Forensics and Assessment of Suicide Risk

USAToday posted a story about a researcher who is working to teach a computer to assess the suicide risk levels of patients; he has a database of suicide notes and is using teams of investigators to tag language in them in order to teach the computer the patterns to look for. You can read the original article here.

The statistics provided near the end of the article are interesting:

The computer was 93 percent accurate -- identifying those with suicidal tendencies over the control group -- while humans were right slightly more than 50 percent of the time with the same groups.

I could see how a computer would be able to move straight to the heart of the matter--looking solely for the triggers it's been trained to catch--which could help improve its accuracy. Human perception is often blinded by too many outside factors. That being said, I am interested in knowing how well the computer performs in continuing trials; I wonder if language is too narrow a scope to identify risk (leaving out other factors, such as history, facial expressions, intonation). The article, of course, does say that the computer is not meant to replace a practitioner/counselor/doctor--it is only meant to aid in diagnosing risk.

Looking at John Pestian's list of publications, I don't see one where he specifically talks about the outcomes of this project--something I'd very much like to see.

A Couple Study for Forensic Students

A student sent me a link to an interesting story shared on Fox News a couple months ago about how couples, when they are happy together, use similar language. The researchers involved in the study claim that even famous poets have a writing style that is similar to their significant other's when they are in a happy period of the relationship. People are taking this one step further to say that it might be possible to make predictions about the success rate of a couple's relationship based on a comparison of their writing styles and language usage. Here is a link to the original story.

What do you think? Is it possible to tell if a couple is happy together based on linguistic analysis? Is your language usage similar to that of your significant other?

I'm off to find a writing sample of my husband's to compare to mine...

Fall Schedule Finally Fixed!

With the flair of alliteration, I'm happy to announce that the fall schedule is finally fixed! Hooray! If you go to the fall schedule, you will now see that ENG 341: Introduction to Linguistics (MWF 10:00-10:50) and ENG 438: Forensic Linguistics (TR 9:30-10:45) are finally on the schedule. And that means you may now register for those courses.

Please get the word out because the only way the courses can run is if they get students registered. So please spread the good word that all linguistics courses are open for registration.

Also, if you're signed up for 344 (Structures of English) for the fall but do not necessarily need to take that course to fulfill your linguistics requirement (I'm specifically thinking of English majors out there), you might want to take a look at the other linguistics offerings for the fall.

The new courses should not have any prerequisites attached to them; that is, you should be able to register for them even if you haven't had 9 or more hours of English. If you have a problem registering because of prerequisites, please let us know immediately so we can get that fixed.

Forensic Linguistics and the Facebook Killer

For anyone asking what forensic linguistics is or what forensic linguists do, you should watch Dr. John Olsson's YouTube video. His video documents the types of comparisons and critical analysis necessary to work with forensic linguistic investigations of language use.

Watching the video will give you a sense of what types of information you can get from language, the types of conclusions you can draw about language use, and how critical analyses of language can help in investigations.



You may have noticed that Dr. Olsson's conclusion was that Chapman "could not be excluded from the authorship of the later texts sent from the teenager's phone." On the surface, that may sound like a vague conclusion, yet it was important to the case. Why do you think such a conclusion was important? Furthermore, why do you think it would not be possible for Dr. Olsson's conclusion to be that Chapman was definitely the author of the text message?

If this video sparks interest for you, you might want to think about taking ENG 438: Forensic Linguistics in the fall.

John Olsson's WORD CRIME

If you are interested at all in Forensic Linguistics (or if you want to find out what it is), you should check out the Forensic Linguistics Institute website.  John Olsson is the founder of the institute and author of the three leading books on forensic linguistics (both are on the website).  I am focusing this blog on his latest book on forensic linguistics, Word Crime: Solving Crime Through Forensic Linguistics.

The inside book flap provides the following information about the book:

"Tell kids not to worry. sorting my life out. be in touch to get some things"

Instead of being a text message from one partner to another, this text message turns out to be crucial and chilling evidence in convicting the deceptive killer of a mother of two.  Sent from her phone, after her death, a few tell tale signs give him away to a forensic linguist... Rarely is a crime committed without there being some evidence in the form of language.

John Olsson is a world-leading expert in forensic linguistics, a science where linguistic techniques are applied to legal processes to solve cases and provide new angles on evidence.  Beginning with a description of exactly what forensic linguistics is, Olsson includes a survey of some of the high profile criminal and civil cases he has worked on where it has been used.  Including the much-discussed dispute between the publishers of The Da Vinci Code and the author of Daughter of God, there are a series of chapters where gripping cases are described--involving murder, sexual assault, hate mail, plagiarism, suspicious death, code deciphering, arson and even genocide.

This is fascinating reading for anyone interested in true crime, in modern, cutting-edge criminology and also about where the study of language meets the law.

Since 1996, John Olsson has operated a world-renowned forensic linguistics consultancy and training service at www.thetext.co.uk.  He is an Adjunct Professor at Nebraska Wesleyan University, USA, where he teaches forensic linguistics online.  He is also Visiting Professor of Forensic Linguistics at the International University of Novi Pazar in Serbia where he runs an annual summer school in Forensic Linguistics, and is a board member of the Language and Law Centre at the University of Zagreb, Croatia where he is also a Visiting Professor.

I am recommending this book instead of his others (which are texts focusing on how to do forensic analysis of language) because this one is split into cases so that one chapter equals one case.  It is a book you can pick and choose from--you don't have to read the chapters in order to understand the whole.  His other books are also interesting and great sources but are not as easily combed through--they work best being read beginning to end while practicing the skills he writes about in each chapter.  They focus on completing analysis instead of on cases.  Reading about the cases will allow you to see the many different applications of forensic linguistics, some of which may surprise you.

I think I need to propose a new TV show: CSI: Linguistics Division.  Any supporters?