Published online 16 February 2009 | Nature | doi:10.1038/news.2009.99


Swedish authorities embroiled in furore over academic freedom

Journal removes paper from website after company threatens legal action.

Lie detection — an emotional issue.

The Swedish Research Council is wading into an escalating row over academic freedom after a peer-reviewed journal removed a published paper — penned by two Swedish academics — from its website following a threat of legal action from the company whose technology the research criticized.

The controversial paper1, entitled 'Charlatanry in forensic speech science: a problem to be taken seriously', was first published in the International Journal of Speech, Language and the Law in December 2007. In it, speech scientists Francisco Lacerda of Stockholm University and Anders Eriksson of the University of Gothenburg, Sweden, criticize different voice-analysis technologies. In particular, the two authors say that there is no scientific basis behind the claims made for the Layered Voice Analysis technology developed by Nemesysco Limited, a company based in Natania, Israel.

On 3 November 2008, the company's lawyers wrote to the journal's publisher Equinox Publishing, based in London, complaining about the paper. The letter said that Amir Liberman, the founder and chief executive of Nemesysco, was considering issuing a claim for defamation, and requested that the publisher retract the article.

The article was pulled from the journal's website within a week and replaced with a note explaining the journal's decision to remove it. But now, the Swedish Research Council, a government agency, is making moves to back up Lacerda and Eriksson. Four out of seven key leaders at the research council have so far signed a statement expressing concern over the retraction of the article from the journal's website.

Freedom fight?

A draft of the statement seen by Nature News says: "Freedom of research would be in great danger if companies and organizations that are not satisfied with the content of scientific articles could get them removed through threats of court action." It adds that the research council considers it to be of the "utmost importance" that the research community demonstrates that it will "not tolerate any attempt to put a lid on the scientific debate".

The paper's authors say they are disappointed the journal "bowed down" to the threat of legal action. "This is an infringement on academic freedom," Eriksson told Nature News.

But Liberman says he "did not object to the publication of a paper that says the technology does not work", but rather that the authors suggest in the paper that he is a "charlatan" by commercializing technology that he knows does not work. "If you are going to call me a liar, you have to be prepared to prove this in court," he says.

On the research council's move, Liberman says "The freedom to research is not a license to slander. It must come with a sense of responsibility, should be addressed in a professional manner and with a minimal respect to the tested topic."

"Nemesysco's voice analysis technology has nothing to do with the science of phonetics, but with psychology and criminology, as well as neurology and other brain science," he adds.

Janet Joyce, managing director at Equinox, told Nature News that the publisher "stands by the scientific findings in the paper" but was legally advised that the paper contained "potentially defamatory" comments and as a result of that advice agreed to remove it.

In search of the truth

Nemesysco says its voice-analysis technology detects emotions such as stress and deception in speech, and can help "truth-detection investigation activities" — but the company emphasizes its products "should not be considered as lie detectors".

To test Nemesysco's voice-analysis system, Eriksson and Lacerda simulated the technology from computer code published in a patent awarded to the company in 2003. The researchers say that the emotion indicators the technology claims to detect in an individual's speech waveform are, in fact, arbitrary artefacts that result from converting the speech signal from analogue to digital form.


The company points out that it has clients in more than 30 countries and that the UK government has been trialling the technology to evaluate citizens' claims for housing and other benefits.

Meanwhile, Arne Jarrick, secretary general of the Swedish Research Council's humanities and social sciences council, who spearheaded the agency's defence of Lacerda and Eriksson, says he expects the remaining council heads to decide whether to sign the statement in the next few days. 

  • References

    1. Eriksson, A. & Lacerda, F. Intl J. Speech Lang. Law 14, 169–173 (2007).
Commenting is now closed.