A. Book
Identity
Title : The Handbook of
Discourse Analysis
Edited by : Deborah Schiffrin, Deborah Tannen and Heidi
E. Hamilton
Publisher :
Blackwell Publisher
Print
publication date : 2003
Page : i-xx, 817
Review part
III page 437 ( Discourse Analysis in the Legal Context
) by Roger W.
B. Introduction
One
of the defining characteristics of discourse analysis is that it is capable of
application in a wide variety of settings and contexts. Wherever there is
continuous text,written or spoken, there is a potential analysis of such text.
The area of law provides an open opportunity for discourse analysis, especially
since law is such a highly verbal field. It is generally regarded as a field
containing written discourse, for care is taken to record in print all oral
interactions that occur in court. Cases are preserved in written form to serve
as the basis for later decisions and to record the cases for later review. Law
libraries, therefore, house immense collections of written text, such as motions,
counterclaims, and judges’ opinions, but they also contain spoken words, transcribed
in writing, such as trial testimony, questioning, and argument. Law, therefore,
is
a fertile field for discourse analysts.
C. Content
Analysis
1. A
Brief History of Discourse Analysis and Law
Forensic
linguistics is a somewhat newly recognized subfield of study, having spawned its
own academic organizations and journal only recently. In the 1990s, forensic
linguistics, in the broader sense, seems to have flowered, with important general
collections of articles on language and law (Gibbons 1994; Levi and Walker
1990; Rieber and Stewart 1990), and books on the language of the courtroom
(Solan 1993; Stygall 1994), bilingualism in the courtroom (Berk-Seligson 1990),
and aircraft communication breakdown (Cushing 1994). Discourse analysis plays a
role in these studies, but
it
is not the centerpiece of these works.
There
were, of course, instances of the application of linguistics to law much
earlier than this. Individual linguists have been called upon to assist
attorneys for many years, but, as far as I can tell, without much
documentation. For example, I know from personal correspondence that the late
Raven I. McDavid, Jr., was used by Chicago area lawyers to help with the
identification of dialects of defendants in law cases. There were probably
other linguists used in the same way throughout the years. During the 1960s,
linguists were called upon to assist both the government and local
and
state school systems in interpreting and evaluating issues related to new laws
on bilingual education and desegregation. Again, official documentation of such
consultation is either nonexistent or spotty.
Before the 1980s, it is
clear that linguists who engaged in such work did so as a side-issue application
of their primary work as dialectologists, phonologists, syntacticians,or, in
some cases, applied linguists in the most general sense.
2. Using
Analysis to Analyze Criminal Cases
2.1
Using familiar tools to analyze criminal cases
2.1.1
Topic and response analyses
One
of the early uses of discourse analysis in criminal cases involving tape
recorded evidence appears to be Texas v. Davis in 1979 (Shuy
1982). T. Cullen Davis was a Fort Worth oil millionaire who was accused of
soliciting the murder of his wife. The government used undercover tape
recordings of conversations between Davis and an employee to attempt to show
that Davis indeed solicited murder. But the tapes had some very odd qualities.
For one thing, topic analysis showed that Davis never brought up the subject of
killing, casting doubt on this as Davis’s agenda in those conversations.
Response analysis showed that when the topic of murder was introduced by the
undercover employee, Davis responded with no agreement and, in fact, no
recognizable interest in the matter. His response strategies were to change the
subject, say nothing at all, or offer only feedback marker “uh-huh” responses.
One battle in court concerned the meaning of these “uh-huhs,” the prosecution
arguing that they signaled agreement with the employee’s offer and the defense
arguing that
they
indicated only that Davis was listening, but not agreeing, to what the other
man was saying.
The
context of the event also shed some light on Davis’s verbal behavior. Davis had
just been acquitted in a trial in which he had been accused of breaking into
his own home, wearing a ski mask, and killing his wife’s boyfriend. After his
acquittal, Davis, perhaps understandably, brought divorce proceedings against
his wife. During these proceedings, Davis heard, correctly or not, that his
wife was running around with the judge in the divorce trial. To obtain evidence
of this, Davis asked that employee to spy on his wife and catch her with the
judge. The employee went to the police and told them that Davis had asked him
to find someone to kill the wife and the judge. The police then wired the
employee with a tape recorder and sent him to get the verbal evidence on Davis.
This produced two brief meetings, both requested by the employee, in which the
two men sat in a car and talked. The employee carried a gun (not uncommon in
Texas) and had a black belt in karate. Davis, a slight man, appeared nervous
throughout.
The
“smoking gun” evidence held by the prosecution was a passage on one of the tapes
in which the employee reports to Davis, “I got the judge dead for you.” To
this, Davis is alleged to respond, “Good,” followed by the employee saying,
“And I’ll get the rest of them dead for you too.” These words do indeed appear
on the tape but not in response to the employee’s statement, as the
government’s own evidence would show. As it turns out, the police not only had
the employee wear a mike but also made a videotape of the meeting, taken from a
van parked across the parking lot. Correlation of the voice tracks of the
audio- and videotapes indicated that Davis was getting out of the car as they
were discussing the employee’s boss, a man named Art. As he got out of the car,
Davis continued to talk about Art while the employee, anxious to get incriminating
evidence on tape, talked about getting the judge and
others
dead.
2.1.2 Speech act and
pragmatic analysis
Speech
act analysis has been especially helpful in cases involving alleged bribery. A
classic example, again in Texas, involved the charge that a state politician
had agreed to accept money in exchange for switching the state employee
insurance program to a new carrier. But was the speech act of offering the deal
what the prosecution said it was? First of all, it was couched in the perfectly
legal context of an offer to save the state money by getting a better insurance
contract. Then suddenly the agents made a second offer, for a campaign
contribution of $100,000 (perfectly legal at that time in that place), to which
the politician replied, “Let’s get this done first, then let’s think
about that.” The agents then upped the ante, saying, “There’s $600,000
every year . . . for whatever you want to do with it to get the business.” To this,
the politician replied, “Our only position is that we don’t want to do anything
that’s illegal or anything to get anybody in trouble and you all don’t either.
This [the insurance plan] is as legitimate as it can be because anytime
somebody can show me how we can save the state some money I’m going to be for
it.” As for the campaign contribution, the politician accepted it as a legal
campaign contribution and clearly said that he would report it. The agent urged
him not to do so.
2.2
Newer areas for discourse analysis
2.2.1
Voice identification
Controversy
over the validity of voice identification led to the modern era of scientific
voice analysis (Tosi 1979). Today, those interested in the field that has come
to be called forensic phonetics can benefit from starting their reading with Baldwin
and French’s Forensic Phonetics (1990), Hollien’s The Acoustics of
Crime (1990), and a special issue of the journal Forensic Linguistics (vol.
3, number 1, 1996). What can discourse analysis add to the issue of identifying
the voices of otherwise unidentified speakers? The need to identify voices on a
tape recording is not always limited to the types of cases normally examined by
forensic phoneticians. For example, in a typical criminal case in which tape
recorded conversations serve as evidence, not only must the words of each
speaker be identified and transcribed, but also the speakers must be identified
accurately. In most cases, this is not too difficult, especially if there are
only two speakers on a tape and those two speakers have distinctively different
voices. But when there are multiple speakers, things get complicated. And when
some of the multiple speakers have, for example, equally deep voices, the same
southern dialect, or other speaker attribution similarities, attention must be given
to other voice identification features. Complicating matters even further for the
use of forensic phoneticians, accurate voice identification usually requires a
tape recording which is of good enough quality to be submitted to sophisticated
spectographic instrumentation. This rules out many, if not most, surreptitious
tape recordings made in criminal investigations, since such recordings are done
under less than optimal laboratory conditions.
2.2.2
Defamation
Defamation
law and dictionaries are in agreement with the definitions of both “opinion”
and “fact,” but both law and lexicography are predictably silent about their
linguistic structure. Yet it is the discourse structure of the language used as
evidence of defamation that is often most crucial to the resolution of the
case.Defamation is an extremely sensitive area in which to cite actual cases.
Therefore, the following examples will protect participants by maintaining
anonymity with pseudonyms. A defamation case was brought by Roy Harris against
a television station which, he claimed, went beyond calling him a suspect to
accusing him of committing the crime in news segments of two different
programs. In most of the two programs, Harris was consistently referred to as
“the only suspect” or “the one and only suspect.”
In
one of the broadcasts, the police investigator said, “The suspect went directly
into the house, into the kitchen, and shot the victim in the head.” Elsewhere
in the programs, Harris was said to be the only suspect. Now we are told that
the suspect shot the victim. Put these together and one can easily understand
the program to be stating as a fact, not opinion, that Roy Harris killed the
victim. This referential definition was overlooked by both the plaintiff and
the defense, until the linguist called it to their attention. Referential
definition is not the only discourse analysis procedure found useful in
defamation cases. Discourse framing, for example, also played a role in the
Harris case. Television news programs characteristically frame their stories
with introductions and conclusions that relevantly focus on the specific news
item. In this case, the introductory frame went a bit beyond this, as follows:
(1)
Female announcer: During the past few weeks, you’ve probably heard about
thelatest in the murder of a suburban Kenmore housewife. Male announcer:
A husband and his one-time girl friend have been indicted for murder in that
case. Well, tonight’s special examines another case where the victim’s husband
is coming under close scrutiny. Here the murder story frame makes use of an
analogy. An analogy is defined as an inference that if two or more things agree
with one another in some respects, they will probably agree in others as well.
Thus by using the analogy of the husband’s indictment in the prior Kenmore
murder with the current Harris case, the discourse frame encourages the
inference that these two separate and unrelated cases are alike even though
Harris was never indicted. The use of the discourse marker, “Well,” uttered
with a lengthened vowel, signals that what follows has semantic cohesion with
the Kenmore murder. The male announcer’s use of “another” strengthens this connection.
The use of analogical discourse framing encourages listeners to infer that Harris,
like the Kenmore husband, is more than just a suspect. As with other areas of
the legal context, discourse analysis has been underutilized in defamation
cases so far.
3
Using Criminal Cases to Address Linguistic Problems
3.1
Discourse analysis and intentionality
One
way to determine the intention of people is to simply ask them about their intentions.
But the possibility of getting an accurate and truthful answer in a court case
is diminished by the fact that participants naturally want to protect their own
best interests. Short of inventing a machine that gets into the mind and
captures actual intentions, the topics one introduces in a conversation come
closer to indicating agendas or intentions than anything else. One should be
very careful not to claim that these clues to intention are actually the real
intentions. But real intentions can certainly be inferred justifiably from
them. Likewise, the responses people make to the topics of others can also
provide clues to their intentions.
3.2
Discourse analysis and ambiguity
Ambiguity
in the use of language is often thought to be the sole province of semantics.
Discourse ambiguity, however, is equally present in both written and spoken
language. The sequencing of discourse can create an ambiguity that is not always
immediately apparent in the individual words or sentences. Many criminal law
cases center on the words used by the participants. Elsewhere I have referred
to such as language crimes (Shuy 1993). That is, there is no physical damage
done to victims, such as robbery, murder, or assault. Such crimes are based solely
on the language used in cases involving bribing, buying or selling illegal
property or substances, illegal soliciting of various sorts, extorting, and
conspiring to do something illegal.
3.3
Discourse analysis and stylistics
Stylistic
analysis is the examination of the characteristic use of language features by a
given writer or writers. The analyst reviews the material presented, written or
spoken, and compares the text of unknown origin with that of the known. Such comparison
focuses on language features of which speakers or writers have little or no
conscious knowledge or control as they speak or write. For example, writers
have rather high levels of consciousness and control over vocabulary choices
but considerably less consciousness and control over their grammar, spelling,
or punctuation
patterns.
Discourse style is another language feature of which most speakers and writers
have little or no conscious awareness or control.
The
contribution of stylistics to the broad field of forensic linguistics is well
documented by Gerald McMenamin’s comprehensive text Forensic Stylistics (1993),
a book which is an excellent resource for this field. The major thrust of
forensic stylistics, however, has been the study of linguistic forms, such as
vocabulary, grammatical categories, syntax, punctuation, and length of
expressions or text, rather than discourse style. The latter has been dealt
with ably in nonforensic contexts (Tannen 1982, 1984; Brown and Yule 1983),
especially with written text, but the application to forensicdiscourse is only
recently beginning.
The
use of discourse markers (Schiffrin 1987), for example, has served as an
identifier of an individual’s style in at least one known legal dispute
(Katherine Thomas, pers.comm.). The analyst was able to determine that the
speaker in question was not theone suspected by comparing that person’s use of
discourse markers in known speech samples with the sample in question. Other
indicators of discourse style are available for similar analysis and
comparison, such as organizational style, patterns and types of register
shifting or mixing, patterns of sequencing given versus new information (Brown
and Yule 1983), the use of cohesion (Halliday and Hassan 1976; Scinto 1986),
and many others.
4.
Directions and Future Connections
The
legal context appears ot be just beginning to take advantage of discourse
analysis to help unravel the complexities of litigation. Whether the language
evidence is written or spoken, whether the case is criminal or civil, and
whether the analysis is done for the defense, prosecution, or plaintiff,
discourse analysis has a bright future in legal disputes. Issues of
intentionality, ambiguity, stylistics, voice identification, defamation,
bribery, solicitation, and many others provide a vast arena for linguists to
explore the uses of these, and other, aspects of discourse analysis. The field
of law seems to be more and more open to such assistance (Wallace 1986). It is
up to linguists to respond.
