Texting study to catch criminals

Researchers will scrutinise volunteers' SMS messages to tease out patterns in the language and style of texts.

The University of Leicester team hopes the work will yield tools that allow identification of a text author.

Analysis of texts proved crucial in convicting Stuart Campbell of the murder of Danielle Jones in 2002.

The 15-year-old Essex schoolgirl disappeared in June 2001. Following a police investigation, her uncle Stuart Campbell was put on trial for her abduction and murder.

As part of his alibi, the defence showed two text messages purportedly sent by Danielle after she disappeared.

Analysis by a forensic linguist highlighted differences in the spelling of words before and after Danielle's disappearance.

The prosecution said the later messages were sent by Mr Campbell after he had abducted Danielle, using her phone.

Mr Campbell was found guilty of her murder in 2002.

Volunteers needed

The new study aims to build on this kind of analysis to develop a forensic toolkit for investigations involving texts.

The six month study will analyse the linguistic conventions of texters.

"The different parts of speech that people leave out seem to be fairly consistent"
Dr Grant, University of Leicester

Forensic authorship analysis, as the technique is known, is already used extensively for written documents like letters and e-mails.

It is often used to analyse abusive or threatening e-mails or in cases of copyright infringement. It has not been widely used to study text messages.

"As texting is both a relatively new mode of communication and a particularly informal way of using language, there is not a strong expectation that texters will follow linguistic conventions," said Dr Tim Grant, one of the team behind the study.

"This freedom therefore allows for significant individual differences in text messaging style, and this can be used to identify the text's authors."

Although the study is still in its early stages, Dr Grant has some idea of what general patterns may emerge.

"There seems to be systematic variation in the words people leave out of text messages, particularly if you use predictive text messaging," he said.

"You won't leave letters out, but to be quicker in the texting you will leave words out."

For some people this will be pronouns, so they will write "went to the shop" rather than "I went to the shop".

Others will leave determiners out, and will write "went to shop".

"The different parts of speech that people leave out seem to be fairly consistent, but the purpose of the study is to find out if that is actually true," said Dr Grant.

The team also hope to identify patterns that give clues to the age and gender of the author.

The team are now looking for at least 100 volunteers who are willing to hand over 10 messages for analysis. They are particularly interested in groups of people who text each other.

"In all languages we know that groups tend to start sounding like one another, said Dr Grant. "Text messaging is likely to be a microcosm of these more general linguistic features."

The study will be conducted anonymously.

RELATED INTERNET LINKS
Take part in the study
University of Leicester
The BBC is not responsible for the content of external internet sites