18 February 2011

Predistortion

Yesterday in Washington, DC I participated on a panel on science and politics in a pre-AAAS workshop on "Responsible Research Practices in a Changing Research Environment."  Also on the panel was former Congressman Bill Foster, of Illinois, one of three PhD scientists in the last Congress. He lost his seat in 2010 to a "Sarah Palin-supported Tea Party candidate."  Foster is presently engaged in a worthwhile effort seeking to start up a political action committee -- Albert's List -- to get more scientists to run for office.  Here I focus on an interesting aspect of Foster's presentation at the AAAS workshop yesterday.

In his short presentation, he explained that politicians often look to experts to provide information that is useful in advancing their agenda, typically information that can be conveyed in SOUNDBITE fashion.  He explained that scientists should expect that the information that they bring to the political process, such as through testimony before congressional committees, will inevitably be "distorted" in the political process.

He then raised what he called "a difficult ethical question" -- if a scientist knows that their message will be distorted in the political process, to what degree should s/he predistort their message in hopes that what comes out the other end is a closer approximation to reality?  Foster cited as an analogy cheap earphones that achieve high quality through software that counterbalances distortion. Foster warned that such predistorion might be "heading down a slippery slope" but he was fairly ambiguous about the tactic.

I am not so much interested in Foster's views on the subject than I am the concept, which I think is very useful for helping to raise issues that often come up at the messy interface of science and politics. Longtime readers will probably guess that I am no fan of predistortion, and in fact, I think it is an enormously problematic practice for various parts of the science community today.

The way to deal with distortion in the political process is not through counterbalancing distortion, but through effective science institutions that can serve as trusted arbiters of knowledge and honest brokers of policy alternatives.  What do you think?  Is predistorion sometimes a justifiable practice from our experts?

14 comments:

  1. Sound bite politics and distortion works on policy's that only require short term support.

    Politicians are disposable from an institutional and governance standpoint. They can spew all the half truths they want, when the 'truth' gets it's shoes on the people will just toss them into the street and we will get new politicians.

    Half truths spewed by the heads of institutions do damage to the institution.

    ReplyDelete
  2. What is missing is trust--- predistortion simply makes the social dilemma worse and the science useless. A lose-lose.

    ReplyDelete
  3. I think you are displaying a bit of "magical thinking" if you think there is such a thing as "effective science institutions that can serve as trusted arbiters of knowledge and honest brokers of policy alternatives". Such a large portion of the published science is funded by the federal government that science institutions will likely tell their benefactors just what they want to hear. So looking for an honest broker may be just as difficult as it was for Diogenese looking for an honest man. As far as pre-distortion goes, I would concur that it is better to tell the story in as straight and honest a way possible. Certainly people will spin, they will pull one way or another but all the warped stories (including the pre-distorted ones) will be screened out prior to the history being written and the honest ones will make it into the permanent record.

    ReplyDelete
  4. Effective communicators need not rely on politicians to relay (and distort) their thoughts. Hansen's inanities get to the public unfiltered (eventually that will be unfortunate for him).

    Any effective thinker can make short, simple summary statements -- even when the statement is "we don't know" or "We think X, but the uncertainty is large."

    I would cite Chris Christie's style as an effective one. Tell the truth. Treat the people as adults. Support you views with the facts.

    ReplyDelete
  5. Pre-distortion seems a very bad idea to me, for many reasons, including the existence of bloggers who are on the lookout for overstatement by scientists.
    The issue you bring up is part of a much larger and growing discussion about the role played by liberal/Green bias in natural and social science. As someone who comes out of a liberal/Green stance, but has become aware of its limits, I see such bias as a serious problem.
    Jonathan Haidt has really gotten things rolling on this issue. Recently, he referred me to the paper at this link for a good presentation of the issue:
    http://www.theatlantic.com/national/archive/2011/02/what-does-bias-look-like/71153/
    The New Yorker and The Atlantic have also carried stories recently about how bias distorts even supposedly "gold standard" experiments, including many in the area of medical drugs. Many such experiments, so it turns out, cannot be replicated, or bring in results that indicate the drug is far less robust than found in the original experiment.
    There is much food for thought here.

    ReplyDelete
  6. My experience is a little different. In my world, the scientists are selected to testify based on their existing views. They are asked specifically to reiterate things they've already said. Otherwise, why would a politician with a point of view take a risk that the selected scientist would not support her/his point of view?

    The use of scientific information to promote policy positions is well understood, at least by poloticians. Getting into a power game with them on their own turf by engaging in distortion does not seem like a good idea from a variety of perspectives.

    ReplyDelete
  7. Predistortion works on the masses. But it turns the knowledgeable against you, unless they already are on your side and understand the predistortion.

    Having seen your dirty tricks, they will then feel the need to undertake their own predistortion, to correct yours.

    ReplyDelete
  8. Hi Roger,

    Can you give an example of "pre-distortion"?

    ReplyDelete
  9. "Predistortion" is just a nice sciencey word for lying.

    ReplyDelete
  10. -8-markbahner

    Sure. Let's say I am going to testify before a committee of Tea Partiers who think global warming is a hoax. I might go in with a stance that it is the end of the world. Then the perceived outcome of the hearing might be something in the middle.

    As Sharon says, one need not be disingenuous in this, the committee staff might simply select a witness that fits their agenda.

    The dynamics can work in exactly the opposite manner of course when the tables are turned.

    ReplyDelete
  11. #8 In White propadanda critical nuance is left out of statements as they are easily used by ones opponents.

    Greenland's Ice Caps are melting. If they melt completely the sea level will rise significantly.

    The missing nuance is the rate of melt and how long it will take. If one is making a case for immediate action on Global Warming talking about melt rate and how long it will take won't help the cause.

    The world has 6,000 billion tons of coal, if we burn it all temperatures could rise 10C

    The missing nuance is burn rate,at 6 billion ton/yr it'll take us 1,000 years to burn it all and economic recover-ability, the vast majority of that 6,000 tons isn't economically recoverable, off shore coal mining is going to be expensive.

    ReplyDelete
  12. Roger,
    The fallacy of your example in post #10 is that enough people can smell untruths, whatever their background. The person pushing apocalyptic hype as a bargaining chip would simply be dismissed as a kook or worse by enough people in a critical venue. The over-the-top hype will work, as we see, in venues dominated by those who agree.
    The net effect of 'predistortion' is nil to negative.
    Harrywr2's examples show how this works from actual 'predistorted' AGW claims.
    That a Congressman sees this as a potentially useful tool tells us more about him and why he lost than it tells about science.
    His constituents picked up on enough of his 'predistortions' that he lost the credulity of his District.
    He can blame the wicked tea party and Palin all he wants. To lace actual responsibility, he only needs to look in a mirror.
    That this topic is getting apparently serious consideration and is not condemned out of hand is symptomatic of why we are in the decline we are in.

    ReplyDelete
  13. "Sure. Let's say I am going to testify before a committee of Tea Partiers who think global warming is a hoax. I might go in with a stance that it is the end of the world. Then the perceived outcome of the hearing might be something in the middle."

    But don't most members of Congress ask to see testimony before it is given, to avoid that sort of surprise?

    ReplyDelete
  14. Roger,

    If you testify at a congressional hearing that AGW means the end of the world and another scientist testifies that climate science is a hoax, only the most hair brained congressman would conclude that the truth is half way between.





    Hey, maybe you're on to something there.

    ReplyDelete