects, on the one hand, the one between countries where English is the native language and, on the other, that between the Common Law and Civil Law philosophies, that is, between those cultures where jurisprudence is based on the principle of the precedence (by means of which “the decision taken by one judge become binding on all subsequent similar cases”²³xGotti 2012a, p. 58.), where the right is argued, negotiated, and established in the courtroom and, on the other, those where principles become statutory laws by means of codification.²⁴xP. Tiersma, ‘A History of the Languages of Law’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 17. For Anglophone cultures, the law is a ‘socially’ negotiated act of jurisprudence (involving different bodies like the prosecution, the defense, the judge, the jury) whereas Civil Law systems hinge on the primacy of written norms and abstract principles established by the civil code. These different philosophies influence consistently the role played by language and the way it is used: in Civil Law contexts, it fixes broad principles into statutes, whereas in Common Law contexts, it is used to discuss practical cases (i.e. the precedents) and abstract principles drawn from or applied to them.²⁵xTiersma 2012, pp. 17-22; Gotti 2012a, pp. 58-59. On the basis of this differentiation, it is possible to claim that, in the first case, language is used to confer authority and enhance the normative character of legal texts, whereas, in the second, “the common law draftsman seems to be more worried about not being misunderstood by the specialist community”,²⁶xV. Bhatia, Analysing Genre: Language Use in Professional Settings, New York, Longman 1993, p. 137. hence language here has (also) a marked referential and informative function: texts need to be clearly understood to acquire performativity, that is, the impact and force of a legal document. For these reasons, PLM and its urge for clarity and unambiguity has been and still is very much influential in Common Law cultures, mainly Anglophone countries (“where English is the language of the law, Common Law is the system most used”,²⁷xJ. Gibbons, Forensic Linguistics, Malden/Oxford, Blackwell 2003, p. 5. whereas in other areas of the world it has a more limited and circumscribed effect for a variety of contextual, historical, political, and cultural factors.²⁸xAdler 2012.

  II. The Corpus

  The analysis presented here is based on CADIS (Corpus of Academic Discourse, compiled at the University of Bergamo,²⁹xM. Gotti, ‘Creating a Corpus for the Analysis of Identity Traits in English Specialised Discourse’, The European English Messenger, No. 15/2, 2006, pp. 44-47; M. Gotti (Ed.), Academic Identity Traits, Bern, Peter Lang 2012b. more specifically on a selection of 80 single-authored LRAs published between the years 2000 and 2010 in leading specialized journals, namely, the European Law Journal (henceforth indicated as ELJ), the European Journal of International Law (EJIL), the Harvard Law Review (HLR), the Yale Law Journal (YLJ), the Stanford Law Review (SLR), the Harvard International Law Journal (HILJ), and the International Review of Law and Economics (IRLE). The texts collected here can be further distinguished between those written by native (40 LRAs totalling over 680,000 running words) and by non-native speakers (40 LRAs totalling over 510,000 types).³⁰xThe operation of dividing authors on the basis of their native language has been made possible owing to the biographical information provided in the footnote which is usually linked to the author’s name and which contains the authors’ affiliation, their biographical, and bibliographical information, the presence in their curriculum of publications written in non-Anglophone languages, etc. This distinction reflects not just the level of linguistic competence that writers possess, which is not at stake here, but rather, and most prominently, their different level of expertise in either Common Law or Civil Law jurisdictional and, most prominently, discursive practices. Given the strong influence of PL in Common Law and Anglophone cultures, it is possible to hypothesize that LRAs authored by native speakers are likely to be more receptive to PL, or markedly PL-oriented, whereas non-native authors operating in other cultures are bound to be less biased by this ideology. The assumption at the basis of this investigation is that the same concern for effectiveness and simplification which is at the core of PL for statutory Common Law documents is likely to be transferred into academic writing by native speakers of English, whereas Civil Law experts and non-natives are expected to privilege authoritative wordings – rather than interpersonal strategies – to guide interpretation on the part of the readers.

  III. Analytical Tools

  Quantitative searches are based on both automatic ad manual analysis. WordSmith Tool 5.0³¹xM. Scott, Oxford WordSmith Tool 5.0, Oxford, Oxford University Press 2007. was used for automated word counts, and the results were then examined through a stage of manual revision to filter out non-relevant instances, i.e. those found in quotations, appendixes, references, tables, etc. As to the qualitative parameters applied to measure attitudinal variation and differences at the interpersonal level of discourse, this analysis concentrates on the use of personalization markers, epistemic modality markers, and code glosses.
  As far as personalization is concerned, first person pronouns both singular and plural (i.e. I, me, we, us, you) and relevant modifiers (i.e. my, mine, our, ours, your, yours) will be counted. The employment of such pronouns – which lexically represent the participants in the communicative event – is symptomatic of the way writers conceptualize their role as ‘guides’ throughout the texts (variably as arguers, researchers, or text-organizers,³²xK. Fløttum, T. Kinn & T. Dahl, ‘“We Now Report on…” versus “Let Us Now See How…”: Author Roles and Interaction with Readers in Research Articles’, in K. Hyland & M. Bondi (Eds.), Academic Discourse across Disciplines, Bern, Peter Lang 2006. thus explicitating their responsibility as facilitators for the processing of the text and their writer-responsible attitude for the decodification of the meaning – a concern which, when lexically and semantically marked, is in line with the principles of PL.
  As to epistemic resources, all the range of boosting and hedging expressions³³xK. Hyland, Metadiscourse, Continuum, London 2005. will be examined, according to their frequency and distribution. Verbal, adjectival, and adverbial formulations which are used to emphasize the level of certainty of a claim and the writer’s commitment towards it will be counted as boosters; conversely hedges are those expressions which are employed to downtone certainty and emphasize possibility. According to the use of such resources, it will be possible to measure the textual interplay between the participants’ stance or, more precisely, between the writers’ persuasive intentions and the readers’ expected attitude and needs. Finally, code glosses, that is, paraphrases, reformulations, synthesis, and exemplifications, will be examined in order to measure the cooperative attitude of the writers, in that revealing their concern for clarity and effectiveness, which is also one of the main criteria of PL.
  Although all occurrences of such linguistic resources will be counted, our searches are not merely designed to measure frequencies, but also the distribution of interpersonal elements throughout the whole corpus, in order to locate and rule out instances of variation due solely to personal style. The significance of the various strategies has been set in their presence in at least 20% of the total number of texts in one of the two subcorpora. Therefore, formulations which are numerically noticeable on the whole, but are found only in a limited number of texts, will not be considered as relevant for the present discussion.

• C. Results and Discussion

  I. First Person Pronouns and Possessives

  Singular and plural first person pronouns and their relevant possessive forms are quite common in our corpus. The results of our word count expressed both in absolute and normalized figures (calculated per 10,000 words) will be listed in Table 1. The left-hand side columns report occurrences found in LRAs written in Anglophone and PL-oriented cultures (PLC), whereas in the right-hand columns we find occurrences counted in LRAs produced in other cultures (OC), that is, by users with a different native language and specific competence in a different juridical system (i.e. Civil Law).

  Table 1. Distribution of first person pronouns.

  	PLC		OC
  	Absolute	Normalized	Absolute	Normalized
  First person singular
  I	2,001	22.1	799	13.7
  Me	69	0.8	22	0.4
  My/mine	288	3.3	69	1.3
  Total	2,358	26.2	890	15.4
  First person plural
  We	899	9.9	449	7.8
  Us	192	2.2	120	1.8
  Our/ours	498	5.6	81	1.5
  Total	1,589	17.7	650	11.1
  Second person
  You	76	0.9	5	0.1
  Your/yours	19	0.3	2	0.1
  Total	95	1.2	7	0.2
  Total	4,042	45.1	1,547	26.7

  By observing the data in the table, the most relevant difference between the two subcorpora resides in the handling of the first person singular, which has a marked identifying function with respect to the authorial self. As we see, writers in PLC tend to lexicalize their persona more consistently than those in OC (26.2 vs. 15.4), thus representing their textual role and function as the organizers of the meaning, that is, those who are in charge of providing informative material and to make the interpretive process easy for the reader, as can be seen in the examples below:

• (1) For these purposes, I use the term ‘legitimacy’ in the sense advanced by Thomas Franck as “a property of a rule … which itself exerts a pull toward compliance on those addressed normatively because those addressed believe that the rule or institution has come into being and operates in accordance with generally accepted principles of right process.” (HLR 59, emphasis added, henceforth e.a.)

• (2) For the avoidance of doubt, I am adopting Fuller’s position only at its most abstract. I do not agree with Fuller that law must be unitary, and thus must have only one purpose. Nor do I necessarily agree with the specific purpose which he allotted to law as being the best available. In fact I would contend (but shall not elaborate here) that Fuller’s deeply Anglo-American Liberal perspective on human agency is in fact the driving force of his work, which in result – though not in methodology – is far more classically natural law than is generally realized. (EJIL 16, e.a.)

• (3) In each of these realms, some of the administrative law procedures and mechanisms that are essential to good governance have been adopted, and, as my theoretical framework would suggest, I find that the regime of administrative law has advanced most where the governance is supranational, formal, and addresses normative issues. (YLJ 115, e.a.)

As can be seen in the excerpts, irrespective of their specific function – whereas portraying the writer’s role as text-organizer (1), arguer (2), or researcher (3)³⁴xFløttum et al. 2006. – by using such formulations the writer is emphasizing his/her relevance as the sole (or main) responsible for the text and its meaning.
The difference between the two subcorpora is drastically less marked as regards the use of the first person plural (17.7 for PLC vs. 11.1 in OC). Whereas the singular clearly points to the writer, the use of the plural is far less face-threatening and writer-responsible. As a matter of fact, individual authors can conceal their role under an exclusive plural (4), as though they were writing on behalf of a larger number of experts in the domain or a virtual research group, or else they use the plural exploiting its inclusive function (i.e. we as the participant to the communicative exchange, cf. (5)), or its discursive function (i.e. we as human beings, cf. (6)), as the texts below show:

• (4) Export promotion is sometimes considered an important macroeconomic policy to counteract unwanted trade deficits. In our example, the connection between official support for the export of a U.S. $10 million turbine and trade deficit reduction may be tenuous, but when the export is […] an entire oil refinery, the impact may be profound and discernable. (HILJ 45, e.a.)

• (5) Let us assume that a multilateral environmental agreement concluded, inter alia, by a and b provides that a is not obligated to allow imports from b, if b does not comply with the MEA, and let us assume that this condition is fulfilled. Then there is no obligation of a toward b not to restrict imports of b’s goods. (EJIL 17, e.a.)

• (6) We are living through a defining moment in international law. The pace of globalization makes cooperation through international law and institutions vital. (HILJ 45, e.a.)

As we can see, by the use of the plural, writers, even disguising their identity, yet remain in control of their text and implicitly represent themselves as facilitators for its appropriate reading, either by calling attention on specific points accessible to the community, by exploiting community acceptance (inclusive we), or by conferring authoritativeness to what is said (exclusive we, i.e. by presenting meaning as ‘not just’ the writer’s individual view).
Interestingly enough, the occurrence of the second person plural is very scarce and entirely unnoticeable in both subcorpora (1.2 for PLC vs. 0.2 in OC), and the very few instances have a discursive function, in that they could easily be substituted by other undifferentiated formulations – that is, the inclusive we in expressions such as as you will see, or adverbial forms like as it were or so to speak for if you will, cf. (7) – or impersonal pronouns – that is, the generic pronoun one in expressions like the richer you become or you need to be clear, cf. (8) and (9) – since they are found in crystallized expressions as those in the following examples:

• (7) The ever-present ‘fact’ of the facticity of law is, if you will, the only repetition – put another way, global means several local places at once. (HILJ 47, e.a.)

• (8) [The] marginal utility of leisure is increasing with income and vice versa, or, in other words, that the richer you become the more you value leisure at the margin, while maintaining the assumption of risk neutrality. (IRLE 25, e.a.)

• (9) Obviously, you need to be clear with your partner on things such as house-cleaning and cooking. Just don’t mix up these day-to-day issues with the bigger legal issues of living together. (YLJ 116, e.a.)

As it is possible to observe, such expressions are not expressly devised to appeal to the actual reader of that specific LAR, but they are taken from a pre-set range of stereotypical linguistic resources used in all conventionalized communicative situations to indicate the general and anonymous receiver. Nonetheless, even though such markers have a purely conventional function, their use automatically confers to the text a marked interpersonal (and almost dialogic) character. The fact that PLC resort to such strategies more frequently than OC, despite their general scarceness, is certainly indicative of a more inclusive rhetorical style.

II. Epistemic Modality Markers

Another sort of interpersonal resource, aimed at enhancing persuasion and stressing evidentiality and inclusion, is represented by markers of epistemic modality, that is, those formulations used by writers to express or imply their attitude both towards the content and the reader, thus signalling their will to disclose their stance and to create a given interactive basis between themselves and their audience. These resources can be distinguished according to the level of commitment that writers express towards the truth of a claim and the degree of certainty which is conferred to a given meaning into boosters and hedges, respectively markers of certainty and uncertainty.³⁵xHland 2005.
By the use of boosting strategies, writers enhance the epistemological validity of a given claim. For the purpose of this study, only the ten most frequent boosters were counted and discussed, namely, the adverbial formulations certainly, clearly, indeed, of course, obviously, necessarily, and surely and the modal verbs must, will, and cannot. Hedging resources are instead employed to downtone assertiveness and imposition by mitigating propositional meaning or circumscribing it within the paradigm of possibility rather than certainty. As far as hedges are concerned, the ten most common instances are represented by adjectival and/or adverbial expressions such as possible/ possibly, probable/probably, likely/unlikely, perhaps, modals like can, could, may, might, and the verbs seem(s) and appear(s). The frequency of such resources can be seen in Table 2.

Table 2. Distribution of epistemic modality markers.

	PLC		OC
	Absolute	Normalized	Absolute	Normalized
Boosters
Certainly	103	1.2	80	1.4
Cannot	404	4.7	345	5.9
Clearly	258	3.0	131	2.2
Indeed	383	4.5	202	3.5
Must	951	11.1	517	8.9
Necessarily	167	1.9	225	3.9
Obviously	50	0.5	44	0.7
Of course	262	3.1	94	1.6
Surely	55	0.6	16	0.3
Will	2095	24.6	1129	19.4
Total	4,728	55.2	2,783	47.8
Hedges
Appear(s)	290	3.5	162	2.8
Can	1,636	19.2	963	16.7
Could	1,385	16.3	506	8.8
Likely/unlikely	567	6.7	145	2.6
May	1,932	22.7	637	11.0
Might	1,053	12.4	293	4.2
Perhaps	312	3.7	46	0.9
Possible/possibly	412	4.8	289	5.1
Probable/probably	89	1.0	68	13
Seem(s)	436	5.2	292	5.2
Total	8,130	95.5	3,405	58.6
Total	12,858	150,7	6,188	106.4

By observing the total of both markers in the corpus, we see that PLC writers in general make a more extensive use of modalization than those stemming from OC, that is, they tend to make claims more effective and easy to understand by emphasizing certain aspects and softening some others. As observed above, the emphatic function is performed by boosters, which represent the content of a proposition as accepted and undisputable in that it is (presented as if it were) based on common or domain-specific acquired knowledge. Such strategies maximize evidentiality, relevance, and importance discursively rather than empirically, that is, by claiming something as an evidence without providing empirical element to support it, as we can observe in the excerpts below:

• (10) Are the Assessors Required to Assess the Risks Specifically or in the Light of General Studies? The answer to that third question must be nuanced. (ELJ 12, e.a.)

• (11) Denmark, Finland, and the Netherlands have revised their originally liberal approach and imposed different types of restrictions on free movement from new Member States. These restrictions certainly deserve critical scrutiny. (ELJ 11, e.a.)

• (12) The most obvious effect of increasing intrinsic deference, of course, is to increase actual deference, which enables agencies to interpret their statutory mandates more aggressively and to promulgate less textually plausible interpretations. (HLR 120, e.a.)

As we see in these examples, the reasons why a third question is to be nuanced (10), or given restrictions deserve scrutiny (11), or increasing defense should increase deference (12), are neither explicitated nor substantiated. The propositional content is asserted in an emphatic way so as to boost its credibility on the basis of its (supposed) self-evidence or community acceptance. The use of such formulations confer trustworthiness to the meaning, authoritativeness to the writer, and mark group membership between authors and their audience.³⁶xJ. Coates, ‘Epistemic Modality and Spoken Discourse’, Transactions of the Philological Society, Vol. 85, No. 1, 1987; K. Hyland, ‘Persuasion in Academic Articles’, Perspectives, No. 11, 1999.
Hedges are epistemic resources of a different polarity, marking uncertainty and rhetorically accounting for the possibility of different and/or discordant interpretation than the one supported by the writer, as the extracts below indicate:

• (13) From this, it can be easily seen that A may possibly prefer not acquiring information regardless of the size of the survey cost and his belief under Rule 2. (IRLE 24, e.a.)

• (14) The terminological difficulty might be indicative of the need for further elaboration and clarification of Tomuschat’s approach. (HILJ 47, e.a.)

• (15) This approach appears to have animated the district court’s rejection of the need for suspension in the Hawaiian Territories in the cases discussed above. (SLR 59, e.a.)

As we notice in the examples, the propositional content introduced by the hedging expression is presented as something possible, probable, or plausible rather than an ascertained truth; thus its acceptability is discursively negotiated with the reader. Such resources, by presupposing and stimulating the recipients’ competence and guiding their alignment to the writer’s own position, have a marked interpersonal function, having the purpose to persuade the audience and bring them to accept the writer’s meaning without sounding threatening, imposing, or arrogant.
As we have already remarked, PLC writers resort more extensively to epistemic modalization, both in terms of enhancing and mitigating markers; therefore their discourse is more interpersonally marked, by calling for and implying the readers’ response, which is not the case – at least not so remarkably – of OC discourse. An interesting quantitative founding is the fact that PLC writers use hedges almost twice as much as boosters, whereas in OC texts we see a drastically different trend, in that the occurrence of hedges and boosters per 10,000 words is very similar. This indicates that boosting formulations here, if numerically contained, are proportionally more significant, in that they balance the presence of mitigators. As a result OC writers are likely to sound more assertive, or, at least, less concerned with interpretive possibilities, possible objections, and unverified truths.

III. Code Glosses

Code glosses are those expressions introducing instantiations, exemplifications, synthesis, simplifications, summarizations, or, in general, alternative ways to make sense of the writers’ meaning.³⁷xHyland 2005. The most frequently found in our corpus are formulations such as for instance, for examples, i.e., e.g., in other words, in sum, that is/that is to say, namely, and specifically. Their quantitative occurrence is expressed in Table 3.

Table 3. Distribution of code glosses.

	PLC		OC
	Absolute	Normalized	Absolute	Normalized
e.g.	1,105	13.0	398	6.9
For example	807	9.5	144	2.5
For instance	95	1.1	121	2.1
i.e.	95	1.1	222	3.8
In other words	75	0.9	87	1.5
In sum	26	0.3	15	0.3
Namely	54	0.6	63	1.1
Specifically	22	0.3	17	0.3
That is (to say)	201	2.4	99	1.7
Total	2,480	29.2	1,166	20.2

This type of gloss is meant to clarify and disambiguate claims by either offering different wordings (16) or linking abstract meanings to possible realizations and realistic scenarios (17, 18), as can be seen in the texts below:

• (16) A person is a slave, he said, when the law fails to declare him a rights-holder, and when its remedial and sanctioning powers are unavailable to him, for then he is without ‘the protection of the law’. A slave, in other words, is just a person who may be beaten, confined, and otherwise abused without the violation of a legal directive, liability, or punishment. (YLJ 115, e.a.)

• (17) Litigation can dislodge information from an otherwise inaccessible private party, for example, through discovery obligations in civil procedure. (HILJ 46, e.a.)

• (18) In contrast, harmonisation takes place on two different levels of governance, that is, the European and the national level. (ELJ 12, e.a.)

As the texts above indicate, these glosses reveal a marked concern for detail, precision, and clarity, all factors which are meant to be beneficial for comprehension. By observing the data in the table, we notice that PLC writers make a much more consistent use of such resources – especially those which are very much typical of academic contexts (like e.g.), on the one hand, or in everyday conversation (like for example, that is), on the other. A closer observation of the frequencies also reveal that OC texts make a more differentiated use of code glosses, alternating among the different resources: as a matter of fact, the range of variation between the most and the least used is 6.9 (e.g.) to 0.3 (in sum, specifically), whereas in PLC the gap is much wider, namely, 13.0 (e.g) to 0.3 (in sum, specifically). Another interesting piece of evidence is that whereas PLC seem to privilege glosses with an exemplifying function (e.g., for examples), OC writers appear to be more prone in using specification glosses (i.e., namely, specifically) and rewording glosses (in other words), that is, privileging clarifications through linguistic reformulation rather than examples and instantiations. This evidences the preference conferred to ‘words’ and their disambiguating function in OC LRAs, which is in contrast with the emphasis attributed to ‘facts’ – more precisely to the linguistic representation of practical or possible cases – used for exemplification purposes in PLC legal writing.

• D. Final Remarks

  The evidence yielded by our word counts reveals two drastically different ways of organizing discourse. On the one hand, PLC texts resort consistently and diffusely to interpersonal resources, both quantitatively (i.e. in terms of frequency) and qualitatively (i.e. according to the variety and range of their use). On the other, cultures which are minimally or marginally affected by PL ideologies – owing to the fact that, for the most part, they stem from Civil Law system, where the establishing of the right and its performative function is granted by written codification rather than its discussion, argumentation, and interpretation – seem to circumscribe the use of inclusive, epistemic, and negotiating strategies, thus privileging the semantic and textual representation of ideational material rather than its communicative effectiveness and impact.
  The tendency of PLC to exploit interpersonal resources much more blatantly and conspicuously than OC does not comes as a surprise. Indeed it fully confirms our expectations, since this way of using the language reflects the main PL concerns about comprehensibility, transparency, and cooperation. However, what is interesting at this point is to observe how this interactive concern is realized in legal studies. As a matter of fact, the interpersonal tenor in our corpus, and specifically in the PLC subcorpus, manifests itself as both writer-responsibility (in cognitive and rhetorical terms³⁸xHyland 2005.) and as considerateness towards the reader’s competence, expectations, and needs. Writer-responsibility is the authorial tendency to lexicalize (through various degrees of explicitness) their persona as being responsible for devising meaning and designing its textualization, through the use of both the first person singular and the exclusive plural. By employing such pronouns, scholars represent themselves not just as authors, but, and most strategically, as reference point and key element for the comprehension of the text: they are rhetorically portrayed as the main source of information and, for this very reason, they acquire the cognitive function of authoritative guides and competent facilitators for the understanding of the meaning. This is particularly significant since, as we have already mentioned at the beginning of this article, traditional views instead considered the first person in legal contexts – both normative and forensic – not just unnecessary but downright threatening, in that believed to presuppose possible lack of objectivity and to imply a form of self-attributed authority and an arrogant attitude. Although it may sound paradoxical, the first person personalization in PLC LRAs has instead a modesty and mitigating function, in that it seems to be primarily used to ‘represent the obvious’, that is to say, to link thoughts, research, and claims to their ‘natural’ source, that is, the writer as thinker, researcher, or arguer.³⁹xFløttum et al. 2006.
  Authors in PLC tend to present meaning as their own, as originating from their own research activity or way of interpreting the world, and at the same time, they rhetorically introduce themselves as the main responsible for the correct comprehension of the content. Through personalization, by framing the validity of what they claim within a personal perspective, they manage to minimize assertiveness and imposition and emphasize cooperation and inclusion.⁴⁰xK. Hyland, ‘Bringing in the Reader. Addressee Features in Academic Articles’, Written Communication, No. 18/4, 2001. Writer-responsibility is also manifested as considerateness for the recipient’s view and takes the form of a reader-friendly style. This interactive attitude is realized through a consistent resorting to metadiscursive features such as mitigation and code glosses, both of which stemming from the same argumentative principle, that is, the need to provide the reader not just with the necessary information but also with either interpretive alternatives or instantiations and clarification, both strategies aimed at stimulating an active, participative, and negotiating response on the part of the audience. This is particularly true in the case of hedges, through which writers represent the potentiality rather than certainty of a given meaning, and its acceptability is rhetorically negotiated by the participants. Along the same line, code glosses are instrumental in this interactive interpretive process in that they are meant to (re-)lexicalize all informative elements needed by readers to comprehend the author’s point and, eventually, to accept it as true. This way of presenting content conceals the writers’ authoritative voice under an appealing and eliciting style, intended to minimize assertive or imposing forms which might elicit possible aversive reaction and rejection.
  In conclusion, from a contrastive perspective we have seen that PLC academic discourse diverges from OC scholarly language in that it significantly exploits personalization (especially the first person singular, but also inclusive plural formulations and second person expressions, cf. Table 1) and privileges dubitative epistemic markers over boosters (95.5 vs. 55.2) and exemplification code glosses over reformulation ones (cf. 3.3.). On the other hand, more traditional legal academic discourse, like the one found in OC, limits personalization (comparatively preferring a vague reference to an undifferentiated we rather than the first person singular or the second person pronoun, thus tending to conceal the participants and their role in the text), balances the use of hedges and boosters (58.6 vs. 47.8, thus resorting to certainty markers more significantly than PLC writers), and privileges glossing through rephrasing than instantiating.
  In the light of the above, it is possible to claim that PLC writers are more concerned with persuasion (appealing to the audience in order to negotiate meanings with them), whereas OC texts are much more focussed on the authoritative wording of the referent: in the first case, the content is dealt with as something to be made accessible, effective, and acceptable for the recipient; in the second case, meaning is made precise and exhaustive so that, eventually, readers can work their way through its (possible) complexity. In extreme synthesis, PLC scholars focus on the reader; OC authors on words. This difference certainly owes to the discrepancies between the legal systems at the basis of PLC and of cultures more resistant to stylistic and register changes, respectively the Common Law system (based on the interpretation of the precedent) and the Civil Law system (based on the codification of the norm) and all the related activities. However, while differences are obvious and reflect the different discursive practices when used in legislative and forensic contexts, it is less obvious why they are so marked also in a genre like the RA, which is not so prototypical of the legal domain. One possible reason for this discursive variation in legal academic writing resides in the fact that “the practices of the academy are not sui generis but are heavily influenced and constrained by personal and professional histories and by professional and occupational requirements”,⁴¹xC. Candlin, V. Bhatia & C. Jensen, ‘Must the Worlds Collide? Professional and Academic Discourses in the Study and Practice of Law’, in P. Cortese & P. Riley (Eds.), Domain-Specific English: Textual Issues: From Communities to Classrooms, Bern, Peter Lang 2002, p. 102. this being especially the case “in Law, where strongly professionalised and organised bodies exist outside the academy, exercising rights of approval and certification”,⁴²xIbid. on community members, since language is the most prominent (if superficial) and accessible aspect that can be measured in terms of appropriateness and acceptability with respect to the required degree of academic literacy.
  Therefore, actors operating within such an ‘institutionalized framing of activities’⁴³xA. Cicourel, ‘The Interpenetration of Communicative Contexts: Examples from Medical Encounters’, in C. Goodwin & A. Duranti (Eds.), Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press 1992. are likely to reproduce existing linguistic practices in order to gain or corroborate community acceptance. By sticking to conventionalized model and replicating recognizable paradigms, writers define their identity, authority, and disciplinary relevance within the community of reference, that is, they are appreciated as expert in the domain owing to the fact that they know how to conceptualize meanings and how to discursively deal with them. Thus, in contexts where PL has become the leading linguistic ideology, PL use is a claim to the relevant PL literacy community and, conversely, the resistance to this is a statement relating the author to more traditional views on language particularly prized in Civil Law cultures.

Noten

• 1 M. Adler, ‘The Plain Language Movement’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012.

• 2 L. Solan & P. Tiersma, ‘Introduction’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 4.

• 3 G. Cornu, Vocabulaire Juridique, Paris, PUF 2007; H. Mattila, ‘Legal Vocabulary’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012.

• 4 F. Bennion, ‘Confusion over Plain Language Law’, The Commonwealth Lawyer, Vol. 16, No. 2, 2007.

• 5 R. Eagleson, ‘Untitled Article’, Sydney Morning Herald, 20 January 1985, p. 8 (reprinted in Plain English and the Law. Report No. 9 by the Victorian Law Reform Commission, 1987).

• 6 N. James, ‘Defining the Profession: Placing Plain Language in the Field of Communication’, Clarity, No. 61, 2009, p. 35.

• 7 M. Cutts, The Plain English Guide, Oxford, Oxford University Press 1996, p. 3.

• 8 M. Gotti, Specialized Discourse. Linguistic Features and Changing Conventions, Bern, Peter Lang 2003, pp. 35-36; P. Tiersma & L. Solan (Eds), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012.

• 9 D. Mellinkoff, The Language of the Law, Boston, Little, Brown & Co 1963; C. Williams, Tradition and Change in Legal English. Verbal Constructions in Prescriptive Texts, Bern, Peter Lang 2007.

• 10 Ibid., pp. 36-38.

• 11 Gotti 2003; M. Gotti, Maurizio & P. Anesa, ‘Professional Identities in Italian Arbitral Awards: The Spread of Lawyers’ Language’, in V. Bhatia & P. Evangelisti Allori (Eds.), Discourse and Identity in the Professions, Bern, Peter Lang 2011, p. 100.

• 12 Gotti et al., 2011.

• 13 Gotti 2003, pp. 35-36.

• 14 Ibid., p. 141.

• 15 A. Trosborg, Rhetorical Strategies in Legal Language: Discourse Analysis of Statutes and Contracts, Tübingen, Narr 1997, p. 25.

• 16 L. Solan, ‘Linguistic Issues in Statutory Interpretation’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 88.

• 17 Even the documented phenomenon of semantic vagueness, which appears to be at odds with this focus on clarity and comprehensibility, very often stems from the concern for semantic transparency, exhaustiveness, all-inclusiveness, and, from the standpoint of perlocution, in view of the wider applicability of a given norm to an extended range of unspecified or underspecified contexts (V. Bhatia, J. Engberg, M. Gotti & D. Heller (Eds.), Vagueness in Normative Texts, Bern, Peter Lang 2005; R. Poscher, ‘Ambiguity and Vagueness in Legal Interpretation’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012).

• 18 Adler 2012, p. 77.

• 19 M. Adler, ‘British Lawyers’ Attitude to Plain English’, Clarity, No. 28, 1993; M. Adler, Clarity for Lawyers, London, The Law Society 2007; J. Kimble, Lifting the Fog of Legalese, Durham, Carolina Academy Press 2005.

• 20 Trosborg 1997, pp. 20-22; M. Gotti, ‘Text and Genre’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press, 2012a.

• 21 Trosborg 1997; P. Tiersma, Legal Language, Chicago, University of Chicago Press 1999; S. Gozdz-Roszkowski, Patterns of Linguistic Variation in American Legal English. A Corpus-Based Study, Bern, Peter Lang 2011; M. Sala, ‘Legal Expertise as a Cultural Identity Trait’, in V. Bhatia & P. Evangelisti Allori (Eds.), Discourse and Identity in the Professions, Bern, Peter Lang 2011; Tiersma & Solan 2012.

• 22 Adler 2012.

• 23 Gotti 2012a, p. 58.

• 24 P. Tiersma, ‘A History of the Languages of Law’, in P. Tiersma & L. Solan (Eds.), The Oxford Handbook of Language and Law, Oxford, Oxford University Press 2012, p. 17.

• 25 Tiersma 2012, pp. 17-22; Gotti 2012a, pp. 58-59.

• 26 V. Bhatia, Analysing Genre: Language Use in Professional Settings, New York, Longman 1993, p. 137.

• 27 J. Gibbons, Forensic Linguistics, Malden/Oxford, Blackwell 2003, p. 5.

• 28 Adler 2012.

• 29 M. Gotti, ‘Creating a Corpus for the Analysis of Identity Traits in English Specialised Discourse’, The European English Messenger, No. 15/2, 2006, pp. 44-47; M. Gotti (Ed.), Academic Identity Traits, Bern, Peter Lang 2012b.

• 30 The operation of dividing authors on the basis of their native language has been made possible owing to the biographical information provided in the footnote which is usually linked to the author’s name and which contains the authors’ affiliation, their biographical, and bibliographical information, the presence in their curriculum of publications written in non-Anglophone languages, etc.

• 31 M. Scott, Oxford WordSmith Tool 5.0, Oxford, Oxford University Press 2007.

• 32 K. Fløttum, T. Kinn & T. Dahl, ‘“We Now Report on…” versus “Let Us Now See How…”: Author Roles and Interaction with Readers in Research Articles’, in K. Hyland & M. Bondi (Eds.), Academic Discourse across Disciplines, Bern, Peter Lang 2006.

• 33 K. Hyland, Metadiscourse, Continuum, London 2005.

• 34 Fløttum et al. 2006.

• 35 Hland 2005.

• 36 J. Coates, ‘Epistemic Modality and Spoken Discourse’, Transactions of the Philological Society, Vol. 85, No. 1, 1987; K. Hyland, ‘Persuasion in Academic Articles’, Perspectives, No. 11, 1999.

• 37 Hyland 2005.

• 38 Hyland 2005.

• 39 Fløttum et al. 2006.

• 40 K. Hyland, ‘Bringing in the Reader. Addressee Features in Academic Articles’, Written Communication, No. 18/4, 2001.

• 41 C. Candlin, V. Bhatia & C. Jensen, ‘Must the Worlds Collide? Professional and Academic Discourses in the Study and Practice of Law’, in P. Cortese & P. Riley (Eds.), Domain-Specific English: Textual Issues: From Communities to Classrooms, Bern, Peter Lang 2002, p. 102.

• 42 Ibid.

• 43 A. Cicourel, ‘The Interpenetration of Communicative Contexts: Examples from Medical Encounters’, in C. Goodwin & A. Duranti (Eds.), Rethinking Context: Language as an Interactive Phenomenon, Cambridge, Cambridge University Press 1992.