Putnam, Hilary (1926–)
Putnam, Hilary (1926–)
Hilary Putnam, after receiving a BA from Pennsylvania (1948) and a year spent at Harvard (1948–1949), studied at the University of California, Los Angeles, taking his doctorate in 1951 with a dissertation titled "The Concept of Probability: An Application to Finite Sequences." He taught at Northwestern (1952–1953), Princeton (1953–1961) and MIT (1961–1965), becoming Walter Beverly Pearson professor at Harvard in 1965. From 1995 to 2000, he served as Cogan University Professor there, becoming emeritus in 2000. He has been influential in most areas of philosophy, particularly in the philosophy of language, of logic, of mathematics, and of science.
Putnam is sometimes thought of as often changing his mind. (See, for example, the Dictionary of Philosophers' Names.) Sometimes he has. But in central respects he has held a single, though developing, position since the mid-1950s, a position that in some aspects resembles the later Ludwig Wittgenstein's. This entry sets out some constant central themes.
Putnam was among those American philosophers to benefit directly from the intellectual exodus from Europe caused by Nazism. He was a student of Rudolf Carnap and of Hans Reichenbach. Though his approach to issues is quite different from theirs, Reichenbach in particular had a lasting and often acknowledged influence on Putnam's thought. Putnam's innovations stand out when it is noted. In Realism with a Human Face (1990, p. 289), Putnam remarks,
In Theory of Relativity and A Priori Knowledge (1922) Reichenbach listed a number of statements … each of which Kant would have regarded as synthetic a priori, and each of which can be held immune from revision …, but which collectively imply statements that are empirically testable, and that Kant would, therefore, have to regard as a posteriori.
Certain principles had, in Immanuel Kant's time, as good a claim as any to fix how particular spatial, temporal, and other concepts are to be applied, and thereby which concepts those were; to be intrinsic to the concepts involved, thus "conceptual truths," thus a priori. Relativity theory allows us to see how they are at least jointly testable, so that some may turn out false. Such, it seems, is a fate to which a priori truths are liable.
Putnam reports Reichenbach as making a related point to his classes. Considering questions such as "How can we show that that blackboard is wider than this ashtray?," he argued that any system of measurement, or of observation, treats some propositions that seem empirical (such as "mere translation does not make things grow or shrink") as axiomatic. One cannot sensibly apply the system while doubting these propositions; they are not subject to confirmation or refutation within the system. But it could prove reasonable to replace the system with another in which these propositions are testable, so possibly false. In that sense they are empirical.
There are two contrasting reactions to these points. One is: What this shows is that every concept commits itself to a particular empirical theory. If the theory proves false, then the concept is incoherent, so without application and to be discarded. This was Paul Feyerabend's reaction, and it is also Paul Churchland's.
The other reaction is: If we are confronted with situations that force giving up what seemed conceptual truths, it may appear that the concepts whose applications seemed to be governed by those principles are, in fact, otherwise governed. Perhaps the application of the concept "straight line" to items in the world is not governed by the Euclidean parallel postulate, but rather in such and such other way. That reaction grants face value to Reichenbach's point that the same proposition that is axiomatic in one system may be testable and false in another. This was Putnam's reaction. He developed the position and drew its implications in a powerful series of papers in the 1950s and 1960s (see Putnam, 1962a–d). Part of the idea is that what principles govern the application of a concept depends in part on how the world in fact is. Putnam defined that role for the world in "The Meaning of 'Meaning'" (1975). This last article, though not published until 1975, was completed by 1968.
By the early 1970s, Putnam had begun to emphasize some new themes. For one thing, he became increasingly impressed with what he calls the "interest relativity" of such notions as explanation and cause. The general point is: What a concept counts as applying to—the correct way of applying it—varies with the circumstances in which it is to be applied. A concept may count, on one occasion, as fitting what it does not count as fitting on another. That is continuous with Putnam's earlier reaction to Reichenbach. The point then was: What it is reasonable to judge as to how a concept operates depends on the conditions in which such judgments are made. The point now is: What those conditions are depends not just on how the world is, but may vary from occasion to occasion, given the world as it is. Not coincidentally, this point went along with other developments in Putnam's thought.
The first of these developments is what he calls "internal realism," first presented in 1976, and amplified in his writings of 1981, 1983, 1987a, 1987b, and elsewhere. The position includes four points. First, there are mundane, true things to say about what our words and thoughts are about: "the word 'gold' means (refers to) gold; this is gold"; "This is a chair; this is what 'chair' refers to," and so forth. Second, there are philosophical dicta that sound much like such mundanities, or their denials, or generalizations of these, but that say, or try to say, something quite different. They are bad answers to the following pseudo-problem. On the one hand, there are thoughts and words—items that purport to represent the world as being thus and so; on the other hand, the items the world in fact contains, which are what and how they are independent of what we think, or do not think, about them. How are our words and thoughts related to these items? How, if at all, does their truth depend on how those items are? And how could they be so related? Internal realism holds that the problem rests on a mistake; hence so do any 'solutions', which take it at face value.
Third, the mundane remarks (point one) are correct because they are a feature of how these words are (or are to be) used. But that formulation depersonalizes things misleadingly. The standard for the correctness of a statement cannot be fixed independently of what users of the relevant words and concepts—that is, human beings—are prepared to recognize as correct: What Putnam identifies as our (human) perceptions of rationality and reasonableness. What it is for a statement to be correct depends on the sorts of beings we are, and is not reducible to some set of principles that would have to hold anyway. Fourth, it is part of what we are prepared to recognize as rational that any concept might be applied correctly in different ways in different circumstances. What sometimes counts as the cause of the explosion may not at other times. It is because human rationality is occasion-sensitive that the problem mentioned in point two is a pseudo-problem. We cannot sensibly take a "God's-eye view" of how we relate to the world, trying to say how our concepts would apply without us.
The occasion-sensitivity of rationality does not mean that truth is relative, or that there are no objective facts—given a framework, or setting, in which concepts are to be applied. Nor does giving up on a God's-eye view mean a deflationist account of truth. Putnam insists that we cannot comprehend what truth is without understanding the role of truth in our lives, notably in our activities of asserting, and of treating assertions in the ways we do; and that deflationism does not help us understand the role of truth in human life.
In arguing against the possibility of a God's-eye view, Putnam has produced what are probably his most discussed arguments. In one he identifies the God's-eye view (what John McDowell has called "the view of the cosmic alien") as one from which we may consider our own language as an uninterpreted calculus with a range of possible interpretations, and then ask which interpretation is the right one. In what he first saw as a generalization of the Skolem paradox, he argues that, in that case, nothing could make one interpretation the right one, so we could not ever be talking about anything (or about one thing rather than others). But we cannot pose serious problems without talking about definite things. This is a reductio of the idea of a God's-eye view (see "Models and Reality," Philosophical Papers, Vol. 3).
In another argument Putnam considers the (apparent) question whether we might be brains in a vat: that we are, and always have been, nothing but brains, kept alive by a bath of nutrients, fed computer-generated stimuli through electrodes. He argues that if the God's-eye view is possible, then that we are, and that we are not such brains should both be possibilities. But for the words of the question to mean what a God's-eye view requires them to means we must be using them in ways that entail that we are not brains in vats. For, as argued in "The Meaning of 'Meaning,'" what our words mean depends, inter alia, on how we are in fact connected to the world, and not just on what we may anyway be aware of. For our "brain" to mean brain, and our "vat" to mean vat, we must be connected to the world as brains in ways vats could not be. So we cannot formulate what, from a God's-eye view, ought to be a possibility, in a way that makes it possible. That is another reductio of the idea of a God's-eye view (see Putnam 1981).
Equally important to internal realism are Putnam's arguments against a causal theory of reference: Arguments, based on the interest-relativity of causation, that our being causally linked to the world as we, in fact, are is not enough in itself to make some one interpretation of our language correct—once it is granted that the language we speak may coherently be viewed by us as less than fully interpreted, so open to interpretation. These arguments appear in many replies to critics, and notably in "Realism with a Human Face" (1990).
At about the time Putnam began to develop internal realism, he also began to change his way of thinking about human psychology, rejecting a picture of it, and with that, a view he once espoused—functionalism. Viewed one way, a human being is an organism constructed in a particular way, a particular battery of mechanisms arranged to interact with each other and the environment in given ways. If, while taking that view, we ask what it is for someone to believe that Mars is a planet, or to have any propositional or other attitude—to be in a mood, experience an emotion, and so on—it is tempting to look for an answer by trying to identify some state(s) of some mechanism(s) such that for someone to believe that is for him to be structured like that. In that frame of mind, for example, one might speak seriously of someone having a "token of a mentalese sentence" in his "belief box." This is the picture Putnam rejects.
Against it Putnam notes that to ascribe belief to someone is to relate that person to the world as we view it, and to ourselves, as on the same side as ours, or a different one, with respect to such and such question as to how the world is, and so on. Given internal realism, this means that there will be different truths to tell on different occasions as to what a given person, as he is at a given time, then believes. So for someone to be as said to be when we say him to believe thus and so, cannot be for him to have some particular mechanism, otherwise identifiable, in some particular state. And so on for other mental states (see Putnam 1989).
Putnam has been refining the ideas discussed above, notably the idea of a distinction between ordinary and philosophic statements, and applying them in new areas, such as philosophy of mathematics. The above indicates a few main themes, omitting Putnam's striking arguments for them.
works by putnam
"The Analytic and the Synthetic" (1962a). Reprinted in Mind, Language and Reality, Philosophical Papers, Vol. 2. Cambridge, U.K.: Cambridge University Press, 1975.
"It Ain't Necessarily So" (1962b). Reprinted in Mathematics, Matter and Method, Philosophical Papers, Vol. 1. Cambridge, U.K.: Cambridge University Press, 1975.
"Dreaming and 'Depth Grammar'" (1962c). In Philosophical Papers, Vol. 2. Cambridge, U.K.: Cambridge University Press, 1975.
"What Theories Are Not" (1962d). In Philosophical Papers, Vol. 1. Cambridge, U.K.: Cambridge University Press, 1975.
"The Meaning of 'Meaning.'" In Philosophical Papers, Vol. 2. Cambridge U.K.: Cambridge University Press, 1975.
"Realism and Reason" (1976). Reprinted in Meaning and the Moral Sciences. London: Routledge and Kegan Paul, 1978.
Reason, Truth and History. Cambridge, U.K.: Cambridge University Press, 1981.
"Models and Reality." Reprinted in Realism and Reason, Philosophical Papers, Vol. 3. Cambridge, U.K.: Cambridge University Press, 1983.
The Many Faces of Realism. LaSalle, IL: Open Court, 1987.
"Realism with a Human Face." In Realism with a Human Face, Philosophical Papers, Vol. 4. Cambridge, MA: Harvard University Press, 1990.
Renewing Philosophy. Cambridge, MA: Harvard University Press, 1992.
Words and Life. Cambridge, MA: Harvard University Press, 1994.
works on putnam
Clark, P., and B. Hale, eds. Reading Putnam. Oxford: Blackwell, 1994.
Ebbs, G. Rule-Following and Realism. Cambridge, MA: Harvard University Press, 1997.
Charles Travis (1996)