It is easy to understand why anyone concerned with linguistics, logic, literature — so with language in a broad sense — shudders when it comes to the problem of meaning (here “meaning” taken to mean ‘anything that may have significance for a human being’): sooner or later meaning will involve mammoth ontological and epistemological problems, i.e., questions of “existence”, of “what there is”, and what, and how, we can know about, and of, the world.☞An early draft of this paper was given in the so-called BuPhoc series of the Department of English Linguistics of ELTE in the April of 2007, at the invitation of Professor László Varga. Yet it is not only this invitation for which I am grateful to him. He is among those who — with his “Introduction to Linguistics” lectures (back in 1979) and his several subsequent seminars in English linguistics, and especially with his example as a scholar, teacher and a great personality — has made a huge impact on my thinking. For a while, I even wished to become a linguist. Ultimately, I did not, yet this paper perhaps indicates that I have never given up my interest in questions of meaning, one of the central problems of linguistics, too. When, in the late 1880s, Gottlob Frege laid the foundations of a new logical semantics, he was soon hailed as consciously and consistently turning (for some: reducing) traditional epistemological and ontological problems into genuinely semantic ones.☜For more on this cf. Dummett (2001: 6–20). All of a sudden age-old riddles of philosophy looked as if at least some of them could be solved through some semantic, especially logical analyses of language: several previous philosophical questions looked as pseudo-questions, i.e. as sheer nonsense that had led the mind astray, while it was through the operations of semantics, more specifically through logical syntax and semantics, and studying the grammar and the meanings of natural human languages that thinkers hoped to gain insight into the workings of human thinking and into what there is (at least in human conceptual-linguistic schemes). Syntactic and semantic theories were supposed to decide philosophical questions, while, of course, there were major steps forward in logic and mathematics independently of the analysis of natural languages.
Thus, on the one hand semantic theories had to “confess” (thematise) their ontological and epistemological commitments, while epistemology and ontology looked hopefully, even yearningly at semantics on the other. This is why no serious semantic theory can say: “I do not have an ontology and epistemology” — it will inevitably have one, even if its ontology is a version of naive realism, or if its epistemology is confused. This is one of the reasons for calling the 20th century “the century of language”, or of “the linguistic turn”. What is significant — even for my present purposes — is that it was not only Analytic or Anglo-Saxon philosophy that expected “salvation” from language, and especially the study of meaning: the Continental (German-French) tradition of thinking soon caught up. For example, Martin Heidegger around the middle of his Being and Time felt the need to give an account of language (cf. Heidegger 1951: 213–230), and later on turned to the interpretation of poetic texts (such as Hölderlin’s) to describe what he meant by “truth” (cf. Heidegger 1981). One of the “fathers” of hermeneutics, Hans-Georg Gadamer famously said that the structure of being is brought into language by the articulation of Logos (cf. Gadamer 2010: 460), and more examples could easily be given. In his admirable book, Aesthetics and Subjectivity from Kant to Nietzsche, Andrew Bowie even argues that the “linguistic turn” happened not in the analytic tradition in the last decades of the 19th and the first decades of the 20th century with Frege, Russell, Moore, Wittgenstein and others but some hundred years before, in German Romanticism after Immanuel Kant, in the works of such thinkers as Hamann, Humboldt, Herder, Novalis and others, whose influence proved especially inspirational in the Continental line of doing philosophy (cf. Bowie 2003: 5–14). It is all the more discouraging that representatives of the Analytic and the Continental school still talk very little to one another, some of the lines of division being their relation to what extent a theory of meaning could or should be formalised; whether there is a “cognitive content” that remains unaffected in the process of translation, reformulation or paraphrase; and whether the language of poetry, fiction and drama should be taken into consideration in a theory of meaning at all.
However interesting the above questions may be, they are for a more extensive and further study. Here, taking my clue from the interrelatedness of semantics, ontology and epistemology, I will examine, in a non-formalised manner, one of the central tenets of any system concerned with meaning: the phenomenon of identity, or, more precisely, the phenomenon that human beings are capable of comprehending what it means that a = a, or a is a, i.e., that something or somebody is identical with itself, herself, himself. In a way, it may be wrong to call this a “problem”, since identity looks perhaps the only unproblematic part of all semantic theories, so much so that some theoreticians have precisely tried to build their analyses of meaning on the identity relation (see below). In philosophy (logic) the identity relation is often called an “analytic”, i.e. (logically) necessary truth, which is supposed to hold true under all circumstances. The interpretation of an analytic sentence — e.g. that “Shakespeare is Shakespeare”, or that “Bachelors are unmarried men” — remains curiously “within” the boundaries of language: it is enough to know the syntactic (logical) structure of the sentence (proposition) and the meaning of its constituents to see that it is necessarily true: one does not need any information coming from the “external world” (one does not have to compare the sentence with “reality”) to see that the sentence is true. Yet, and as a result, we pay a heavy price for the certainty of analytic truths: they do not convey any useful piece of information about the world; analytic truths are tautologies, tautology defined by Wittgenstein in the Tractatus Logico-Philosophicus as being “unconditionally true” (4.461). Tautologies are “not pictures of reality” (4.462), they “admit all possible situations” (4.462), a “tautology leaves open to reality the whole — the infinite whole — of logical space” (4.463).☞ Throughout, I quote the Tractatus according to Wittgenstein (1961). My references, as it is the custom, are to paragraph, and not to page numbers.
The riddle that there are identity statements which do report valuable pieces of information about the world was noticed by Gottlob Frege in his famous article published in 1892, Über Sinn und Bedeutung (On Sense and Reference/Nominatum) (Frege 1992: 150–180). Frege’s by now well-known riddle was: what do we wish to express with an identity relation the general form of which can be a = b, when the two parties on the left and the right hand side of the equation are even visibly different? To say, as Frege’s example goes, that “the morning star” is “the evening star”, that the morning star is identical with the evening star, or, as Kripke later put it, that Hesperus and Phosphorus are identical, sounds at least strange, since we surely neither wish to communicate the identity of the signs themselves (we can see or hear they are not identical), nor do we wish to produce a tautology. Frege’s celebrated solution was that when we say “the morning star is the evening star”, we say nothing else than that the two names (descriptions), “morning star” and “evening star” refer to the same external object in the external world, namely to the planet Venus.☞ On the problems of “washing names (“Venus”) and descriptions (“the morning star” / “the star we see in the evening”) together”, and especially on the question whether Frege ever claimed that sense would “determine” Bedeutung, cf. Heck and May (2008: 3–39) but I cannot pursue this argument any further here. Of the planet Venus several names or descriptions are possible (one of these is precisely “the planet Venus”, or: “Morning Star”, or “Phosphorus”, or “Evening Star”, or “Hesperus”, but even, under the right circumstances, “the star I saw yesterday in the sky”, etc.) and with identity statements like the above we wish to establish that they have the same referent: they refer to one and the same thing (nominatum, denotatum, designatum; they have the same “Bedeutung”, they have the same extension). Frege called the referent or thing, or object, i.e., to which we refer “Bedeutung”, while the “content” of the descriptions (names) like “morning star” or “evening star”, i.e., the content of that with which I refer, he called “Sinn”.☞ It is often pointed out that although Sinn is very close to English sense, Bedeutung is a rather unfortunate term since it means precisely ‘meaning’ in German, while Bedeutung in Frege’s use of the term in fact means the object referred to itself. But the root of the word Bedeutung is Deut, deuten auf means to ‘point at’, deuten ‘to explain’, Deutlichkeit ‘clearness’, and thus it seems that Frege uses Bedeutung in a more literal sense, which might be translated as: ‘that which has been clearly, unambiguously pointed at, and thereby explained from the point of view of what we are talking about’. So the sense (intention) of a description is the “road”, the “path” “on” which I get to the object (and there are several ways to get to something, or somebody, as there are several ways, e.g. to get to the Department of English Linguistics at ELTE).
Bertrand Russell, in the early years of the 20th century, worked out a similar theory (cf. Russell 2001), although without explicitly dividing meaning into sense and reference. Russell agreed that, besides proper names, we refer to objects and persons with descriptions such as: “the other side of the moon”, or “the present King of France”. Based on the theories of Frege and Russell, it became customary to give reference by way of descriptions. However, some problems soon became obvious. A description (the sense, the meaning of the referring expression) was claimed to pick out the object or person from among all the others unambiguously, since descriptions were supposed to function as names. Therefore, from Frege on it was claimed that a description determines the referent (the Bedeutung). Yet how is determination possible? There is nothing necessary in giving an object or person through this or that name or description. Nothing has a “natural” or “right” designation; if it had, we would not bother about naming: it would be automatic and we would all speak the same language (this is a dream-world Socrates fancies, as one possible alternative, and more in terms of a parody, in Plato’s dialogue, Cratylus). A description is nothing more or less than a piece of knowledge or belief about the object; for example, I may know (believe) that there is a star which appears in the sky both in the morning and in the evening, and I may use both descriptions to refer to it. Yet, first of all, successful reference may occur through an imprecise, or even false piece of knowledge. The morning and evening “star” is an excellent example because Venus is not a star but a planet, yet who cares, if we all know what we are talking about. Several philosophers and semanticists, including Strawson (1959, 2001), Searle (1958) or Donnellan (2001) claimed that like everything else in language, naming is also based on convention. They added a few very useful refinements to Frege’s and Russell’s theory: e.g. Strawson pointed out that we borrow descriptions from one another. There surely have been an “initial act of baptism” but from that moment on people simply follow the practice of the name-giver in applying the same name to the referent in an imitative and repetitive manner. Referring — like many other activities in and with language — occurs along the lines of a social chain. Even further, we might know very little about a person, and yet we are still able to successfully refer to him: a single expression we have just heard about him, e.g. “the boy standing with the empty champagne glass”, also exhausting all our knowledge about the person with the same stroke, is enough for successful reference under the right circumstances. Further, Strawson and others held that although it is true that various people may have different pieces of information about a referent — for example, a Shakespeare scholar has, say, a hundred and twenty ways to refer to Shakespeare, while ordinary people three or four — we usually fix referents in fact through “clusters” of descriptions, and that will ensure determining the referent more or less unambiguously. We will always look for, in terms of knowledge, the common denominator when referring to people and if, talking about Shakespeare, for example the expression “member of the Lord Chamberlain’s Men” will not work, we will resort to “the Swan of Avon”, or the “the author of Hamlet” (though not “the author of Pericles”, since it is less widely known that we attribute a play to Shakespeare under this title, too). What we know about Shakespeare might be given in the form of clusters of descriptions which we measure against the descriptions of others. Strawson, largely following the pragmatic approach to language introduced by Wittgenstein in Philosophical Investigations (2001), envisaged referring as a language-rule following, convention-based activity. Keith Donnellan and others, to ensure the determination of reference through description, insisted on supposing a causal relationship between description and referent, yet this causal nexus is also based on convention.
Then came Saul Kripke, in 1970, with a series of lectures held at Princeton University, entitled Naming and Necessity (1980), which put the problem of reference into an entirely new light. Since it remains true that every semantic theory will imply epistemological, as well as ontological questions, Kripke offered some real challenges to philosophy, in a — to my mind — highly original way.
Kripke’s main objection to descriptive theories was on the level of both epistemology and ontology. On the level of epistemology — ontology will be discussed later — he claimed that none of the items of knowledge, given in the descriptions, are necessary facts of our world. The problem is not that items of knowledge, true or false, or even on an ad hoc basis, could not do the job of referring — Kripke is very much aware that this is done all the time. The real problem is that descriptive theorists treat proper names on the same level as descriptions, in other words they regard the proper name “Shakespeare” to be exactly synonymous with “member of the Lord Chamberlain’s Men”, or “the Swan of Avon”, or “the author of Hamlet”. But suppose that Shakespeare never became an actor and playwright, suppose he was not born in Stratford, suppose he was too lazy to write Hamlet, and still we would be able to successfully refer to Shakespeare with the name “Shakespeare”. Of course, William Shakespeare could have been named otherwise by his parents, e.g. “Christopher”, or “Ben”, or even “Voldemort”, although this last one is not very likely. The fact that Shakespeare happened to be named William is, in itself, not a necessary fact of the world. But once his name was decided on, the name, as Kripke puts it, rigidly designates (cf. Kripke 1980: 48) — refers to — the person called William Shakespeare: there is a necessary relationship between the name “William Shakespeare” and William Shakespeare, the person, while all we can “predicate” of Shakespeare and thus give also in the form of descriptions (that he wrote Hamlet, etc.) could have been otherwise, and thus are contingent facts of our world. (cf. Kripke 1980: 62). However, after the naming process had taken place, that William Shakespeare is William Shakespeare is not a contingent fact of the world, since this sentence expresses that William Shakespeare is identical with himself. Thus, for Kripke, only “Shakespeare is Shakespeare”, that is, only a genuine identity statement is an analytic truth, and thus a necessary truth in the strict logical sense. By contrast, “Shakespeare is identical with the author of Hamlet” is not an analytic and, thus, a necessary truth.
The problem, then, with the proponents of the descriptive theory of reference, such as Frege or Strawson is that they treated proper names and descriptions as synonymous. Here, of course synonymy is meant not as poetic, or rhetorical, or stylistic synonymy but as strictly cognitive synonymy. Poetically, no two expressions will ever be totally synonymous (cf. Quine 1963: 28). But Kripke’s claim is that not even cognitively will a description of somebody and his or her proper name be synonymous because the criterion of cognitive synonymy is that you can change the two terms — the description and the proper name — in the same proposition, i.e. in exactly the same context salva veritate, i.e., without changing the truth value of the proposition. But while “Shakespeare is Shakespeare” is a necessary truth, “Shakespeare is identical with the author of Hamlet” is not.
This, of course, needs further refinement. The sentence “Shakespeare is not Shakespeare” can make perfect sense in certain contexts: for example, imagine a scholar who, after having done serious research on Shakespeare’s life and work, arrives at the conclusion that everything there is in books, documents, etc. about Shakespeare is wrong, he has been mixed up with somebody else from the start (this claim, as it is well known, has been made in “real life” more than once). That scholar, going up to the pulpit at a conference might start his revelatory lecture by telling his audience: “Ladies and Gentlemen, Shakespeare was not Shakespeare.”☞ Tautologies like “Shakespeare is Shakespeare” can be expressive of something else than identity, too. E.g. somebody enthusiastically tells me about an excellent Hamlet-performance he saw and I may respond: “Well, Shakespeare is Shakespeare”, meaning something like ‘Shakespeare is still (one of) the best playwright(s), so what did you expect?’ These uses of tautologies are in the “Boys will be boys” category. But the scholar will precisely wish to communicate that all items of knowledge and beliefs humankind has so far associated with the name Shakespeare is false and not that Shakespeare, if there was such a person, is not identical with himself. The scholar will wish to say: Shakespeare did not do this or that, did not write the plays attributed to him, etc.; e.g. somebody else did. It will be some kind of knowledge the scholar will challenge, not that a person was identical with himself.
Thus, Kripke pays special attention to identity, treating the name as somehow being “expressive” of the identity of the person or thing; the name as rigid designator is something the designated person or object simply cannot “lose” but is “attached” to each object or person with the force of logical necessity. We may understand what the force of logical necessity is if we look at the definition of necessary truth: a proposition is necessarily true if and only if (=iff) it is true given the way the world (our “real world as we know it”) actually is, and it would have been true, even if the world had been in any other possible state it could have been in (cf. Kripke 1980: 62–63 and Soames 2003: 338). Yet this is not without problems, either. Who could precisely tell what possible state the world “could have been” in? Is it, for example, a possible state of the world that there are no human beings, or that there is no language, in it? These are clearly metaphysical (ontological) questions I will not go into here. Kripke’s definition of rigid designation claims identity for something or somebody in and across all possible worlds, whatever possible worlds may be: for a term X to be a rigid designator is for it to designate (refer to) the identical (the same) person or object in every possible world where the term designates at all (cf. Kripke 1980: 102–105). I just note here that “in every possible world” sounds to me very much like “the whole — the infinite whole — of logical space” tautologies leave open in Wittgenstein’s Tractatus — more on this later.
The concept of possible worlds was introduced not by Wittgenstein but by David K. Lewis (1968), yet Kripke does not conceive of possible worlds the way Lewis does (cf. Kripke 1980: 44–47). Instead of going into lengthy comparisons, I will give an example. Suppose that for the role of Michael Corleone in the film Godfather, two actors competed: Al Pacino and Robert de Niro. Actually, i.e., in our world, it was Al Pacino who got the role but this is a contingent fact of the world that he did; it could have been otherwise, so there is a possible world where the role was played by Robert de Niro. Kripke’s point is that Al Pacino remains Al Pacino, through the rigid designation of his very name, even in the possible world where Robert de Niro played the role. In the possible, alternative world it was not somebody “similar” to the real Al Pacino who did not get the role (as Lewis thinks); Kripke’s proof is that Al Pacino could not have cared less about a “similar” Al Pacino not getting the role; it would have been the real, this-world Al Pacino who may have mourned not to have been able to play Michael Corleone, and would have envied Robert de Niro for playing it.
Let us suppose even further that after the auditions somebody, say Marlon Brando (who played Michael’s father, Vito Corleone in part 1 of Godfather) starts to lecture to de Niro on under what conditions he would have been given the role: “If you had paid more attention to your partner”, “if you had tried to please the director a little bit”, “if you had studied the script more carefully”, etc. Then Marlon Brando would in fact be giving (at least some of) the truth conditions that would make the sentence “Robert de Niro succeeded in getting the role of Michael Corleone” true in our (real) world. But would Marlon Brando, lecturing to de Niro, have said: “if you had tried to please the director a bit, etc., and if you had been identical with yourself?” No, Brando took (would have taken) for granted that de Niro is identical with himself, both in the real world and in the possible world where de Niro got the role.
Similarly, Brando, lecturing to de Niro would not have added: “you would have been given the role had you existed”, either: that is also taken for granted. I think — and this is now my interpretation of Naming and Necessity — that Kripke’s whole theory about rigid designators revolves around the idea that identity is not a predicate. It is an age-old insight (though still contested, of course) that existence is not a predicate. It was relying on this thesis that Immanuel Kant demonstrated why Descartes’s (and, previously, several other philosophers’) ontological argument about the existence of God was at fault: they treated existence as a predicate, i.e., as an attribute, a quality we may claim about a being (cf. Kant 1956: 504). Descartes’s argument was that if we have the concept of God in our minds and we see in that concept that, for example, God is perfect, then it would be absurd, i.e., a logical contradiction to say that he does not exist: the idea of perfection includes or implies existence. But Kant claims that we cannot treat the predicate “exit[s]” on the same level as, say, “is [be] perfect”. If I characterise, for example, my neighbour and say that he is ninety years old, he has white hair, he is six feet tall, he likes apricots, and so on, shall I add, somewhere in my account: “and, oh, by the way, he exists”? Not at all: you and I have taken for granted that it only makes sense to give the attributes of someone if the person talked about exists, and thus existence will not be given among the attributes. In a sense, in a way, everything we talk about, exists: we may call this “mention-existence”. Imagining the existence of something is also a kind of mention-existence: “My financial position” — Kant writes — “is, however, affected very differently by a hundred real thalers than by the mere concept of them (that is, of their possibility). For the object, as it actually exists, is not analytically contained in my concept, but is added to my concept (which is a determination of my state) synthetically; and yet the conceived hundred thalers are not themselves in the least increased through thus acquiring existence outside my concept” (1956: 505). This is to say that I can easily imagine that there is a hundred thalers in my pocket; from the act of imagination there will not — unfortunately — be a hundred thalers in my pocket.
One of Kant’s fundamental insights was — and Frege and his followers whole-heartedly agreed as well — that from the logical structure of language we cannot tell what exists and what does not because language will endow everything with what I call mention-existence. Looking at the logical structure of language, or the meanings in language, or the grammar of language, or at anything in language will not decide for me whether the things I talk about exist in reality or not. Bertrand Russell put this insight in the following way: “In one sense it must be admitted that we can never prove the existence of things other than ourselves and our experiences. No logical absurdity results from the hypothesis that the world consists of myself and my thoughts and feelings and sensations and everything else is mere fancy” (1976: 10). No wonder that lots of philosophers decided that the material world is only my idea — but I will not go into that. The important thing to see is that existence is not a predicate and nothing in language decides whether something does exist in the external world or not. Therefore, from Kant’s argument against Descartes it does not follow that “God does not exist”. What follows is that with the existence of God (and, I wish to claim, with the existence of anything) we are not in a knowing relationship: God’s existence is something we cannot decide about on the basis of knowledge. One of the fundamental problems of Western philosophy has been that it put knowledge on the highest pedestal among the human faculties, and philosophers tended to discard things we do not know but rather feel, intuit, surmise, (aesthetically) appreciate. However, this question cannot be pursued here any longer.
How about identity? Kripke claims that to deny that people or things are identical with themselves is a logical contradiction (1980: 53). This sounds difficult but only until we think of identity as one of the “properties”, “qualities”, “attributes” of a thing, i.e., until we think about identity as if it were a predicate, something we state about an object. But why is it proper names that — as Kripke claims — are most likely to become rigid designators? I may also put the question this way: why was it proper names through which Kripke encountered the identity–problem? Because proper names are more typical of naming particular beings (persons or things) than other words. Particularity is one of the most important “features” or “characteristics” of identity; “feature”, or “characteristic” is, I admit, not the most fortunate term because identity is something, as it will become perhaps clearer below, that cannot be analysed any further. Let us say that particularity “goes along” with identity: identity is always particular. But the problem is that, on the one hand, practically anything can be a carrier of identity. On the other hand, the fact that a proper name is expressive of identity is often shrouded, veiled by several factors.
Practically anything can be expressive of identity because unfortunately anything can be used as a proper name and, thus, become a rigid designator: Kripke at one point acknowledges even demonstratives like this or that as potential rigid designators (cf. 1980: 49). Very confusingly, even descriptions like “the Avon of Stratford”, or “the author of Hamlet” can be used as proper names (I think it was precisely this that confused Strawson and others). Further, genuine proper names may sound or look like descriptions: we may think of the great Native American chief, Sitting Bull, or the chief of the Blackfoot tribe, The-One-Who-Carries-The-Pot or, in the Harry Potter-saga, the evil magician, Voldemort is also called You-Know-Who and even He-Who-Must-Not-Be-Named (a nice rigid designator for Kripke!). Even Voldemort (“[who] wish[es] [the] death”) is a “speaking name” for those who know some Latin. It is also very true that e.g. John Smith is the name of lots of men in the English-speaking world: one (relatively common) proper name may pick out several individuals. These factors all give less chances to us to see Kripke’s point but he does not insist on this or that form of a name; what he insists on is that there is a point when, with a rigid designator, which is often a proper name, we give expression to the identity of a person or thing.
The Kripke-thesis runs as follows (although not with Al Pacino but with Richard Nixon): proper names like Al Pacino are rigid designators, for although the man (Al Pacino) might not have been several things (he might have not become an actor etc.), it is not the case that he might not have been Al Pacino, i.e., identical with himself (cf. Kripke 1980: 49). He of course might have been called something else, had his parents called him otherwise or had his father’s surname been something else, or had we another culture where children do not usually inherit their father’s surname.☞ Illegitimate children usually inherit their mother’s surname but Al Pacino was not an illegitimate child. Kripke’s point is that we need something which is expressive of the necessity of the identity of a thing or person, once that thing or person has been identified. And again it might even be a contingent fact of the world that we identify things and persons through names.☞ Which implies, again, the non-negligible metaphysical questions asked above: is the fact that humans use language a contingent fact of the world as well? Would we have a concept of identity if there was no language, would we bother about it at all? This is tantamount to asking: is a world without language a possible state in which the world could have been? But it is precisely identity itself, more precisely that an object or person is identical with itself (herself, himself), which we need in order to attribute anything to it at all, to describe it, to give it qualities (after identity has been “established”).
It has often been suggested that Kripke thinks rigid designators are somehow anchored in essential properties of things: that in a proper name, earlier than the naming, some essential properties of the named object or person are “dormant” or, later than the naming, some essential properties “get coded” in the name and Kripke thinks it is through being attached to essential properties that a rigid designator becomes rigid (cf., e.g., Soames 2003: 336 and 347–357). But Kripke openly denies this: “Some properties of an object may be essential to it, in that it could not have failed to have them. But these properties are not used to identify the object in another possible world, for such an identification is not needed. Nor need the essential properties of an object be the properties used to identify it in the actual world, if indeed it is identified in the actual world by means of properties” (1980: 53). Rigid designators (often proper names, with the above qualifications) do not “name” or “grasp” a “bundle of properties” in persons or objects. Objects or persons do have properties, of course, essential and accidental, i.e., properties without which they would not be what they are, and, in turn, properties without which they would remain what they are but are still characteristic of them. (That human beings have a heart seems to be an essential property of them, while the colour of their eyes is accidental.) But identity — like existence — is not a characteristic feature, or quality, or property of a thing or person we may predicate of it, or him, or her, either as an essential, or as a an accidental “attribute”. If it were — and this is Kripke’s ingenious insight, I think — it would be a piece of knowledge or belief about the thing or person, with respect to which we may be right or wrong, and thus it would be a contingent fact of the world that could have been otherwise. Yet that something or somebody is identical with itself, himself, herself is a necessary fact of the world (and of all possible worlds as well). Kripke’s thesis seems to imply that — as we are not in a knowing relationship with existence — we are not in a knowing relationship with identity, either.
That “identity is not a predicate” does not, of course, mean that I cannot use identical with predicatively. That identity is not a predicate means that identity is not something I attribute to a thing or person as being among the other properties I know, rightly or wrongly, about the thing or the person. Or perhaps it should be said that I “know” identity in a very special sense, as I know that something “cannot and could not be otherwise”.☞ As, I think — following Wittgenstein — that it is wrong to say that “I know I am in pain”, or “I know I have hands”. These are “closer” to us than we could “know” these, we are somehow “one” with our pain or our hands. But I suggest we reserve “know” to cases where we can go wrong. And e.g. “Shakespeare is identical with Shakespeare” is not such a case because the denial of this sentence is a logical contradiction (unless one means it as the scholar does on the pulpit but that was discussed above). That “Shakespeare is Shakespeare”, or “Shakespeare is identical with Shakespeare” is, thus, not stating a fact about the world. It is a tautology, an analytic truth. And, as Wittgenstein said in the Tractatus: tautologies are “unconditionally true”, they are “not pictures of reality” (4.462).
Here is another example, based on Kripke (1980: 47–51) to explain identity. I have this table I am writing on right now in front of me: this is a particular table. Now let us not ask: what could a table be in a possible world? We are talking about this table. I can physically grasp it, I can refer to it but I am not grasping or referring to an abstract “it”: I am referring to it here and now. Could this table be red in a possible world? (In our ordinary world, it happens to be brown.) Of course. Could it be in another room (and here “another room” is taken as a “possible world”)? Of course. Being red or being in another room are all attributes, qualities of an object. But it would still be it, this particular table which could be in another room or could be red, so could be in a possible world; even in the possible world I would be talking about this table. Here this it-ness, this-ness is expressive of the table’s identity, something it cannot lose. The table must retain its identity in all possible worlds because it may change as many of its qualities as we like, we will still need identity, expressed as “this table” or “it”, to be able to tell what has changed with respect to it. The table must have identity in order that it may have qualities (whatever these qualities may be); otherwise what has changed?
Undeniably, the problem is that what is notoriously ambiguous. To the question: what has changed? I can answer both: “the table”, or “the table’s colour”. But it is its colour that has changed, and — here is Kripke’s point — by it I can not only mean “one or other qualities of the table that have remained unchanged” but it can also refer to the table’s identity, which is not one of the attributes. But let us suppose that I change all the attributes of the table: I cut it up into pieces, and make part of the floor of a room from it. Have I changed, with all the attributes, the identity of the table as well? If I say yes, then what is it all the attributes of which I have changed? I have changed the identity of the table which was so-and-so, and have created a new identity I am expressing with another name, namely, “floor”, which again has all sorts of properties. Of course we identify things, so “this table”, too, through its qualities. But, for Kripke, these qualities are not the bundle that gives the thing identity. Qualities rather “hide” the fact that there is “something”, perhaps “in” the object which is separate and strictly different from all qualities and makes it identical with itself. What is that “something”? It seems as if we were looking for the “soul” of the object, which “flickers” dimly inside, like the flame of a candle, making the thing what it is. But what makes an object what it is, is still not its identity: it is its essential qualities. Yet essential qualities are still qualities and identity is not a quality. Identity is something the object will have until I call it by the name I have learnt about it. The name — as we saw — in its form, as a part of language is arbitrary with respect to the object. But it is precisely with respect to the thing’s “nature” (its qualities etc.) that a name is arbitrary. It is true, as I have already pointed out, that there is nothing in the nature of the thing that would predestine that the object should be called this or that. Even motivated names, e.g. metaphors will carry a fair amount of arbitrariness. There is motivation behind calling the lowest part of a mountain the “foot” of the mountain but there was nothing necessary about the metaphorical extension going this way: perhaps the “saucer” of the mountain, or the “sole” or “toes” of the mountain would also do. There is nothing necessary about seeing, even by a whole speech-community, some analogical relationships which can become the basis of metaphors. If a boy is named after his father, we can see the motivation quite clearly but the decision is not a necessary one: the parents could have decided otherwise; nothing compelled them to name the boy after his father with the force of necessity.☞ The problem of rigid designators involves the problem of free will and determinism, too but here these cannot be dealt with. And the same name can be used to fix the identity of several people (this follows from the very fact that names are arbitrary). The name is not expressive of the thing’s “nature”: it is expressive of its identity (perhaps this is exactly why names are arbitrary). The name fixes the thing for us so that we may identify it as such and such, yet for Kripke identification comes “first” and “then” comes the list of attributes. To speak about what comes “first” and “later” is not a historical, chronological account: identity is so notoriously difficult to talk about because all these happen in “one moment”: the naming and the possible realisation of the thing’s attributes. As for Kripke, this fixing, this designation, this naming (not only in the act of baptism but also when I use a name for reference later on) and the identification of the thing comes about in the same moment as well: one cannot be without the other (cf. 1980: 96). Once an arbitrary name has become the name of the object, it necessarily fixes its identity, or else I use a different name because I have — or I think I have — identified a different thing.
Then what is identity? What is that “something” which is perhaps “in” the object as its “soul”? Identity is so hard to grasp — in fact, to identify — because it is not a thing; if it were, we would have a firm grip on it and get to know it. Identity is a referential relation which we seem to take for granted when we use a name. Identity, I would like to claim — and this again is purely my claim — is part of our logical attitude to the world.
To make this clearer I would like to point out some significant similarities in Kripke’s conception of logic and Wittgenstein’s standpoint in the Tractatus, although Kripke has often been used to repudiate Wittgensteinian insights (e.g. cf. Soames 2003: 13–15). Among other things, the very term rigid designator points towards some affinity between the two positions. Both Kripke and Wittgenstein seem to hold that it is logic, or, more precisely, the logical structure of language which contains some fundamental, unshakeable, unalterable, unconditional truths with absolute and necessary certainty, yet these truths are precisely not facts of the world and not facts we “know” because in the world nothing is unalterable; in the world everything could be otherwise: everything could be true or false and thus these “absolute” truths are not part of the world. For Kripke, it seems to me, such an unalterable truth is that things and persons are identical with themselves, for Wittgenstein, in the Tractatus, among other things such an absolute truth is that there is a logical structure (logical form). Wittgenstein even says that the logical structure of language and of the world cannot be talked about: it remains in the realm of the ineffable, the unsayable, the inexpressible but this does not mean that there is no logical structure as it is not the case that what cannot be talked about would be unimportant or non-existent; on the contrary: it is what we hold to be most precious that lies in the domain of the ineffable. Logical structure is not something we can put into words and further analyse or interpret with language, either.☞ Cf. “Propositions can represent the whole of reality, but they cannot represent what they must have in common with reality in order to be able tot represent it — logical form. I order to be able to represent logical form, we should have to be able to station ourselves with propositions somewhere outside logic, that is to say outside the world. ” (Tractatus, 4.12) “Propositions cannot represent logical form: it is mirrored in them. What finds its reflection in language, language cannot represent. What expresses itself in language, we cannot express by means of language. Propositions show the logical form of reality. They display it. (4.121) “What can be shown, cannot said” (4.1212) (emphasis throughout original). I interpret Wittgenstein’s logical structure in the Tractatus as our very attitude to the world, to the world around us, it is our constant and unalterable way in which we relate to the world. To put logical structure, i.e., our logical relation to the world into words in order to, for example, comment on it would require another standpoint than the one we have, namely a standpoint from which we could see and scrutinise our very attitude. But this attitude is a part of us (it is a pair of irremovable spectacles everyone has on their noses, as it were): we always already relate to everything with this very attitude, we cannot get, so to speak, “before” it so that we may then comfortably compare, from the “outside”, this attitude and the world as two independent phenomena.☞ On the Tractatus see further Kállay (2012)
I would like to interpret Kripke’s notion of identity as part of the logical attitude Wittgenstein, I think, talks about: epistemologically, we do not have a hold, a firm grip on identity – I mean identity itself, the very relation that e.g. a person is identical with her- or himself — because we are not in a knowing relationship with it: if we express it, we express it in a tautology, leaving the whole of logical space open; identity is not something we could analyse any further because it is something with respect to which we analyse everything else. Thus, it appears to us as tautological, and, hence, as trivial but trivial things seem to be the most evident for us; they literally “go without saying”. Ontologically, however, it is an unshakeable part of our being in the sense that it is, so to speak, a part of our primary, instantaneous relation to the world, a relation we always already take for granted. Thus, identity is not “in” the things or persons but rather “in” us as part of the way we logically relate to the world. Identity can be put on display — in the form of tautologies — but cannot be further analysed and — as Wittgenstein proposes it in his “Lecture on Ethics” — we can only resort to similes and allegories to illustrate them (cf. Wittgenstein 1993: 42). I propose that our relation to identity (which is a relation itself) is similar to being absolutely determined or convinced about something, “somewhere deep down inside”, for example — somehow — “in our guts”, something which will never and nowhere change in us, come what may; this is why I consider rigidity in the term rigid designator such a fortunate metaphor.
I think with identity Kripke revived something very significant in philosophy. He revived, among other things, the Kantian insight that with lots of things we are not in a knowing relationship and the Wittgensteinian insight about the nature of necessary or absolute truths: that there are such truths yet they can only be necessary if they are not “reached by language” which could thematise, interpret, or analyse them because if they were, they would cease to be necessary truths, since language can only thematise things about which we may disagree, which can be true or false. (Let me make this clear: a tautology does not thematise, or interpret, or “analyse” identity; it expresses it, it puts identity on display). And, at the same time, and very curiously, these ineffable truths are the ones on which we build when we relate to the world, for example when we wish to get to know the world, when we talk, when we do anything.
Most of the ideas that follow may sound weird; the best is if they are treated as indices of the various directions I would like to go with the problem of identity and, of course, meaning; the two are inseparable.
I take it to be a wonderful gem of wisdom that the Old Testament author, whoever he was, put this sentence into God’s mouth when Moses asks about God’s name: “I am that I am”, a tautology.☞ “And Moses said unto God, Behold, when I come unto the children of Israel, and shall say unto them, The God of your fathers hath sent me unto you; and they shall say to me, What is his name? what shall I say to them? And God said unto Moses, I AM THAT I AM: and he said, Thus shalt thou say unto the children of Israel, I AM hath sent me unto you” (Exodus 3:13–14). I take this to be expressive of the insight that, first of all, the name of God is nothing else but He stating His identity with Himself, which is, at the same time, a necessary truth. Second, if God is the Lord of creation, i.e. that He is the source of all beings — as I think the Old Testament author believed this to be the case — then identity is being, and also the source of being. I do not wish to raise theological issues, I am merely asking: is it possible, now philosophically, that identity precedes being, that is, existence (as the Biblical author seems to imply)? When philosophers, such as Heidegger, wrestled with the problem of being, they insisted that the problem of being should be understood from being itself. Yet could it be that, following the Kripkean and Wittgensteinian path, we could approach the question of being through identity?☞ At one point, Kripke says something very interesting: “Once we’ve got the thing, we know that it existed” (1980: 29, emphasis mine). It seems to me that identity includes the question of being and not the other way round. Or let me put it this way: when we identify something then, with the same stroke, we grant it being as well, identity being the “source” of being, as it were. Perhaps our most fundamental, non-predicative but logical relationship with the world is not being, but identity (which is, as I have tried to argue, not a predicative relation, either).
How does Kripke’s insight that the name is expressive of the identity of a particular thing or person relate to personal identity? Paul Ricœur, who wrote a whole book on personal identity (Ricœur 1992), distinguishes between identity in the sense of sameness on the one hand, and selfhood identity on the other. Sameness is e.g. that I consider myself to be “the same” as I was yesterday; Ricœur calls this idem-identity. Selfhood-identity, called ipseity by Ricœur, is to mean that I am an autonomous, unique self precisely not identical with anybody else. I am not completely identical even e.g. with my yesterday’s self, so this is the self which is capable of changing. It is ipseity which is capable of recognising him-or herself in the Other too; it is the self as ipseity who realises that his or her identity is, at least partly, given in other people, as if others were “mirrors” of the self. Ricœur — who otherwise was one of the few Continental philosophers who built insights coming from the so-called Analytic, Anglo-Saxon tradition into his thinking, too — does not mention Kripke in this book but he does use Strawson’s theory of identity and even the Tractatus appears at one point (cf. Ricœur 1992: 539) . Yet I do not think that some of Kripke’s and Ricœur’s insights would not be compatible; they might even be mutually fertilising. Much depends on to what extent we interpret Kripke’s rigidity in designation, i.e., his insistence that the name is expressive of an identity the person cannot lose in any possible world. Should we say that, indeed, the rigid identification of identity is also expressive of one’s uniqueness? It is, as I pointed out, precisely not the “bunch of essential qualities” of the human being “as such” which is in question; Kripkean identity, I think, can be interpreted precisely in terms of personal identity in the sense of uniqueness, identity being expressive of the fact that each and every personality is a non-repeatable, separate being, different from everyone else and it is in this uniqueness that congenial character, individuality (in which the self is true precisely to him- or herself) is anchored. This leads me to another question.
Could it be that names are expressive of identity because, in one way or another, we grant identity to everything and this granting is based on our very awareness of our selves? Let me put it this way: when we grow conscious of things around us, i.e., we are able to reflect on things, we are also capable of reflecting on our selves. When we become aware of the world, we also become aware of ourselves and vice versa: gaining self-consciousness surely goes hand in hand with growing conscious of what is “outside of us” (“over there, in the world”). I know this is a very difficult question and I will of course not go into it. But provided the above account is not too incredible, I would like to ask: is it possible that we grant identity to persons and things around us using ourselves, our identity, as a “model”? Granting identity might be further described as acknowledging the Other as a being and that he, or she, or it is a unique personality. Let me illustrate this in terms of a credit transaction: granting identity to the Other is like giving the Other a cheque which is already signed by me but the figure, the amount the Other can have access to, has been kept blank: any amount can be written there. The space for the amount is not filled in because I do not have access to the content of the Other’s uniqueness but with the handing over of the cheque I grant, I acknowledge that he, or she, or it is unique. (As I am far from “knowing myself”, too but I am aware that I am unique, I am like nobody else.)
What happens to my identity when I read a novel, watch a play or film, etc, and, as we say, I identify, more or less, with one or more of the characters? Let us take the perhaps crudest case: when an actor (here I will stick to a “he”) personifies somebody on the stage. The age-old question is: does the actor, lending his identity to, say, Hamlet, lose his personal identity while he is Hamlet because for three hours he is not, say, Lawrence Olivier but Hamlet? And how about the identity of Hamlet himself, the role, the role being, after all, first and foremost a text? But the text implies movement, postures, gestures, etc. So is then Hamlet all these, ceasing to exist when he is not personified? Or is Hamlet’s identity anchored in the text and/or the person reading the text? Or was it Shakespeare who gave identity to Hamlet when he named him Hamlet? But we know that Hamlet already existed in Danish chronicles: was it Saxo Grammaticus, author of “the first connected account of the hero whom later ages know as Hamlet” (Jenkins 1982: 85), in his Historia Danicae, who identified Hamlet (with, in fact, not the name “Hamlet” but “Amleth”)? However, Saxo wrote his piece at the end of the 12th century but it was only published for the first time in 1514. Does Hamlet have an identity from the end of the 12th century, from 1514, or from 1600 when (most probably) Shakespeare wrote Hamlet (cf. Jenkins 1982: 85–86)? Or is naming a fictitious character a different business than naming a real being? But what if Saxo considered Hamlet to be a historical, i.e., “real” figure?
From the point of view of the actor we may perhaps claim that if we treat the stage or drama as a “possible world”, then, on the basis of Kripke’s famous dictum, we should say that the person personifying Hamlet does not lose his identity while he is Hamlet; he will remain e.g. Lawrence Olivier for those roughly three hours he needs in order to act Hamlet out. But how does his identity, now in the sense of uniqueness, relate to his interpretation of the role? Will his uniqueness be the “core” of Hamlet’s identity? I think acting differs from granting identity in the fact that acting is also lending identity. But how is that done? And does not the author, or even the viewer, or reader lend some of his or her identity to Hamlet?
Philosophers often like to treat the problem of fiction, acting etc. as something totally different from everyday life, hence willy-nilly implying that what happens in fiction, on the stage etc, cannot inform the questions we are concerned with in real life and, thus, in philosophy. I do not think this would be true. At least some of the things that happen in fiction and at least some of the ways in which we relate to fiction may help us to genuinely philosophical insights and are applicable in everyday life as well. If we consider this question from the point of view of the author, we may find that the author of the fictitious character named Hamlet, did, I claim, exactly the same thing we do when we name a real person. When the — arbitrary — name “Hamlet” became the name of Hamlet the character, the name became expressive of Hamlet being identical with Hamlet. Yet — and here I think there is some benefit for philosophy from fiction — the case of the author naming a fictitious character makes it more obvious, the case of the author displays more perspicuously, what we do in “real life”. We say the author creates his characters, hence the author creates their identity, too. But do we not do exactly the same thing with “real-life” characters as well? I offer the following analogy: if I child is taken from an orphanage, it is obvious the child was adopted. But is it not also true that parents have to adopt even their “natural” (biological) child (and, as a matter of fact, the child his or her parents)? Fiction, in this analogy, plays the role of the orphanage: fiction only sharpens, magnifies or amplifies what the case in “real life” happens to be. We do not only see ourselves, as in a mirror, in the Other. We are also creators: creators of the identity of the Other, including fictitious characters.
Identity seems to me to be a battlefield where ultimately only questions remain standing. One last of these may be put this way: is the author of a text identical with his text? Am I identical with the text you have read? Am I identical with the text’s meaning? I would say no; the text may be typical of, but not identical with, the author. But please read below a very short text, made up of only two words, which, even for Kripke, is expressive of my identity; here it is, with all my gratitude for your having read all this above:
Bowie, Andrew. 2003. Aesthetics and subjectivity: from Kant to Nietzsche. 2nd ed. Manchester and New York: Manchester University Press.
Donnellan, Keith. 2001 [1966]. Reference and Definite Descriptions. In: A. P. Martinich (ed.), The Philosophy of Language. New York: Oxford University Press. 247–259.
Dummett, Michael. 2001. Gottlob Frege. In: A. P. Martinich and David Sosa (eds.), A Companion to Analytic Philosophy. Blackwell Companions to Philosophy. Malden, Mass., and Oxford: Blackwell Publishers Ltd. 6–20.
Frege, Gottlob. 1992 [1892]. On Sense and Reference. In: Michael Beaney (ed.), The Frege Reader. Oxford: Blackwell. 150–180.
Gadamer, Hans Georg. 2010 [1960]. Warheit und Methode. Grundzüge einer philosophischen Hermeneutik [Truth and Method. The Basics of a Philosophical Hermeneutics]. Gesammelte Werke. Band 1. Heremeneuik I. Tübingen: Mohr Siebeck.
Heidegger, Martin. 1951. Sein und Zeit [Being and Time]. Seventh ed. Tübingen: Max Niemeyer Verlag.
Heidegger, Martin. 1981. Erleiterungen zu Hölderlins Dichtung [Explanations for Hölderlin’s poetry]. In Heidegger Gesamtaufgabe. I. Abteliung: Veröffentliche Shriften 1910–1976. Band 4. Frankfurt Am Main: Vittorio Klostermann.
Jenkins, Harold (ed.). 1982. William Shakespeare: Hamlet. The Arden edition. 2nd series. London and New York: Methuen.
Kállay, Géza. 2012. Nonsense and the Ineffable: Re-reading the Ethical Standpoint in Wittgenstein’s Tractatus. Nordic Wittgenstein Review 1: 103–130.
Kant, Immanuel. 1956 [1787]. Critique of Pure Reason. Trans. by Norman Kemp Smith. London and New York: Macmillan and Co. Ltd. and St. Martin’s Press.
Kripke, Saul. 1980. Naming and Necessity. Cambridge: Harvard University Press.
Lewis, David K. 1968. Counterpart Theory and Quantified Modal Logic. Journal of Philosophy 65: 113–126.
Quine, Willard Van Orman. 1963. Two Dogmas of Empiricism. In: Quine From a Logical Point of View. New York and Evaston: Harper and Row. Harper Torchbooks. 20–46.
Ricœur, Paul. 1992. Oneself as Another. Trans. by Kathleen Blamey. Chicago and London: The University of Chicago Press.
Russell, Bertrand. 1976. The Problems of Philosophy, Oxford: Oxford University Press.
Russell, Bertrand. 2001 [1905]. On Denoting. In: A. P. Martinich (ed.), The Philosophy of Language. New York: Oxford University Press. 212–220.
Searle, John R. 1958. Proper Names. Mind 57: 166–173.
Soames, Scott. 2003. Saul Kripke on Naming and Necessity. In: Soames Philosophical Analysis in the Twentieth Century. Vol. 2. The Age of Meaning. Princeton and Oxford: Princeton University Press. 333–456.
Strawson, Peter. 1959. Persons. In: Strawson Individuals. London: Methuen. 87–116.
Strawson, Peter. 2001 [1950]. On Referring. In: A. P. Martinich (ed.), The Philosophy of Language. New York and Oxford: Oxford University Press. 228–242.
Wittgenstein, Ludwig. 1961 [1921]. Tractatus Logico-Philosophicus. Trans. by David Pears and Brian McGuinness. London: Routledge.
Wittgenstein, Ludwig. 1993 [1929] A Lecture on Ethics. In: James Klagge and Alfred Nordman (eds.), Ludwig Wittgenstein Philosophical Occasions, 1912–1951. Indianapolis and Cambridge: Hackett Publishing Company, 1993. 37–44.
Wittgenstein, Ludwig. 2001 [1951, 1958]. Philosophical Investigations. The German text, with a Revised English Translation, trans. by G. E. M. Anscombe, Malden, Oxford, Carlton: Blackwell Publishing Ltd.
ucsc.edu
VLlxx, ISBN 978-963-284-315-5