Uncertain Principles

Toward the end of the post, Janet writes:

The point here is that science engages in a certain problem-solving strategy that works really well for understanding certain kinds of systems. And that problem-solving strategy is continuous with common sense rather than completely alien to it. Cultivating habits of observing carefully, identifying different controllable variables, and attending to what happens when those variables are controlled in various ways is thinking like a scientist.

In comments, Jonathan E. objects:

Science is not the continuation of common sense. There is nothing that I would call common sense in theories like quantum mechanics, evolutionary theory and so on. There is nothing I would call common sense in the method of science: the simple fact is that no one had succeed in supplying a reasonable model of "how we do science".

I'll leave the comment about evolution for someone with a larger stake in that fight, and just object to the quantum mechanics part.

While I'll be one of the first to tell you that many of the predictions of quantum theory defy our ordinary common-sense picture of how the world works (in fact, my students are probably sick of hearing about that), that's not the kind of common sense that Janet's talking about (at least, that's not the way I read her comments). Quantum mechanics is a tremendously counter-intuitive theory (gloriously so, I would say), but as a procedural matter, it absolutely and rigorously follows the rules of common sense.

When Janet describes the "problem-solving strategy" of science as "continuous with common sense," I think that's a procedural matter. That is, she's not saying that the predictions of science are obvious, but that the process of getting there follows normal, recognizable rules of logic. In fact, science relies on that.

One of the bedrock principles of science is repeatability: If you perform the same experiment the same way a hundred times, you can confidently expect to get the same answer every time (or, possibly, a statistical distribution over the set of possible answers). That's as true for quantum systems as classical ones-- the answers you get may not fit with a common-sense view of the world (particles may behave like waves or vice versa), but they'll be the same every time. And what's more common sense than that-- one of the definitions of insanity, after all, is "doing the same thing twice and expecting different results."

When an experiment fails to show repeatable results, there's a reason for it. There's some parameter that isn't properly controlled, or some new effect that wasn't anticipated popping up to change the possible outcomes. These parameters or effects may be unexpected, but they're explainable. There's no room for "it's just one of those things..." in the normal practice of science. And again, that's entirely consistent with common sense-- when something happens, there's a reason for it (in a causal sense, not a metaphysical one).

And, again, there's internally consistent logic to science. Some of the rules may seem a little foreign, and there's occasionally a sort of Sherlock Holmes twist (Planck's quantum hypothesis being a great illustration of "whatever remains, however improbable"). You don't get to make up new rules in the middle of the game, unless you've got the evidence to support the existence of those rules. The famous Sydney Harris "Then a miracle occurs" cartoon is funny because it's not science, not because it's an accurate reflection of how we work.

I won't attempt to claim that the postulates and predictions of quantum mechanics align with common sense. But as a procedural matter, the process used to apply those rules, and the way we explore and interpret those results is absolutely consistent with the common-sense picture of how the world works.

It's easy to forget that, and drift off into superstition (and I'll dig up a Classic Edition post on just that topic for later today), but to lowest order, science is nothing but common sense put into practice.