2. Being That Can Be Understood Is Language
RICHARD RORTY
In a book called Reason in the Age of Science, Hans-Georg Gadamer asked the question: Can “philosophy” refer to anything nowadays except the theory of science?[1] His own answer to this question is affirmative. It may seem that the so-called “analytic” tradition in philosophy—the tradition that goes back to Frege and Russell and whose most prominent living representatives are Quine, Davidson, Dummett, and Putnam—must return a negative answer. For that tradition is often thought of as a sort of public relations agency for the natural sciences.
Those who think of analytic philosophy in this way often describe Gada-mer's own work as a sort of apologia for the humanities. In this view of the matter, each of what C. P. Snow called “the two cultures” has its own philosophical claque. Those who accept Snow's picture of the intellectual scene think of the quarrel over science versus religion that divided the intellectuals of the nineteenth century as having evolved into the contemporary quarrel between the kinds of people whom we Californians call the “tech-ies” and the “fuzzies.”
This crude and oversimplified picture of the tension within contemporary philosophy is not altogether wrong. But a more detailed account of the history of philosophy in the twentieth century would distinguish between a first, scientistic phase of analytic philosophy and a second, anti-scientistic phase. Between 1900 and 1960 most admirers of Frege would have agreed with Quine's dictum that “philosophy of science is philosophy enough.” But a change came over analytic philosophy around the time that philosophers began reading Wittgenstein's Philosophical Investigations side by side with Kuhn's Structure of Scientific Revolutions. Since then, more and more analytic philosophers have come to agree with Putnam that part of the problem with present-day philosophy is a scientism inherited from the nineteenth century.
Putnam urges us to give up the idea that natural science has a distinctive “method,” one that makes physics a better paradigm of rationality than, for example, historiography or jurisprudence. He is joined in this appeal by philosophers of physics like Arthur Fine, who asks us to abandon the assumption that natural science “is special, and that scientific thinking is unlike any other.”[2] Putnam and Fine both ridicule the idea that the discourse of physics is somehow more in touch with reality than any other portion of culture. Post-Wittgensteinian Anglophone philosophy of language, of the sort found in Putnam, Davidson, and Brandom, has collaborated with post-Kuhnian philosophy of science, of the sort found in Latour, Hacking, and Fine. The result of this collaboration has been a blurring of the lines between the sciences and the humanities, and an attempt to make Snow's techie-fuzzie controversy seem as quaint as the nineteenth-century debate over the age of the earth.
This is not to say that scientism is dead. There are many distinguished analytic philosophers, particularly admirers of Kripke like David Lewis and Frank Jackson, who are unabashed physicalist metaphysicians. They think of themselves as continuing the struggle against mystificatory nonsense that Thomas Huxley waged against Bishop Wilberforce, Russell against Bergson, and Carnap against Heidegger. These philosophers still award a special on-tological status (“fundamental reality”) to the elementary particles discovered by the physicists. They believe that natural science gives us essences and necessities. They think that Wittgensteinian philosophers of language are dangerously irrationalist in saying that all distinctions between essences and accidents, or between necessities and contingencies, are artifacts that change as our choice of description changes. They think that Kuhnian philosophers of science are equally misguided in refusing to grant natural science any metaphysical or epistemological privileges.
This quarrel over whether natural science is special presently dominates analytic philosophy. I want to suggest that a much-quoted sentence from Gadamer might serve as a slogan for those philosophers of language and science who follow Putnam and Fine rather than Kripke and Lewis. The sentence is: “Being that can be understood is language” (“Sein, das verstanden werden kann, ist Sprache”). That claim encapsulates, I shall argue, both what was true in nominalism and what was true in idealism.
Let me define “nominalism” as the claim that all essences are nominal and all necessities de dicto. This amounts to saying that no description of an object is more true to the nature of that object than any other. Nominalists think that Plato's metaphor of cutting nature at the joints should be abandoned once and for all. Proponents of nominalism are often described as “linguistic idealists” by the materialist metaphysicians. For the latter believe that Dalton and Mendeleev did indeed cut nature at the joints. From this Kripkean perspective, Wittgensteinians are so infatuated with words that
Nominalism, however, is a protest against any sort of metaphysics. To be sure, it was misleadingly associated with materialism by Hobbes and other early modern philosophers, and is still so associated by Quine. But these thinkers are inconsistent in holding that words denoting the smallest bits of matter cut nature at the joints in a way that other words did not. A consistent nominalist will insist that the predictive and explanatory success of a corpuscularian vocabulary has no bearing on its ontological status, and that the very idea of “ontological status” should be dropped.
This means that a consistent nominalist cannot countenance a hierarchical organization of the kingdom of the thinking mind that corresponds, as Plato's organizational charts did, to an ontological hierarchy. So struggles for priority between metaphysics and physics, or between techies and fuz-zies, look ludicrous from a nominalist perspective. So does Heidegger's distinction between metaphysics and Thinking, as well as his claim that “in the end, philosophy's business is to safeguard the power of the most elementary words.” For a nominalist, Heidegger's favorite words such as phusis (Greek for “nature”) or Wesen (German for “essence”) are no more “elementary” or “primordial” than words such as “aubergine” and “baseball.” The more resonant words have no philosophical privilege over the rawest neologisms any more than the elementary particles over the latest human artifacts.
To defend my suggestion that nominalism can best be summarized in Gadamer's doctrine that only language can be understood, I shall take up the obvious objection to that claim. Techies are quick to expostulate that the paradigm of achieving greater understanding is modern science's increasing grasp of the nature of the physical universe—a universe that is not language. The nominalist riposte to this objection is: we never understand anything except under a description, and there are no privileged descriptions. There is no way of getting behind our descriptive language to the object as it is in itself—not because our faculties are limited but because the distinction between “for us” and “in itself” is a relic of a descriptive vocabulary, that of metaphysics, which has outlived its usefulness. We should interpret the term “understanding an object” as a slightly misleading way of describing our ability to connect old descriptions with new. It is misleading because it suggests, as does the correspondence theory of truth, that words can be checked against nonwords in order to find out which words are adequate to the world.
In a nominalist account, the progress made by modern science consists
The central thesis of idealism is that truth is determined by coherence among beliefs rather than correspondence to the intrinsic nature of the object. This doctrine suggests, though it does not entail, the central thesis of nominalism: that we should replace the notion of “intrinsic nature” with that of “identifying description.” For the notions of real essence and of truth-as-correspondence stand or fall together. Gadamer's slogan gives us a way of sweeping both aside. For it is not an announcement of a metaphysical discovery about the intrinsic nature of being. It is a suggestion about how to redescribe the process we call “increasing our understanding.”
From the Greeks to the present, this process has usually been described with the help of phallogocentric metaphors of depth. The deeper and more penetrating our understanding of something, so the story goes, the further we are from appearance and the closer to reality. The effect of adopting Gadamer's slogan is to replace these metaphors of depth with metaphors of breadth: the more integration between these descriptions, the better is our understanding of the object identified by any of those descriptions.
In the natural sciences, the obvious example of such better understanding is the integration of a macroscopic with a microscopic vocabulary. But the difference between these two sets of descriptions is of no more onto-logical or epistemological significance than the difference between a description of the Mass in the terms of orthodox Catholic theology and a description in the terms of comparative anthropology. In neither case is there greater depth, nor a closer approach to reality. But in both there is increased understanding. We understand matter better after Hobbes's corpuscles are supplemented by Dalton's atoms, and then by Bohr's. We understand the Mass better after Fraser, and better still after Freud. But if we follow out the implications of Gadamer's slogan, we shall resist the temptation
The latter distinction has legitimate, unphilosophical uses in describing perceptual illusions, financial chicanery, government propaganda, misleading advertising, and so on. But intellectual progress is only occasionally and incidentally a matter of detecting illusions or lies. The appearance-reality distinction is no more appropriate for describing the advances made between Priestley and Bohr than the advances made in our understanding of the Iliad. We pride ourselves on our ability to fuse Homer's own descriptions of his poems with those used by Plato, by Virgil, by Pope, by nineteenth-century philologists, and by twentieth-century feminist scholars. But we do not, and should not, say that we have penetrated the veil of appearances that originally separated us from the poem's intrinsic nature. The poem has no such nature any more than matter does.
The fuzzie-techie debate, like the religion-science debate of the nineteenth century, is a quarrel about which area of culture gets us closer to the way things “really” are. But as the twentieth century wore on, proposals for the peaceful coexistence of religion and science proliferated. Debate about the respective merits of the two has come to seem jejune. With luck, the quarrel between the techies and the fuzzies will, in the course of the next century or two, gradually dissipate in the same way. For the attempt to find a philosophically interesting difference between techies and fuzzies was a symptom of the attempt to preserve a certain picture of the relation between language and nonhuman entities. This is the picture that Wittgen-steinian nominalists and Kuhnian philosophers of science are helping us to give up. If they succeed, we shall no longer find it paradoxical to assert that being that can be understood is language. This slogan will be taken as a commonsensical account of what understanding is, rather than as a contrived attempt to improve the image of the humanities.
Gadamer has often been accused of inventing a linguistic variety of idealism. But, as I suggested earlier, we should instead think of him as keeping the gold in idealism and throwing out the metaphysical dross. Idealism only acquired a bad name because it was slow to abandon the appearance-reality distinction. Once this distinction is set aside, idealism and nominalism become two names for the same philosophical position. The ill effects of that distinction can be seen in Berkeley. Having said that “nothing can be like an idea except an idea,” Berkeley went on to infer that only ideas and minds are real. What he should have said was that only a sentence can be relevant to the truth of another sentence, a nominalist claim that is devoid of metaphysical implications.
Berkeley's metaphysics is a typical result of the idea that thoughts or sentences
Our ability to shrug off this question increased when we took what Gus-tav Bergman called “the linguistic turn”—a turn taken more or less simultaneously by Frege and by Peirce. For that turn eventually made it possible for logical positivists like Ayer to de-metaphysicize a coherence theory of truth. They urged us to stop talking about how to cross the abyss that separates subject from object and to talk instead about how assertions are justified. The positivists saw that once we substitute language for “experience” or “ideas” or “consciousness,” we can no longer reconstruct Locke's claim that ideas of primary qualities have some sort of closer relation to reality than ideas of secondary qualities. But it was precisely this claim that the Kripkean revolt against Wittgenstein resurrected. In doing so, the Kripkeans were proclaiming that the linguistic turn had been a bad, idealistic idea.
The current quarrel between the Kripkeans and their fellow analytic philosophers is one way of continuing the old debate about what, if anything, was true in idealism. But a more fruitful way to approach this quarrel may be to take up a suggestion of Heidegger's. Heidegger viewed the series of great metaphysicians from Plato to Nietzsche as control freaks: people who thought that thinking would let us achieve mastery. In a Heideggerian account, the metaphysicians' phallogocentric metaphors of depth and penetration are expressions of the will to take possession of the inner citadel of the universe. The idea of becoming identical with the object of knowledge, like that of representing it as it really is in itself, expresses the desire to acquire the object's power.
The scientism of the nineteenth century mocked both religion and idealist philosophy, because natural science offered a kind of control that its rivals could not. This movement saw religion as a failed attempt to achieve control. It saw Absolute Idealism as an escapist, self-deceptive attempt to deny the need for control. The ability of natural science to predict phenomena, and to provide technology for producing desired phenomena, showed
The strong point of this scientistic line of thought is that although understanding is always of objects under a description, the causal powers of objects to hurt or help us are unaffected by the way they are described. We shall get sick and die, no matter how we describe disease and death. The Christian Scientists are, alas, wrong. The weak point of scientism is the inference from the fact that a certain descriptive vocabulary enables us to predict and utilize the causal powers of objects to the claim that this vocabulary offers a better understanding of those objects than any other. That non se-quitur is still put forward by the Kripkeans. Whether or not one sees it as a non sequitur depends on whether one is willing to redescribe understanding in the way that Gadamer has suggested.
To follow up on Gadamer's redescription, we should have to give up the idea of a natural terminus to the process of understanding either matter, or the Mass, or the Iliad, or anything else—a level at which we have dug down so deep that our spade is turned. For there is no limit to the human imagination—to our ability to redescribe an object, and thereby recontextualize it. A descriptive vocabulary is a way of relating an object to other objects—putting it in a new context. There is no limit to the number of relations language can capture, nor of contexts that descriptive vocabularies can create. Whereas the metaphysician will ask whether the relations expressed in a new vocabulary are really there, the Gadamerian will ask only whether they can be woven together with the relations captured by previous vocabularies in a helpful way.
As soon as one uses a term like “helpful,” however, those who believe in real essences and in truth as correspondence will ask “helpful by what criterion?” To think that such a demand for criteria is always reasonable is to imagine that the language of the future should be a tool in the hands of the language of the present. It is to become a control freak—someone who thinks that we can short-circuit history by finding something that lies behind it. It is to believe that we can now, in the present, construct a filing system that will have an appropriate pigeonhole for anything that might possibly turn up in the future. Those who still hope for such a filing system will typically select some single area of culture—philosophy, science, religion, art—and assign it “the first rank in the kingdom of the thinking mind.” But those who follow Gadamer, like those who follow Habermas, will drop this project of ranking. They will substitute the idea of what Habermas calls a “domination-free” (herrschaftsfrei) conversation that can never come to an end, and in which the barriers between academic disciplines are as permeable as those between historical epochs.
Such people hope for a culture in which struggles for power between bishops and biologists, or poets and philosophers, or fuzzies and techies, are
A culture of this sort will seem to materialist metaphysicians like one in which the fuzzies have won—a culture in which poetry and imagination have finally gained the victory over philosophy and reason. So my little sermon on a Gadamerian text will probably look to them like one more public relations exercise on behalf of the humanities. I shall end by saying why I think that this is not the right way to look at the matter.
In the first place, a Gadamerian culture would have no use for faculties called “reason” or “imagination”—faculties that are conceived as having some special relation to truth or reality. When I speak of “capturing the imagination,” I mean nothing more than “being picked up and used.” In the second place, a Gadamerian culture would recognize that everybody's filing system will need to have pigeonholes into which to fit everybody else's filing system. Every area of culture would be expected to have its own parochial description of every other area of culture, but nobody will ask which of these descriptions gets that area right. The important thing is that it will be herrschaftsfrei; there will be no one, overarching filing system into which everybody is expected to fit.
My sermon on the text “Sein, das verstanden werden kann, ist Sprache” obviously has not been offered as an account of the real essence of Gadamer's thought. Rather, it is offered as a suggestion about how a few more horizons might be fused. I have tried to suggest how Gadamer's own description of the movement of recent philosophical thought can be integrated with some alternative descriptions currently coming into use among analytic philosophers.
I suspect and hope, however, that once another century has passed, the distinction I have just employed—the distinction between analytic and non-analytic philosophy—will strike intellectual historians as unimportant. Philosophers in the year 2100, I suspect, will read Gadamer and Putnam, Kuhn and Heidegger, Davidson and Derrida, Habermas and Vattimo, Theunissen and Brandom, side by side. If they do, it will be because they have at last abandoned the scientistic problem-solving model of philosophical activity with which Kant burdened our discipline. They will have substituted a conversational model, one in which philosophical success is measured by horizons fused rather than problems solved, or even by problems dissolved. In this philosophical Utopia, the historian of philosophy will not choose her
Gadamer once described the process of Horizontverschmelzung as what happens when “the interpreter's own horizon is decisive, not as the standpoint of which he is convinced or which he insists on, but rather as a possible opinion he puts into play and at risk.” He went on to describe this process as “the consummatory moment of conversation [Volhugsform des Gesprachs] in which something is expressed [erne Sache zum Ausdruck kommt] that is neither my property nor that of the author of the text I am interpreting, but is shared” (TM388). To replace the appearance-reality distinction with the distinction between a limited and a more extensive range of descriptions would be to abandon the idea of the text or thing we are discussing (the Sache) as something separated from us by an abyss. It would be replaced by a Gadamerian conception of the Sache as something forever up for grabs, forever to be reimagined and redescribed in the course of an endless conversation. This replacement would mean the end of the quest for power, and for finality, that Heidegger called “the history of metaphysics.”
That tradition was dominated by the thought that there is something nonhuman that human beings should try to live up to, a thought that today finds its most plausible expression in the scientistic conception of culture. In a future Gadamerian culture, human beings would wish only to live up to one another, in the sense in which Galileo lived up to Aristotle, Blake to Milton, Dalton to Lucretius, and Nietzsche to Socrates. The relationship between predecessor and successor would be conceived, as Gianni Vattimo has emphasized, not as the power-laden relation of overcoming (Uberwindung), but as the gentler relation of turning to new purposes (Verwindung). In such a culture, Gadamer would be seen as one of the figures who helped give a new, more literal sense to Holderlin's line, “Ever since we are a conversation …” (Seit wir ein Gesprach sind …).
NOTES
This essay first appeared in the London Review of Books 22, no. 6, Mar. 16, 2000, and is reprinted here with permission. It was given as an address at the University of Heidelberg on the occasion of Gadamer's hundredth birthday.
1. “Philosophy or Theory of Science?” in Reason in the Age of Science (Cambridge, Mass.: MIT Press, 1983), 151. [BACK]
2. Fine, “The Viewpoint of No-one in Particular,” Proceedings and Addresses of the American Philosophical Association, vol. 72 (Nov. 1998), 19. [BACK]