LawyersUSA

Wife killer Brian David Hummert never intended to rat himself out. But in the end, his own written words – picked over by a forensic linguist – did just that.

Hummert was convicted of first-degree murder by a Pennsylvania jury in 2006 after Hofstra University linguistics professor Robert Leonard, serving as an expert for the prosecution, provided evidence that undermined a grand ruse that Hummert had created to cover up the crime. Leonard established striking syntactical similarities between Hummert's writing and a series of letters supposedly written by a man stalking his wife and a "serial killer" who wrote letters to police claiming credit for the murder.

"We noticed, after a lot of study, that there was a fairly interesting – and very rare – rhetorical device used in both the serial killer and stalker letters," Leonard told Lawyers USA.

The author of those letters had a "grammatical tic" of never contracting positive verbs. That is, instead of saying "I'm," he always wrote "I am." It was a practice shared by Hummert, the fictional stalker and the fictional serial killer. Based in part of Leonard's expert testimony, the jury Hummert convicted of murdering his wife and he is now serving a life sentence.

The Hummert case exemplifies the increasing role that professional linguists have in resolving legal disputes – civil as well as criminal.

Ronald R. Butters, who retired last year after 40 years as a linguistics professor at Duke University, has seen a marked change since he first began assisting lawyers 20 years ago.

"It's more of a serious endeavor now," he said. "It's not just a sideline where we made a few bucks on the side helping out some lawyers and getting to do something a little different."

Carole Chaski, a longtime forensic linguist in Georgetown, Del., ties the birth of the modern forensic linguist to the 1993 Daubert decision, which raised the bar on scientific admissibility and forced linguists to clearly establish the scientific legitimacy of their field. As a result, lawyers began using them as experts in addition to their previous role as investigators.

"If you look at the history of the field, in the 1980s this was used mostly as an investigative tool," she said. "But now forensic linguists are appearing in court as expert witnesses."

Chaski traces her start as a forensic linguist to 1992, when she was still a professor at North Carolina State University, "teaching linguistics to bleary-eyed engineers." One day she received a phone call from a police detective who needed to determine the authenticity of computer-generated suicide notes purportedly left by Michael Hunter, a young man who died from the lethal injection of three drugs.

Hunter lived with two other men and all of them used the same computer.

"I was able to analyze the syntax of all the documents from the three people and was able to show that the decedent had about a one in 10,000 shot of writing the suicide notes," Chaski recalled.

She was convinced that Hunter was murdered by one of his roommates, Joseph Mannino, who was only weeks away from received a medical degree and who had access to the drugs found in the dead man's body.

"I just could not get any significant difference between his writing and the suicide notes. I [told the attorney], 'I think he's the one.'"

Mannino was charged with first-degree murder after he admitted he wrote the phony suicide notes and was later convicted of involuntary manslaughter and sentenced to seven years in prison.

Chaski was intrigued by her first taste of working in the criminal justice system and soon left academia to work full-time as a forensic linguist. Today she is considered one of the leaders in her field.

"I study syntax, which is the way words get combined to make phrases," she said. "I measure syntactical structures based on their complexity. I categorize them in terms of their degree of complexity and then I run those numbers through statistical procedures."

Her statistical procedures have grown ever more complex and refined. She now claims that she is accurate 95 percent of the time in identifying the authorship of a computer document when there are 10 potential authors – and she said it is precisely that kind of mathematical proof that judges conducting Daubert hearings look for in determining whether forensic linguistics should be admitted into evidence.

Forensic linguists point out, however, that much of their work for lawyers is more advisory.

"And sometimes we just teach the attorney enough linguistics to present the evidence himself," Leonard said.

They also emphasized that it is not the linguist's job to help the attorney win the case.

"The role of the linguist is not to get the best deal for the client," said Butters. "The role of the linguist is to tell the lawyers what the linguistic limits of the case are. Then they can go from there."

Roger Shuy, considered by many to be the dean of forensic linguists in the U.S., agrees.

"Winning and losing is not an issue to me," said Shuy. "I'm not an advocate; I just analyze the data. If the lawyers can use it, then great."

The language of trademarks

Forensic linguists are involved in a broad range of practice areas.

"One area where the linguist's testimony can be helpful is in determining similarities and differences of trademarks," said Butters. "Is the similarity or the difference of the mark such that people are likely to be confused?"

To aswer that question, Butters said, the linguist refers to the smallest components of language for clues: how they sound.

Several years ago, lawyers for the European pharmaceutical company Aventis retained Butters in a trademark suit they initiated against a pharmaceutical start-up called Advancis. In addition to the similarity of sound, Butters also examined whether the two companies' names "connote similar kinds of ideas. 'Adventure' and 'advance' both seem to be words that are relatively close in terms of meaning, and meanings can also be something that would help to distinguish between the two or cause a likelihood of confusion."

Aventis won the case. So did another Butters client in the 1990s, Ty Inc., which believed that its Beanie Baby trademark was sufficiently strong to bar other companies from using the word "beanie." The competitors argued that the word was generic. But "the linguistic evidence is that it is not," said Butters.

Analyzing police tapes

Shuy, who retired from teaching at Georgetown Law School 12 years ago and is now working as a consultant in Missoula, Mont., has been a forensic linguist on more than 500 cases and written seven books on the subject.

In recent years, he's gravitated toward criminal matters, helping defense lawyers analyze tape recordings made by undercover police officers and cooperating witnesses.

Shuy's specialty is "discourse analysis" – determining what is on a person's mind based on what subjects he raises in the course of a conversation. It is used most often in criminal cases involving secretly recorded conversations.

He listens to the tapes and lists the topics that are raised, who raised them and what the responses are. He then subjects them to linguistic analysis from a variety of perspectives.

"Just as medical experts are used to describe and define what is on an X-ray, so linguistic experts describe and define what is in a tape-recorded conversation," he wrote in his chapter of the 2001 book, Forensic Linguistics.

"The linguist cannot know for sure what the speakers' intentions really are, but a careful examination of the topics they bring up gives a useful snap-shot of what they are thinking about, what is foremost in their minds and, perhaps even more important, what is not on their minds. Likewise, a careful analysis of the responses that given persons make to the topics introduced by others offers a similar clue to their agendas and intentions."

For example, Shuy was hired to analyze a taped conversation between an informant and a man who was accused of hiring someone to kill his wife. An analysis of the conversation showed that while the informant was persistent in raising the topic, responses of the accused showed he had little interest in the subject.

Shuy said that his focus on taped conversations is not a passion shared by many forensic linguists.

"Large chunks of conversation are things that most linguists don't like to fiddle with. It's big and messy and takes a lot of time. But I love it."

He said that he rarely performs document analyses anymore, but one of his most famous cases involved a kidnapping in which the police asked Shuy to look at a ransom note. When he was done, he asked them if they had any well-educated suspects from Akron, Ohio.

They did, and Shuy's analysis led to the man's arrest.

Shuy's seemingly Sherlock Holmes style conclusion was based on the fact that the note contained misspellings, such as "kops," while more complicated words were spelled correctly. Shuy deduced that the note was written by an educated person trying to portray himself as poorly educated.

As for his Akron conclusion, Shuy was struck by the kidnapper's demand that the ransom money be placed at a specific intersection "next to the devil's strip." He happened to know that "devil's strip" – an area of grass between the road and the sidewalk – is a term used only in Akron, Ohio.

It's the kind of information that good linguists file away for possible future use.

Questions or comments can be directed to the writer at: dick.dahl@lawyersusaonline.com