The Realism vs. Anti-Realism Debate
The mid-1980's saw a transformation of the debate over "scientific rationality" which had been unleashed by Kuhn's perceived challenge to the traditional claim that scientific belief is determined by evidence and reasoning. Kuhn undoubtedly sought to differentiate his view from that attributed to him by his most radical defenders, but his apparent endorsement of the "sociological thesis," in effect transformed the traditional philosophical task of accounting for scientific rationality into a sociological account of the behavior of concrete individuals in specific social situations. Since the goal of the pro-sociology camp is in effect to put philosophy of science out of business (at least as traditionally conceived), philosophers of science sought to distance themselves in varying degrees from the extreme relativism implied by the most radical reading of Kuhn. However, among the sociologists of science there was general enthusiastic agreement on the relativity of scientific beliefs, even though there was considerably less agreement as to what they were relative to.

Thus philosophers found themselves disagreeing less and less over whether or not science exhibited a "rationality" -in that sense they were almost all "pro-rationality"- in alliance over the relativist sociological foe, but they found themselves in serious disagreement over the nature of that "rationality." On one side there emerged a group who defended scientific rationality in the name of what has been called "scientific realism" or, in this context, simply "realism."

What is Realism?

Realist justifications of scientific rationality are anchored in the claim that scientific knowledge aspires to discover the truth about how things really are.  Moreover, realists claim that if it is rational to accept a theory, it is equally rational to believe in the existence of the "theoretical entities" posited by such a theory, even though such theoretical entities cannot be directly observed.

"Realism" can be thought of as a philosophical theory answering the old question which we called the "Problem of Authority": how can we justify the claim that it is rational to believe scientific explanations?  The realist answers by saying the ultimate authority which justifies the rationality of scientific beliefs is simply that they are true in the sense of "truth"as a relation of correspondence between what we believe to be the case and what in reality is the case.

Alternatively realism can be thought of as a theory about the aim of science: scientific theories aspire to tells us the truth about the world.  Thus it is an Axiological theory about science which holds that all science has one fixed goal: finding out the truth about the nature of reality.

Scientific methodology -the realist claims- is adopted to the extent that it proves an effective method of attaining the truth, and all other goals to which scientific knowledge may aspire are ultimately dependent upon how things really are. The realist may well admit that there may be more than one way to describe the world, but the ones which are selected are selected because they are -of competing rivals at any point in history- the closest to the truth about the nature of reality. Furthermore, the history of science is also rational in the sense that it represents progress in getting closer to the truth about how things really are; today's beliefs are closer to that goal than were the discarded beliefs of former times.

Who are the realists' opponents?

Sociologists who accept some form of the sociological thesis of Kuhn reject realism and its story about progress as "approaching the truth" as a philosophical myth (and no doubt account for the myth making tendency of philosophers by some social/psychological explanation). It is typical of some philosophers who hold this sociological outlook to speak of "Philosophy" as having been terminated with the end of the "modern" period, so these "postmodern" thinkers disown the label "Philosopher." Their viewpoint might be labeled a "non-philosophical anti-realism" or, more descriptively, a "social constructivist anti-realism." This outlook is most commonly found in programs or departments which are typically called "Science Studies" programs, by which is intended a cluster of different social sciences applied to the study of scientists and their activities, including not only "sociology" but also psychology, economics, linguistics, cognitive science, gender and ethnic studies, etc. This group is one of the opponents of realism.

On the philosophers' side, however, there also flourished in the 1980's a wide variety of anti- or non- realisms which opposed realism in a way quite differently from the social constructivists. These philosophical anti-realists saw the realist defense of rationality as anchored in truth as tantamount to handing the sociologists victory. It is because scientific rationality has been misconstrued by realists in terms of "mirroring nature" (i.e. the correspondence theory of "truth") that the relativists are able to gain easy victories. For these "philosophical anti-realists" scientific rationality needs to be de-coupled from talk of "truth" and "reality" and re-expressed in terms of "empirical adequacy" and a variety of "pragmatic" criteria.

These anti-realist philosophers agree with the (non-philosophical) sociologist anti-realists that scientific beliefs are largely a "construction," but -in opposition to the sociologists- they deny that this construction is explained by social factors "external" to scientific knowledge.  Instead they seek to justify a scientific rationality according to which a belief is adopted because it makes correct predictions about observable outcomes and displays other pragmatic virtues such as simplicity, breadth of scope, fertility, etc.

The philosophical anti-realists are today's descendants of older empiricists, and they perceive themselves as such. Their empiricism makes them wary of "metaphysical" claims about the "nature of reality" and enthusiastic for "observational claims" which can be terminated in an actual experiential state in which some piece of data is recorded. Only this latter class of observational statements need be accorded "truth" and their truth is settled directly by an appeal to experience. They are said to be "directly verified" by experience. All other statements which scientists make about what is not directly observed need not be thought of as "true" (or "false") but only as "successful" as a means for predicting or deducing "observation statements." Any terms in theories which appear to refer to "unobservable" entities, states, or processes, should not be understood as referring to real events behind the screen of phenomena which we observe. Such "putatively referring terms" (appear to refer but don't) are constructs of our theories, the acceptance of which is solely a function of their empirical adequacy (getting the numbers right), with the usual pragmatic virtues providing further back up when there's a choice to be made between empirically equivalent theories. Thus these anti-realists may also be called "empiricist/pragmatist anti-realists."

This empiricist/pragmatist anti-realism is pro-rationality (unlike sociological anti-realism) but anti-truth; throughout the eighties and well into the nineties philosophy of science centered on debates between these anti-realists and realist defenders of scientific rationality.

What reasons do realists give for their view?

Realists see philosophical anti-realists as making the classical pragmatists' blunder of confusing the process by which human inquiry leads to a particular belief as the rational belief to accept with what makes that belief acceptable. The pragmatist/empiricist position of the anti-realists provides a candidate for addressing the first issue (how we decide what to accept), but ignores the second (what makes that belief acceptable).

From the empiricist/pragmatist anti-realists' point of view, there is no difference between these two; what makes a belief acceptable is just that: the process by which human inquiry leads to a particular belief as the rational one to accept. The realist disagrees: what leads a scientist to accept a belief is the empirical evidence (combined with the pragmatic virtues) -with that much of what the anti-realists say the realists agree- but the realist wants to add: what makes the process of inquiry settle on that particular belief is the fact that out of  currently competing beliefs, it is closest to the truth.

Why, asks the realists, would a particular theory be more successful than its rivals at "getting the numbers right"?  In other words what causes this particular theory to be so "empirically adequate"? The best way to explain the marvelous empirical adequacy of our best theories -the realist contends- is to claim that (of competing rivals), they are closest to the truth. So, by "inference to the best explanation" realism is demanded to explain the "success" of science. The anti-realists say we accept a theory because of its success, and realists do not dispute that. But the anti-realists do not address what makes it so successful. Therefore, on their anti-realist account, so the realists tell us, that success is "miraculous." But, the realist continues, there are no miracles, so the anti-realist account is unacceptable. (This is known as the "no miracles" argument.)

Realists see scientific inquiry as discovery while anti-realists sees it as invention.  For the realist there is a "way things really are" and science is trying to find out what it is; it endeavors to discover the "truth."  For the anti-realist there is no way things are apart from how our theories construct them.  All "worlds" are constructions of how we view the world, of our theories.   Therefore there is no "way things are" to discover the truth about.  To think of theoreis as "true" or "false" descriptions of an unseen world "beneath" or "behind" the phenomena we observe is to mistake what theories are.  Scientific theories are not attempts to describe what is (allegedly) the real cause of phenomena by "representing" or "mirroring" an independent "reality" as it exists apart from the phenomena we experience.  They are invented by theoreticians to serve as tools for making observational predictions about empirical phenomena.

The anti-realist is unimpressed by this realist "explanation" of the success of science simply because -given an anti-realist understanding of scientific theories- scientific theories really do not "explain" anything at all; they are just successful predictors, i.e. "tools" or "instruments" for, as scientists often say, "getting the numbers (in other words the observational data) right." The realist craves "explanation"; thus the realist regards the extraordinary accuracy of contemporary scientific theories as a phenomenon that needs an explanation; and, so the realist argues, the best explanation we can provide for this marvelous "success" is that such theories are (at least close to) the truth about reality. The anti-realist regards that realist craving for "explanation" as something we need to learn to do without.

Defenders of scientific realism are by no means all agreed upon a single doctrine.  There are many ways of distinguishing different realist views.  One such distinction is between "entity realists" and "theory realists":

"Entity realists" defend the "reality" of the entities to which at least some purely theoretical terms refer and de-emphasize the problems of "truth" as correspondence to reality; they will agree with anti-realists that theories may be only "useful instruments" more than "approximately true descriptions" of an unobservable reality, but these theories enable us to discover real hidden (unobservable) processes and entities causing observable phenomena. Entity realists tend to be naturalists and may emphasize experiment over theory as the road to discovery.

"Theory realists" on the other hand pursue theories of "correspondence" and reference and theories of approximate truth, and will stress logical and linguistic arguments for truth.

Contemporary realists may be choosy about what theories or entities to regard "realistically"; in a naturalistic vein one may argue that a realistic interpretation is itself a function of empirical evidence and thus as evidence changes what we may rationally regard as real also changes.