Forensic Linguistics

According to an article published in The Star on October second, linguistic forensics was recently used as evidence leading to conviction in a high-profile South African court case.
(see_http://www.iol.co.za/index.php?set_id=1&click_id=15&art_id=vn20071002001320177C866364 for the complete article.)

Apparently, because each person makes unique language errors when speaking a language (especially a secondary language), forensic linguistic analysts can use these errors to link suspected criminals to crimes. In addition to these errors, the analysts also use instances of concomitant words to connect texts. Because the probability of anyone using any three given words in the same sentence is remarkably low, if the same three words are used in two different sentences in two different texts, analysts take this as evidence that the two texts were written by the same author. In the field of forensics, this new method is considered far more reliable than the old method of handwriting analysis.

Personally, I found the concept of linguistic forensics particularly intriguing because I had no idea that manners of speech really varied so much from person to person. Linguistic forensics is evidence that no two people really speak the exact same language, but that we all have our own unique idiolects.

Looking into the background of the field, I found that linguistic forensics is still not widely accepted. Apparently, a man named Robert W. Shuy wrote a book focusing largely upon difficulties he had encountered in trying to convince judges of the reliability his forensic linguistic evidence. When compared with the precision of DNA evidence, linguistic forensics looks shady at best—based on probability rather than concrete evidence—yet when compared to other abstract forensics, forensic linguistics seems more reliable. It is based on probability and is thus far more mathematical and precise than observation based forensics such as handwriting analysis. Yet because linguistic forensics is a relatively new field, it will still be some time before it gains acceptance in the broader field of forensics.

I also found that the field of written forensic linguistics is far more reliable than its oral component. I encountered one document from the North Carolina State University detailing the misconceptions which led to one man’s wrongful conviction in a case based on tape recorded evidence. While greater credence should be given to written forensic linguistics, courts should still exercise caution surrounding oral evidence. Voices are far more difficult to analyze than words, although oral evidence could still be used if the forensic scientists analyzed the actual words spoken and attributed only secondary importance to the words’ intonation.