Wednesday, May 20, 2009

From Foolishness to Fraud

One of my favorite popular science books is Voodoo Science by Bob Park. Like Martin Gardner’s earlier book Fads and Fallacies, it takes a critical and entertaining look at many things that superficially appear to be science but aren’t. Park’s targets include perpetual motion machines, cold fusion, and human spaceflight.

Most cases of voodoo science involve some level of fraud. Yet the people responsible almost always start out with good intentions. I’m always intrigued by this apparent paradox: How can “good” people commit fraud? Park summarizes the answer in his subtitle: The Road from Foolishness to Fraud.

In other words, what ends up as outright fraud usually begins as mere foolishness. We all make mistakes, and we all want to believe we’re right even when we’re wrong. At first, such a belief is merely foolish. But each time we exaggerate the evidence in our favor, or ignore contrary evidence, we take one more step down the road from foolishness to fraud. Each step down that road makes it harder to turn back and admit we were wrong all along. Eventually, in the worst cases, our dishonest attempts to maintain our position become fraud.

(My students make mistakes all the time, so I frequently remind them that the good scientists aren’t the ones who don’t make mistakes--they’re the ones who routinely check their work and fix their mistakes.)

Park limits his examples to the periphery of science, but it’s easy to find examples elsewhere. Most of us have had the experience of telling a well-intentioned “white lie” and then, faced with further questions, having to choose between telling a bigger lie to cover for it, or coming clean.

Politics abounds with examples of foolishness and fraud. The Iraq WMD allegations come immediately to mind. Here in Ogden, our biggest recent example was the “gondola” proposal. Others that come to mind include (at least in some aspects) Ernest Health, the ice climbing tower, and Envision Ogden.

In science, fortunately, there are mechanisms to correct foolish mistakes and root out fraud. It’s standard practice for scientists to brutally scrutinize new claims, and there are professional rewards for scientists who show that someone else’s results were wrong.

In politics we also have checks and balances: multiple levels of government with multiple branches; law enforcement agencies that investigate the most serious allegations; and the media who are free to conduct independent investigations and report multiple viewpoints. But the mechanisms for correcting mistakes are less effective than in science, for two reasons. First, the underlying subject matter is harder to probe for objective truth. And second, too much of politics is driven not by truth but by power.

Much of the inspiration for our democratic political system came from science during the Enlightenment, and the system continues to improve over time. The more we all participate, the better the system will work.

5 comments:

  1. One of my favorite books on fraud in science [often, but not always, driven by the foolishness and good intentions] is Stephen J. Gould's The MisMeasure of Man. I recommened it to my students often. Fine piece of history too.

    ReplyDelete
  2. Thanks for reminding me of that book. I read it over 20 years ago, along with most of Gould's other books, when I was in grad school and should have been working on my thesis. Thought Mismeasure of Man was one of his best. As I recall, it includes some good lessons in interpreting statistics. I'll probably be blogging on that subject before too much longer.

    ReplyDelete
  3. TheLovelyJenniferMay 25, 2009 at 9:56 PM

    A little off-topic (science-wise), but an interesting read anyway - Freakonomics, Steve Levitt & Stephen J Dubner. Off color look at how economics theory & principles apply to unlikely activities, from crack cocaine business hierarchy to Ku Klux Klan to the naming of children.

    TLJ

    ReplyDelete
  4. TLJ

    Freakonomics sounds like the study of allometry, which a colleague [an econometrician] described to me as "the occurrence of invariants -- dimensionless (that is, unit-less) ratios between two characteristics of a system, living or non-living, such as between a measure of organism size and energy consumption or between engine horsepower and engine displacement."

    Felt sure you'd want to know.... [grin]

    ReplyDelete
  5. TheLovelyJenniferMay 26, 2009 at 6:36 AM

    Now that's something that would make you go

    "Hmmm ...."

    Though, allometry is concerned with growth - so the ratios may not necessarily be dimensionless ...

    TLJ

    ReplyDelete

Not registered? Just choose "Name/URL" and enter any name you like; you can ignore the URL field.