Tuesday, August 31, 2010

Science over Politics

The media tend to present scientists as super-rational beings, who stand outside the cognitive biases they note in others. For example, the paragon of rationality Carl Sagan famously wrote:
In science it often happens that scientists say, "You know that's a really good argument; my position is mistaken," and then they would actually change their minds and you never hear that old view from them again. They really do it. It doesn't happen as often as it should, because scientists are human and change is sometimes painful. But it happens every day. I cannot recall the last time something like that happened in politics or religion.

I think this is profoundly misleading, because I can't remember any occasion where some scientist, on the spot, changed his mind about something important. They might concede a minor plank, but never the big idea. Inconsistent data points are seen as fatal or noise depending on what side you are on. Almost all interesting economic phenomena do not have definitive proof, so basically most of what we argue about comes down to common sense, just as in political or moral debates. After all, we only have a handful of recessions, and the effects of deficits (say) on unemployment (say) depend on whether the economy was at full employment, and whether the spending was an automatic stabilizer or exogenous increase, etc. Basically, we are left with about 2 datapoints to extrapolate from, and these also had their own idiosyncracies (especially if one is old enough to remember them personally).

The objective standard errors on important policy disagreements basically allows one to have any belief they want on fiscal policy. That doesn't mean everyone is equally right/wrong, just that it's essential to have good prejudices, opinions on things that can't be proven from axioms. This is why the academics aren't much help, because IQ and education just allows one to distinguish what is true, false, or indeterminate, and not very good at establishing probabilities. That is, a good statistician like Joshua Angrist will have a chapter on nonlinear relationships, and how important they are, and then argue that a 2% wage increase from when kids started school has implications for spending more on college; or Paul Samuelson will continually beat the drum for more deficit spending, and when someone mentions he is encouraging a debasement of the currency he will mention that he acknowledged that governments could spend too much, so he has done no such thing.

Priorities, not truths, underlie common sense.

The main advantage of scientific as opposed to political debates is that you don't have to be democratic, the mob does not rule.


  1. Anonymous9:25 AM

    I hope you mean Krugman and not Samuelson...otherwise it's a pretty faint drum.

  2. Paul Samuelson is considered one of the greatest economists of the 20th century, so, I think he's an appropriate anecdote.

  3. "...I can't remember any occasion where some scientist, on the spot, changed his mind about something important."

    I think Sagan is talking about the daily work of science. And if so, then it shouldn't be too surprising that you can't think of such an instance because the daily work of scientists is not highly visible. It's kind of like saying that you can't think of a time when a manufacturing engineer changed his mind on the spot about something important in manufacturing engineering.

    But you can see that in science (at least in the physical sciences, where Sagan worked) ideas do rise and fall according to evidence as it is developed. If it's happening in the field altogether, then it probably is happening with individuals in the field at least some of the time.

  4. Robert: can you give me an example of something? All I can think of, are minor facts that are themselves ambiguous.

  5. Anonymous11:50 AM

    When Russel published 'the foundation of mathematics', Dedekind was in the process of final review of the second edition of its own book on set theory.

    It was too late to stop the publishing process, so Dedekink simply added a famous preface stating that Russel's work made all of his own false.

  6. Anonymous11:53 AM

    but then, mathematics is not a science.

  7. Anonymous12:14 PM

    I was told at one point that the phenomenon known as "Newton's rings" caused Newton to (somewhat abruptly) abandon his corpuscular theory of light in favor of a wave theory. Unfortunately, I'm having trouble finding a good reference. (The fact that he changed from a particle view to a wave view is clear, the timescale less so.)

  8. Anonymous9:16 PM

    Completely wrong about scientists changing their mind. Eric, I think you are thinking of economics, not real science. Yeah scientists will not change their mind on whether they believe in God or something like that, but strong evidence always changes minds. Also, real science is not so much about opposing ideas - it's about adding to the body of knowledge. New things don't usually contradict old things, they fill in details that were previously blurry. But there are examples of rapidly changed minds. When Watson and Crick published their DNA structure Pauling immediately abandoned his triple-chained model. When Dolly the sheep was cloned from adult cells the people who said that wouldn't be possible (mostly) relented quickly.

  9. anon: I would call those minor concessions. Think in science whether someone changed their mind on global warming , string theory, evolution, ethanol-based energy, or stem cells.

  10. Anonymous8:00 PM

    Well, since I was in college ('70s) the cosmoological constant went from Einstein's biggest mistake to accepted, the neutrino ceased to be a massless particle, etc. (In science, the changes are not overnight, but do occur within a career. In animal behavior, the nonviolent chimpanzee has disappeared, and in anthropology, I did read an anthropologist admit he was wrong about the San.
    As for your examples: 1)one of the leading warmists was preaching the return of the ice age during the '70s, 2. no one can solve the string theory equations, 3) a lot of scientists switched to evolution when it first came out, and again when radioactivity showed the earth was old, but there hasn't been any real data changes since then, 4) no real scientific question, you can make ethanol and run a car on it, the only question "does it make any sense?", and 5) the argument is moral, "you can get stem cells from a fetus, but should you?"