11Big Questions: Philosophy of Science Handout 3Dr. H. BeebeeScientific Realism & Anti-RealismIntroductionScientific theories claim, or at least seem to claim, that the universe ispopulated by a host of entities that we cannot observe in any obvious sense: we havegenes, quarks, curved space-time, the superego (if you think psychoanalysis is ascience) etc. etc. Do we (including scientists) have any right to believe in suchentities? Scientific realists say yes - such entities really do exist - while anti-realistssay no. There are three basic positions which one can adopt on this issue:Realism: We have very good reason to believe that the unobservable entitiespostulated by well-confirmed theories exist.“Constructive empiricism”: We have no good reason to suppose that such entitiesexist. The evidence which supports scientific theories supports only the claim thatsuch theories are “empirically adequate” - that what they say about observable entitiesis true. We have no reason to suppose that what they say about unobservable entitiesis true.Instrumentalism: This is a thesis about the meaning of “theoretical” terms (i.e. termswhich appear to refer to unobservable entities). Instrumentalists claim that such termsdon’t really refer to any such entities. A theory employing theoretical terms is reallyonly “about” the observable world: what makes the theory true is the observable factsbeing the way the theory says they are. Theoretical terms are introduced into a theoryonly to make it simpler or more elegant. Their presence does not indicate anyontological commitment to unobservable entities “referred” to by the terms, since theterms don’t, despite initial appearances, refer to such entities.For example, suppose that Quantum Mechanics makes a claim about the quantumstate of a particular photon. None of the realist, constructive empiricist orinstrumentalist need deny that Quantum Mechanics is a hugely successful theory forpredicting and explaining the outcomes of all sorts of experiments. None of themthink there’s anything wrong with Quantum Mechanics. Rather, they disagree over:(a) what the theory is really telling us about the world when it saysthat, say, the photon is in state S,(b) whether we have any good reason to believe this, and(c) what the purpose of a theory like Quantum Mechanics is.Realism and constructive empiricism agree on (a): the theory is telling us exactlywhat it appears to be telling us, namely, that the photon really is in quantum state S.But they disagree on (b): the realist thinks we have every reason to believe what thetheory says about the quantum state, whereas the constructive empiricist thinks wehave no reason to believe it.Does it follow that the constructive empiricist takes rather a dim view of QuantumMechanics? After all, if the aim of science is truth (and this is what the realist thinks),and we have no reason whatever to suppose that what current science tells us aboutthe unobservable is true, it follows that we have no reason to suppose that QuantumMechanics is doing what a scientific theory ought to be doing, namely, telling us thetruth about the universe.22But the constructive empiricist does not take such a dim view of Quantum Mechanicsbecause she does not believe that the aim of science is truth. Rather, she thinks theaim of science is empirical adequacy: what we want and need from scientific theoriesis not the whole truth, but merely the truth about observable phenomena. It doesn’tmatter very much what a theory says about the unobservable, so long as it gets itspredictions right. So we shouldn’t really be concerned that we’ve got no reason tobelieve what Quantum Mechanics says about the photon’s quantum state - so long asthe observable consequences of this claim are borne out.Instrumentalism, on the other hand, disagrees with both realism and constructiveempiricism over (a). According to instrumentalism, what Quantum Mechanics isreally saying when it says that the photon is in quantum state S is merely that, if Iwere to make such-and-such a measurement, I would get so-and-so observable results.The terms “photon” and “quantum state” are not intended to refer to real features ofthe world; rather they have a purely instrumental role. “The photon is in state S” isjust a more convenient and more elegant way of saying “doing this experiment wouldget you these results”.Both forms of anti-realism (instrumentalism and constructive empiricism)thus agree,against the realist, that the primary aim of science is to get to the truth about theobservable world - that is, empirical adequacy.Two arguments for scientific realism1. The “no-miracles” (or “Ultimate”) argumentMany realists regard this argument as the strongest motivation for scientific realism:It would be a miracle, a coincidence on a near cosmic scale, if a theory madeas many correct empirical predictions as, say, the general theory of relativityor the photon theory of light without what that theory says about thefundamental structure of the universe being correct or “essentially” or“basically” correct. But we shouldn’t accept miracles, not at any rate if there isa non-miraculous alternative. If what these theories say is going on “behind”the phenomena is indeed true or “approximately true” then it is no wonder thatthey get the phenomena right. So it is plausible to conclude that presentlyaccepted theories are indeed “essentially” correct. (Worrall, p.101)Put simply: It would be a miracle if the universe behaved by and large - as it does - asif there were quarks and fields and photons, if in fact there were no such things. Butwe shouldn’t believe in miracles. The best explanation of why scientific theories areso successful - of why the universe behaves as if there were quarks and fields andphotons - is that there really are such entities. Hence we ought to believe that realismis true.An anti-realist responseThe no-miracles argument basically says that the only sensible explanation of thepredictive success of scientific theories is that what those theories say about theunobservable entities which give rise to the predicted phenomena is true. But Bas vanFraassen, the inventor of constructive empiricism, has another explanation ...... science is a biological phenomenon, an activity by one kind of organismwhich facilitates its interaction with the environment. And this makes methink that a very different kind of scientific