"Linguists Could Provide Insights into Abscam Tapes."
31 August 1981, p. 12.
As a lawyer who is also trained as a linguist, I am concerned about the evidence that went to the jury in the Abscam cases. Those cases in which the defendants were convicted are now before the trial judge again, on post-trial motion, or are on appeal. They will certainly come in due course before the Supreme Court. I have the greatest respect for the attorneys who represented the defendants in the Abscam cases. I respectfully suggest, however, that linguistic analysis of the taped conversations in those cases might raise new issues of due process. Indeed, it might raise new issues of scienter and mens rea, and of simple innocence or guilt.
The facts of Abscam are by now well known. Agents of the Federal Bureau of Investigation, aided by a convicted swindler named Mel Weinberg and others, ranged up and down the East Coast acting the parts of rich Arab sheiks and their men, eager to pay American officials for help with various investment projects. If the officials were congressmen and senators, the Arabs offered to pay for legislative help with immigration problems.
Although there have been a few acquittals, 17 persons, including seven members or former members of Congress, have been convicted on one or more of the charges brought. These include bribery, 18 U.S.C. § 201(c), accepting a gratuity, 18 U.S.C. § 201(g), aiding and abetting, 18 U.S.C. § 2, and conspiracy, 18 U.S C. § 371.
The prosecution depended on videotapes and audiotapes for the proof of its cases. On each videotape, it was argued, the defendant could be seen and heard accepting a bribe or agreeing to accept a bribe or both. Without those tapes, and the transcripts made from them, the government's cases would have been wholly different.
None of the defendants called a linguist as a technical witness. It may be indicative of the potential usefulness of a linguistic defense, however, that the only congressional defendant to be acquitted on even one charge, John Murphy, made several arguments of a linguistic type to the jury. One must assume. however, that the defendants will present to the courts of appeals legal arguments based largely on theories of entrapment and due process.
Technical linguistic analysis of the taped evidence would have been advisable, first and principally, because these conversations were not
These government-instigated conversations are more like high-
pressure interviews or sales sessions.
normal conversations. The fact that the government videotaped or audiotaped and had an agenda for the critical sessions makes these conversations different from normal conversations in specific, measurable ways. These differences distort the perception of these conversations by untrained persons unaided by expert analysis.
As a linguist I know that it is relatively easy for one party not only to impose his own agenda on the conversation, but to produce almost any impression on a transcript of the conversation. Few of us speak with one eye on what a resulting transcript might look like. Nor have we come so far from the civilized view that gentlemen do not eavesdrop that we must take the invasion of privacy for granted in daily life.
People merely hearing tapes of artificial conversations, however, apply normal standards of interpretation to those tapes. The simplest and the most common false impression that a speaker with an agenda can convey is that everyone in a conversation agrees with the other speakers, when in fact there is no agreement at all. Simply to bring up a topic a number of times can produce the impression on a transcript that everyone agrees about that topic. Linguists call this contamination.
Let me illustrate contamination with a type of example not found in the Abscam tapes, to my knowledge. Suppose that there are two participants in the conversation, one with an agenda, the other without. If the speaker with the agenda vigorously attacks and slanders an absent third party, and the other party remains silent, the venom of the agenda will contaminate the entire conversation: someone overhearing the conversation or reading the transcript, but subjecting it to no analysis, may very well conclude that the two speakers agree about the defects of the one who is absent. In fact, the silent participant may be silent not out of agreement, but out of politeness, boredom, torpor, or good judgment.
One of my fellow linguists, Dr. Roger W. Shuy of Georgetown University and the Center for Applied Linguistics, has analyzed government tapes and transcripts in numerous criminal cases, including the Texas Brilab cases. Dr. Shuy has kindly provided me with his unpublished research papers on the role of a linguist as an expert in a criminal case. The following summary is taken from those papers, in which he describes the types of technical analyses he has used. (Dr. Shuy has changed the names or the parties.)
© Copyright Mary Campbell Gallagher 1981. All rights reserved.
Topic-comment analysis. Dr. Shuy points out that in normal conversations participants are not only expected, but indeed required, to introduce approximately equal numbers of topics. Conversations instigated by government agents rarely have that characteristic. In normal conversations, moreover, one does not keep bringing up the same topic over and over again if the listener does not respond to it.
Again, since these government agents have definite agendas of what they must accomplish in each conversation, principally getting the other party to say that he will commit a crime, these conversations differ from normal conversations. The government agents must often bring up the same topic over and over again, trying to get the other party to respond.
In one case, the government alleged that the defendant had tried to persuade an informer, Foster, to murder not only the defendant's wife but the judge in their divorce case. The government equipped Foster with a body microphone and sent him off to get the defendant to commit himself on tape to plans for the putative crime.
The defense, using Dr. Shuy’s analysis of the resulting tapes, argued at trial that if the defendant were the instigator of a criminal plot, it was odd that of the 22 topics on a 20-minute tape, it was Foster who brought up 13 of the topics. Only Foster, furthermore, brought up the topic of "doing Gwendolyn and the judge." Moreover, he brought it up over and over again. The argument for the defense was persuasive.
Response analysis. When the speaker who has an agenda for the conversation dominates and controls the conversation, it is particularly important to analyze the responses of the defendant. It is easy to confuse vague or ambiguous responses or placemarkers ("uh-uh") with something totally different--resolutions of the topic. This is particularly true where the entire conversation is dominated—and contaminated--by the sole topic that the government agent keeps bringing up over and over again, namely, the commission of a crime.
Referencing analysis. Dr, Shuy’s "Arthur Jones" case illustrates his use of still a third tool. An FBI agent disguised as a foreigner, Herb Colier, introduced almost all of the topics at the crucial videotaped meeting. To Colier's apparent chagrin, whenever he mentioned his offer to pay Jones money, Jones responded off-topic. Of the 23 introductions of topics in the tape, according to Dr. Shuy's analysis, 11 are Colier’s introduction and repeated reintroduction of the money topic. Despite Jones's obviously wishing not to deal with the money topic, however, and despite his constantly changing the subject--saying, even, that he wanted nothing to do with the money--Arthur Jones took $25,000 in his hands, and the videotape plainly shows his doing so. Dr. Shuy, however, has analyzed the references of pronouns throughout the conversation, and he argues cogently that although Jones in fact did touch the money, he believed that he was taking it for someone else, namely, his friend Earl Walker, and not accepting it for himself.
I have other doubts about Abscam. I wonder about the effect of videotape evidence supporting the prosecution’s case, when there are no videotapes of the events the defense might find useful. Jury members, after all, are largely persons who long ago gave up listening to conversations in favor watching television. I have serious political doubts about Abscam--doubts that are not wholly assuaged by the present FBI guidelines designed to guard against entrapment. If no demonstrated predisposition to crime is required to bring an innocent person within the scope of such an investigation, why not subject your political enemies to overwhelming temptation, with the cameras and tape recorders grinding all the while? Consider the fact that, guilty or not, former Rep. Richard Kelly was offered money nine times before he touched it. Keep at it and you, too, can bring your enemies to ruin. This is the lesson of Abscam.
My doubts as a lawyer and linguist, however, concern the distortions produced in conversations when one speaker has a definite, irrepressible agenda. These government-instigated conversations are less like conversations than like high-pressure interviews or sales sessions. Linguistic analysis of the evidence could cast light on questions of criminal guilt or innocence and produce results in the interest of justice.
© Copyright Mary Campbell Gallagher 1981. All rights reserved.