L
N
C
O
Journal Name
0 0 2
Operator: Simon Wang
Dispatch: 05.01.07
PE: Andrew Humphries
M anuscript No.
Proofreader: W u YM
No. of Page: 15
Editor: M andy Tang
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
1Language
in the American Courtroom
Roger W. Shuy*
Abstract
PR
O
O
F
Georgetown University
U
N
C
O
R
R
EC
TE
D
One of the most promising connections that linguistics can make to other fields is
to the legal arena, primarily because much of the work in law is done in language
and it is often largely about language, either spoken or written. For example, lawsuits,
indictments, pleadings, briefs, legal opinions, and, of course, laws and statutes are
all revealed and preserved in written language. Courtroom appearances and
testimony, although presented orally, end up in written form and even oral evidence
gets transformed into written transcripts. We begin with the observation that
language and law are close bedfellows.
1. The beginnings. The development of any new area of any discipline
usually begins when practitioners of that field get involved with it in a
noticeable way. In forensic linguistics, this development started when various
linguists made appearances in the courtroom setting. But even “courtroom
setting” has different implications and meanings. For example, in Europe,
it is often judges who call upon experts to analyze evidence. Expert reports
are read by judges, who may, but usually do not, ask the experts to testify
about their reports at trial. This contrasts with the common procedure in
the United States, where experts are called by one side of the case and give
testimony on behalf of that side.
Whether the European or American system obtains, linguists appear to
have begun getting involved in court cases in earnest sometime around the
early 1980s. Hannes Kniffka wrote expert linguistic reports in Germany
during that period (Kniffka 1996), as did Malcolm Coulthard in England.
Roger Shuy has testified at US trials since 1979. Public records are not
available about the use of forensic linguists in other parts of the world but
it is likely that they came later than the early 1980s.
History is unclear about exactly when linguists began to work with lawyers
on specific law cases, but we have written evidence that one of the first
Western linguists to get involved was Jan Svartvik, who published an analysis
of the altered police statements in the case of Timothy John Evans, who
was hanged 15 years earlier (and posthumously pardoned) for murdering his
wife and baby (Svartvik 1968). By the 1970s and 1980s many linguists had
gotten involved in law cases in both Europe and the United States, and
today this group is growing rapidly all over the world. There is now an
© Blackwell Publishing 2007
2 . Roger W. Shuy
PR
O
O
F
International Association of Forensic Linguists (http://www.iafl.org/), which
has a biannual journal, The International Journal of Speech Language and the
Law. Regional associations are appearing as well, as in Europe and China.
Commonly, when a new field of study develops, it takes several decades
for a comprehensive introductory text to be written. Sociolinguistics, for
example, now has several good introductions in print but it took some 20
years for the field to sort itself out enough for this to happen. Today there
are three introductions to forensic linguistics available (McMenamin 2002;
Gibbons 2003; Olsson 2004). All contain useful introductions to parts of
the field but by no means cover all of it. Because the field is still relatively
new, it would appear that the breadth of forensic linguistics has not yet been
fully represented.
U
N
C
O
R
R
EC
TE
D
2. Definitions. The term forensic linguistics has grown to be the common
current name for such work. One could argue that the field is really applied
linguistics or, for that matter, simply linguistics, because in order to be an
expert witness, as the courts require, one should be an expert in the field
upon which one testifies. The field is linguistics. An expert is someone with
advanced training and skill, usually with a publication record as well. To
some, it can seem a bit odd to call the field forensic linguistics since this
identifies only the content area to which linguistics is applied. But the name
has stuck just the same.
Forensic linguistics covers a wide range of topics, including the language
used in trials by judges, lawyers and witnesses; the language of the law itself;
the language used in civil cases; and the language used in criminal cases.
Linguists have testified and consulted in virtually all types of civil cases,
including trademark disputes; product liability (usually on warning labels or
hazard statements); deceptive trade practices; discrimination by age, gender,
race, and employment; defamation (slander and libel); business fraud; contract
disputes (including the fertile area of insurance policies); and copyright
infringement. Criminal cases in which linguists have testified or consulted
include confessions, police interrogation, child sexual abuse interviews,
perjury, adult sexual misconduct, price fixing, money laundering, illegal
purchases and sales, and trade secrets. Some cases overlap the usual
civil/criminal categories, such as authorship analysis and speaker
identification.
Linguists call on their training in the tools of phonetics, phonology,
morphology, syntax, semantics, pragmatics, speech acts, discourse analysis,
sociolinguistics, dialectology, lexicography, language learning, and the
processes of language change, depending on the nature of the law case.
In civil cases, these tools are used to unravel or identify ambiguities, to
analyze words and expressions revealing discrimination, to determine
potential meanings of words and expressions used, and to show linguistic
similarity or differences in cases involving trademarks and alleged copyright
infringements.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 3
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
In criminal cases in which tape-recorded oral language is part of the
evidence, the linguist’s first job is to be certain that any written transcripts
accurately represent the spoken language. Because many such tapes are of
conversations, confessions, and police interrogations, discourse analysis is
helpful to reveal language structures and features that the jury might easily
miss; to assist the jury keep track of the themes, topics and agendas of the
speakers; to point out ambiguities; to identify conversational strategies of
the speakers; and to show how clues to the intentions of speakers are revealed
through what they say (Shuy 19942, 2005). For example, the topic of
“investment” was crucial in the case of US v. John Z. Delorean. Analysis
of the 63 tapes made by undercover agents in an effort to get Delorean to
discuss his alleged “investment” in drug trafficking showed that when
Delorean used this term, his reference was to getting investors for his car
company but when the agents used this term, they referred to Delorean’s
possible investment in their ostensible drug enterprise. At the end of the
government’s most crucial tape Delorean had agreed that investment “was
a good thing.”The prosecution, believing that they now had him, indicted
him and brought him to trial. At court, discourse analysis of all 63 tapes
showed that Delorean and the agents were on separate topics.
3. General uses. Why do the courts need a linguist in the first place? Here
is where the “applied” part of applied linguistics makes sense. Lawyers,
judges and juries speak and write the language but cannot see it the way
specialists can. The analogy of a dentist’s X-ray suggests itself. The patient
sees the same X-ray that the dentist sees but the dentist is trained to see
structures and features that go unnoticed by the patient. Those who use
language expertly, such as lawyers and judges, simply do not know its
complexity. They can easily miss important features that go by quickly in
normal conversation or interrogation, and they often pay no attention to
things that linguists know about intonation, syllables, grammatical scope,
variability, or felicity conditions. The forensic linguist’s first task is to make
these and other things clear to the attorneys with whom they work. It is a
lot like teaching introduction to linguistics to intelligent students. When it
comes to teaching these things to jurors, the task can be even more difficult.
One has to do this from the listener’s perspective, starting from where they
are and not talking down to them while explaining such things as vowel
fronting, distinctive features, conversation strategies, or negative scope in
ways that they can understand.
4. Uses of linguistics in the courtroom. As mentioned above, linguistic
expertise is relevant and useful in three areas of the legal context: (1) the
language used by lawyers, judges and witnesses, (2) the language used as
evidence in civil disputes, and (3) the language used in criminal case
evidence. These categories are not always that discrete, however, because
some types of law cases can be categorized as either civil or criminal.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
4 . Roger W. Shuy
Defamation cases in Germany, for example, are often treated as a criminal
matter, whereas in the United States, they are almost exclusively a civil tort.
Likewise, authorship and speaker identification can occur in either civil or
criminal cases. Following is a brief and incomplete summary of the types of
work linguists do in these areas, limited only to books on the subject.
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
4.1 The language of law. An early book dealing exclusively with this topic
is David Mellinkoff ’s The Language of Law (1963), who is among the first
to point out that the legal lexicon is made up of a great deal of jargon (he
calls it argot) that labors under the false assumption that legal language is
precise. If it is truly the case that the language of laws and statutes is precise,
one wonders, along with Mellinkoff, why there are so many lawsuits over
meanings in such documents as contracts, agreements and jury instructions,
all of which are written by lawyers.
Mellinkoff opened the door to later analysts and critics of legal language,
including William O’Barr, whose 1982 book, Linguistic Evidence: Language,
Power, and Strategy in the Courtroom (1982) describes his research on lawyers’
use of language in legal tactics, their presentation style, and how they acquire
and amass power. Just Words: Language and Power, by John Conley and
William O’Barr (1998) extends the study of courtroom power further,
pointing out lawyers’ dominance strategies through the use of silence,
question forms, topic management, evaluative commentary and challenges
to witnesses. The authors also deal with the disadvantage of women in
mediation hearings and the masculine control of law in general. Peter
Tiersma (a law professor with a PhD in linguistics) wrote Legal Language
(1999), a current classic in legal language history and analysis. Tiersma gives
a descriptive history of how Anglo-American legal language developed and
continues to this day, including the efforts to try to bring about clarity and
precision of meaning, often unsuccessfully. It also contains valuable
suggestions for reform in this area.
4.2 The language of judges and lawyers. In addition to these important
overviews of legal language, several linguists have written books about the
language of judges. Conley and O’Barr’s book, Rules Versus Relationships: The
Ethnography of Legal Discourse (1990), describes research on the efforts of
litigants to make their case in small claims courts. Those who follow the
rules that reflect the language of the courts are successful. Those who describe
their problems using the language of broad social terms tend not to succeed.
Lawrence Solan (also a law professor with a PhD in linguistics) wrote his
The Language of Judges (1990) to show how various linguistic features and
terms are handled by judges, including “plain language,” the “and/or” rule,
and how antecedents and pronouns are used and interpreted by the courts.
For example, Solan outlines the confusion about the words “and,” “or,”
and “and/or” in the laws of various US states (45 –55). Susan U. Phillips’
book, Ideology in the Language of Judges (1998), is an ethnographic study of
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 5
PR
O
O
F
the power of judges who apparently believe that they simply follow and
apply the law in a neutral and apolitical way but who still fall deeply into
individual interpretation of it.
The language of lawyers during trials has also been researched by forensic
linguists. Gail Stygall’s Trial Language: Differential Discourse Processing and
Discursive Formation (1994) analyzes an Indiana trial, including juror
qualification, preliminary instructions, opening statements, questions and
answers, final arguments, and jury instructions. In a similar manner, Janet
Cotterill’s Language and Power in Court (2003) describes how story telling,
framing, cross examination, and reframing worked in the famous O. J.
Simpson trial.
U
N
C
O
R
R
EC
TE
D
4.3 The language of jury instructions. If the language of law itself and the
language used by lawyers and judges during trials is different from that to
which the nonlegal world is accustomed, judges’ instructions to the jury
should provide a golden opportunity for them to communicate clearly these
unfamiliar expressions and concepts to laypersons so that they can properly
do their job as triers of the facts. Unfortunately, such instructions are still
far from clear. For example, in American capital murder cases involving
death penalty statutes, juries have to consider both the aggravating and
mitigating circumstances.Tiersma (1999:234 –40) shows how this process is
seriously flawed, since most “courts assume that jurors understand these
terms in the legal sense.” Tiersma notes that in at least ten capital murder
cases in which jurors requested definitions of these terms the judges told
them to use the commonly accepted and ordinary meanings of these
words. The legal definition of aggravating includes acts such as torturing the
victims and the defendant’s past history of violent crime. The legal definition
of mitigating includes evidence of remorse, an unhappy childhood, or mental
problems. Even less helpfully, when juries request such definitions, the judges
often simply repeat the formerly given instructions. Gibbons (2003: 174)
points out that jury instructions are intended for two different audiences at
the same time. That is, they have to communicate to jurors but still be
worded in ways that will withstand legal criticism and appeals based on
that wording. Stygall (1994: 92) points out that jurors have problems
understanding instructions because of the form of the instructions themselves
and because of the lack of exemplification of difficult legal concepts. Jurors
even have trouble distinguishing one instruction from another and have
trouble discovering any kind of relevant ordering principles. A few linguists
have carried out research on jury instructions (Charrow & Charrow 1979;
Elwork et al. 1982) but there is much more work that linguists can do in
this area.
Although the work is certainly a helpful beginning step, research on the
language used by legal professionals, such as lawmakers, judges and lawyers,
is far from complete, as the authors of these works admit. But at least some
of the parameters for future studies have been well set.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
6 . Roger W. Shuy
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
4.4 The language of civil cases. One might think that hundreds of years of
trying to accomplish clarity and precision in the writing of documents such
as business contracts, warning labels on products, employment agreements,
product names, and industrial work proposals would lead to fewer rather
than more lawsuits. This has not been the case. It might be argued, perhaps
with some justification, that society has become more litigious, or that we
now have more lawyers than ever before to seek out new lawsuits, or that
the art of clear writing has descended to depths hitherto unknown. But these
hypotheses do not give us the whole picture. However, expert lawyers may
be skilled in the law and in using language to persuade juries to accept their
positions, they do not have the expertise in linguistics to help them to analyze
and discover how to fully address topics that linguists know about, such as
language ambiguity, references, and inferences. Most civil cases are duels
over words, yet many lawyers come to these duels with blunted swords. And
even when they recognize such language phenomena, they are ill equipped
to explain how they work. For this, they can use the help of linguists, for
a scientifically sound explanation trumps, amplifies, or at least assists persuasive
rhetoric and understanding the language used in the legal context.
Eventually humans discovered that writing down their spoken language,
stories, history, religious ideas, and legal affairs and actions could make them
more consistent and permanent. In the legal arena, written language was
first used to record what previously had been agreed upon orally. Later,
shortly after about 1600, the focus was turned to the written document
itself, leading to the parole evidence rule, which, as Tiersma (1999: 37) points
out, prohibited any reference or evidence about discussions that led up to
the written contracts. Some positive effects of this ensued, such as reducing
the opportunity to commit perjury about how the contract was understood
as well as the unstable reliance on memory (Solan 2001, 94). But can we
always rely on the written word in such cases? Solan (2001: 94) argues that
in many cases it may be necessary to discover the context in order to
determine sensibly any language that at first glance seemed clear, since
meaning is inevitably context dependent to some degree. The debate over
parole evidence continues to this day. When a contract is judged to be
ambiguous or incomplete, for example, lawyers can call on linguistic analysis
to help unravel meaning. It is obvious that linguists cannot, and should not,
claim to get into the minds of those who wrote and read the contract in
order to try to fathom what the writers intended or what the readers
understood. But analysis of the text of a message enables linguists to
determine the range of possibilities that the text could mean and the range
of possible understandings that readers could have of that language, no matter
what the signers thought or believed when they signed it.
Tiersma (1999) outlines some of the major areas in which improvements
can be made in contracts. Words and sentences can be shortened, active
verbs can replace passives, technical terms and jargon can be avoided, Latin
and foreign language words and expressions can be scuttled, definitions can
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 7
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
be added, sentences with more than one conditional clause can be eliminated,
and expressions that cite exceptions to exceptions can be clarified and
simplified. Legal language, like other forms of jargon, is perfectly acceptable
when used internally, but it can become very troublesome when used outside
the context of the inner group.
The major issue for linguists in a court case is not necessarily to improve
the language of contracts, but to figure out what the meaning is. One thing
a linguist easily can say about this is that the text is often written in a legal
register that nonlawyers are not likely to comprehend, one that may have
led to the dispute in the first place. For example, the Montana Department
of Revenue recently asked me to help the staff revise that state’s tax forms
and instructions. The department had spent many hours answering taxpayers’
questions about the meanings of various items on the forms. The state was
also involved in litigation about what some of the tax forms intended, a sure
signal that clearer texts were needed. A brief examination made it apparent
that a major problem was that much of the text was taken directly from the
statutes, resulting not only in legal problems but also the expenditure of
departmental time and effort trying to clarify things for confused taxpayers. As
a result, this was not a cost-effective situation for the state. At the federal
level, many similar examples of writing in the wrong register and the lack
of textual clarity that lead to litigation against the Social Security
Administration, Medicare/Medicaid, the US Food and Drug Administration
and the US Occupational Safety and Health Administration are discussed
in Shuy, Bureaucratic Language in Government and Business (1998a)3. Forensic
linguistics also has a role to play in preventing litigation from happening in
the first place.
Civil cases involving contracts often call for the use of such linguistic tools
as semantics, grammatical referencing and scope, and discourse structure. The
language of insurance policies is one frequent area where problems in these
areas occur. Writers of such policies may work very hard to make them
clear and precise, but sometimes they are unsuccessful, often because the
writers try to cram into few words what longer texts might make even clearer.
Contract disputes, of course, are not the only area of civil law in which
linguistics can play an important role. In product liability cases the language
of warning labels offers many opportunities for the use of linguistic expertise.
Such warnings often fail to be explicit about the dangers of the products,
instead containing indirectness or inferences. They also sometimes place the
major hazard deeply within the discourse, defusing its potential effectiveness
as a warning to consumers. Some are written in the legal or medical registers,
containing unfamiliar lexicon and deeply embedded syntax, rather than
being consumer friendly. Others fail to make clear the association of the
dangers with the product itself, choosing instead to describe the potential
illness or disease in detail but leaving out its explicit association. The linguistic
tools of semantics, discourse analysis, pragmatics and speech acts can help
identify such lack of clarity in the courtroom.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
8 . Roger W. Shuy
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
Lawyers handling trademark cases have been turning more and more to
linguists to help them with disputes about whether a trade name can be
protected. The basic categories of trademarks are: generic, descriptive,
suggestive, and fanciful or arbitrary. Generic refers to common dictionary
words used to describe a class of objects, like cheese, oil, salt, or cereal.
Descriptive words describe the qualities, ingredients or characteristics of a
product (e.g., Holiday Inn, Tasty Bread, or Tender Vittles). Suggestive
words are said to require some operation of the imagination to connect the
name with the product (e.g., Greyhound bus or Tide laundry soap). Fanciful
marks are words coined for the express purpose of functioning as a trademark
(e.g., Kodak cameras, Clorox bleach, or Xerox printers), arbitrary marks are
words commonly used in the language but which are arbitrarily applied to
a product in such a way that it is not descriptive or suggestive (e.g., Nova
television program, Camel cigarettes, or Apple computers), most trademark
cases involve the sometimes- murky boundaries between generic and
descriptive or between descriptive and suggestive. The basic questions involve
whether the name of the product that held the mark first (senior user) and
that of the more recent product (junior user), such as the senior Healthy
Choice and the junior Health Selections, sound alike, look alike and convey
the same meanings. As illustrated in Shuy, Linguistic Battles in Trademark
Disputes (2002), linguists who are asked to work on these cases can make
effective use of their expertise in phonetics, phonology, morphology, syntax,
semantics, pragmatics, and lexicography.
4.5 The language of criminal cases. Prior to the 1960s it was extremely difficult
for the courts to prove certain types of criminal activities unless the law
enforcement officer was present when the crimes took place or unless people
who witnessed them reported and testified about them. Tape recording
actual events as they happened changed all this. Federal wiretapping and
electronic surveillance became law in 1968 (Blakey 1993) and federal and
local law enforcement agencies began setting up sting operations in which
undercover police or cooperating witnesses (usually people who had already
been caught and then agreed to wear a body mike to tape others) engaged
in both face-to-face and telephonic conversations with targeted suspects.
Shortly thereafter, surreptitious videotaping began to be used for the same
goals and more recently other electronic language, such as Email messages,
have provided written evidence in criminal cases.
The original targets of such undercover operations were potential
white-collar criminals. Suspects were often public officials thought to be
giving, taking or conspiring to give or take bribes, engaging in business fraud
or deceptive trade practices, and involved in other crimes that are committed
by using language rather than physical violence to perpetrate the offense.
Shuy’s book, Language Crimes (1993), was among the first to refer to such
crimes with this expression.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 9
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
It was not long, however, before law enforcement agencies, especially at
the local level, began to tape record other types of language crimes, including
the purchase and sale of illegal substances, solicitation of various types,
especially sexual contacts, drug deals, prostitution and murder. Not
surprisingly, they found that taped evidence was more persuasive than
eyewitness accounts. In the early days, when audio recording equipment
and techniques were somewhat primitive, many of these recordings were
badly done. Even early videotapes were sometimes less than useful, because
of lighting problems in hotel rooms where the meetings took place and due
to camera angles, which often showed only part of the scene or only one
of the speakers. When linguists are called on in such cases, the first task
always is to use their skills in phonetics and language variability to make
sure that an accurate, jury-ready transcripts of the conversations were
made. After accuracy of the text is assured, linguists can determine such
things as whether an offer was actually made or accepted, whether a
legitimate threat was uttered, and whether a promise was involved. Linguists
have made good use of speech act theory in such analyses (Shuy 1993).
Videotapes offer more information than audiotapes, of course, since they
can provide evidence of who was actually speaking and whether the target
was within hearing distance – or even present – when incriminating
statements were made.
Identification of the conversation strategies of undercover police and
cooperating witnesses also calls for discourse analytic procedures (Shuy 2005).
It is common in undercover cases for agents to try to elicit evidence from
their targets of past or present criminal acts, or their predisposition to commit
such acts in the future. To do so, law enforcement agencies suggest that
their agents first let their targets talk uninterruptedly, thus incriminating
themselves. If this does not work, the next step is to drop hints suggesting
illegality, letting targets infer from these and incriminate themselves. If this
still does not yield incriminating information, the next step is to try to get
targets to retell some alleged past criminal action about which the agent has
allegedly “forgotten” the details and “needs to be reminded.” Failing in this
strategy, the last hope is for the agent to be clear and unambiguous about
the nature of the illegal enterprise. However, effective these four strategies
may be in theory, agents seldom follow them in this order. Impatient to nab
the suspect, undercover operators often skip the suggested first step and go
straight to one of the others, despite the fact that the best evidence is
self-reported. But some agents consider this step too time consuming. Even
more troublesome is when there is no representation of illegality at all,
leaving only the agents’ hints for the jury to hear on the tape. From these,
the jury also is then expected to infer illegality, even if the suspect showed
no up-take on the hints.
Gradually, as law enforcement agencies continued to record conversations
with targets, they also began to tape record events that occurred after suspects
were apprehended. Police interrogations, along with alleged confessions,
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
10 . Roger W. Shuy
began to be used in court proceedings. This practice opened the door to
linguistic analysis of these events as well (Shuy 1998b)4. Again, speech act
analysis was prominent, helping linguists address such questions as whether
the suspect actually confessed to the crime and whether the interrogation
procedures influenced what the suspect said. Linguists have also analyzed
how the police give suspects their Miranda rights as well as the nature of
responses to them.
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
4.6 The language of authorship analysis and speaker identification. A separate
category is used here for writer and speaker identification of unknown texts
and tapes because they do not fit neatly into the conventional categories of
civil and criminal cases. Identifying authors based on anonymous texts and
speakers based on unattributed speech samples can be an important part of
either civil or criminal proceedings, depending on the nature of the case.
Authorship analysis has a long history, primarily relating to literary texts.
In the forensic context, it is also very common, especially in Europe and
America (Kniffka 19955, Chaski 2001, McMenamin 2002; Coultard 2004;
Solan & Tiersma 2005). Here the source of texts is frequently threats, hate
mail, and suicide notes. In most cases these texts are very short, making the
certainty of identification difficult. In an investigative mode, sociolinguistic
research can help identify expressions that may suggest a writer’s region,
gender, age, occupation, and education level, but without a larger corpus
to work with, even these suggestions are necessarily tentative. The practice
is known to have helped law enforcement agencies narrow down lists of
suspected writers, but the handicap of limited amounts of evidence can make
it difficult, if not impossible, to be certain enough about the results to testify
about authorship at trial.
Standard procedure is to compare the unknown text with other
comparable samples of written texts of known authorship. Unless such
comparisons can be located, there is even less hope of certainty. Finding
comparable texts is made difficult by the fact that the register of threat
messages often differs considerably from the register of the comparison
documents. In cases where there is no comparable text available, all the
linguist can do is point out whatever indications there are about regional or
social variables in the writing.
The diagnosticity of linguistic features is also somewhat unsettled. Some
forensic linguists focus on such features as punctuation, spelling, vocabulary
choices, and grammatical aberrations in the writing. Others believe that
these features are not diagnostic enough, assigning more weight to syntactic
features, aspects of written text about which the writer is less likely to be
consciously aware. Even when the comparison texts are large enough to
yield reasonable statistical comparisons, it is not always clear exactly what
frequency of occurrence of a given feature distinguishes it from the same
feature in the unknown document. Suppose, for example, that the known
text shows a frequency of 65% and the unknown document shows 50% for
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 11
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
the same feature. Is that enough difference to show same or different
authorship? Where is that line drawn? Also, there is the problem of unknown
texts in which it appears to be likely that the writers were trying to disguise
their language. Little is known about how this works.
Speaker identification has been common since the invention of recording
equipment but the practice has met opposition in American courts for some
of the same problems that plague authorship testimony (Baldwin & French
1990; Hollien 1990, 2001; Solan & Tiersma 2005). Phoneticians have found
some acoustic characteristics that can help identify individual speakers in a
small set of data but the results have not been as successful for a large set of
speakers or when voices are disguised or recorded under different conditions.
Nor has the question been resolved about the statistical distribution of known
characteristics across the population at large. In comparison with DNA data,
with its huge data base for such comparison, a huge voice data sample has
not been compiled. The courts have generally agreed that “voiceprints,”
(the popular term for spectrograms) are not warranted at trial. This does not
mean that no phonetic evidence can be used at trial, because there is a great
deal that phoneticians can say about voices that untrained listeners cannot
know. At this point, speaker identification is primarily useful in the same
way that authorship analysis can be used – in the investigative stage, to help
narrow suspect lists and to advise about certain features outside of the
courtroom itself.
5. Testifying at trial. To qualify as an expert, the linguist must first be
one. This means having proper and extensive training in linguistics (usually
with a PhD) and being able to demonstrate this expertise according to the
standards of the field (usually with a relevant publication record). It does
not mean that the expert should have legal training, although such is not
harmful. The point here is that forensic linguistics is a field that requires
practitioners to take their linguistic knowledge, expertise and experience
into the legal arena and to apply the relevant tools of the field to the
particular language evidence in a civil or criminal case. For a practical guide
to linguistic expert witness process, see Shuy (2006).
As noted above, working with lawyers is not the same as working with
other people, because lawyers have considerable knowledge about how to
use language effectively but usually do not know how to analyze it the way
linguists can. And most linguists, of course, know little about law. So that
means that linguists and lawyers need to be frank, clear, and honest with
each other. Both need to be teachers. Both need to be students. Both need
to nurture each other about their contributions to the case.
In American civil cases, experts are usually expected to write a report
about their analyses. This report, which sometimes takes the form of an
affidavit, is given to the opposing side’s lawyer, who reads it and then calls
the linguist in for a deposition. This is a questioning event in which the
lawyer tries to understand what the report says and does not say. The expert
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
12 . Roger W. Shuy
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
is sometimes challenged, because one thing the opposing lawyer wants to
do is to impeach the expert’s method, findings or even integrity. The
deposition is not common in many other countries.
After the deposition comes the trial. First is the expert’s direct testimony,
in which the lawyer with whom the expert works asks the questions. This
is followed by cross examination by the other side’s lawyer. Because a judge
is present, trial testimony can be more controlled and civilized than many
depositions turn out to be.
A serious issue that American forensic linguists face is in getting
admitted as an expert witness. For 70 years, beginning in 1923, the common
standard of acceptability was that the proposed testimony had to be relevant
to the dispute, the theory or technique used had to be generally accepted
by the scientific community, and the witness had to be an expert. This
changed in 1993, when the US Supreme Court ruled in Daubert v. Merrell
Dow Pharmaceuticals; http://www.law.cornell.edu/supct/html/92-102.ZS.html
replacing the general acceptance test with a reliability assessment.
Four factors were now to be considered: whether the theory or technique
has been tested and found to be sound; whether it was subjected to peer
review; whether there is a high known or potential rate of error; and whether
the technique is generally accepted by the relevant scientific community
(Solan & Tiersma 2005; Shuy 2006). In 1999, in the case of Kumho Tire v.
Carmichael; http://www.law.cornell.edu/supct/html/97-1709.ZS.html the US
Supreme Court expanded these factors to apply to all forms of expert
testimony, not just to the narrow category of scientific tests. It is clear that
experts at one end of the spectrum of scientific “objectivity,” such as those
who do DNA testing, are judged acceptable. It is equally clear that experts
at the opposite end, such as astrologers and palmists, are rejected. Debate
continues about experts who fall between these extremes, including many
experts with social and human science expertise. Because some courts are
not clear about how to deal with experts in linguistics, linguists have to
educate the lawyers they work with about the scientific nature of the field.
6. The reaction of judges. While an increasing number of lawyers have
realized the need to get assistance from linguists in their cases, many judges
have been less receptive to the idea. It is difficult to learn why, but a number
of factors have been suggested.
For one thing, linguistics is still an unknown field to them. Lawyers have
invited linguists to their conferences and workshops at a far greater rate than
have judges. Many simply do not know what linguistics is and, even worse,
what it is not. This does not stop them, however, from sometimes claiming
that proposed linguistic testimony is likely to “go beyond the scope of
linguistics.” Judges seldom define what they mean by this but it is certainly
not based on most judges’ knowledge of linguistics. One possibility is that
the judges feel that the linguists are getting close to legal argument rather
than linguistic analysis. An equally common reaction of some judges is that
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 13
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
linguistic testimony is not necessary because the attorneys are perfectly
capable of introducing the findings of linguistics analyses by themselves.
Linguists clearly offer useful assistance to trial lawyers but some judges feel
that a linguist will cause them to have to deal with still one more expert
that will slow down their trials.
Second, judges are deeply concerned, as they should be, that linguistic
testimony might venture into the forbidden territory of usurping the role
of the triers of the fact. This is a serious problem, of course, but one which
competent forensic linguists know how to avoid. Nevertheless, the fear that
this might happen is sometimes quite enough for a judge to rule out
testimony by linguists, even without first hearing it.
Third, one of the greatest fears of judges, especially in bench trials, is the
possibility that their decisions might be overturned on appeal. The fewer
“unusual or new” types of evidence or expertise that are brought to bear in
a trial, the more likely their decision can stand. Part of this fear is probably
based on the fact that lawyers are actually very good at using language but
they confuse this competence with the ability to analyze it scientifically.
Medical, engineering, and other expert witnesses have been around a lot
longer and do not seem to be affected by the “newness” phenomenon. There
is the looming specter of violating precedent that permeates all of law. If in
a previous case a judge has ruled out using a linguistic expert, other judges
like to rely on this decision.
7. Final words. When lawyers call on linguists to help them with their
criminal or civil cases, they usually want an expert who will be ready and
willing to testify at a forthcoming trial. However, it is estimated that some
90% of civil cases are settled before trial and that an equal percentage of
criminal cases end with a plea bargain, guilty plea, or the indictment being
dismissed. This means that the forensic linguist must prepare in the same
way, whether or not the case goes to trial. At any rate, being a consultant
to lawyers can be every bit as useful as being an expert witness at trial. If
the linguist teaches the lawyer what is linguistically important about the
case, and if the lawyer represents this information effectively during cross
examination and arguments, the linguists’ tasks may be considered as
important as when they actually testify.
For many years, applied linguistics has focused primarily on language learning,
language teaching, and language testing. These are fine things to do but
there are many more areas of life where linguistics is relevant and useful to
society. Forensic linguistics is certainly one of these promising areas and it
is not surprising that more and more linguists are beginning to do this work.
Short Biography
Roger W. Shuy is Distinguished Research Professor of Linguistics, Emeritus
at Georgetown University. After receiving his PhD from Case Western
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
14 . Roger W. Shuy
PR
O
O
F
Reserve University, he taught briefly first at Wheaton College, then at
Michigan State University, and the final 30 years or his teaching career at
Georgetown, where he created and headed the sociolinguistics program.
His major research interests are in sociolinguistics and forensic linguistics.
His most recent books include Language Crimes (1996), The Language of
Confession, Interrogation and Deception (1998), Bureaucratic Language in
Government and Business (1998), A Few Months to Live (co-authored, 2001),
Linguistic Battles in Trademark Disputes (2002), Creating Language Crimes (2005),
Linguistics in the Courtroom: A Practical Guide (2006), and Fighting Over Words
(in press, 2007).
Endnotes
* Correspondence address: Roger W. Shuy, 629 Beverly Avenue, Missoula, MT 59801, USA.
Works Cited
U
N
C
O
R
R
EC
TE
D
Baldwin, John, and Peter French. 1990. London: Pinter Publishers.6
Blakey, G. Robert. 1993. Foreword to Roger W. Shuy. Language Crimes. Oxford: Blackwell.
Charrow, Robert, and Vida Charrow. 1979. Making legal language understandable: a
psycholinguistic study of jury instruction. Columbia Law Review 79.1306–74.
Chaski, Carole. 2001. Empirical evaluations of language-based author identification techniques.
Forensic Linguistics 8.1–65.
Conley, John, and William O’Barr. 1998. Chicago: University of Chicago Press.
——. 1990. Rules versus relationships: the ethnography of legal discourse. Chicago: University
of Chicago Press.7
Cotterill, Janet. 2003. Language and power in court. Houndmills, Basingstoke: Palgrave Macmillan.
Coultard, Malcolm. 2004. Author identification, idiolect, and linguistic uniqueness. Applied
Linguistics 25.431–47.
Elwork,Amiram, Bruce D. Sales, and James J. Alfini. 1982. Making jury instructions understandable.
Charlottesville: Michie Press.
Gibbons, John. 2003. Forensic linguistics: an introduction to language in the justice system. Oxford:
Blackwell.
Hollien, Harry. 1990. The acoustics of crime: the new science of forensic phonetics. New York:
Plenum Press.
——. 2001. Forensic voice identification. New York:Academic Press.
Kniffka, Hannes (ed.). 1996. Recent developments in forensic linguistics. Frankfurt: Peter Lang.
McMenamin, Gerald. 2002. Forensic linguistics: advances in forensic stylistics. Boca Raton: CRC
Press.
Mellinkoff, David. 1963. The language of the law. Boston: Little Brown.
O’Barr,William. 1982. Linguistic evidence: language, power, and strategy in the courtroom. New
York:Academic Press.
Olsson, John. 2004. Forensic linguistics: an introduction to language, crime and the law. London:
Continuum.
Phillips, Susan U. 1998. Ideology in the language of judges. New York: Oxford University Press.
Shuy, Roger W. 1993. Language crimes. Oxford: Blackwell.
——. 1998a. Bureaucratic language in government and business. Washington DC: Georgetown
University Press.
——. 1998b. The language of confession, interrogation and deception. Thousand Oaks CA: Sage.
——. 2002. Linguistic battles in trademark disputes. Houndmills, Basingstoke: Palgrave Macmillan.
——. 2005. Creating language crimes. New York: Oxford University Press.
——. 2006. Linguistics in the courtroom: a practical guide. New York: Oxford University Press.
Solan, Lawrence. 1990. The language of judges. Chicago: University of Chicago Press.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 15
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
——. 2001. The written contract as safe harbor for dishonest conduct. Chicago-Kent Law Review
77.87–120.
Solan, Lawrence, and Peter M.Tiersma. 2005. Speaking of crime: the language of criminal justice.
Chicago: University of Chicago Press.
Stygall, Gail. 1994. Trial language: differential discourse processing and discursive formation.
Amsterdam: John Benjamins.
Svartvik, Jan. 1968. The evans statements: a case for forensic linguistics. Gothenberg: Gothenberg
Studies in English 20.
Tiersma, Peter M. 1999. Legal language. Chicago: University of Chicago Press.
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
16 . Roger W. Shuy
Author Query Form
Journal: Language and Linguistics Compass
Article: LNCO_002
PR
O
O
F
Dear Author,
U
N
C
O
R
R
EC
TE
D
During the copy-editing of your paper, the follow ing queries arose.
Please respond to these by marking up your proofs w ith the necessary
changes/additions. Please w rite your answ ers on the query sheet if
there is insufficient space on the page proofs. Please w rite clearly
and follow the conventions show n on the attached corrections sheet.
If returning the proof by fax do not w rite too close to the paper's edge.
Please remember that illegible mark-ups may delay publication.
M any thanks for your assistance.
Query
Refs.
Query
1
Is this short title OK? Please
change another if not acceptable.
2
The bibliographic details for Shuy
1994 is not included in the list,
please supply.
3
A letter 'a' has been added to the
year of 1998 both in text and
bibliographic details, please
confirm that if this change is OK.
4
A letter 'b' has been added to the
year of 1998 both in text and
bibliographic details in the list,
please confirm that if this change
is OK.
© Blackwell Publishing 2007
Remarks
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
Language in the American Courtroom . 17
Query
Refs.
Query
Remarks
The bibliographic details for Kniffka
1995 is not included in the list,
please supply.
6
Please supply the book title for
Baldw in & French 1990.
7
Please supply the book title for
Conley & O’Barr 1990.
U
N
C
O
R
R
EC
TE
D
PR
O
O
F
5
© Blackwell Publishing 2007
Language and Linguistics Compass 1 (2007): 10.1111/j.1749-818x.2007.00002.x
MARKED PROOF
Please correct and return this set
Please use the proof correction marks shown below for all alterations and corrections. If you
wish to return your proof by fax you should ensure that all amendments are written clearly
in dark ink and are made well within the page margins.
Instruction to printer
Leave unchanged
Insert in text the matter
indicated in the margin
Delete
Textual mark
under matter to remain
New matter followed by
or
through single character, rule or underline
or
through all characters to be deleted
Substitute character or
substitute part of one or
more word(s)
Change to italics
Change to capitals
Change to small capitals
Change to bold type
Change to bold italic
Change to lower case
Change italic to upright type
under matter to be changed
under matter to be changed
under matter to be changed
under matter to be changed
under matter to be changed
Encircle matter to be changed
(As above)
Change bold to non-bold type
(As above)
Insert ‘superior’ character
Marginal mark
through letter or
through characters
through character or
where required
or
new character or
new characters
or
under character
e.g.
Insert ‘inferior’ character
(As above)
Insert full stop
Insert comma
(As above)
Insert single quotation marks
(As above)
Insert double quotation marks
(As above)
over character
e.g.
(As above)
or
or
(As above)
Transpose
Close up
Insert or substitute space
between characters or words
Reduce space between
characters or words
linking
and/or
or
or
Insert hyphen
Start new paragraph
No new paragraph
or
characters
through character or
where required
between characters or
words affected
and/or