Not infrequently, police officers will use curse - that is, use extremely vulgar or obscene language - in the course of their encounters with criminal suspects. This practice has been defended as not only unobjectionable but as positively necessary to law enforcement. In this paper, I will first analyze the purposes of the use of such language generally and then go on to address its specific use by police officers in field interactions with suspects. I will discuss both the illocutionary and perlocutionary aspects of such language in this context.

Swearing - particularly by those in authority is a powerful signal of role-transgression, and as such, signals the possibility of other forms of role-transgression as well, including the possible use of physical force against the suspect. Finally, I will argue that legal doctrine must be cognizant of the impact of police swearing in determining such legal issues as whether a suspect validly consented to a search or validly waived her right not to answer police questions. I will use US caselaw to suggest that legal doctrine gives inadequate attention to this problem.

The society of knowledge is the label that has been given to the first years of the third millennium. This lecture claims that in this society, the language of the Spanish Civil Administration has changed in its connection with citizens. The first thing that citizens in this society of knowledge know is their rights, and they are prepared to defend them in court against the unfair or arbitrary decisions of the Administración. An analysis of the language, especially the terms used by the Spanish Civil Administration, is carried out in this paper in its relation with law and power. It also offers a comparison of the procedures followed in Spain and in the United States to take the Civil Administration to Court. The starting statement of the paper, which was "The king is not answerable to his subjects", has noticeably changed to La Administración sí debe dar cuentas de sus decisiones a los ciudadanos in the closing statement.

The presentation will attempt a brief overview of the peculiarities of American trademark litigation and the linguist's role therein and will then go on to engage three issues:

1. The increased importance of Spanish in the United States and its subsequent influence on trademark litigation;
2. The availability of internet search engines as research tools and the consequent fundamental alteration in the nature of expert reports;
3. Attention to ethical issues in the light of cross-national differences in standards, the availability of the internet tools discussed in (2), and recognition that some linguistic subspecialties are more relevant than others to expert opining concerning trademark litigation.

This paper aims at presenting the current status of our bilingual corpus-based research of binomial expressions in legal agreements.

Binomial expressions are formed by two words belonging to the same grammatical category and joined by a conjunction (Tagnin). Some examples are: terms and conditions, any and all, executed and delivered, due and payable, successors and assigns, representations and warranties.

Binomials are an integral part of English legal language (they are present in statutory provisions, opinions, boilerplate, contracts and agreements, etc.), and as such have deserved a considerable amount of study (Mellinkoff, Bhatia, Vystrcilova, Kwok, Rossini, among others). On the other hand, the same does not apply to Brazilian Portuguese legal language in which binomial expressions are much more rare. Therefore, when dealing with binomial expressions, studies show that Brazilian translators tend to translate all the elements of a binomial literary (Corrêa Pinto) and, consequently, in many cases fail to attain the 'intended legal effect' (Sarcevic´) of the communicative event.

For this reason, to study binomials in English legal language and determine their translation equivalents in Brazilian Portuguese, we are compiling and exploring a bilingual comparable corpus made up of authentic 'agreements' and 'contratos', totaling, approximately, 600 000 words. Exploring such a corpus greatly depended on the tools and methods of Corpus Linguistics. The software used was Scott's WordSmith Tools.

Finally, the choice of text type is due to the fact that 'agreements' and 'contratos' are among the most difficult documents to translate (Mayoral Asensio), they also seem to have an incredibly high concentration of binomial expressions (Corrêa Pinto), and are among the most translated documents in the Brazilian market. In addition, certain provisions in translated agreements have been the cause of long litigation in the Brazilian Legal System.

The July 22, 2005 resolution of the Spanish Government Cabinet approved the new Guidelines for Bill Drafting, thus modifying those Guidelines passed in 1991. In this paper we analyse these new Guidelines, the majority of which attach great importance to good practice and use of legal language. The main aim of these new Guidelines is "to attain a closer relationship with the constitutional principle of the Rule of Law, by means of increasing the technical and linguistic quality of norms". The idea is to seek linguistic correctness and norm comprehensibility at the same time.

These Guidelines establish the criteria with which the organization and display of the normative texts will have to comply. The basic divisions of the legal norms are established and very precise instructions as to the content of each of those are included. Uniform criteria as to the spelling used in each case are also proposed.

In particular, we will address section IV of these Guidelines, on "General linguistic criteria", which among other things recommends using a clear and precise language and respecting the word order of sentence elements, and appeals to linguistic correctness.

The aim of this communication is to present an on-line computer application that can assist the forensic phoneticians in several areas of their profession, which has been conceived together by ForensicLab and the Laboratory of Language Technologies, both from IULA. This application, called SERP (Electronic System of Speakers Recognition), presents several couples of samples of voice that can belong to a same speaker or to different speakers. The user, by means of a scale of opinion, has to identify the samples of voice. The application registers the result of the identifications and the time that the user has been late to complete the test. Moreover, the SERP allows to manipulate the speed of reproduction, the direction and the intensity of the voice, among other parameters that can influence on the perception.

The applications of this tool in the forensic phonetics field are, among other:

• Evaluation of the ability of the users in the auditory identification of speakers.
• Checking of the progression of the phoneticians regarding their ability in the auditory identification of speakers.
• Construction of perception tests with naïve users to measure the level of confusion among determinate voices.
• Evaluation of the effect of the level and type of background noise in the identification.
• Evaluation of the effect of the speed of reproduction of the voices in their identification.
• Calculation of the correlation between the time of decision and the degree of success in the identification.

The communication will include the presentation of the design of an experiment of validation made with an extensive sample of potential users.

Writing challenges abound for pre-law and law students at our U.S. Colleges and Universities. Among the many types of written expression, the legal register is fraught with obfuscation and dryness. How much of this is syntactic? This paper will raise two questions. First, is there continuity over time in good legal syntax? Is there syntactic similarity between good opinions and good briefs?

In this paper, Deakins and Cochran will explore the syntax of the opinions of three US Supreme Court Justices from each of three centuries-- the 19th, the 20th, and the 21st. What characterizes the syntax of those justices considered by contemporary lawyers to be skilled writers? Their syntax will then be compared to early and final drafts of three legal briefs, considered well-written in their final drafts, to answer the question of whether revisions of the briefs move toward the syntax of the opinions.

In July last year an internationally renowned British pediatrician was severely censured by his professional body and then struck off the medical register following damning criticism of expert evidence he had given in several cases of sudden infant death syndrome. He continued to assert the correctness of his opinions, but did concede that he may have been wrong to express them in terms of exaggerated and over-persuasive probabilities.

In this paper I will use examples from real court cases to illustrate the problems facing the forensic linguist when s/he is asked to rate the strength of the evidence s/he is giving. I will relate what I have to say to the two levels of proof in British Courts - 'on the balance of probabilities 'and 'beyond reasonable doubt'.

KISTE as a German acronym stands for forensic database "texts". KISTE (in the latest version completed in 2003 but nevertheless in a ongoing process of improvement) is intended to contribute to linguistic work on incriminated texts. Incriminated texts are those that appear in criminal contexts or that constitute a criminal offence by themselves, e. g. letters of extortion or threat. It is the aim of text analysis supported by KISTE to find rela-tions between cases and to apply statistical methods to linguistic features.

This paper will present

1. some aspects of forensic linguistics at the BKA and I will focus on what we call text analysis as well as on our text corpus of incriminated texts KISTE
2. I will show some details of KISTE, essentially the grammatical processes of deflection and word determination on the one hand, as well as research strategies that help us to evaluate the statistical relevance of a certain language feature on the basis of the distribution of linguistic findings on the other hand. I will discuss authentic examples of in-criminated texts in order to demonstrate how these texts are processed with the help of KISTE and to show further functions of KISTE available to the linguist.

We wish to provide a linguistic-pragmatic solution to the seeming contradiction between 'silence as consent' and 'the right of silence'. Our contribution sets out from the following:

1. Roman Jakobson's model of the functions of language. In his paper of 1960 Jakobson shows that making salient one of the six components partaking in communication yields a unique language function. Thus, emphasizing the world - the referent - as the outstanding component of discourse results in the referential function, locating the addresser as the center fosters the emotive function and locating the addressee results in the connotative function (speech act s being the typical case thereof). The phatic function takes place when the contact the actual means of communication is at the center. When emphasizing the organization of the message the poetic function is at work. Finally, the metalanguage function is the one in which the discourse is concerned with the code itself.
2. Clear differentiation between pause: an insignificant lack-of-speech, a pause in the flow of speech and silence: a meaningful means of communication along side speech.

Typically - bona fide - use of silence as consent assumes that the accused will make his at most to appear innocent. Thus, keeping silent is taken as a speech act of consent. Discoursewise, due to its parsimony: not repeating each and every fact or claim but acting as a pro to all that was said before, silence may act as the preferred expositive means for expressing such consent (connotative function). The right of silence as a metalinguistic function must be declared right before any interaction between the suspect and authorities since it is the only possible immunity to secure that the suspect's idiolect: style, accent, vocabulary, fluency etc will not directly or indirectly result in self-incrimination. This is why explicit statement of the Miranda warning but not "Qui tacet consentire videtor" is imposed in many judicial systems.

We shall support our interpretation of the two functions of silence with cases and rulings brought to court, including the matter of the eligibility to the right of silence regarding public representatives.

This paper reports the results of a study on the prosody of two groups of Western Canadian suspects' speech as it occurs during the course of investigative interviews with police. The suspects under examination are first-time offenders and repeat offenders. Based on interviews conducted by the Vancouver Police Department, I offer an examination of the prosodic characteristics of pitch range, average pitch, speech rate and pause durations across various categories of responses, comparing these two groups.

Walton (2003) characterizes the police interview as a "subspecies of information seeking dialogue" (p. 1775). Whereas other interview types, and indeed many other conversation genres, are information seeking and sharing exchanges, the police interview is non-collaborative. The investigator alone drives the turn-taking with questions and assertions (prompts) which elicit continuous suspect response. Their responses are categorized according to the type of information they contain, and include the following:

• Confirm information - suspect confirms the information contained in the investigator's prompt (Hill 2003).
• Reject information - suspect rejects the information contained in the investigator's prompt (Hill 2003).
• Offer information - suspect offers information related to the investigator's prompt.
• Irrelevant information - suspect offers information that is not relevant to the investigator's prompt.
• Concession - suspect concedes to committing an act, although not the same act alleged by the investigator.
• Confession - suspect confesses to a crime or crimes.

It will be shown that the various response types pattern consistently with respect to prosodic behavior within groups, and that the two groups show significant differences. On the basis of my findings, profiles composed of prosodic characteristics can be created for each response type, for each of the two groups. It will also be shown how these prosodic profiles align with previous research on the prosody-emotion interface.

• Hill, Martin. (2003). Identifying the source of critic details in confessions. Forensic
  Linguistics 10 (1), 23-61.
• Walton, Douglas. (2003). The interrogation as a type of dialogue. Journal of Pragmatics
  35, 1771-1802.

Voice imitation and disguise are possible threats to the performance of a speaker recognition system and to the accuracy of earwitness descriptions. One common disguise is the modification of the own dialect or accent. In this paper, this kind of disguising is explored, using recordings from a well-known actor with considerable experience of dialect and accent imitation.

In order to see how successful his dialect imitations are and how the process of speaker discrimination is influenced by accent disguise, two sets human perception tests were constructed. One set focused on American and British English dialects, and one set on American and London English accents and Spanish-accented English. Each set consisted of three parts: a same-different speaker test, a same-different accent test, and a select the accent from a closed-set of options test. The results show that Johnny Depp is successful without his visual props and demonstrate a high correlation between the quality of the accent imitations and the failure of the human listeners to recognize that the voices come from the same speaker. The third parts of the experimental sets suggest the importance of familiarity with the accent that feeds into parts one and two. Spanish listeners, for example, are less accepting of the Spanish-accented English than non-Spanish speakers. Furthermore, the results afford a detailed understanding of voice behaviour in this specific kind of imitation, locating the importance of prosody and dialectal phonetic features to a non-initially expected level.

Finally, the same speech segments used in the perception test were used in an automatic speaker recognition experiment in order to compare the results and to check the robustness of the system in front of the voice changes. The results showed, once again, a low correlation between human and automatic speaker recognition.

A public corpora is defined as a body of data which can serve as the basis for linguistic analysis and description and which is available to linguists in general either as an identifiable whole or from easily accessible materials.

Databases/corpora in law enforcement are not new and several have been in existence for a decade or more. For example, the FBI's "Anonymous Letter File" is located at the FBI Laboratory in Quantico, Virginia, (U.S.). Its corpus contains bank robbery demand notes, high profile kidnapping and extortion notes/letters. In this database, only keywords are entered, and any subsequent searches will result in only those enteres lexical feature being recognized.

Other corpora include the British National Corpus and the American National Corpus.

The FBI's Behavioral Analysis Unit-1's responsibilities include behaviorally oriented investigative assistance in all matters relating to counterterrorrism and threat assessment and the analysis of textual material. In conjuction with the BAU-1's responsibility in these areas, the Unit designed and recently implemented of the Communicated Threat Assessment Database (CTAD). The CTAD is a computerized query-oriented database/software program designed to be the primary repository for all communicated threats and other criminally oriented communications within the FBI. Its purpose is to assist the BAU-1 Agents and analysts in the categorizing, analyzing and assessing of the various communications. Text from the entire document, to include transmittal envelopes (if applicable) are entered, not just keywords.

The CTAD is presently the only database in the United States which catelogues all varieties of threatening communications, searches them for commonalities, and provides numerous analytical and research tools.

It was designed from a forensic linguistic and behavioral perspective, and includes numerous components related to each. The submitted threatening communications are placed into 22 separate categories, and issues relating to sex, age, race, nativeness, register shifting, etc., are captured both in the "Linguistic Profile" section, and "Case Data Confirmed" section. There are presently 1,500+ threatening communications contained within the CTAD. It is expected to double in size in the next year.

The scope of this paper is to facilitate authorship identification by applying corpus based techniques on electronic textual resources. Our domain of application is proclamations by terrorist groups in Greek. The use of computational techniques can provide efficient, systematic and precise information which is not possible by human judgment alone. Our application study is the testimonies of the "17th November" ("17N") Greek terrorist group. In the summer of 2002, members of this group were arrested and the case is currently in court. We focus on the textual processing of their testimonies, trying to establish authorship of the proclamations. Proclamations were sent to Greek newspapers, immediately following a 17N terrorist attack. By 2002, 84 proclamations have been gathered. However, none of those accused to be members of the 17N admitted authorship of the proclamations. Based on a set of linguistic and extra-linguistic features, we match information from the testimonies with the proclamations of the 17N.

We have applied corpus linguistics analysis on each of the 17N members' testimonies, to acquire linguistic and extra-linguistic information for characterizing a person's use of language (idiolect). Features regarding linguistic information include frequency word lists, frequency expression lists, number of different words, etc. Attention is given to rare co-occurrences as well since they can depict idiolectic usage. Regarding extra-linguistic information, we have examined content words and expressions, i.e. those that give information on the content of the text. We have produced linguistic and extra-linguistic profiles for each member based on the weighted features extracted from each testimony. A comparison between the testimonies and the proclamations, in terms of their feature set, provides information on the degree of language similarity between them. Future work will include automatic multilingual processing of terrorist group proclamations in Arabic and English.

Following a police traffic stop, a deaf couple brought suit against a police department in an American city, charging several officers with unfair treatment. The plaintiffs claimed that the officers, rather than providing a deaf interpreter, insisted on using handwritten notes to explain the details of the situation to the couple and to ask them questions, thereby making it impossible for the parties to communicate effectively. I was asked by an attorney for the defendant to examine the notes, written by both the police officers and the couple, to determine whether or not effective communication had taken place.

My analysis involved examining 1) the clarity of the police officers written communication and 2) evidence for comprehension and communicative competence on the part of the deaf couple, as well as markers of their competence as English users--including lexis, morphology, syntax, and semantics. My presentation will show the materials and methods used, explain the determination I made, and give the outcome of the case. If time allows, I will also open a discussion of some of the ethical questions involved in advocating for government agencies in their legal conflicts with handicapped populations.

Oracles, puppets and social engineers: some thoughts on the language of judges The controversy about legal language and its target public literally dates back centuries and it is fair to predict that is shall drag on for the foreseeable future. While some reasonable measures to reform this language are proposed, other solutions often advanced in this area do not seem feasible and can be merely considered wishful thinking. This is probably due to the fact that they do not distinguish clearly between different texts and documents and their target readers.

Its is evident that to speak about legal language implies understanding law as a communicative process (Nelken, 1996). At the same time, judgments, the prototypic judicial text, reflect better than any other documents the concept of legal culture and all conditioning factors implied therein. By legal culture we understand, among other things, the use of a certain language and a certain communicative style that can modulate the intelligibility of the issued text. Other factors, such as social origin, mentality and awareness of the special role played in society, have a bearing on the question. The title of this paper refers to the different roles assumed by, or attributed to, judges in the course of history.

A comparative analysis of the communicative style of judgments issued by different courts will bring some insight on the weight of legal cultures in texts of different origin. At the same time, the decisive role of international courts is increasingly showing the possibilities and limitations of a common legal (and therefore linguistic) ground that is being created sep by sep. The area is of particular relevance for all actors involved in the process of analysis, construction or translation of this type of legal texts.

Chilean courtroom procedure includes obtaining statements from the witnesses, both for and against the defendant. The data for this study are tape recordings of Chilean court testimony by witnesses, and the official written statements of this testimony. They provide different accounts of the same events, not elicited element by element as in much Common Law testimony, but in a largely open narrative style, which encourages the expression of the witness's view of events.

This paper compares and contrasts the competing views of the same event as presented by different witnesses, and also examines the way in which the 'relator' who takes the testimony may modify the narratives in the written statement. It examines both the linguistic resources and the discourse elements used to construct the competing versions.

In asylum application interviews in Belgium, applicants are asked to relate their reasons for leaving their country of origin. The goals of these interviews are to assess the applicant's credibility and to assess whether the reasons correspond to those specified in the Geneva Convention. Due to the existence of fake accounts that are handed on from one applicant to another, the asylum agencies appreciate detailed and personalized accounts. During the interviews, the alleged reasons for leaving the country of origin are recorded in a written text by the interviewers. This text is transmitted to the decision-makers who base their decisions solely on the information recorded in it.

The rendition of a non-institutional oral first generation genre (the narrative of escape) to an institutional written second generation genre (a factual account of the claimed events) entails transformations and omissions from the original text. This paper analyses seven admissibility interviews conducted in Brussels and discusses the relations between the said transformations and the inadmissibility reasons given by the asylum agencies.

The analysis yields the following results. First, there is evidence that the decision-makers do not base their decisions solely on the written account but also on interviewers' comments about the applicant's performance during the interview. Indeed, some inadmissibility reasons cannot be seen to emerge from the information recorded in the written report. Secondly, the categorization of the applicant's experience in the interviewer's terms by means of the introduction of legal terminology, the replacement of terms from African varieties of French by Standard French, the use of distancing devices such as passivization, the formalization of register and the increased level of abstraction through nominalizations all contribute to a standardization of the account. Finally, some transformations clash with the criteria used by forensic psychology to detect deception through the analysis of verbal content with techniques such as Criteria based content analysis.

When language analysis is used in asylum cases to provid

e evidence regarding an applicant's claims of his national or ethnic origin, his utterances recorded for this purpose can be analysed on two levels, which are quite different but nonetheless closely related. Firstly and ideally, the phonological, morphological, syntactic and lexical features are identified in their ideolectal configuration as belonging to one language or one or more dialects of a language. Secondly, the semantic or conceptual content of the applicant's utterances is evaluated to ascertain whether or not it reflects what is considered "true" and "sufficient" knowledge of his alleged country of origin or some special geographical and/or social settings therein.

While it is becoming more and more generally accepted that evidence concerning phonological or grammatical features ought to be established and presented within a scientifically acceptable linguistic framework, the further evaluation of the applicant's knowledge relies exclusively on the intuition of the analyst and the relevant practical experience he is assumed to have.

Any content analysis, however, operates on the level of the applicant's linguistic representations of his concepts. Cognitive Linguistics has developed models of conceptualisation as well as experimental procedures and elicitation techniques to uncover the structure and content of such concepts.

Drawing on actual cases and research conducted independently in the African field, this paper aims to explore the feasibility of applying models and methods developed in Cognitive Linguistics to the content analysis of an applicant's utterances in asylum cases. Special attention is given to their implications in the planning of interviews, the use of pictures as stimuli and cognitive semantics.

Analyses of attorneys' arguments to juries in adversarial courtrooms have long recognized that the outcome of a criminal case depends not only on the strength of the evidence, but on the persuasiveness of competing narratives about the events in question and about the defendant's specific role in those events, in deciding guilt or innocence.

In capital trials, sentencing juries must, in addition, weigh narratives concerning the defendant's character and background, in deciding a life or death verdict. Post-trial interviews with capital jurors have shown show that in the deliberation room, these competing stories may take on a life of their own which drives decision making. (Fleury-Steiner, B. 2004. Jurors' Stories of Death.) The wording of the judge's instructions and jurors' incomprehension of these guidelines may also contribute to the influence of the larger society's biases in their sentencing decision, as jurors struggle for unanimity.

This paper compares the language and themes of the competing story lines in a small number of California capital cases as they were presented in the courtroom, the news media, and the reviewing court's opinion, examining the manner in which the story of the case (and the defendant) is shaped and directed in its passage through the trial and postconviction period.

In the first part of the talk the range and variety of work carried out by forensic speech scientists will be discussed. This includes speaker identification, speaker profiling and disputed utterance analysis. Each of these areas will be exemplified with real case material and the analysis techniques employed will be presented. Some more unusual and interesting cases will also be discussed.

The second part of the talk will examine how the results and conclusions of speaker identification examinations are presented. Until recently nearly all scientists in this field presented their conclusions in a logically-flawed way which is known as the 'prosecutor's fallacy'. This will be explained and exemplified. A new method of presenting speaker identification results, which avoids the logical flaw, will be examined together with a modified version tailored to forensic authorship analysis.

This paper involves an analysis of the role of police interview discourse throughout the English judicial process. The form and function of interview data is traced through the system from the original interview itself through to its use as evidence at trial. This process is illustrated by using the trial of Dr. Harold Shipman as a case study. I will also draw on my own corpus of recent English police interview data.

In terms of its function, interview data is used for various purposes by a variety of different audiences: the police, Crown Prosecution Service, trial lawyers, judge and jury. Using a multi-method approach combining CA, CDA and pragmatics, I examine the effect of the different future audiences and purposes on the dynamics of the interaction in terms of the 'audience awareness' and goal orientation of the interview participants. I also consider the consequences at trial of the participants failing to take such factors into consideration.

In terms of its form, I assess the transformation of interview data through the judicial process. The original interaction is audio-recorded, and the tape is converted to a written transcript by police clerks. Features such as discourse markers, interruption and overlap are routinely omitted. The data is then introduced as evidence in court by the transcript being read out as a 'script' with a police witness acting as the interviewer and, more disturbingly, with prosecuting counsel taking the defendant's turns. I assess the effect this has on the data and hence on the legitimacy of the subsequent uses to which it is put.

This paper aims to illustrate the ways in which linguistics can contribute to improving and enhancing the use of interview data throughout the judicial process. The conclusions reached are based partly on the author's own experiences as a practising barrister.

Over the last decade linguistic studies have focused on manual and computationally assisted detection of plagiarism with particular reference to lexical similarities in authorship in student writing (Coulthard, 2004; Johnson, 1997; Woolls 2003 & 2006). This paper addresses some of the pedagogic and practical issues that arise when we examine what students actually do with the material they share between each other or take from electronic sources. The authors demonstrate an action research approach that is methodologically responsive to pedagogical and disciplinary needs whilst being developmental in the design of software that is of practical effect in academic and commercial contexts.

Data includes two case studies: one includes examples of three students' writing that illustrates instances of acknowledged failure to cite web sources in an essay question which discusses Achebe's novel: Things Fall Apart. We examine the different strategies used when taking material from web sources: cut-and-paste, re-arrangement, insertion, omission and 'patchwriting' (Howard, 1999; Pecorari, 2002). We then look at the inter-relationships between texts, sources and writers, citation practices and challenge presented by the material for detection and academic writing support. A second case considers the rewriting of extended unacknowledged passages from a published book, examining the problematic adaptive strategies used by the student.

The implications of these case studies are considered, suggesting that computational and manual analysis of cases of plagiarism has preventative potential. By examining what happens when some students fail to write it is possible to equip future leaner-writers for the challenging task of successful academic writing.

Silence in speech is a critical part of expression, and pause realization, and the variation therein, can carry differences in both speaker-generated (subconsciously or consciously) and hearer-perceived meaning. While pause needs to receive further focus in linguistics generally, it is especially crucial that investigations under the category "Language & Law" deal with pause in an explicit and thorough way, since these analyses often have real-world implications.

Although our understanding of the psychological and cognitive factors behind pause realization is still being developed, studies focusing on pause have begun to yield important findings. In the forensic linguistic literature, a few scholars have fruitfully examined the role of pause in such areas as hearing testimony (e.g., Mendoza-Denton 1995) and deposition (e.g., Walker 1985). These studies have focused on the role of silence in the generation and interpretation of meaning. At the same time, there is evidence that pause varies in socially derived ways, beyond intentional (or subconscious) action. Campione and Véronis (2002) found that pause duration varies in statistically significant ways between languages. My recent research (Kendall 2006) has found evidence that pause realization varies along social axes (e.g., gender and/or ethnicity). In short, pause can carry speaker-generated meaning, but can also be misinterpreted by listeners based on cultural and social differences. These studies highlight the need to re-examine our thinking about and treatment of pause.

In this paper, I focus on the importance of accurately transcribing pause. I compare pauses in the videotaped deposition of an expert witness with the treatment of those same pauses in transcripts prepared by court reporters. I show how the practice of not indicating pause in the deposition transcription often creates opportunity for misinterpretations, and I argue that even simple changes to deposition transcription practices will result in better understanding of these statements.

We all encounter situations where we receive some kind of text and find that even though it is written in a language we are familiar with, we are unable to get to the heart of the message. Indeed, we find examples of complex language in almost every aspect of day to day life- websites, technical instructions and even in newspapers.

The importance and necessity of plain language is often underestimated. Whether it is a legal brief, a procedure, a brochure, or a newsletter; whether you are managing a business department, maintaining a website, or training workers you need plain language. It is language which everyone can easily understand. Gone are the days when publications were written with a vocabulary and syntax modified in the pursuit of precision. Today, the comprehensibility of a text outweighs the need for its precision.

Because plain language is communication designed to meet the needs of the intended audience, this audience has to be identified first. Therefore, plain language is not a certain style of writing; it is not text-based but reader based. A document which is written in plain language should adequately convey the writer's intentions. The best way to communicate intentions is through the organisation of information.

An example of plain language which I wish to present is from the United Kingdom's Home Office and its brochures and application forms for visas.

My reasons for choosing this example is based on the principle behind plain language: the audience and the organisation of information. In the first part of my presentation, I will show how both have been taken into account in the drafting of the information brochures and application forms. The second part of my presentation will be an evaluation of the work based on my own survey.

With an emphasis on the linguistic experiences of rape victims in the UK legal system, this paper is a consideration of recent innovations in cognitive semantics that offer important methodological tools for the analysis of legal encounters. The particular concepts to be discussed in this paper include conceptual frames and mental spaces theory to shed light on how perceptions of guilt and victim-hood are cultivated by the questioning barrister's lexical choices. Frames are culturally accepted information sets surrounding every lexical item used in discourse. The appropriateness of the chosen frame is of key importance to a frequent phenomenon in courtroom discourse, 'smuggling' information (cf. Luchjenbroers, 1993), where non-asserted (and often negative) information is covertly inserted into a witness's testimony. For example, when a rape victim 'corrects' a barrister's characterization of the type of establishment she had been to on the night in question (e.g., 'bar' is replaced with 'club'), she is rejecting more than just the name; she is rejecting the entire social scene and conduct association with a 'bar'. Similarly, when a barrister asks, 'and did he tell you about his love of duck-shooting?' the frame associated with 'love' requires an activity that happens often and with a passion. Given that defendant is on trial for murder (with a shotgun), this is clearly an undesirable presentation of the defendant's character. Additionally, the required mental spaces needed to process these propositions require hearers to conceptually separate or juxtapose information in a way that is contrary to the witnesses' interest or intent. Using UK police interviews and cross-examination testimonies with child witness claimants in sexual abuse cases, we investigate alternate realities involving varying representations of abuser or victim. These representations emerge from alternate conceptual frames of reference and the mental spaces constructions they require.

In contemporary law, there appears to be a tendency to enact provisions which describe rather than sanction (Amselek 1986), which are principles rather than rules (Dworkin 1977, 1986). Thus, since the 1992 Treaty on European Union (Treaty of Maastricht), a growing number of provisions functioning merely as principles have been added to European Community primary law. The 2004 Treaty Establishing a Constitution for Europe was the culmination of this development. These provisions typically concern culture, education, human rights, and the general values of the European Union, i.e., subject areas largely outside the jurisprudence of the Union. The question thus arises as to what actually is the primary function of these provisions.

Through a textual analysis of the founding treaties and the constitutional treaty of the European Union, this paper argues that legal provisions consisting of principles can be considered as originating from genres other than law, thus rendering the EU's constitutional law an essentially polyphonic genre. The paper therefore inquires whether the illocutionary force residing in law's genre and discourse in fact defines the contextual meaning of contested concepts, i.e., expressions which lack a universally accepted meaning. Examples of such concepts include "Europe" and "regional or minority languages and cultures." And if law is understood as the ultimate discourse of truth (cf. Foucault 1971), fixing the meaning of such contested concepts indirectly defines European identity, that contested concept par excellence.

The paper aims at combining methodologies of (the French tradition of) discourse analysis, the theory of énonciation, and systemic-functional grammar.

REFERENCES:

• Amselek, Paul 1986. "Philosophie du droit et théorie des actes de langage." Théorie des actes de langage, éthique et droit. Paris: PUF, 109-63.
• Dworkin, Ronald 1986. Law's Empire. Cambridge, Mass.: Belknapp.
• Dworkin, Ronald 1977. Taking Rights Seriously. London: Duckworth.
• Foucault, Michel 1971. L'ordre du discours. Paris : Gallimard.

The aim of this paper is to present structural interpretation of pleadings in English and in Polish. Pleadings (also termed 'statements of case') will be taken to mean documents, such as a claim form (or a petition), particulars of claim, defence, a counterclaim or third party proceeding (now both known as 'a Part 20 claim') and reply to defence, containing the written allegations of facts that each party is required to communicate to the opponent before trial. Parallel pleadings will refer to comparable persuasive texts, which are all originals in the mother tongue, and which are particularly useful for examining differences in conventions in text and sentence structure between legal languages as well as legal cultures they mirror.

It is assumed that cognitive structuring reflects conventionalised social knowledge available to a particular discourse community (Bhatia 1993: 21). The analysis of structural organization of the pleadings should reveal preferred ways of communicating intention in written legal discourse. An attempt will be made to define the cognitive structure of pleadings and guidance notes in English and in Polish with the concepts of relevance and intertextuality uppermost throughout. In doing so, we will take account of the general rules about the format of all court documents laid down by the Practice Directions (PD) made under the Civil Procedure Rules 1999 and other linguistic guidelines given in drafting manuals. In this paper we will also examine the positions taken by parties to civil proceedings: legal representatives (i.e. lawyers), and litigants in person as well as target recipients of persuasive texts (judges, defendants/respondents). The implications for achieving desirable clarity and preciseness (i.e. plain language) will also be discussed.

• References: V. K., Bhatia (1993) Analysing Genre. Language Use in Professional Setting. London

The general principle of Multilingualism for drafting of legislation within the Community institutions guides:

"...acts shall be framed in terms and sentence structures which respect the multilingual nature of Community legislation; concepts or terminology specific to any one national legal system are to be used with care."

How do translations correspond to the English text? Are these precise, equally understandable for everybody?

TERMS in Swedish are more like in English, as both are Germanic languages. Estonian and Finnish as Balto-Finnic languages have certain common wordstock, but not as many international terms in use as Swedish has. Both Finns and Estonians have a long tradition of derivating terms from their own genuine stems and have been active in creating neologisms. In 2002, a contest was organized in Estonia to get clear native terms for 12 EU-concepts.

E.g. EN subsidiarity principle - SV subsidiaritet, närhetsprincipen - FI toissijaisuus - ET lähimus, lähimuspõhimõte

Balto-Finnic SYNTACTIC STRUCTURES are quite different from Germanic ones. There are also some differences between English and Swedish, like between Estonian and Finnish.

The question to be examined in this paper is, which linguistic means have been used for communicating one and the same message in different languages, e.g. in these excerpts from the Treaty establishing a Constitution for Europe (Article III-394):

• EN
  projects of common interest to several Member States
  In carrying out its task, the ... Bank shall facilitate the financing ...
• SV
  Projekt av gemensamt intresse för flera medlemsstater
  När ... investeringsbanken fullgör sina uppgifter skall den underlätta finansieringen ...
• ET
  mitmele liikmesriigile ühist huvi pakkuvad projektid
  Oma ülesannete täitmisel hõlbustab ... Investeerimispank ... rahastamist ...
• FI
  useiden jäsenvaltioiden yhteistä etua koskevat hankkeet
  Tehtäväänsä suorittaessaan ... investointipankki helpottaa ... rahoittamista ...

Atkinson and Drew (1979) observe that, during cross-examination, the witness is sometimes able to infer that the lawyer is implicitly criticising them or their direct-examination testimony. Drew (1990, 1992) explores this phenomenon further, noting some of the linguistic ways lawyers imply criticisms of witnesses to juries.

The ability to implicitly criticise the witness is an important weapon in lawyers' rhetorical arsenals since, as Amsterdam and Bruner (2000) observe, it allows them to deny having made an assertion if it is challenged and also allows them to subtly persuade the jury (p. 175-176). However, despite the importance of this phenomenon, previous studies into it have been tentative. This is the reason for my research.

In this study I will report on an analysis of the different ways that lawyers use their questions to implicitly criticise witnesses during cross-examination. In this analysis I employed a qualitative pragmatic methodology, referring to the work of Grice (1975) to identify lawyers' use of implicature in cross-examination, and a corpus methodology to enable the quantitative surveying of different types of lawyers' questions that generate implicature.

The data I analyse are official courtroom transcripts of recent, criminal cases that I have extracted from electronic sources - including the trials of Dr Harold Shipman and of Oklahoma Bombing conspirator Terry Nichols.

I conclude that lawyers embed implicit criticism of witnesses in their questions during cross-examination not only for the reasons listed above, but also because a witness cannot refute a criticism that is not made explicitly due to their constrained speaking rights, and also if they do challenge an implicit criticism, they risk bringing it to the attention of the jury. Keywords: implicature; inference; pragmatics; corpus; cross-examination questioning.

This talk deals with the use of speaking metaphors in legal discourse. For instance, one may speak of the victim's body as "telling us what the defendant won't." Also, it is not unusual for attorneys to introduce material evidence - or the lack thereof - as the deceased victim testifying in open court. Indeed, legal evidence - whether direct or circumstantial - is often presented as speaking in a similar fashion as an actual witness might. In my view, the frequent presentation of legal evidence as speaking up reflects the importance of language in the law on the one hand, and the conceptual relation between speaking, believing and the 'truth' on the other hand.

First, in the Western legal system, only evidence presented in witness testimony is to be considered by the ultimate evaluator(s) in the judicial decision-making. Given the importance of the witness testimony phase, it should not be too surprising that attorneys turned what might be diffuse circumstantial evidence into concrete direct testimony evidence, if only rhetorically. Second, I believe such speaking metaphors also have a common heritage in the cultural model according to which our experience of ordinary inter-subjective conversation is used as a general frame for the conceptualization of verbal as well as non-verbal processes. Since we commonly learn about facts and events through some sort of verbal interaction, attorneys - as well as all of us - often present the source of new knowledge as speaking to the individual who attains it. To put it differently, 'factive' courtroom interaction is reproduced in metaphorical language in what could be called 'fictive interaction' (Pascual 2002).

Pascual, E. 2002. Imaginary Trialogues: Conceptual Blending and Fictive Interaction in Criminal Courts. Utrecht: LOT.

Thanks to research in forensic linguistics by scholars such as Coulthard, Chaski, and Grant, there have been considerable advances in the areas of authorship attribution and detection of plagiarism. However, research into the detection of deception in written text has so far failed to benefit from these advances. This presentation will report current work, part of a larger research agenda, directed at a more systematic approach to the detection of deception.

Deceptive statements are rarely entirely deceptive. Lying authors prefer to imbed their deception within essentially truthful accounts. This blurs the boundaries between the truth and the lie and makes the deception, and its linguistic markers, harder to identify. However, DePaulo's comprehensive review of research into cues to deception concluded that certain verbal behavioural cues do occur during deception, and that these cues are likely to occur when deceivers are more motivated to succeed in their lie (DePaulo et al 2003), a motivation that studies in laboratory environments fail to replicate.

This study hypothesises that certain linguistic markers are more likely to be confined within the boundaries of deceptive sections in narratives, and are more likely to be excluded from truthful sections. If this is the case, then certain deception markers may be considered more likely to be present when actual deception is occurring, rather than simply be the result of fear of being disbelieved.

This study uses original witness statements given to the Police, independently authored by suspects or witnesses directly involved in serious crimes. In analysing the data, episode boundaries within the narratives are first identified using marked sentence structures (Prideaux 1999). Then, linguistic markers characteristic of distancing and negation (i.e. pronouns and function words) are identified within the episodes.

Preliminary results support the hypothesis that these linguistic markers predominate within deceptive episodes.

Tolerance and diversity are threatened in many ways today. A basic clue to tolerance and diversity is a good understanding of the other and the other's otherness. Within the field of law and forensic linguistics, however, what does understanding or comprehensibility mean? In which ways could an enhanced comprehensibility contribute to an enhanced tolerance and diversity?

In this paper, I will try to give new answers to these questions. It has often been claimed that legal texts are incomprehensible for the layman. This global judgement has been made concrete in various ways, psycholinguistic experiments have been set up to examine the actual degree of comprehension. Drawing on these experiments, attempts have been made to optimize legal texts for the layman. However, the results of this research are unsatisfactory in several respects. First, there are different sorts of laymen, some understand more (e.g. a law student), some less. Text optimization usually aims at meeting the needs of a layman understanding next to nothing. Second, there is no principled way to predict comprehension (for whatever kind of layman). Instead, comprehension is just measured empirically. We demand more from a linguistic theory. Third and maybe most important, the results of this research are not applicable in everyday life. A lawyer who wants to check a letter to his client in terms of comprehensibility does not have anything at hand to do this. He still has to use his intuition.

The paper is structured as follows. Section 1 presents linguistic approaches to comprehensibility in general. Section 2 is devoted to comprehensibility within forensic linguistics. The theoretical and practical shortcomings of this research will be shown. Section 3 presents a new approach to comprehensibility within forensic linguistics. Frame Semantics as a shallow semantic approach shall be used to predict the comprehensibility problems of a legal text.

In this paper we present a very simple yet effective algorithm for authorship attribution. By this term we mean the act of telling wether a certain text was or was not written by a certain author. We shall not discuss the advantages or applications of this activity, but we propose a method for doing it in an automatic and instantaneous way, without considering the language of the texts and without undertaking any kind of text preprocessing like tokenization or part of speech tagging. We conducted an experiment that shows how authorship attribution can be seen as a text categorization problem. That is to say, each author represents a category and the documents are the elements to be classified.

Text categorization has became a very popular issue in Computational Linguistics, and it has grown up to great complexity, motivating a huge amount of literature. Although, in this article we conducted the most basic method applied to authorship attribution. The program is language independent because it uses pureley mathematical knowledge: an n-gram model of texts. It works in a very simple way and is therefore very easy to modify. In spite of this simplicity, it is capable of classifying, hundreds of documents by author obtaining more than 90% of accuracy. As an example, we present an experiment carried out with a rather homogeneous corpus composed of 100 short newspaper articles (around 400 words each) from 5 different authors. This corpus is divided on a training set of 20 texts (five of each author) and the rest as the test set. Once the algorithm has "learned" the style of each author, then it begins to classify each of the remaining 80 texts by each author-category.

As a final comment, it is worth to say that this program can be used as general purpose document classifier, for example by content instead of authorship, because it only reproduces the criteria that it learned during the training phase.

Keywords: text categorization by authorship; quantitative stylistics; n-grams analysis; language-independent text categorization.

Establishing a rule of law based on rights and obligations expressed in language works only if speakers of the law's language understand it the same way. If I understand "You must stop at a red light" as meaning that you must stop at a green light, or meaning Spanish anchovies are the best in the world, then it is difficult to see how I can organize my life around the law's requirements. Nonetheless, people, including legal actors, do talk past each other. The more they do so, the greater the threat to rule of law values.

This presentation explores situations in which different players in the legal system are likely to interpret the same language differently. When the difference is easily discoverable before anything of consequence occurs, no harm is done. But at times, not only do people understand the same language differently, but it is not easy to determine that fact until after something unfortunate happens, and it is sometimes impossible to determine the fact at all. I call this undiscovered miscommunication pernicious ambiguity. The presentation will explore the sources of pernicious ambiguity. It results largely from a predictable lack of consensus in decisions we make about how to categorize non-prototypical events into categories we routinely use. The presentation will then discuss some preliminary results of experimentation that shows people to overestimate the degree of consensus around their own understanding of legally-relevant terms, suggesting that pernicious ambiguity may be unavoidable.

Since the early days of Forensic Linguistics there has been an incessant search for idiosyncratic features that would allow us to distinguish one author from another on the basis of their linguistic production. Studies in Authorship Attribution constantly report on the latest findings and newly developed techniques, applying different linguistic units as measures, in forensic stylometric analysis. In this quest for valid and reliable identification markers, syntactic structure has been shown to be less appealing. That is easily explained by the fact that syntactic constructions are notorious for their structural complexity and processing difficulty. However this does not seem to be a problem if we consider syntactic structure to be a simple sequence of categories grouped together and whose function analysis is set apart.

This paper presents the preliminary results of a series of experiments in author identification aimed at evaluating the discriminatory capacity of sequences of linguistic categories (Morpho-syntactically Annotated Tag Sequences (SMAT)) and demonstrating which one of the tested three is the best candidate marker of authorship. The hypothesis tested in the experiments is that the most frequent tag sequences will discriminate effectively between authors in limited-size samples of texts. All the experiments were carried out on a morfosyntactically annotated corpus consisting of 15 newspaper articles and 15 novel fragments written in Spanish by 3 authors, five of each genre per author. The occurrences of every type of SMAT were extracted by means of the programming language AWK and later their frequencies were calculated using the environment for statistical computing and graphics R. An innovative combination of traditional statistical methods was applied to determine the authorship of a pseudo anonymous text. Preliminary results show that SMAT sequences can become valid and reliable markers of authorship.

Article 6 of the European Convention on Human Rights states that an individual is entitled to a "Fair Trial." A fair trial may be compromised when the individual charged or tried does not speak or understand the language of the system or the country in which the prosecution is taking place. Without appropriate interpreters, non-native speakers, even those whose language proficiency is - or appears to be - reasonably good, will almost inevitably be placed at a disadvantage when they come into contact with the police or courts.

Drawing upon a number of case studies from the UK, Hong Kong and Australia, this paper describes and analyses some common linguistic assumptions and misconceptions, (such as, "a fluent speaker who says he does not need an interpreter really does not need one") and makes some observations and further recommendations regarding the training of specialist forensic linguists, interpreters and interviewers.

It is a well-known fact that lawyers make use of their language skills to exercise control over witnesses' testimonies. It is a part of their professional training to master the use of specific types of questions for particular purposes. Lay people usually do not have any command of cross-examination techniques and legal jargon in general. Ordinary citizens representing themselves in court can therefore be seriously disadvantaged.

This study looks at cross-examinations performed by lawyers and lay people representing themselves and contrasts the two approaches from the linguistic point of view. The formal analysis of questions asked by lawyers and lay people shows that lay people's questions are less elaborate and closer to questions used in everyday conversations. The nature of cross-examination presupposes that the cross-examiner has to flout Gricean maxims. But it seems that lay people flout them in a slightly different way, which allows them to be more polite and yet exercise power over witnesses.

The data is mainly drawn from the libel case McDonald's Corporation v. Helen Steel and David Morris, which was tried in Great Britain. While McDonald's hired a legal team of leading libel lawyers, Helen and David had to represent themselves and the burden of proof lay on them. Neither of them had any previous experience with legal proceedings. In fact, their work experience ranged from being unemployed to various low-skilled jobs and voluntary work. During the trial they were occasionally assisted with legal matters by a volunteer lawyer Keir Starmer. The case lasted for two and a half years. Such a long period of time makes it possible to trace whether lay people are prone to gradually adjust their language to a 'lawyer-like' style as the trial progresses.

The study aims to contribute to current discussions on the advantages and disadvantages of lay advocacy.

Plain language guidelines typically are used for educating writers and for setting standards for revising communications. However, they can also be tools for expert consultants/ witnesses in forensic linguists. Guidelines provide linguists with a set of accepted standards required in doing analyses in legal cases.

This is a report on how guidelines were applied in a civil case (contract) and in a criminal case in the United States. The role of such guidelines may vary by state/ province and country. Also the requirements on expert witnesses may also vary. However, it is hoped that these two cases will suggest strategies that could be adapted for different contexts.

The civil case involved a state law on the use of plain language in contracts and its application to a housing construction contract. At issue were several sections of the contract which the client found extremely difficult to understand and felt the company had not met its obligations under the contract.

The criminal case involved a non-native English speaking merchant. The relevant Plain Language guidelines were tied to a Presidential Memorandum on writing by federal government agencies. The documents were from a federal agency. Had the merchant adequately understood the procedures and his obligations? Was he negligent or had he conspired?

The analysis procedures for both cases are introduced along with practical analysis problems encountered and their solutions.

The Spanish Criminal Prosecution Law states in Article 520 that any detained person will be informed of his rights in a comprehensible way. In the last two decades, the content of this article was submitted to revisions and the particular reference to foreigners with comprehension difficulties of the Spanish Language has been ambiguously established. When Spanish police officers proceed to arrest a person, they must provide him with a form conveying the basic rights he is entitled to and which must signed, this being a crucial part of the detainee's police record. Since any police force can freely adapt and translate the statements of Article 520, detainees are given a particular version of rights depending on the police force responsible for their detention.

Thus, by comparing some English translations of rights available at police stations, the aim of this paper is to explore to what extent determined linguistic choices can influence the comprehensibility of the person interviewed.

The contextualization process is approached at the discourse level, revealing significant lexical and syntactic complexities ( Halliday and Hassan 1976; Halliday 2004 ) which far from enhancing the intelligibility of rights may contribute to inaccurate legal implications.

The paper discusses law on minority languages in Poland with reference to the problem of linguistic relativism and the mutual influence of law on language and language vitality.

The focus is on Kashubian, a language spoken in the North-West region of Pomerania. Genetically, Kashubian, spelt with the Latin alphabet, belongs to the West Slavic group and is a Lechitic language. Under the Polish law it has recently been recognized as the only regional language in Poland (acc. to the Statute 'Ustawa' of 6 January 2005 on national and ethnic minorities and the regional language, binding from 2nd May 2005 (Dz. U. 2005 r., Nr 17, art. 141)).

Having been suppressed for years under the communist rule, now Kashubian appears to experience a period of revival. Although 53,000 citizens of Poland declared Kashubian as their first language in 2002 and there is a growing number of people who claim Kashubian nationality and report Kashubian as their first language, there is still an ongoing theoretical debate whether Kashubian is in fact a language or a dialect, further complicated by the presence of its many varieties.

The discussions includes comments on the function of linguistic form in creating extralinguistic reality as well as in reflecting the real world. These issues are of special importance in the language of the law, which is naturally operative. The inherently performative nature of legal language is commonly emphasized, but it is more readily associated with its directive function than with creation of new attitudes and beliefs, which is the focus of the present approach. The paper further includes comments on the claim according to which (language) identity is formed by the repetition of pre-defined linguistic patterns.

The translation of EU-produced legal texts presents multifarious problems for practitioners from Member States voiced in private, as well as at seminars of the Translation Service and in the on-going "Fight the fog" campaign. For the countries seeking accession to the European Union problems are multiplied, since they have to harmonize their existing institutions, create new ones and find the language to communicate adequately within unified Europe.

The presentation will focus on some of the specific functional, linguistic and communicative problems arising from translating EU Directives and other EU official documents from English into Bulgarian. A functional approach is indispensable for the adequate rendering of terms that should be evaluated according to the conceptual system of the particular field under codification. We shall also touch upon some more general issues related to the future language policy of an expanding EU from the point of view of a "small language".