Balneus

Australian Lefty on Politics, Governance, Science and Info Management

Science journals, commercial censorship, law and social security benefits

Posted by Dave Bath on 2009-02-17


Nature News reports something very troubling in doi:10.1038/news.2009.99 (2009-02-16), entitled "Swedish authorities embroiled in furore over academic freedom" and with the teaser "Journal removes paper from website after company threatens legal action".

In this post I’ll give the abstract of the original paper, some background, and discuss the serious consequences of what will happen if this censorship by corporate pressure on scientific journals is permitted.

The original paper [citation: Eriksson, A. & Lacerda, F. Intl J. Speech Lang. Law 14, 169– 173 (2007)] was entitled "Charlantry in forensic speech science: A problem to be taken seriously", and in it were explicit statements that claims of "lie-detector" manufacturers had no scientific basis.

It wouldn’t surprise me if someone has died or committed suicide already because of these machines.

Here is the abstract (my bolding):

A lie detector which can reveal lie and deception in some automatic and perfectly reliable way is an old idea we have often met with in science fiction books and comic strips.  This is all very well.  It is when machines claimed to be lie detectors appear in the context of criminal investigations or security applications that we need to be concerned.  In the present paper we will describe two types of "deception" or "stress detectors" (euphemisms to refer to what quite clearly is known as "lie detectors")." Both types of detection are claimed to be based on voice analysis but we found no scientific evidence to support the manufacturers’ claims.  Indeed, our review of scientific studies will show that these machines perform at chance level when tested for reliability." Given such results and the absence of scientific support for the underlying principles it is justified to view the use of these machines as charlatanry and we argue that there are serious ethical and security reasons to demand that responsible authorities and institutions should not get involved in such practices.

In the publisher’s note that has replaced the paper is a hint that while they guarantee online publication of a letter from the manufacturer, there is no guarantee that the scientific article will be again made available:

We have received complaints from Mr Liberman and Nemesysco Limited about the content of this article and particularly that the allegations made against them in it were highly defamatory, containing many inaccuracies and misleading statements.  In addition, they complain that it was prepared without reference to them and without giving them an opportunity to comment upon it.  The Journal accepts that Mr Liberman and Nemesysco Limited were not asked to assist in the preparation of the article and further that they were not invited to comment on the content of the article prior to its publication where, in view of the content of the article, it would have been appropriate to invite them to do so.  We have agreed to publish a letter from Mr Liberman and Nemesysco Limited setting out their objections to the article in more detail in a future issue of the journal.

Personally, if you are doing a scientific analysis of a commercial product, you shouldn’t need to seek comment from the manufacturers.  Those commercial interests can always make a media release, and seek to have their refutation published alongside the original paper.

This is not a theoretical squabble, as the products have been purchased in more than 30 countries, including insurance companies in the US, the UK government for assessing validity of housing and pension benefits, and various security services.

The UK has already stopped 300,000 pounds of payments based on this machine that performs no better than a coin-toss according to the academic research.

Consider the following claims and scenario:

  • The machine claims to assess truth or falsity of claims based on voice stress.
  • A person comes into (or phones) a social security office about payments that directly affect whether they and their family have a roof over their heads and food in their bellies.
  • The claimant will almost certainly be stressed by this threat.
  • This stress will be reflected by voice patterns.
  • If the stress is flagged by the machine, the person doesn’t receive payments
  • The person (and possibly their children) may become homeless or not have enough money for food.
  • If the lack of money causes disconnection of power, then in extremely cold weather (such as the UK has experienced recently), this may cause severe illness, and possibly, directly or indirectly, death.
  • Even the mere added stress of loss of social security benefits may be enough to push someone over the edge, resulting in suicide.

Remember, even evidence from traditional polygraphs, administered by an expert with lots of effort to establish baselines, are not admissible as evidence in any court of law I know about.

So, we have the following issues:

  • Academic freedom;
  • use of highly questionable technology as part of regulatory processes (that could result in legal action against those claimed by social security agencies to have made false declarations for benefits); and
  • a precedent for commercial interests to stifle inconvenient research that can directly affect people’s welfare.

If this particular incident, and any others like it, do not receive public attention and the commercial interests are not pushed aside, it cannot be too long before businesses start doing this as a matter of course.

Imagine if this practice had been common in the past.  Big Tobacco would have caused withdrawal of medical research that showed a link between smoking and ill-health.

The current case is directly equivalent to Big Pharma suppressing evidence of lack of efficacy or significant side-effects of a drug going through a regulatory process, be it for approval for use, or for subsidies from government.

I can also imagine the deep pockets of Big Carbon (including Australia Inc that depends so much on coal sales) suppressing publication of climate data and models.

This makes the fight to be able to teach Darwinian thought in Biology classes look trivial.

What next, we’ll see the claims of clairvoyants used in courts to determine liberty or incarceration?

Technology used in legal and regulatory systems should be openly discussed, and open to academic criticism without fear of litigation.  I’m a fan of DNA evidence, but I’m well aware how such information can be abused by barristers, especially to an uneducated jury.

… but, when dubious machinery is used to make decisions that can put the most vulnerable members of society in deep trouble, without the ability to protest in a reasonable timeframe (for how many weeks can such people do without social security benefits… probably two at the most), threats to sue academic journals by commercial interests are unconsciable, as are the actions of academic publications when caving in to such pressure.

Such publishers deserve no respect, and no subscriptions.  At least there is a fair alternative these days in open refereed journals like PLoS.


See Also/Notes:

  • More quotes from the Nature News article (my bolding):

    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".

    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.

  • You can request a private copy (not for publication) of the original paper from frasse at ling dot su dot se
     
  • Other mentions of the paper
  • Nemesysco on Wikipedia
     
  • From the blog of the Linguistics Department at the University of Stockholm: (you’ll need to go down the pages a bit to see the English version)

     

  • "Brouhaha Over Controversial Forensic Technology: Journal Caves to Legal Threat" (AAAS ScienceNow Daily News, 2009-02-10)
     
  • Comment thread at antipolygraph.org
     
  • "No such thing as a voice-print" (2008-12-04) Discovery News
     
  • "Lie detectors extend their reach to social security helplines " (2008-05-07) The Inquirer
     
  • A 2006 analysis of this sort of technology by University of Florida reports against two datasets and an unacceptably high false conclusion rate, first, straight from the technology, and the second, where the human expert and the technology agreed:

    Figure 2: Identification of stress in speech samples from the VSA database (IASCP team).

      Actual Condition
    High Stress Low Stress
    Machine Analysis High Stress (blocking) 57%
    True Positive
    62%
    False Positive
    Low Stress (no blocking) 43%
    False Negative
    38%
    True Negative

    Figure 4: Identification of deception and truth in speech samples from the VSA database (IASCP team – Operator Agreement Condition). These rates correspond to a subset of the samples – only those in which there was operator agreement on the presence/absence of blocking and only when those judgments were made with high confidence.

      Actual Condition
    Deception Truth
    Machine Analysis Deception (blocking) 75%
    True Positive
    75%
    False Positive
    Truth (no blocking) 25%
    False Negative
    25%
    True Negative

    In other words, even with expert humans verifying the individual tests, the basic technology is unworthy for serious use in regulatory environments, either for identification whether there was stress or not, or the inferences about true or false statements.


Advertisements

One Response to “Science journals, commercial censorship, law and social security benefits”

  1. jammy said

    great job .simply awesome

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: