Language in the Judicial Process

International Association of Forensic Linguists

9-12 July 1995

University of New England/Armidale New South Wales/AUSTRALIA

For financial and practical support we thank the following sponsors:

Department of Linguistics, University of New England Department of Sociology, University of New England Department of Law, University of New England The Vice-Chancellor, University of New England National Institute of Forensic Science, British Council (supported Professor Coulthard”s visit to Australia)

Department of Linguistics
The University of New England
Phone: 067 73 3185
Fax: 067 73 3735

Abstracts of Papers Accepted for Presentation

Sue Blackwell, University of Birmingham

Taking a Closer Look at “look”: Discourse Markers in Disputed Texts

It has often been argued that the words which bear the least lexical information are those least subject to conscious control by the speaker. If this is the case, then “function” words and discourse markers may be of more diagnostic value to forensic linguists than “content” words in evaluating claims about the authenticity of alleged utterances.

However, the potential usefulness of such words is complicated by issues of transcription. Discourse markers may be missing from a disputed text because they were not transcribed or even perceived by the “scribe” rather than because the text did not originate in spoken discourse. This paper will examine the distribution of discourse markers such as “look”, “well” and “you know” in some disputed texts and their undisputed counterparts, and will compare these distributions with those encountered in spoken corpora of English. It will attempt to assess the degree to which various discourse markers are perceived and transcribed by the British police and by British court reporters, and the significance which can consequently be attached to them by the forensic linguist.

Mark Brennan, Charles Sturt University

Speaking pragmatically: Police and the communicative needs of people with an intellectual disability

The “Cleartalk” materials (Brennan & Brennan, 1994) have been created to support police work at three levels. The first is a report providing background and research information which is intended to inform policy. One set of policy outcomes that can be drawn from this report is expressed in terms of the educational needs of police officers and this represents the second level. The education modules have been designed for use either during “on-the-job” training days or together as a whole unit of pre-service study. Of course there are other manifestations of policy that the report can be used to generate. Also the educational implication will be useful for providers other than police officers in the NSW Police Service. The third level is presented as publicity material; a poster and pamphlet for wide distribution to police generally. All of these materials are derived from a single research and development process. The aim has been to create three sets of materials in which policy, education, and publicity complement each other.

The “Cleartalk” report provides a challenge to the centrality of the verbal record. The verbal record is not necessarily the best or most complete record of an interaction, especially for those with less than articulate verbal skill. For so many reasons the bald written or spoken word is not an adequate encapsulation of communicative quality or dimension. But other records, like drawings, acting out, puppet manipulation, picture pointing, are seen to be too negotiable or interpretable for the purpose of procedure and administration.

There are two kinds of responses to this dilemma.

The first is to challenge the validity of this latter claim and possibly to find that the position is in fact based on a notion of majority; “It”s good enough for most of the population and for the rest Ü they simply miss out!” If we take the issue of language rights seriously, this position is untenable and it is incumbent on those in control to devise better ways of facilitating the voice of all.

The second is to accept the centrality of the verbal record and to undertake to make it as complete as possible and to also incorporate into that record an assessment of just how reflective of participation and comprehension that record is.

Michael Cooke, Batchelor College

A different story: Comparing Aboriginal testimony given in narrative versus question/answer form

The elicitation of evidence from a witness usually proceeds by way of question-and-answer (Q/A). Police and courtroom lawyers generally adopt this pattern. Yet for Yolngu, the indigenous people of north east Arnhem Land, this format is foreign. It does not exist as a Yolngu register while the narrative does. The Q/A style impedes the telling of the “whole truth” as frequent interruptive questions serve to fragment a story and inhibit its telling.

For Yolngu the problem is compounded by

comprehension difficulties arising from formal English communicative contexts. Yolngu, for whom English is a foreign language, also tend to utilise the grammatical structures and lexicon modelled by the inquisitor in his/her questions, leading to a moulding of testimony through this “scaffolding”.

These issues are exposed in a recent trial where a Yolngu was charged following a killing and interviewed by police without an interpreter and in the typical Q/A style. The testimony which resulted supported a charge of murder. At the ensuing trial the defence lawyer conducted his examination of his client (she was the only defence witness) by eliciting and guiding a narrative given with occasional assistance from an interpreter. The account of the killing provided in this familiar style resulted in the charge of murder being abandoned. It was not that the two accounts were contradictory Ü rather that the Q/A account was deficient.

These two accounts by the same person of the same events, but provided under different conditions, are linguistically analysed for their judicial implications.

Malcolm Coulthard, University of Birmingham,
President of the International Association of Forensic Linguists

Restating requests: The role and techniques of the forensic linguist

(Keynote address)

This plenary address will be concerned with the kinds of requests for help which linguists regularly receive from lawyers, with the ways in which they reformulate these requests and the theoretical bases and practical techniques with which they attempt to solve the resulting problems. After briefly mentioning the range of areas in which linguists in England work, the address will concentrate on disputed authorship and the possibility and usefulness of linguistic fingerprinting. Illustrations will be drawn from real cases though names may be changed to protect some of those involved.

Bethany Dumas, University of Tennessee

The Inclusivity of Terms for Documents:
Problems in “The Search Warrant in the Matter of 7505 Derris Drive”

A question arose in a an exclusionary hearing in a federal criminal case in a United States District Court in 1994 as to whether specific items seized under a search warrant executed by the Federal Bureau of Investigation fell within the categories listed in the Search Warrant, dubbed “The Search Warrant in the Matter of 7505 Derris Drive.” One set of categories listed in the search warrant referenced specific types of written documents: “bills of sale for automotive parts and rebuilt automobiles sold to businesses and to private individuals; automobile titles, including completed titles, blank titles, and open titles; payroll records, accounts receivable and accounts payable documents, which will disclose the sale and receipt of automobiles, both rebuilt and salvage; receipts from salvage yard and auction companies; daily receipts and Federal Express mailing receipts; telephone logs, address books, diaries, handwritten notes and memoranda, which will reveal the identities and location of co-conspirators; travel records and itineraries.” Those terms fall generally into two categories, those denoting financial records and those of a more general nature.

A linguist was called to testify about the current meanings of the latter terms, “telephone logs, address books, diaries, handwritten notes and memoranda, which will reveal the identities and location of co- conspirators; travel records and itineraries,” as well as the scope of the modifying clause, “which will disclose the sale and receipt of automobiles.” This paper reports the analysis and testimony of the linguist as well as the judicial response to that testimony.

Diana Eades, University of New England

Forensic linguistics and Australian Aborigines: Reactions of the legal profession

Australian Aborigines are possibly the most over- imprisoned people in the world. Many studies in recent years have examined ways to address the complex factors which give rise to this over-imprisonment. Forensic linguistics has begun to play a role, in two distinct areas:

  • firstly, in cases in which an Aboriginal person has been convicted on the basis of a confession which has been allegedly fabricated, and
  • secondly, in cases where cross-cultural communication differences prevent an Aboriginal person from effectively participating in the speech events which are crucial to the way in which they are dealt with by the criminal justice process (such as police interviews, lawyer interviews, examination-in-chief and cross-examination).

This paper focuses on reactions of the legal profession to forensic linguistics in relation to Aboriginal people in five particular cases. In two of these cases, an Aboriginal person”s conviction was eventually quashed, thus securing their release from prison, where they had each been serving a life sentence. (These cases each illustrate one of the areas of concern outlined above.)

John Favretto, Office of the Director of Public Prosecutions, NSW

The right to an interpreter

The paper examines the law relating to the right to an interpreter and recent developments in the High Court, (NSW) Court of Criminal Appeal and (NSW) Court of Appeal and other superior courts. The recognition by the High Court of the fundamental right to a fair trial in Dietrich v R (1992) 177 CLR 292 now means that the right to an interpreter is not to be simply viewed in the contest of a person”s language capabilities alone, rather, factors such as the person”s age, ethnicity, legal representation and particular circumstances in which the need for an interpreter arises must all be considered.

Helen Fraser, University of New England

Auditory Identification of Familiar and Unfamiliar Voices

This paper reports on an experiment designed to test people”s ability to discriminate, identify and describe voices. Subjects are presented with 4 sets of stimuli. In the first three, pairs of voices are to be discriminated. The voices occur both plain and disguised, and some of the speakers are familiar to some of the subjects. In the fourth set, single stimuli are heard and subjects are asked to identify these; they include some of the same voices and some different voices, and a range of disguises and recording conditions.

Results are discussed with respect to issues of the relative reliability of auditory and machine-aided speaker identification, the construction of voice line-ups, and the validity of comparing identification performance by strangers and familiars in a recent case.

Muhammad Gamal, University of Queensland

Vox et pritorai nihil: The high cost of linguistic injustice

Forensic interpreting in multicultural Australia is a complex area which encompasses the triangle of language, culture and the law. This pyramid-like structure has at its base the linguistic element followed by the cultural aspect and is topped by an adequate legal knowledge. Such pyramid, needless to say, reveals an inner structure reflecting the same albeit “foreign”, linguistic, cultural and legal systems.

The paper looks at the process of legal interpreting for Arabic-speaking clients within the Australian legal context. Although the issues involved are not exclusive to Arabic, Middle Eastern cultural or legal systems, they have not been addressed in the literature on forensic interpreting.

Thus the “experience” of “going to” court, talking to the police, giving a statement, taking an oath, being asked questions by counsel and/or cross-examined, addressing the jury, answering questions and attempting to be convincing reveal a multi-faceted cognitive process by the interlocutors behind the box, the bar and the bench.

Such process is judged only on one aspect, namely, the linguistic. Since the linguistic aspect involves two persons, a monolingual speaker and a bilingual interpreter, the paper will particularly focus on the traductological relation between another triangle of communication: Barrister/Client/Interpreter.

The linguistic aspect will be examined with the view of exploring not only the power-relation between the interlocutors, and particularly the interpreter”s precarious part, but also the role of the very medium they employ, and manipulate, to achieve their respective purposes: Language.

John Gibbons, University of Sydney

Evidence of Miscommunication between Police and Second
Language Speakers during Police Interrogation

In a previous paper (Gibbons, 1995) I documented the possibility of vital information being lost during police interrogation of low proficiency speakers of English, when police used only English for the purpose of interrogation. In that paper it was shown the police “Records of Interview” are inaccurate records of such interaction. Since that paper was written, the NSW police have introduced the videotaping of interviews, to avoid some of the pitfalls of the “Record of Interview”. Consequently we now have available video-recordings of police interrogation. These enable a much more precise documenting of miscommunication between police and second language interviewees. Such miscommunication can serve as a basis for court testimony which discredits video-taped evidence. This paper will describe this process, illustrating from transcriptions from video-tape, and exemplifying from a particular case the contribution that this type of expert testimony made to the discrediting of police evidence.

Cliff Goddard, University of New England

How do judges know what they mean (and can linguists help them)?

This paper considers how modern linguistic semantics can assist judges in determining the “plain meaning” of words for the purpose of statutory interpretation. It describes the main schools of contemporary semantics, showing that leading experts in the field differ enormously in their basic assumptions and methods. It gives a detailed critique of surveys as a research method in semantics, concentrating on a recent American study (Cunningham et al., 1994) which has been proposed as a model of how linguists can help judges. Although I advocate Anna Wierzbicka”s reductive paraphrase approach and demonstrate its value for conceptual analysis in legal contexts, I argue that in view of the fragmented and under-developed state of lexical semantics it would be ill-advised for courts to recognise linguists as experts on word meanings.

John Hajek, University of Melbourne

Starting small in forensic phonetics: a first case report

Forensic linguistics is still an undeveloped field in this country Ü a result of limited opportunity in the past and reflected in its absence from the structure of linguistics courses around the country. However, recent developments suggest an increase in interest in forensic linguistics amongst legal professionals and hence demand for the services of forensic linguists. Despite changing circumstances, most linguists are still unprepared for forensic linguistics. In this paper I discuss my own sudden, unexpected and extremely rushed introduction to the legal world and to forensic phonetics. The circumstances surrounding the case in question, as well as the problems faced by a forensic first-timer are detailed. Some suggestions for the development of forensic linguistics as a professionally recognized field by linguists and legal professionals are also made.

Terry Hillcoat, University of New England

Investigating acoustic parameters for forensic speaker identification

A preliminary experiment was performed to investigate whether a number of acoustic parameters obtained from sibilant and nasal consonants can be used in the identification of speakers. Using statistical techniques, the experiment tested whether parameters extracted from speech samples from six adult male subjects exhibited significantly greater variation between speakers than within speakers. It also tested whether it was possible to use the parameters to correctly classify each of the samples. Included in the study were samples containing whispered and disguised voices, and also samples recorded over a telephone line, as these types of distortion are often present in forensic samples. The results showed that some of the parameters appear to have potential for speaker identification under forensic conditions.


John Ingram, University of Queensland

Telephone transmission line effects on formant and extraction for forensic speaker identification

Acoustically based methods of speaker voice identification under forensic recording conditions require robust speech parameters and parameter extraction methods. Target recordings are typically made over telephone lines, whereas comparison recordings are typically obtained from police interviews. Apart from difficulties of matching voices gathered under conditions likely to elicit very different speech styles or registers, there is the problem of determining transmission line effects upon the speech parameters used for voice identification.

This paper reports an experiment on transmission line effects upon the extraction of formant trajectories and fundamental frequency contours, which are the main parameters that are used by the author for text- dependent speaker recognition. The effects of standard and mobile phone transmission lines are compared with analysis of recordings made under typical free field recording conditions.

Marie-Thurse Jensen, Monash University

Determining the authenticity of confessions: Linguistic and legal constraints

What can the linguist say in court about the authenticity of confessions made by non-native speakers of the language involved?

Cases involving non-English speaking background defendants in Australia are used as the basis for a discussion of two kinds of constraint. The first has to do with the indirectness of the linguistic evidence, given that the only evidence of the original discourse is the disputed text itself. The linguist typically elicits another speech sample from the defendant and makes inferences about the plausibility of the disputed text on the basis of this. A number of problems with this methodology are discussed.

The second constraint involves the legal rules governing the presentation of the linguist”s expert evidence, which may prevent the admission of evidence which the linguist considers relevant. The author”s experience of preparing to give evidence in court in such cases forms the basis for the paper.

Hannes Kniffka, University of Bonn

Forensic Linguistic Author Identification: Limits and Chances

This paper reports symptomatically on two cases of author identification representing “maximum” and “minimum” degree of what linguists can contribute to help to determine anonymous authorship of written texts. A detailed discussion of data will be in focus as a prerequisite of a more general discussion. The overall objective is a realistic assessment of achievements, possibilities, precautions, … applying to our work, from various ends in various respects for various people. This report is somewhat a summa and aftermath of some 23 years of expert testimony in criminal court cases in Germany.

Ann Laubstein, Carleton University

Building voice lineups: Problems

It is readily apparent that accurate voice recognition and misidentification are common phenomena, the former the more common of the two. Which features of the voice are instrumental in this (mis)recognition, how much speech is enough to identify a familiar speaker, or to make an unfamiliar speaker familiar, are among the questions to which definitive answers are still unavailable (Hollien).

Nonetheless voice identification is being used by the legal systems of both the United States and Canada as evidence in suspect identification. There is no well established method for getting witnesses to identify suspects” voices but the most common method (Mayor) involves the presentation of a voice lineup to the “ear witness”. This paper describes an experimental study which investigates the viability of the procedure currently in use in the preparation of voice lineups both in Canada and the US. In this procedure a suspect”s voice is recorded and a transcript is made of his speech. This transcript is then given to six or seven foils (e.g. actors) to read. Witnesses are then asked to determine whether any of the taped voices in the resulting “lineup” is that of the criminal. There are a number of problems with this “transcript” method Ü in particular, it is not clear that it is capable of producing an unbiased lineup. Two experiments were run (n=72, n=90) testing whether two lineups built by the Ottawa Police were biased. The suspect was judged as significantly different from the foils on a number of parameters Ü parameters that appear to be intractable using the transcript method.


Hollien, H. (1990) The Acoustics of Crime The New Science of Forensic Phonetics, Phenum Press, New York.

Mayor, D. and Eeva Komulainen (1989) Subjective Voice Identification ms. Calgary Police Service, Calgary, Alberta.

Neil McLeod, Murdoch University

Psycho-Linguistic Analysis of Tax Judgements

Neuro-Linguistic Programming, a school of psychotherapy drawing heavily on Chomsky, holds that a client’s distorted or inappropriate “world view” will manifest itself in linguistic structures. These structures serve to delete or disguise sentence content so that the client avoids having to recognise the inappropriateness of the premises on which he or she is operating.

The author believes that Australian income tax jurisprudence represents a dysfunctional “world view”, especially in so far as it is based on wholly inappropriate trust law concepts. Lord Green MR noted that “in many [tax] cases it is almost true to say that the spin of a coin would decide the matter almost as satisfactorily as an attempt to find reasons”. Judges are imbued with this inappropriate jurisprudence, and forced to work within its parameters. They, too, tend to use linguistic devices to mask its inadequacies. These devices create the illusion that the court has arrived at its decision through the logical application of what are inherently irrational “principles”. In this way judges are able to preserve the plausibility of an unstable tax law jurisprudence.

It is not suggested that judges are contriving their speech at the conscious level. Rather, linguistic formations at the unconscious level enable them to convert their intuitive reactions to tax problems into respectable judicial discourse.

The linguistic structures considered in this paper include:

  • the substitution of active verbs for passive ones as a means of transforming reality (and its attendant taxation consequences);
  • validating an invalid statement by conjoining it to a valid but logically unrelated, or even contradictory, one
  • corroboration through repetition, and the assigning of causative force to synonyms for the outcome;
  • the use of negative structure to avoid facing uncomfortable facts head on; and
  • mere verbalisation of a factor as a substitution for its logical consideration.

Yon Maley and Chris Candlin, Macquarie University.

Questions in court: What can the expert witness expect?

In a common law courtroom, expert witnesses are linguistically privileged. Unlike lay witnesses, expert witnesses are not only permitted but also obliged to provide explanations and offer opinions. In return for such privileges, they must, however, expect to have their evidence challenged and to have its value and accuracy put into doubt by opposing counsel.

In this paper, we shall argue that (i) court processes subject witness testimony, and expert witness testimony in particular, to a complex and continuing process of interpretation and transformation; and that (ii) the final meaning of evidence and its significance to the court derive from an interrelationship of these and other institutional and contextual factors.

In order to substantiate this view, we propose to:

  • summarise current research on the linguistic practices of lawyers in courtroom examination sequences;
  • analyse the lexico-grammar and semantics of questions directed to expert witnesses in recent hearing before the NSW Land and Environment Court; and
  • show how the information received was interpreted by supporting and opposing counsel and the judge.

Carmel O’Shannessy, University of Melbourne

Pre-court barrister and client interactions: an investigation

This study takes a case study approach to examining pre- court barrister-client interactions. Within an ethnomethodological framework, the norms of interpretation which the participants are orienting to, are examined. In particular, conversation analysis concepts and techniques are employed to investigate preference organisation in this context. It is found that, in the context under study, when an utterance contains inaccurate detail, it is corrected. When an inaccuracy is not corrected, the correction is relevantly absent; that is, it is noted and is a basis for inference. The inference is that the information produced was correct. However, when a statement is not confirmed, it does not lead to the inference that the information produced was false. Thus, there is a structural preference for “other- correction” in this context.

Mami Okawara, Takasaki City University of Economics

The Amish and the criminal courts: the Samuel D Hochstetler case

This paper linguistically analyzes a seven page transcript of a criminal case in State of Indiana vs. Samuel D. Hochstetler (February, 1948 Term). Samuel D. Hochstetler, 75, Amish bishop in Clinton township, Indiana, was arrested on the charge of assault and battery against his insane daughter Lucy, 41. The sheriff accused Samuel Hochstetler of keeping Lucy forcibly chained in her bed in an unlighted, unventilated room for several years. Lucy was, in fact, not chained all the time; she was chained only when her condition was violent. It is normal that the mentally ill are cared at home rather than in a mental hospital in Amish society.

On the day succeeding his arrest, at the Elkhart circuit court Samuel Hochstetler made a few statements which were interpreted as a plea of guilty and was sentenced to serve a six-month term at the state penal farm in Putnumville accordingly. However, nearly three months later Gov. Ralph F. Gates granted a petition for Samuel Hochstetler on the recommendation of the state clemency board where new facts had been brought out in a review of the case. In this presentation I will focus on the “new facts” which were not used as a defence in court and will examine the cause of misjudgment shown in the transcript from cross-cultural perspectives such as Amish rejection of litigation; Amish characteristics of verbal interaction; self-sufficiency as Amish culture; disadvantaged old Amish from English (second language) legal language.

Hugh Potter and Mee Wun Lee, University of New England

Speech and action outside the court: Can forensic linguistics be applied to issues of hate speech?

Forensic linguistics takes as one of its primary objects of study the use of language in the courtroom. However, the topic of language as an offence is becoming more common in the law itself. This paper will examine the status of language as a category of harmful behaviour in the criminal law, as well as in the civil law. Specifically the focus will be on vilification and hate speech as it is constructed in the law. The relationship between speech and action will also be explored through a focus on pornography as a form of hate speech.

Bruce Rigsby, University of Queensland

Aboriginal evidence and the transcript in two Queensland land claims

Claimants in two Queensland land claims gave evidence in several venues and settings. They came from a variety of communities and backgrounds, and they used several indigenous languages, Cape York Creole, Aboriginal English and Australian English. Their examiners varied in professional background, prior experience with Aboriginal people and communicative competence.

The paper discusses settings, participants and the claimants” evidence generally, focuses on some problems of preparing an accurate transcript, and suggests some ways in the future for improving the situations for claimants to give rich evidence and for the cooperative preparation of transcripts that can serve cultural heritage and other purposes beyond their legal functions.

Phil Rose, Australian National University

Between- and within-speaker variance in acoustic parameters of similar voices.

A prolific literature exists on speaker recognition using acoustic features. However, few if any studies have approached the problem of speaker recognition from a forensically realistic viewpoint. One way in which speaker recognition tests are forensically inadequate is in the lack of control for prior similarity in voices. This paper looks at how successfully certain acoustic features can discriminate between voices that have demonstrated auditory similarity.

Six male speakers were chosen as subjects: the same six who were shown in Rose and Duncan (1995) to have voices similar enough to be misidentified even by listeners who knew them very well. Subjects were recorded saying several tokens of hello as they thought they might say it under different conditions (e.g. answering the telephone, passing someone in the corridor). This provided a set of utterances with realistic within-speaker variation. The magnitude of within-speaker variation for a given feature, e.g. F1 in /l/, is estimated in order to specify under what conditions, and with what probability, an observed difference in a feature can be said to reflect provenance from different vocal tracts. The magnitude of between- speaker variation is then estimated, which, in conjunction with the within-speaker estimates, allows us to say whether it is practical to use an acoustic feature approach in forensic speaker identification.

P. Rose & S. Duncan (1995) “Naive Auditory Identification and Discrimination of Similar Voices by Familiar Listeners “to appear” in Journal of Forensic Linguistics 2(1).

Kate Storey, Monash University

The history of a transcript

This paper describes the various stages involved in the preparation of 2 independent transcripts of a very poor quality covert tape recording of a conversation between several people and conducted in a number of difficult- to-identify languages. Some observations are made about the problems of forensic foreign language transcription, and some approaches to dealing with these problems are suggested.


James Stratman and Patricia Dahl, University of Colorado at Denver

Readers’ Understanding of Temporary Restraining Orders enforced in Domestic Violence Cases: an Empirical Study

The purpose of this study was to explore how ordinary readers comprehend the language in Temporary Restraining Orders (TRO”s) recently used by the State of Colorado in its efforts to curb domestic violence.

A common courtroom defence for violating TROs is the defendant”s claim that he or she “didn”t know” the order restricted certain actions (such as contacting the complainant through a family member or friend), or otherwise did not comprehend the language of the TRO form. While judges and members of the Colorado bar have been asked to review the current TRO forms and to make suggestions for improving their clarity (i.e., thus constituting an “expert-only” test audience), no independent comprehension tests of the TRO forms on a population of ordinary, non-legally trained readers has taken place. Indeed, though the texts of the Colorado TROs we examine attempt to be exhaustive in identifying prohibited actions, the prohibitions are conveyed in very broad language, and obviously not all prohibited actions or scenarios can be listed on a two page form. Persons served the form therefore are expected to make accurate inferences about actions which, though not explicitly described or listed in writing, are nevertheless ones that courts and complainants seek to prevent.

To investigate ordinary readers” ability to correctly draw these inferences and apply them to actual situations, we have collected both “scenario” comprehension data and concurrent think-aloud data. Our results suggest that ordinary readers have quite specific difficulties in extending the broad TRO terms regarding permissible “contact” between parties to concrete situations, and that some of these difficulties may be gender based.

Michael Walsh, University of Sydney

Transcripts and traditions: Problems in recording land claim proceedings in Northern Australia

This presentation will detail some of the inadequacies of the transcripts of Aboriginal land claim proceedings in the Northern Territory. It will be shown that land claim proceedings rely on a partnership of Aborigines, lawyers and anthropologists in which there are many opportunities for misunderstanding. Some transcript “repair” can be carried out immediately (by, for instance, rephrasing a question to a witness) but much of the transcript is produced by people unfamiliar with the situation at the time of recording and physically far removed from it. Some transcript inaccuracies result from the difficult recording circumstances since much of the evidence is collected on site visits so that there is not only background noise but a degree of informality that would rarely be tolerated in a courtroom. But the more interesting problems arise because of the range of expectations brought to the land claim proceedings by the range of players: Aborigines, lawyers and anthropologists. A detailed examination of the transcript reveals how these differing expectations are reflected in the language used.

Weiping Wu, Center for Applied Linguistics, Washington

Language and law: A data approach to linguistic issues in the legal field

As an academic discipline, the study of language and law seems to have come of age after more than a decade of growing. Now that there is the organization (IAFL), the journal (JIAFL) and the fairly regular conference on the international level, the grope for a systematic approach, even if just for the benefit of those who are interested but are not yet “in”, would be worthwhile.

Described in this paper is a data-approach in studying linguistic issues in the legal field. Among the few systematic studies of issues in language and law, Levi”s categories in her bibliography provides a fairly comprehensive view of the field from a linguistic perspective. Since the field by definition involves both linguistics and the law, it is also conceivable that one can approach the same issues as seen from the eye of law, though we are still waiting to see a systematic approach from the legal perspective to the field comparable to Levi”s approach.

The data-approach as described in this paper offers an alternative to studies of languages problems in the legal field and provide a relatively easy access to this complicated field of study. Such an approach divides the field into three parts according to the data being analyzed, which is broadly defined as the object of research. Thus we have written data, which includes the study of written language for legal purposes in terms of its comprehensibility, clarity and style; spoken data, which includes any oral communication in the legal process, presentations by attorneys, witnesses and judges, conversations presented as evidence in court, and so on; and bilingual data, which covers the special area at which two or more languages intersect in a legal setting. It is due to the special complications involved in dealing with bilingual data in legal issues that one single category, bilingual data, is set aside, and not merely for the convenience of categorization.

Compared with the open-ended lists in the linguistic and the legal approaches, the data-approach offers more closure because all issues can be put under one of the three categories. Important studies in each division, as well as other pros and cons of such an approach, will also be discussed in this paper.




The University of Tennessee, Knoxville. Big Orange. Big Ideas.

Knoxville, Tennessee 37996 | 865-974-1000
The flagship campus of the University of Tennessee System