(This paper appeared in Philosophical Perspectives, 1999. It derives from Chapter 5 of my hopefully soon forthcoming book, The Magic Prism: An Essay in the Philosophy of Language.)

           Howard Wettstein

           University of California, Riverside

          

 

 

“Don’t complain; don’t explain.” (Attributed to Henry Ford.)

 

 

I

Introduction: Wittgenstein and the Anti-Fregeans

 

           When I was a graduate student in the late 60’s, Wittgenstein was very fashionable. Remarks like “meaning is use” rolled off one’s tongue as easily as “Hell no, we won’t go,” or “It’s not the case that necessarily the number of planets is greater than seven.” I vowed to avoid the Philosophical Investigations, and I was true to my vow until some years later when a friend commented that my approach to indexicals[1] exhibited what he called a social perspective. Difficult and quirky as Wittgenstein’s text might be, I reluctantly concluded, it might well be a source of insight concerning the social character of language.

There was a second reason for taking the plunge into the Investigations. Wittgenstein, it was well known, defended a variant of the description theory of names, specifically the cluster theory. Moreover, Wittgenstein opposed in a radical sort of way making naming any sort of key to language—this was additionally well known. A study of the Investigations, then, would provide an excellent test of the anti-Fregean, “direct reference,”[2] approach I’d been developing in my dissertation and in several subsequent papers.

Imagine my surprise when I found that the orientation towards language and thought that I had been identifying and criticizing as Fregean was for Wittgenstein almost an obsession. It is the grip of just such ways of thinking that, according to Wittgenstein paralyzes us, that creates puzzles and quandaries. To break its hold, we need to attend closely to actual practice. “Look at our practices,” urges Wittgenstein, “don’t think about what they must be like.” But this, I reflected, is just what contemporary anti-Fregeans—Kripke and Donnellan, for example—had been doing, the outcome being a host of examples that strike at the heart of traditional philosophy of language. I was finding, contrary to what I expected, something of a convergence of views.

           Even more startling, Frege himself was one of Wittgenstein’s central targets. This didn’t emerge right away but by the time I hit Wittgenstein’s discussion of Frege on concepts without boundaries in §79 it was clear. Indeed Wittgenstein often sees Frege as the foremost advocate of the targeted traditional views.

And while Wittgenstein roundly opposed the assimilation of other forms of speech to names—he opposed any sort of equation or even close connection between meaning and naming—he certainly took naming to be of central interest. Names and name-like pieces of language are focal, for example, in the elementary language games. The idea that Wittgenstein advocated a descriptional account of names didn’t fit very smoothly, moreover, with remarks like

 

It will often prove useful in philosophy to say to ourselves: naming something is like attaching a label to a thing. (PI, §15)

 

Not to speak of

 

And the meaning of a name is sometimes explained by pointing to its bearer. (PI, §43)

 

           It’s not as if Wittgenstein advances all sorts of views characteristic of the later anti-Fregeans. Surely not: no rigid designation, possible worlds, or propositions with objects as constituents. Nor should we infer from the two passages just quoted that Wittgenstein is a Millian on proper names.[3] Perhaps most important, Wittgenstein thinks about reference itself in a very different way than do the anti-Fregeans. Still, Wittgenstein not only anticipates important features of the later anti-Fregean approach, he often provides a deeper and more satisfying rationale than in recent work. And where Wittgenstein sharply diverges from the anti-Fregeans, it often seemed to me that Wittgenstein was pointing the way forward.

           That Wittgenstein might be something of an ally was surprising enough. Even more so was anti-Fregean assistance in understanding Wittgenstein. Wittgenstein’s work not only represents a radical departure from traditional philosophy, it is also quite obscure, difficult to penetrate. Wittgenstein maintains—to mention some ideas that will be pivotal here—that meaning is use, that he has no interest in explaining anything, that philosophical puzzles do not require solutions. How to understand any of this, not to speak of all of it! The antecedent likelihood of finding the anti-Frege literature helpful was very slight. Nevertheless, that literature provided considerable assistance, as I’ll explain.

           The convergence of views to which I’m drawing attention seems to me almost universally unappreciated. It is unappreciated by Wittgenstein sympathizers, whose vision is obscured by the rigid designation/possible worlds aspect of the anti-Fregean literature. It is unappreciated by the anti-Fregeans, who tend to see Wittgenstein sometimes as an arch-anti-theorist who is happy to leave matters muddy, sometimes as a sort of description theorist of proper names, an obscure one at that.

Not recognizing the Wittgenstein link has been costly to my fellow anti-Fregeans. As I argue in “Turning the Tables on Frege,” anti-Fregeans have been all too conservative in their criticisms of traditional semantics, often departing only from the letter of the Fregean law while maintaining its spirit. Their conservatism has been facilitated by the sometime tendency of analytic philosophers to make philosophical issues into technical problems, and by an attendant insensitivity to the larger philosophical stakes at issue. Of these Wittgenstein, whatever his foibles, is not guilty. Philosophical Investigations highlights many substantial issues that lie just under the surface of the reference debate.

In this essay, by locating and developing important points of Wittgenstein/anti-Fregean congruence, I hope to redirect our focus to these large questions, and in doing so to develop further the social practice conception of my earlier work. The shift from the traditional orientation that I will advocate is considerably more radical than suggested by the direct reference literature.

 

 

II

Convergence: Meaning ain’t in the Head

 

The anti-Fregean critique has underscored examples in which speakers refer in the absence of beliefs that uniquely identity referents. Sometimes their beliefs about their referents are very meager, sometimes mistaken, and so on. Frege’s problems are not, however, limited to such examples. Consider an assertion involving the name of someone you know very well, or one involving an indexical reference to something with which you are quite familiar, something that you can identify in any number of ways. In many such cases, there will be no way to select some discrete bit of identifying information from the available information—some particular description or set of descriptions—and conclude that this bit of informa­tion functions as the sense of the name. And Frege’s view—that on such occasions one asserts a distinct, definite proposition—requires that it not be arbitrary which bit of identifying information is to play this role. So even where there is no problem with the speaker’s identifying information—even where he possesses a what I have called a cognitive fix on the referent—it is implausible that this knowledge plays the role assigned to it by the Fregean orientation.[4]

This line of reasoning played a pivotal role in my earlier work. But its consonance with the tenor of Wittgenstein’s work, while it now seems positively striking, was only recently pointed out to me. Think of my “too many descriptions” point in connection with the Donnellan-Kripke idea that the use of a name requires very little identifying information. The result of putting these together is this: Reference does not require a cognitive fix, and even where there is one the cognitive fix does not do the work it was supposed to do, the work assigned to it by the Fregean orientation. Turning to the Philosophical Investigations, with its more general concerns, I read one central theme thus:

 

 There is likely to be considerably less in the head[5] than traditional philosophy supposes. And what is in the head (for example, mental images, occurrent intentions—in the sense of conscious acts of decision) is likely to be playing less of a role—or at least a different role—in speech and thought than philosophers have assumed.

 

It seems to me remarkable that such fundamental points of contact have received so little attention.

           Hilary Putnam, in his early anti-Fregean discussion of natural kind terms, remarked that, “meaning ain’t in the head.”[6] Putnam’s remark, although he may or may not have been thinking of Wittgenstein, nicely focuses this important point of contact with Wittgenstein. But it isn’t only this shared negative proclivity, but also a positive tendency. (I had been led to believe not only that Wittgenstein was something of a description theorist, but also that he had no positive views.) Wittgenstein, more than anyone else in recent times, brings to life what I want to see as the central idea of the direct reference revolution, that of a public language, a set of shared social practices.

          

III

The Life of the Sign: Wittgenstein’s (Difficult) Naturalism

 

           That linguistic significance is not driven by associated representations is pivotal for both Wittgenstein and the direct reference advocate. Yet the contrary idea has both the weight of tradition and powerful intuitive support.

In The Blue Book (p.4), Wittgenstein criticizes the traditional view:

     Frege ridiculed the formalist conception of mathematics by saying that the formalists confused the unimportant thing, the sign, with the important, the meaning. Surely, one wishes to say, mathematics does not treat of dashes on a bit of paper. Frege’s idea could be expressed thus: the propositions of mathematics, if they were just complexes of dashes, would be dead and utterly uninteresting, whereas they obviously have a kind of life. And the same, of course, could be said of any proposition: Without a sense, or without the thought, a proposition would be an utterly dead and trivial thing. And further it seems clear that no adding of inorganic signs can make the proposition live. And the conclusion which one draws from this is that what must be added to the dead signs in order to make a live proposition is something immaterial, with properties different from all mere signs.         

But if we had to name anything which is the life of the sign, we should have to say that it was its use.

                     

We need help, of course, with Wittgenstein’s positive alternative to representationalism, his idea that “the life of the sign” is its use. He provides a bit more assistance (but only a bit more) in a subsequent passage.

 

The sign (the sentence) gets its significance from the system of signs, from the language to which it belongs. Roughly: understanding a sentence means understanding a language.

 

To understand the distinctive way in which linguistic symbols are alive for us, maintains the Fregean, look past the symbols to their meanings, to the intrinsically alive representations. Wittgenstein redirects our focus to the symbols themselves, not in isolation, but “as part of the system of language.”

           But this hardly explains the matter. For, first, to see a particular sentence as situated in a system of language seems to be to see it in relation to other sentences of the language. But, as Wittgenstein says, “no adding of inorganic signs can make the proposition [come to life].” Second, what we set out to understand is Wittgenstein’s notion of use. How does it help to tell us that what vitalizes the sign is its inclusion in a system of sentences? This doesn’t sound much like an explanation of use, but rather a new and different—even if related—idea.

So what’s going on? The answer is that linguistic vitality is a matter of embeddedness of symbols in social, communicative practice. When Wittgenstein speaks of the inclusion of a sentence in a system of language, he means the inclusion of a sign, among other systematically related signs, in a living language, in a system of communicative practice.

           This is, of course, not more than a beginning, for we still need an account of how embeddedness in practice breathes life into a symbol. But one thing that does clearly emerge from these passages, as well as from the famous quotations about meaning and use—“Meaning is use,” “Don’t ask for the meaning….”—is a general explanatory direction. Significance is not a matter of associated ideas in the mind or concepts in a third realm, but rather of what we do with symbols. While it is difficult to know how to work out such a view, its naturalism has always seemed to me very attractive. Just as in a naturalistic spirit it would be preferable to characterize human beings without notions like that of a purely spiritual substance, so it would be preferable to account for language without employing any notion like that of Frege’s sense. A sense, you might say, is the soul of a word.

The appeal to Fregean senses seems to violate naturalism. Yet one may develop a Fregean approach along naturalistic lines, for example, by somehow putting the significance-giving representations into the brain instead of construing them as abstract or mental. There is an important parallel in the philosophy of mind, where one may develop—as many have—a fundamentally Cartesian approach to mentality in a physicalistic direction.[7] Such naturalisms seem to me unnatural, the products of arranged marriages, as it were. What I see as Wittgenstein’s naturalism, by contrast, proceeds from a wholly different direction. Don’t look to representations (whether mental, abstract, or physical) to understand significance; attend instead to our practices.

Still, the idea of significance-without-representations is very difficult to get under control. How, one wants to ask, could linguistic significance not be a matter of something like association with intrinsically representative entities? Doesn’t our ability to use symbols, our saying things with them, presuppose that they are significant? Doesn’t the significance have to come first, before the use? How then can significance be explained in terms of use, or of what we do with the symbols? How can mere human action accomplish this magic? Nor does going social, bringing more people into the picture, resolve anything. As if a bunch of people could accomplish the magic simply by behaving and responding in coordinated ways.

            A colleague once commented to me that he found Wittgenstein’s approach to these matters tantalizing, even if ultimately unacceptable. They were unacceptable for just the sorts of reasons just mentioned. Unlike my colleague, I had the sense that Wittgenstein was on to something, that something more sound, indeed less magical, was in the offing. But how, given all these questions, can we get further inside Wittgenstein’s picture? How can we bring it to life as a real alternative?

IV

Gaining a Foothold  

 

 

When we think about significance abstractly, about what makes it possible, it’s very difficult to conceive an alternative to a broadly Fregean outlook. Nor does Wittgenstein clearly articulate such an alternative, at least not in a way that makes it readily available. It is indeed very difficult to bring clearly before one’s mind what Wittgenstein is after with his talk of use, role in the language, and so on.

What is perhaps most helpful in Wittgenstein is his working through many examples, and many kinds of examples; his exploring the dead ends engendered by traditional modes of philosophical thinking. But Wittgenstein’s approach to the course of the intellectual therapy is notoriously difficult, and not unlike the other sort of therapy, is almost always painfully slow and, until one gets the hang of it, quite frustrating.[8] I want to recommend another mode of access to Wittgenstein’s themes, an alternative or supplementary route not only to the rejection of the Fregean orientation, but even—dare I say it—to meaning as use. Enter direct reference.

A first step is to avoid the lofty plane at which discussions of meaning—Fregean and even anti-Fregean—often proceed. My approach to direct reference avoids discussion of significance in general or in the abstract. This might seem strange in light of the tendency among some direct reference advocates to make names the paradigm, and further—this is a distinct move—to think of meaning as reference. Or alternatively, to make indexicals the paradigm, and to think of meaning as character (in Kaplan’s sense). These quite general theses are far from what I am after. Instead, let’s narrow our focus to particular categories of linguistic expressions—for example to proper names, or indexical expressions, definite descriptions, predicate expressions—and scrutinize actual practice. The question is: What does significance come to with respect to this practice?[9]

When one asks this question for the case of, for example, proper names—understood not as bearers of Fregean sense but as Millian tags—Wittgenstein’s ways of talking about significance seem natural, while representationalism seems all wrong.

To begin on the negative side, consider the traditional idea that a expression's being significant is a matter of its possessing a meaning, where close to the heart of “meaning” is “grasped by the competent speaker.” Proper names just don’t fit this model. One does not have to grasp any such meaning-entity to be competent with a name. Simply acquiring a name in some appropriate way (like conversing with someone who is using the name) puts one in a position to use the name. This is not just a contention of the anti-Fregeans; it is a datum. People are regularly judged competent with names even when they lack familiarity with the bearers of the names.

Notice also that we do not ask what names mean. When we lack familiarity with an expression like ‘lugubrious’ we inquire about what it means; this is not the case for ‘Aristotle’ or ‘Gell-Mann’. Nor do we—except perhaps in the grip of a theory—think of the name “Harry”—as ambiguous, just because there are many Harrys. “Meaning” as a substantive, or “the meaning,” seems like the wrong notion for proper names.

This is not to say that if one comes to Millianism with a traditional conception of meaning in hand, thinking that meaningfulness requires meanings, that one cannot lay one’s hand on a candidate. The function of names, after all, is to stand for things. Their significance consists, we might say, in the fact that they stand for things. It may seem like a short distance between this last formulation and the idea that the significance of the name just is the thing stood for. Russell thus maintained that the meaning of a (real) name just is the bearer. But there is no need to say this, and one will not naturally do so unless one feels the need to find something to be the meaning of the name.[10]

So the traditional way of thinking about significance has no purchase here. But Wittgenstein’s fits naturally: When we say that names are significant, meaningful, we gesture to their role in the public practice, their use or function: making things subjects of discourse, as Mill says. What brings names to life, to use Wittgenstein’s metaphor, is their function, their role in our practice. In terms of Wittgenstein’s (probably unhelpful) slogan one might say that for proper names their meaning (that is, their significance) just is their use (that is, their function of standing for things).

           In §1 of the Investigations, Wittgenstein is attacking one variant of the traditional way of thinking about meaning:

—But what is the meaning of the word “five”? —No such thing was in question here, only how the word “five” is used.”

Substitute the name “Aristotle” for “five,” and you will see the confluence of views of which I’ve been speaking.

We have been struggling with Wittgenstein’s approach to significance, trying to find a way into the picture, so to speak. The anti-Fregean treatment of names, in modeling key Wittgensteinian themes, provides a good beginning. In addition, anti-Fregean successes—substantial in my view—constitute a powerful argument for Wittgenstein’s approach. Indeed, we have isolated one clear domain in which Wittgenstein’s intuitions about significance seem right on the mark.

 

 

V

Explaining Significance?

 

           One of aims of Wittgensteinian therapy is to loosen the hold of our almost genetically inherited Fregean impulses about significance. I have tried to accomplish some of the same by modeling Wittgensteinian themes in direct reference terms. Even so, it is very difficult to lose a sense of uneasiness about significance-without-representation.

One way to highlight the discomfort is to focus on explanatory adequacy. Descartes says that he found it amazing that bodies, mere pieces of nature, could move themselves. If locomotion can seem miraculous, what about reference? That mere pieces of nature can mean, or symbolize, or stand for something, really seems extraordinary. It cries out for a philosophical account. Frege’s approach seems inadequate, empirically inadequate, insensitive to actual linguistic practice. But at least Frege tries to explain, to get behind or underneath, reference. My favored alternative seems worse.

           Mill’s remarks about proper names hardly constitute an explanation of how it is that names signify. What, after all, does the Millian tell us about how it is that ‘Aristotle’ refers? Mill’s remarks are largely negative: Names do not refer by means of associated connotations, senses or anything of the like. His positive contention is that names are assigned ad hoc, as it were, and then used to make their bearers subjects of discourse. If there is anything like an explanation here, it is a matter of explaining the reference of a particular name in terms of the general name-using practice. But this is pretty obviously circular. We wanted to know how it is possible for one piece of nature, the word ‘Aristotle’, to stand for another, the man Aristotle. And we are told that in general we use such symbols to stand for things.

Wittgenstein sometimes talks as if use is somehow to play the explanatory role of Frege’s sense.

 

But if we had to name anything which is the life of the sign, we should have to say that it was its use. (P. 4, “The Blue Book”)

 

           But again, what we want to understand is how it is possible that a name stands for a thing. So it does not help to be told that what makes this possible is that the name is used to refer to the thing. Indeed, as noted above, the use of a linguistic expression seems to presuppose that the expression is significant. How then can use explain significance?

It would seem that both accounts, Millian and Wittgensteinian, need to make good on the idea that somehow socially coordinated activity can confer significance on pieces of language. But, as noted above, moving to the social level doesn’t seem to resolve anything. What the intelligibility of linguistic practice seems to require is something like the association of concepts with the words.

           So it looks like with respect to the most important question of all, Frege’s view—problematic as it may be—has a leg up. Frege has a story to tell about how the Red Sea parted, as it were; we remain mute.

But has the Fregean really provided an explanation of significance? Never mind empirical inadequacy; does it explain? The Fregean proposal has only the form of an explanation, as Tom Blackburn once commented to me. It would represent a genuine explanatory advance if we understood its essential ingredient, the intrinsic aboutness (intentionality) of the representations. But how exactly are representations significant; how do they manage to stand for things? Why isn’t this as problematic as the aboutness of words, even more so, given how unclear we are about senses? The Fregean explanation, unless further developed, seems like positing God to explain how it all got here, but having nothing to say about how God did it, and little helpful to say either about how God got here or about why that’s no problem.

           In the absence of an account of the matter, the Fregean is in an embarrassing position. With the aboutness of the representations unexplained, their theoretical utility is cast into doubt. But the representations are a theoretical posit—it’s not, after all, that attention to ordinary practice makes plain the presence of suitable representations accompanying our words. Why bother with them?[11] Why not leave things where we found them? This suggests—and this might seem at least mildly depressing (but not to worry, it grows on one)—that perhaps the best we can do is to describe our ways with language, making no attempt to get beyond or behind.

Direct reference advocates—me among them—have emphasized Frege’s empirical failures. But quite apart from the direct reference critique, it is far from clear that positing intrinsically alive representations to ground the life of the sign constitutes an advance in our understanding. Where does this leave the Millian? My idea is to turn what looks like a vice into a virtue, not by substantive transformation, but by declaration. The vice is the Millian’s begging the questions about intentionality, his helping himself to aboutness. Maybe that’s a good idea. Maybe the attempt to provide an explanation is the mistake.

           Let’s begin with Wittgenstein’s remark:

 

We must do away with all explanation, and description alone must take its place. (PI, §109)

 

Even allowing for its sound bite quality and attendant overstatement, this is another one of those remarks that inspires Wittgenstein’s detractors. I certainly found it frustrating. And then I made the connection with Mill. I now think I see something of the first importance in this passage, as well as in Wittgenstein’s frequent admonishments that philosophy needs to stay at the surface and avoid hypothesizing intermediate entities and processes. Think of §109 in connection with §18 of Philosophical Grammar:

 

In philosophy one is constantly tempted to invent a mythology of symbolism or of psychology, instead of simply saying what we all know.

 

Perhaps at work in §109 is not hatred of theory, but rather a critique of one central way in philosophy, the attempt to get behind or underneath what would otherwise seem miraculous. Wittgenstein might well agree that philosophy is all about providing intelligibility—you might have to catch him on a good day. However, the urge to provide intelligibility drives philosophers to desperate measures, measures that often involve no explanatory progress.

Positing a god to explain the origin of things provides a model of this sort of pseudo-explanation. Of course, we don’t need Wittgenstein’s radical-sounding rejection of explanation to see the explanatory deficiencies in such a posit. But what I see as the suggestion of Wittgenstein’s explanation-description remark may help us take a substantial step forward. Perhaps it is not only the theistic answer to the question of ultimate origins that is the problem. The question itself is suspect; perhaps the very attempt to explain is out of place. It is one thing to explain the origin of particular parts of nature, even the entire physical universe in its present form, which may well derive from earlier forms. It is quite another thing to explain the origin of everything, anything. “Where does it all come from?” is a question that perhaps we ought to resist. It is not clear that there is here any explanatory space, that anything could constitute explanatory progress.[12]

           However, it is one thing to consider that explanation may not be appropriate to such a global matter, quite another with regard to something like the aboutness of language. How might we motivate the “no explanatory space” idea for the latter case?

           Consider the following picture. Creatures of a certain neurological complexity, appropriately socialized, can use pieces of nature—our interest here is in names—as symbols for other pieces of nature. People, that is, use symbols to stand for things. Think of this as primitive for philosophy. It is the datum with which we begin, not something for which philosophy owes or might provide an explanation in simpler or more primitive terms. This is not to deny that philosophy and related disciplines can augment our understanding of this ability in all sorts of ways. We can, for example, provide detailed characterizations of the ways this ability is implemented, our practice with proper names being a case in point. It may also be helpful to study the evolution of such practices, the question of how they derive from more primitive forms.[13]

Nor is increasing the intelligibility of this fundamental ability strictly a task for philosophy, linguistics, and the like. The natural sciences may provide a characterization of organisms that possess such abilities, and someday (now, for all I know) specify the facilitating neurological structures and functions. That is to say, we will know that such structures/functions support this ability and that others do not, that still other related structures/functions support related abilities and the like.

It might appear that what I’m saying comes to the idea that philosophy is stymied or limited here, and that the real explanatory work is to be done by science. But the question that my imagined scientist is answering is not the same one the Fregean sets out to answer. The scientist wants to know what structures support this ability and what needs to go on neurologically. How could anyone, why would anyone, even Wittgenstein, argue with this? The Fregean seeks quite a different sort of understanding. She wants to understand what Colin McGinn once called “the mechanism of reference.”[14]

Once we have answers to all the sorts of genuine questions mentioned—we hear, for example, from the scientist, and we have an adequate account of the ways the ability is implemented in linguistic practice, and so on—there is no further question of explaining how these signs come to life for us. There is nothing further to explain, no explanatory space.

When the Millian is asked why ‘Aristotle’ refers to this particular person, he cites our name-using practice, and remarks further about how ‘Aristotle’ came to be attached to the person in question. Were the Millian trying to explain the intentionality of the name ‘Aristotle’—the miracle of its aboutness, how this is possible—then he would indeed be begging the crucial question. In fact, the Millian seeks to explain no such thing. The Millian merely seeks to situate this particular name in the general practice. The Millian thus stays at the surface. He says what he knows and resists the temptation to invent a mythology of symbolism or psychology.

Let’s come at this from another direction, one suggested by conversations with Joseph Almog. In “Why does ‘Aristotle’ refer to that individual?,” traditional philosophy of language hears a question of great philosophical interest:  What connects the name with the referent? What, to use McGinn’s expression, is the mechanism of reference? Frege’s sense-reference picture tries to supply an answer to this question, the needed mechanism.

There are other questions one might hear in “Why does ‘Aristotle’ refer to that individual?,” questions about the general linguistic practice invoked by a use of the name, questions about the how the particular name came to be associated with this particular individual. To these other questions, the Millian is happy to respond. But the Millian hears no request for a mechanism that connects name to referent.

“All I can tell you is that we have this general practice, and I can of course tell you something about how ‘Aristotle’ entered this practice as a name for this particular person. In telling you these things, I realize that I’m not supplying what the Fregean is seeking. But there is nothing more to tell.”

To say that there is nothing more to tell is to join forces with Wittgenstein. For it is to say that there is no further explanatory space, no genuine additional question to be answered. Intelligibility in this context, the kind that philosophy can provide, is a matter of describing our name using practice, and explaining, really describing again, how this particular name comes to fit in.

Philosophers nowadays distinguish externalist approaches from internalist ones, in any number of domains. An important question in epistemology is that of the vexed third condition on knowledge. (Knowledge is true belief + what?) Internalists have tried to provide an account in terms of features that are internal to the knower, for example, in terms of the justificatory structure of her beliefs. Externalists look to external features, for example to causal relations between the agent and the environment.

It might be assumed that the anti-Fregean advances an externalist account of intentionality in opposition to Fregean internalism. It is sometimes suggested—incorrectly in my view—that in Naming and Necessity Kripke advanced a causal theory of reference. On such a view, Kripke is indeed seeking to provide a mechanism of reference, specifically a causal chain that connects the name to the referent. This would be an externalist account of the intentionality at least of proper names.[15]

Whether or not this is Kripke’s considered view,[16] some passages in Naming and Necessity suggest it. It is in any case not the view defended here. The envisaged externalism accepts the explanatory project and provides an externalist answer. Wittgenstein and my Millian reject the explanatory project.

 

 

VI

Concluding Remarks: The Miracle of Reference

 

Carnap emphasized that semantics, the study of the relations between symbols and the items in the world that the symbols are about, proceeds by abstraction from the use of language, by abstraction from linguistic practice. Strawson, going a step further, maintained that the use of language needs to be the primary focus of the philosophy of language; that a more abstract or abstracted view of the relations between symbols and the world may mislead at crucial points. And surely, something like this is Wittgenstein’s view. Of course, there may be both semantic, and perhaps especially syntactic, questions for which such abstraction may be just the ticket—I speak here for myself, not Wittgenstein. The point is not to discourage such approaches, nor to discourage formal treatments. The point is to keep our collective eyes on the ball. There is some danger, especially in thinking about the connections between language and the world, that a lack of attention to the full-blown practices may skew the kinds of questions that we ask and ultimately the way we see language vis-à-vis the world.

Case in point: the miracle of name reference. Abstract the name and its semantic properties from the practice. Then stare at a name, and then at its referent, and keeping looking back and forth. The connection between these two pieces of nature, that one is about the other, can seem dazzling. What is this magical aboutness? It’s as if—or maybe it is—that words possess non-natural, sui generis, properties, or stand in non-natural relations. Or maybe words are conventionally connected with other things mental or abstract that really have something like non-natural relations to things. It might seem like the only alternative to such spookiness—short of eliminitivism—is reduction; perhaps the suspect non-natural property or relation can be reduced to something physicalistically acceptable, a causal chain or something of the like.[17]

           A great deal changes when we step back from the abstraction, and attend to the use of names in practice. We may then see this practice as implementing an ability, a potentiality of creatures so neurologically equipped and socially involved, an ability that constitutes not a problem but a datum for philosophy. Armed with this perspective, my Millian is happy to characterize the function of names in our practice. He is less happy to explain the miracle, how it is possible that mere pieces of nature signify.

           And when Wittgenstein identifies the life of the sign as its use, it is not that uses, or linguistic practices, now play the explanatory role that senses play for Frege. Words are the paradigm intentional entities, not shadows of the genuine ones. At the same time, our ability to use symbols is something that evolves over time, something that stands in need of implementation, that awaits social practice for its implementation. So Wittgenstein can say that the life of the sign is its use, drawing attention to the idea that it is only in the context of social practice do pieces of nature come to semantic life; only in such a ways do such abilities get implemented.[18]