Page 63 - Encyclopedia of Philosophy of Language

The instrumentalist distinguishes between obser- vational and theoretical sectors of language and main- tains that only sentences of the former sector are objectively true or false. Theoretical sentences are instrumentalistically useful to the extent that their observational consequences turn out to be true but truth for the theoretical sector consists in nothing more than such predictive utility. For the scientific realist, by contrast, even if one can draw a significant distinction between observational and theoretical sec- tors, which some doubt, there is no difference in respect of truth: sentences from either sector are rend- ered true or false by a mind-independent world. Scien- tific realism is also often taken to include the antiskeptical view that science inevitably progresses, if correctly pursued, closer and closer to the truth, and that the theoretical terms of current science are not empty but refer to mind-independent entities with natures of roughly the same type as we ascribe to them.
The extreme instrumentalist denial that the concept of truth applies to scientific theory (found in some logical positivists) did not survive Tarski's dem- onstration of the definability of truth for certain for- mal languages. But instrumentalistic views are common amongst philosophers influenced by the pos- itivists or by pragmatism. W. V. O. Quine, for instance, draws a fairly sharp observation/theory dis- tinction and denies there is any mind and theory- independent distinction of fact not reflected in a difference in observational data. His views incline towards a relativism regarding truth for those theories compatible with, but transcending, the empirical facts. Similarly relativistic views on scientific truth can be found in many other influential philosophers (in the early writings of T. S. Kuhn, for instance) and are incompatible with the realist view of a mind- independent realm of facts against which theories can be measured absolutely for truth or falsity.
As well as scientific realism, 'realism' is also cur- rently used as a name for certain fallibilistic doctrines. The fallibilist about a certain class of beliefs holds that such beliefs, even if arrived at in optimal conditions for belief formation, can be whollyfalse. Thelinguistic turn of modern philosophy has led to fallibilism being interpreted as a semantic doctrine: that truth for sen- tences transcends the evidence for them. A tension thus emerges between the skepticism of fallibilism and the antiskepticism of the scientific realists (who often claim the semantic doctrine of fallibilism has little to do with traditional realism). Nonetheless the idea of mind-independence needs explicating—antirealists agree the world is independent of, for example, the will—and fallibilism offers one fairly clear way of doing so. On the other hand, traditional realism did not involve radical skepticism towards sentences such as 'there are physical objects,' the very reverse in fact. Even if the fallibilist evinces no actual doubt regarding
this sentence but holds merely that our opinions and the truth on this matter might have come apart, realists with a naturalistic approach to cognition might demur. Conversely one might take a fallibilist view towards, say, mathematics while denying that it deals with mind-independent entities. A compromise is to characterize realism towards a class of entities as the view that some beliefs (typically linguistically expressed) about them are fallible. Antirealist fal- libilism in mathematics then consists in denying that mathematics is about anything. Explicating just what 'aboutness' comes to, is the major difficulty with this proposal.
Fallibilism is taken to be the key realist notion by philosophers skeptical of its truth such as Hilary Put- nam and Michael Dummett. They agree of course that one is often wrong; that, for example, the objects one applies terms to in actual linguistic practice are often not those one ought to apply them to if one is to respect the objective content of the expression. But they insist objectivecontent is rooted in our linguistic practices too. Empiricist views of sentence meaning— the meanings transmitted and learnt in language- learning cannot transcend the empirical circumstances of the learning—then lead them to suggest the notion of truth is identified or replaced with epistemic notions such as verifiability or justified assertibility and to deny that beliefs arrived at in optimal circumstances can be false. Thus metaphysical doctrines such as realism are held to depend on theories from the phil- osophy of language.
Critics of such views counter that they rest on poor epistemology. Skeptics about the notion of non- deductive justification will find the notion of justified assertibility dubious. And Chomskyans will charge the antifallibilists with a crude empiricist theory of the acquisition and transmission of linguistic meaning, holding by contrast that the empirical circumstances of language learning may act only as a catalyst acti- vating a largely innate cognitive state of under- standing which can transcend its experiential inputs. Even if the general negative arguments against realism fail, however, the task of establishing it as correct for particular cases, for example, for microphysical objects, quantum mechanical states, abstract objects, and so forth, is still a substantial and unrealized one.
See also: Instrumentalism.
Bibliography
Devitt M 1991 Realism and Truth. Blackwell, Oxford Dummett M 1978 Truth and Other Enigmas. Duckworth,
London
Kuhn T S 1970 The Structure of Scientific Revolutions. Uni-
versity of Chicago Press, Chicago, IL
Putnam H 1978 Meaning and the Moral Sciences. Routledge
and Kegan Paul, London
Quine W V O 1992 The Pursuit of Truth. Harvard University
Press, Cambridge, MA
Realism
41