Atul Gawande on the mistrust of science

The New Yorker published Atul Gawande’s commencement address at the California Institute of Technology. He calls upon the graduates to defend science from pseudo science. I don’t think he’s really defending science so much as basic reasoning. He writes:

To defend those beliefs, few dismiss the authority of science. They dismiss the authority of the scientific community. People don’t argue back by claiming divine authority anymore. They argue back by claiming to have the truer scientific authority. It can make matters incredibly confusing. You have to be able to recognize the difference between claims of science and those of pseudoscience.

Science’s defenders have identified five hallmark moves of pseudoscientists. They argue that the scientific consensus emerges from a conspiracy to suppress dissenting views. They produce fake experts, who have views contrary to established knowledge but do not actually have a credible scientific track record. They cherry-pick the data and papers that challenge the dominant view as a means of discrediting an entire field. They deploy false analogies and other logical fallacies. And they set impossible expectations of research: when scientists produce one level of certainty, the pseudoscientists insist they achieve another.

To be precise, all five of those moves are logical fallacies–well most of them anyway. And this speaks to the broader point–it’s not just science, but basic reasoning that he’s defending. The trouble is, however, that the enemies, as it were, of reason take themselves to be its defenders. In fact, calling them out on their sorry reasoning, as Gawande has just done, is, as Gawande notes, not advisable:

The challenge of what to do about this—how to defend science as a more valid approach to explaining the world—has actually been addressed by science itself. Scientists have done experiments. In 2011, two Australian researchers compiled many of the findings in “The Debunking Handbook.” The results are sobering. The evidence is that rebutting bad science doesn’t work; in fact, it commonly backfires. Describing facts that contradict an unscientific belief actually spreads familiarity with the belief and strengthens the conviction of believers. That’s just the way the brain operates; misinformation sticks, in part because it gets incorporated into a person’s mental model of how the world works. Stripping out the misinformation therefore fails, because it threatens to leave a painful gap in that mental model—or no model at all.

To put this another way. Science teaches you a lot of truths and techniques that don’t matter to people who most need them. Invoking these truths and techniques not only does not convince them, it makes it worse. By analogy, the truths and techniques of critical thinking 101 don’t matter to the people who most need them and invoking them only serves to make matters worse.

Read the rest of Gawande’s piece.  At least he’s optimistic.

12 thoughts on “Atul Gawande on the mistrust of science”

  1. I was very disappointed with Gawande’s piece on the mistrust of science. It’s like an idealized discussion of Catholicism that considers doctrine, but doesn’t mention the Borgia popes. Where is the discussion of the corruption in medical research including the failure to discipline Harvard faculty members for failing to disclose monies received from drug companies? How about the non-reproducibility of results in psychology research? How about the fact that doctors in the US now compose 40% of the top 1%? He doesn’t see any connection between the financial power of doctors, the poor state of US healthcare, and the distrust of science? Really?

    Gawande wants us to think of science as an abstract and disinterested quest for knowledge, but in the real world science is carried out by individuals and corporations who have their own stake to defend in society. His piece shows clearly where he sees his own stake.

  2. Hi Harold,

    Those are some interesting points. I think your argument goes like this. Gawande addresses reasons x and y for mistrust in science, but fails to address reason z. Reason z’s being true, however, doesn’t preclude that he’s right about x and y.

  3. Not exactly. The objection is not based on logic. The scope of his inquiry is wrong. Gawande has taken “Science” to mean the abstract endeavor of expanding knowledge by applying hypothesis testing. But he has ignored the sociology of science, i.e. an enterprise undertaken in society by real people who are rigorously pursuing their own self-interest like everyone else. No doubt he would like to point out that the abuses of the health care industry are not “Science” just as the sins of the popes should not invalidate Catholic dogma. But he would be wrong in that case. The public mistrusts “Science,” because those who march waving the banner of “Science” have been taking advantage of us. Gawande complains that over the course of two generations trust in science has eroded. Well, it was during those same two generations that doctors as a group managed for the first time, probably in history, to lodge themselves entirely in the top 2% of the richest country in the world. And he doesn’t make the connection.

  4. Hi Harold,

    Two points. First, I think your focus (Rich American Doctors) is narrower than Gawande’s. He was addressing students at Cal Tech in general, not a medical school. Second, what is the Sociology of science, but just more science? How do we know these claims are true, if not using the methods of science Gawande is proposing? One more point. To push the Catholic analogy somewhat. Rich American Doctors are not equivalent to the whole of the clergy. They’re not even primarily clergy (medicine is an Art, after all). Let’s say it’d be like confusing Opus Dei for Catholicism.

  5. This is a key to my own opinions about “skepticism” lately. It’s one thing to say that knowledge is NEVER complete, that science is a process not a destination. It’s quite another to draw conclusions from scientific evidence and then use those conclusions as justification to take political action. If it’s true that knowledge is never complete, and scientific investigation always has more steps to go, then a reasonable conclusion would be that the law should NEVER be exclusively based on science … or perhaps also that the law should not overtly contradict science.

    Consider the following statement: “Scientific studies have demonstrated that keeping a gun in the home is associated with approximately double the risk of dying from a gunshot, therefore laws should be passed to prevent people from having guns in their homes.”
    Consider the following statement: “The preponderance of the scientific evidence is that the trend in global warming is due to human use of fossil fuels for energy, therefore human use of fossil fuels must be curtailed in order to prevent further global warming.”

    So, it seems to me, the mistrust of science is not only due to the way that the human brain works, or that most people have not been adequately trained in a scientific view of the universe; or solely due to the misuse of science by scientists for their personal gain; but also due to the sociopolitical use that politicians, officials and activists have put scientific evidence to that it was not meant to be used for.

  6. Hi Marshall,

    Thanks for the comment.

    I’m afraid not sure I get what you’re saying. Are you saying that the necessary incompleteness of science (i.e., knowledge) requires we shouldn’t use it to frame policy? That view seems problematic. Two reasons. First, if we act (or fail to act) at all, we use information. We cannot but do this. Second, it seems preferable to act on the best information we have. The best information suggests that lead, for instance, is generally bad. Seems like a good reason to make policy about lead in the environment. But maybe I misunderstand you.

  7. The question is what does Gawande mean by “Science.” which he doesn’t define. Probably he means hypothesis testing via controlled experiments followed by publication of the results in journals. But that narrow definition is self-serving for the persons and organizations who profit by wrapping themselves in the flag of “Science,” most of whom are not scientists. What is Catholicism? Is it the sacred texts plus the writings of Thomas Aquinas or does it include the institutional history of the Catholic Church from the Crusades, simony, to clerical child abuse? Even the term, “the mistrust of science” constitutes artful misdirection from what might be called “the mistrust of scientists and people who would like you to think they are scientists.”

  8. Hi Harold,

    I think you’re making a very similar point to Gawande.

    The problem with the demarcation problem is that everyone thinks they’re on the “science” side of the divide. They are not. I think this is obvious. The tricky thing about Gawande’s address is that he does the very thing he says he shouldn’t do: i.e., he calls attention to the demarcation issue.

  9. You couldn’t be more wrong. For Gawande science is “a systematic way of thinking.” Not an enterprise or an establishment pursuing its own self-interest. People don’t trust science partly because the groups that profit from science, such as doctors and drug companies, are not very trustworthy.

    I don’t why in your comments you are so eager to see agreement. I don’t agree with Gawande and I don’t agree with you. What’s wrong with that?

  10. I myself prefer clarity to charity. I think you neither read very well nor think very well. The minimum qualification for participating in a discussion of this topic, which I take to be your purpose in posting, is to articulate your views for which the phrase “read the part about demarcation” fails to qualify.

    Gawande is a skilled writer, but not, at least in this piece, a good thinker. His topic is not a philosophical one about the nature of “Science,” although he wants to make it that, but why more Americans don’t believe all the preachments of the “scientific” community. Philosophy is an abstract discussion of timeless truths, but his topic is the change in public opinion from 1974 to 2010. The place to look for the answer then is in social developments during that period. What changed? Nothing that Gawande wants to consider. And did trust in the scientific establishment decline similarly in other countries during the period? Was this a general trend in world opinion or an American phenomenon? My guess is that in countries where the health care industry did not succeed in sequestering a huge percentage of the GDP during the period in question, the population did not show such a decline in trust of “Science” as we se in the US.

    And Gawande makes other glaring errors. He reports that vaccinations dropped in the period 1989 to 1991 which he claims resulted in 55,000 cases of measles. That doesn’t make sense. What was the base rate of measles infections during the period before vaccinations declined? It’s the number of excess infections, above the base rate, caused by fewer vaccinations that matters, not the total number, unless the base rate was zero.

    Perhaps we cannot hold commencement addresses tothe very highest standards of discourse, but I found Gawande’s piece lazy, sloppy, and, in an indirect way, self-serving.

  11. Hi Harold,

    Sorry that you think I’m a poor thinker and reader.

    Sadly, your criticism of Gawande is an illustration of the point Gawande is trying to make. You’ve cherry-picked a failing of some group of people with scientific backgrounds and used this as a basis for doubt in scientific practice in general.

    My sense is that if your criticism of the failing of this particular branch of scientific failing has any basis in evidence, then Gawande agrees with you–at least about your approach. If, however, your claims have no basis in empirical evidence, and we have no reason to think it extend to the practice of frog biology, physics, or astronomy, then again you’re the one he’s criticizing.

    Finally, Gawande’s mistake, as I think I mentioned, is to mention this at all. No one likes being scolded. And the people who most need to learn the lesson he’s teaching–you in this case–are least likely to learn it when put in this way.

Comments are closed.