Scientific Realism

Annual Question: 

Rebecca Newberger Goldstein

Has science discovered the existence of protons and proteins, neurons and neutrinos? Have we learned that particles are excitations of underlying quantum fields and that the transmission of inherited characteristics is accomplished by way of information-encoding genes? Those who answer no (as opposed to dunno) probably aren’t unsophisticated science deniers. More likely they’re sophisticated deniers of scientific realism.

Scientific realism is the view that science expands upon—and sometimes radically confutes—the view of the world that we gain by means of our sense organs. Scientific theories, according to this view, extend our grasp of reality beyond what we can see and touch, pulling the curtain of our corporeal limitations aside to reveal the existence of whole orders of unobserved and perhaps unobservable things, hypothesized in order to explain observations and having their reference fixed by the laws governing their behavior. In order for theories to be true (or at any rate, approximations of the truth) these things must actually exist. Scientific theories are ontologically committed.

Those who oppose scientific realism are sometimes called scientific non-realists and sometimes, more descriptively, instrumentalists. Their view is that scientific theories are instruments for predictions that don’t extend our knowledge of what exists beyond what is already granted to us by way of observation. Sure, theories seem to make reference to new and exotic entities, but bosons and fermions don’t exist the way raindrops on roses and whiskers on kittens do. Quantum mechanics no more commits us to the existence of quantum fields than the phrase “for our country’s sake” commits us to the existence of sakes. The content of a theory is cashed out in observable terms. A theory is a way of correlating observable input with observable output, the latter known as predictions. Yes, between the report of what has been observed and the prediction of what will be observed there is a whole lot of theory, complete with theoretical terms that function grammatically as nouns. But don’t, as Wittgenstein warned, let language “go on holiday.” These theoretical nouns should be understood as convenient fictions, to be spelled out in operational definitions. Science leaves ontology exactly as it finds it.

Instrumentalism is so deflationary a view of science that one might think it was conceived in the bitter bowels of some humanities department, determined to take science down a notch. But in fact in the 20th century instrumentalism became standard in physics for a variety of reasons, including the difficulties in solving the stubborn measurement problem in quantum mechanics. Then, too, there was strong influence wafting from the direction of logical positivism, the program that, in an effort to keep meaningless metaphysical terms from infiltrating our discourse and turning it into fine-sounding gibberish, had proposed a criterion of meaningfulness that pared the meaning of a proposition down to its mode of verification.

The thrust of these pressures drove many of the most prominent scientists towards instrumentalism, by which scientists could both wash their hands of an unruly quantum reality, rife with seeming paradox, while also toeing the strict positivist line (as evidenced by the frequent use of the word “meaningless.”) The Copenhagen Interpretation, which was accepted as standard, dismissed the question of whether the electron was really a particle or a wave as meaningless, and asserted that to ask where the electron was in between measurements was likewise meaningless.

There were, of course, scientists who resisted—Einstein, Schrödinger, Planck, de Broglie, and, later on, David Bohm and John Stewart Bell, staunch realists all. Said Einstein: “Reality is the business of physics,” which is about as simple and direct a statement of scientific realism as can be. But Einstein’s realism marginalized him.

The Copenhagen Interpretation is no longer the only game in town, and the main competition—for example, the many-worlds interpretation, in which what quantum mechanics describes is a plethora of equally realized possibilities, albeit existing in other worlds, and Bohmian mechanics, in which the unobserved particles are nonetheless real particles having actual positions and actual trajectories—are realist interpretations. Though they differ in their descriptions of what reality is like, they unflinchingly commit themselves to there being a reality that they are attempting to describe. So far as their empirical content goes, all these interpretations are equivalent. They are, from the standpoint of instrumentalism, indistinguishable, but from the standpoint of scientific realism vastly different.

The questions that press up behind the concept of scientific realism are still very much in play, and how we answer them makes a world of difference as to what we see ourselves doing when we’re doing science. Are we employing a device that churns out predictions, or are we satisfying, in the most reliable way that we have, our basic ontological urge to figure out where we are and what we are? Are we never carried, by way of science, beyond the contents of experience, or does science permit us to extend our reach beyond our meager sensory apparatus, enabling us to grasp aspects of reality—the elusive thing in itself—that would be otherwise inaccessible?

What a different picture of science—and of us—these two viewpoints yield. What then could be more central to the scientific mindset than the questions that swirl around scientific realism, since without confronting these questions we can’t even begin to say what the scientific mindset amounts to. 


[ Fri. Dec. 23. 2016 ]