On a Dallas-bound flight in1979, Roger Shuy, a professor of linguistics at Georgetown University, struck up a conversation about his work with a lawyer sitting next to him. Shuy had spent a couple of decades studying dialect differences, the effects of context on meaning, and how social norms shape language. He explained that, essentially, his job was to tape-record how people talked. “Wow,” the lawyer said, “my partner has a case involving tape-recorded evidence. Would you be willing to look at the tapes?”
Shuy agreed- and became an expert witness in a case of attempted murder-for-hire. A Texas oil millionaire was accused of arranging to kill his estranged wife and the judge overseeing his divorce proceedings. While the oilman claimed he had hired one of his employees to tail his wife and collect evidence of an affair, the employee told police he had been asked to find someone to kill the pair.
Shuy analyzed recorded conversations, and found that the oilman’s answers of “good” and “all right” to suggestions about wanting people dead -crucial evidence for the prosecution- had strange intonation and timing. Shuy showed, with the help of an FBI video, that the “good” and “all right” weren’t responses to his employee at all, but discourse fillers related to a different thread of the conversation. The oilman was acquitted. The experience led Shuy into a career as a forensic linguist.
Language has always been an important part of law. But it wasn’t until the 1980s that forensic linguistics came into its own as a profession. Before that, linguists occasionally offered testimony about the meaning of slang terms, voice identification, and inconsistencies in confessions or police statements. But through the 1980s and 1990s, as technology made recorded language a bigger part of crime investigation, language experts became a staple in the courtroom.
It’s easy to assume that if words are caught on tape, the evidence is clearly there. But in order to determine whether a bribe has been offered or accepted, whether an “uh-huh” means yes, whether a speech act counts as a threat, whether a suspect has agreed to a search, and even which speaker is talking at any point on a recording, careful analysis of language and context is required.
In one case where an insurance company blamed a plane crash on a faulty engine that released toxic gas and incapacitated the pilot, Shuy looked at sentence structure, pause fillers, topic relevance, and pronunciation in the pilot’s speech on the flight recorder. He found that there was no chemical-induced disorientation before the crash. In another case, Shuy used the relationship between sentence and syllable structure to show that a phrase on a scratchy tape transcribed as “I would take a bribe, wouldn’t you?” was actually “I wouldn’t take a bribe, would you?”
The field doesn’t revolve around recordings; forensic linguists can study any form of language. Criminals have been identified through linguistic patterns in text messages and letters they have written. In one famous investigation, a single phrase in a low-tech, pencil-scrawled ransom note cracked the case.
Shuy was called in to examine a note that had been left on the doorstep of an Illinois home where a girl had disappeared. It instructed the family to stuff $10,000 into a diaper bag, then: “Put it in the green trash kan on the devil strip at corner of 18th and Carlson. Don’t bring anybody along. No kops!! Come alone! I’ll be watching you all the time. Anyone with you, deal is off and dautter is dead!!!”
Shuy determined that, despite the misspellings, the suspect was probably well-educated. Kan and
dautter are not the kind of misspellings uneducated people tend to make, and the letter was otherwise well punctuated and spelled difficult words like diaper correctly. The clumsier “mistakes” were probably intended to deliberately mislead.
The more important conclusion had to do with the term devil strip- a moniker for the strip of grass between the sidewalk and the street, otherwise known as a parkway, tree lawn, verge, or berm. Shuy knew it was the term used in Akron, Ohio, because that’s where he was born and raised. But he also knew it was exclusive to Akron because he was a linguist who had studied American dialects. He asked the Illinois police if they had any well-educated suspects from Akron. One of them fit that profile, and when confronted with the evidence, he confessed. He tried to mislead with language- but in the end, with the help of forensic linguistics, it was his own words that gave him away.
The article above by Arika Orent appeared in the November-December 2016 issue of Mental Floss magazine. It is reprinted here with permission.