Skepticism of science, necessary & needless

Just a quick addendum to my previous post where I advised caution about skepticism of science.  A biomedical scientist recently told me that the journal Virology had a statistician audit all their papers within a 1 year interval with statistics to see if they were using them correctly. Turned out that 2/3 of the papers which had statistics made basic elementary errors!  The moral here is to be very cautious of, and therefore skeptical of, new science, especially sexy new science.  Junk statistics are especially an issue with medical science because of the incentive structure of these research. But when it comes to 100 year old science, skepticism is far less warranted. If you are hearing about it now, chances are that it has made it through the gauntlet of generations of skepticism.  Almost all scientific hypotheses initially conjectured are wrong, but those which last centuries and are still in circulation are almost certainly of great heuristic utility and predictive power.  It is important to respect and conserve knowledge which has been accumulated through experience and has proved useful.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

10 Responses to Skepticism of science, necessary & needless

  1. A Different Matt says:

    Sometimes it’s difficult to differentiate between science, junk science, and old wives’ tales. That’s why I trust Cracked.com to explore what is and what isn’t a fundamental underpinning of the human experience.

  2. Andrew T. says:

    An addendum to your addendum: when a supposedly scientific organization spends more of its budget on PR than research, that’s a pretty good sign you’re not dealing with scientists.

  3. Conan the republican says:

    I agree with the points made in your post. And, in keeping with earlier proscriptions, I don’t want to argue for/against evolution. However, evolution seems suspiciously immune from the points you have made on two flanks: 1) the main issue of past evolution into the many branches of life appears to have little chance in predicting future events in any manner that one would find helpful. In fact, it seems more effective in p*ssing off others who have alternative theories about our origins, and 2) evolution as a fact has had little chance of being disproven because it discusses evidence in terms of millenia and ancient genetics–both being unavailable for observation and review by critics. That said, I agree with an earlier comment. Any comments on evolution is a “third rail” best left untouched in this blog.

  4. John Quiggin says:

    Interestingly, the science of global warming just passes your test. The first analysis of the greenhouse effect was by Arrhenius
    “On the Influence of Carbonic Acid in the Air Upon the Temperature of the Ground”, Philosophical Magazine 1896(41): 237-76
    Arrhenius estimated the sensitivity of global climate to a doubling of CO2 at 5-6 degrees C, about twice the current best estimate. It was spelt out in a popular book published in English in 1908 (a bit more of a lag in those days!).

    Global warming, and not evolution, is the big test of credibility for the secular right. If you are going to reject science when it has implications that are politically unpalatable, you might as well accept the leadership of those who reject it when it has implications that are unpalatable on religious grounds.

  5. David Hume says:

    Conan, see TalkOrigins. Like I said, I’m not too interested in arguing about this topic, so instead of me responding to you, if you’re in good faith you will just expend your own marginal time if you are sincerely interested in science. Though I’m not holding my breath, your comment sounds suspiciously like bizarro-world talking points 🙂 Point #2 is just what they call *wrong*!

  6. Pingback: Take research papers with a grain of salt « David Kirkpatrick

  7. David Hume says:

    If you are going to reject science when it has implications that are politically unpalatable

    That is the question, are its implications politically unpalatable? What norms scaffold the cost vs. benefit analysis? I think it is totally expected that people will agree on the facts, but disagree on the inferences from those facts.

  8. Andrew T. says:

    Hume: exactly. It’s distressing that for too long, the “mainstream” Republican response to global warming has been to deny the facts, rather than point out that, say, the social costs of the Kyoto Protocols exceed even the wildest-best-case-scenario for its benefits.

    As far as I can tell, tradeable pollution licenses appears to be a market-based solution that adequately incorporates the true, real costs of environmental damage INTO the market rather than imposing external controls.

  9. John Quiggin says:

    AT, acceptance of tradeable pollution permits represents a big win for market-based thinking. Global warming is the issue that converted lots of environmentalists to a belief in market incentives as a central component of any policy response. Kyoto is consistent with tradeable permits, as witness the EU.

  10. Thanatos Savehn says:

    First, I hope your site takes off – I became a conservative because I was in college and graduate school during the assault on the Enlightenment by the postmodern crits and saw that Leftists were ultimately all about replacing Enlightenment beliefs (those of Hume, Locke, Smith, etc) with their own (Marxism and other forms of anti-individualist / anti-liberty collectivism).

    Second, I must take issue with your premiss. If true, then you would predict that Newton’s explanation of gravity was superior to Einstein’s. Ooops. Indeed there are a great many such black swans and they pop up all the time (see e.g. the emerging science of epigenetics in case you ever thought DNA was “all that”). So, I’d encourage you to travel the line from Hume to Popper and conclude that ALL skepticism is ultimately good and that every belief we hold, however cherished, whether in God or in Darwin, should be again and again dipped in the acid bath of skepticism to see if it withstands the truth of our observations.

    By the way, the reason statistical evidence so commonly births “junk science” is that it’s so easily manipulated and the rules of causal inference are so arbitrary. For example, if you understand probability you’ll understand why a data dredge is highly likely to produce proof of a causal nexus that does not exist. Similary, and speaking of “heuristics”, if you have a view about how the universe works and then set about proving it, the so-called Texas Sharpshooter trick will prove that blogging indeed causes belly-button cancer and, the more you study it, at ever more shocking rates.

    So go ahead and drop your dogmas in the acid. If they don’t make it, so what? What’s the point in believing 2+2=5? If they survive, you’ll cherish them even more.

Comments are closed.