Abstract
I hope that some people see some connection between the two topics in the title. If not, anyway, such connections will be developed in the course of these talks. Furthermore, because of the use of tools involving reference and necessity in analytic philosophy today, our views on these topics really have wide-ranging implications for other problems in philosophy that traditionally might be thought far-removed, like arguments over the mind-body problem or the so-called ‘identity thesis’. Materialism, in this form, often now gets involved in very intricate ways in questions about what is necessary or contingent in identity of properties — questions like that. So, it is really very important to philosophers who may want to work in many domains to get clear about these concepts. Maybe I will say something about the mind-body problem in the course of these talks. I want to talk also at some point (I don’t know if I can get it in) about substances and natural kinds.
In January and February of 1970, I gave the three talks at Princeton University transcribed here. As the style of the transcript makes clear, I gave the talks without a written text, and, in fact, without notes. The present text is lightly edited from the verbatim transcript; an occasional passage has been added to expand the thought, an occasional sentence has been rewritten, but no attempt has been made to change the informal style of the original. Many of the footnotes have been added to the original, but a few were originally spoken asides in the talks themselves.
I hope the reader will bear these facts in mind as he reads the text. Imagining it spoken, with proper pauses and emphases, may occasionally facilitate comprehension. I have agreed to publish the talks in this form with some reservations. The time allotted, and the informal style, necessitated a certain amount of compression of the argument, inability to treat certain objections, and the like. Especially in the concluding sections on scientific identities and the mind-body problem thoroughness had to be sacrificed. Some topics essential to a full presentation of the viewpoint argued here, especially that of existence statements and empty names, had to be omitted altogether. Further, the informality of the presentation may well have engendered a sacrifice of clarity at certain points. All these defects were accepted in the interest of early publication. I hope that perhaps I will have the chance to do a more thorough job later. To repeat, I hope the reader will bear in mind that he is largely reading informal lectures, not only when he encounters repetitions or infelicities, but also when he encounters irreverence or corn.
The erratum of this chapter is available at http://dx.doi.org/10.1007/978-94-010-2557-7_26
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
In January and February of 1970, I gave the three talks at Princeton University transcribed here. As the style of the transcript makes clear, I gave the talks without a written text, and, in fact, without notes. The present text is lightly edited from the verbatim transcript; an occasional passage has been added to expand the thought, an occasional sentence has been rewritten, but no attempt has been made to change the informal style of the original. Many of the footnotes have been added to the original, but a few were originally spoken asides in the talks themselves. I hope the reader will bear these facts in mind as he reads the text. Imagining it spoken, with proper pauses and emphases, may occasionally facilitate comprehension. I have agreed to publish the talks in this form with some reservations. The time allotted, and the informal style, necessitated a certain amount of compression of the argument, inability to treat certain objections, and the like. Especially in the concluding sections on scientific identities and the mind-body problem thoroughness had to be sacrificed. Some topics essential to a full presentation of the viewpoint argued here, especially that of existence statements and empty names, had to be omitted altogether. Further, the informality of the presentation may well have engendered a sacrifice of clarity at certain points. All these defects were accepted in the interest of early publication. I hope that perhaps I will have the chance to do a more thorough job later. To repeat, I hope the reader will bear in mind that he is largely reading informal lectures, not only when he encounters repetitions or infelicities, but also when he encounters irreverence or corn.
Given a chance to add a footnote, I shall mention that Rogers Albritton, Charles Chastain, Keith Donnellan, and Michael Slote (in addition to philosophers mentioned in the text, especially Hilary Putnam), have independently expressed views with points of contact with various aspects of what I say here. Albritton called the problems of necessity and a prioricity in natural kinds to my attention, by raising the question whether we could discover that lemons were not fruits. (I am not sure he would accept all my conclusions.) The apology in the text still stands; I am aware that the list in this footnote is far from comprehensive. I make no attempt to enumerate those friends and students whose stimulating conversations have helped me. Thomas Nagel and Gilbert Harman deserve special thanks for their help in editing the transcript.
Keith Donnellan, ‘Reference and Definite Descriptions’, Philosophical Review 75 (1966), pp. 281–304.
See also Leonard Linsky, ‘Reference and Referents’, Philosophy and Ordinary Language (ed. by Caton) Urbana, 1963. Donnellan’s distinction seems applicable to names as well as to descriptions. Two men glimpse someone at a distance and think they recognize him as Jones. ‘What is Jones doing?’ ‘Raking the leaves’. If the distant leaf-raker is actually Smith, then in some sense they are referring to Smith, even though they both use ‘Jones’ as a name of Jones. In the text, I speak of the ‘referent’ of a name to mean the thing named by the name — e.g., Jones, not Smith — even though a speaker may sometimes properly be said to use the name to refer to someone else. Perhaps it would have been less misleading to use a technical term, such as ‘denote’ rather than ‘refer’. My use of ‘refer’ is such as to satisfy the schema, ‘The referent of ‘X’ is X’; where ‘X’; is replaceable by any name or description. I am tentatively inclined to believe, in opposition to Donnellan, that his remarks about reference have little to do with semantics or truth-conditions, though they may be relevant to a theory of speech-acts. Space limitations do not permit me to explain what I mean by this, much less defend the view, except for a brief remark: Call the referent of a name or description in my sense the ‘semantic referent’; for a name, this is the thing named, for a description, the thing uniquely satisfying the description. Then the speaker may refer to something other than the semantic referent if he has appropriate false beliefs. I think this is what happens in the naming (Smith-Jones) cases and also in the Donnellan ‘champagne’ case; the one requires no theory that names are ambiguous, and the other requires no modification of Russell’s theory of descriptions.
Strictly speaking, of course, Russell says that the names don’t abbreviate descriptions and don’t have any sense; but then he also says that, just because the things that we call ‘names’ do abbreviate descriptions, they’re not really names. So, since ‘Walter Scott’; according to Russell, does abbreviate a description,’ Walter Scott’ is not a name, and the only names that really exist in ordinary language are, perhaps, demonstratives such as ‘this’ or ‘that’, used on a particular occasion to refer to an object with which the speaker is ‘acquainted’ in Russell’s sense. Though we won’t put things the way Russell does, we could describe Russell as saying that names, as they are ordinarily called, do have sense. They have sense in a strong way, namely, we should be able to give a definite description such that the referent of the name, by definition, is the object satisfying the description. Russell himself, since he eliminates descriptions from his primitive notation, seems to hold in ‘On Denoting’ that the notion of ‘sense’ is illusory. In reporting Russell’s views, we thus deviate from him in two respects. First, we stipulate that ‘names’ shall be names as ordinarily conceived, not Russell’s ‘logically proper names’; second, we regard descriptions, and their abbreviations, as having sense.
When I speak of the Frege-Russell view and its variants, I include only those versions which give a substantive theory of the reference of names. In particular, Quine’s proposal that in a ‘canonical notation’ a name such as ‘Socrates’ should be replaced by a description ‘the Socratizer’ (where ‘Socratizes’ is an invented predicate), and that the description should then be eliminated by Russell’s method, was not intended as a theory of reference for names but as a proposed reform of language with certain advantages. The problems discussed here will all apply, mutatis mutandis, to the reformed language; in particular, the question, ‘How is the reference of ‘Socrates’ determined?’ yields to the question, ‘How is the extension of ‘Socratizes’ determined?’ Of course I do not suggest that Quine has ever claimed the contrary.
Gottlob Frege, ‘On Sense and Nominatum’, translated by Herbert Feigl in Readings in Philosophical Analysis (ed. by Herbert Feigl and Wilfrid Seilars), Appleton Century Crofts, 1949, p. 86.
Ludwig Wittgenstein, Philosophical Investigations, translated by G. E. M. Anscombe, MacMillan, 1953, p. 79.
John R. Searle, ‘Proper Names’, Mind 67 (1958), 166–73.
Paul Ziff, Semantic Analysis, Cornell U.P., 1960, esp. pp. 85–89, 93–94, 104–105, 173–176.
Those determinists who deny the importance of the individual in history may well argue that had Moses never existed, someone else would have arisen to achieve all that he did. Their claim cannot be refuted by appealing to a correct philosophical theory of the meaning of ‘Moses exists’.
By the way, it’s a common attitude in philosophy to think that one shouldn’t introduce a notion until it’s been rigorously defined (according to some popular notion of rigor). Here I am just dealing with an intuitive notion and will keep on the level of an intuitive notion. That is, we think that some things, though they are in fact the case, might have been otherwise. I might not have given these lectures today. If that’s right, then it is possible that I wouldn’t have given these lectures today. Quite a different question is the epistemological question, how any particular person knows that I gave these lectures today. I suppose in that case he does know this is a posteriori. But, if someone were born with an innate belief that I was going to give these lectures today, who knows? Right now, anyway, let’s suppose that people know this a posteriori. At any rate, the two questions being asked are different.
The example I gave asserts a certain property — electoral victory — to be accidental to Nixon, independently of how he is described. Of course, if the notion of accidental property is meaningful, the notion of essential property must be meaningful also. This is not to say that there are any essential properties — though, in fact, I think there are. The usual argument questions the meaningfulness of essentialism, and says that whether a property is accidental or essential to an object depends on how it is described. It is thus not the view that all properties are accidental. Of course, it is also not the view, held by some idealists, that all properties are essential, all relations internal.
David K. Lewis, ‘Counterpart Theory and Quantified Modal Logic’, Journal of Philosophy 65 (1968), 113–126. Lewis’s elegant paper also suffers from a purely formal difficulty: on his interpretation of quantified modality, the familiar law (y) ((x)A(x) ⊃ A(y)) fails, if A(x) is allowed to contain modal operators. (For example, (∃y)((x) ◊ (x ≠; y))is satisfiable but (∃y) ◊; (y ≠ y) is not.) Since Lewis’s formal model follows rather naturally from his philosophical views on counterparts, and since the failure of universal instantiation for modal properties is intuitively bizarre, it seems to me that this failure constitutes an additional argument against the plausibility of his philosophical views. There are other, lesser, formal difficulties as well. I cannot elaborate here. Strictly speaking, Lewis’s view is not a view of ‘transworld identification’. Rather, he thinks that similarities across possible worlds determine a counterpart relation which need be neither symmetric nor transitive. The counterpart of something in another possible world is never identical with the thing itself. Thus if we say “Humphrey might have won the election (if only he had done such-and-such), we are not talking about something that might have happened to Humphrey but to someone else, a ‘counterpart’.” Probably, however, Humphrey could not care less whether someone else, no matter how much resembling him, would have been victorious in another possible world. Thus, Lewis’s view seems to me even more bizarre than the usual notions of transworld identification that it replaces. The important issues, however, are common to the two views: the supposition that other possible worlds are like other dimensions of a more inclusive universe, that they can be given only by purely qualitative descriptions, and that therefore either the identity relation or the counterpart relation must be established in terms of qualitative resemblance. Many have pointed out to me that the father of counterpart theory is probably Leibniz. I will not go into such a historical question here. It would also be interesting to compare Lewis’s views with the Wheeler-Everett interpretation of quantum mechanics. I suspect that this view of physics may suffer from philosophical problems analogous to Lewis’s counterpart theory; it is certainly very similar in spirit.
Another locus classicus of the views I am criticizing, with more philosophical exposition than Lewis’s paper, is a paper by David Kaplan on transworld identification. This paper has unfortunately never been published and no longer represents Kaplan’s position.
Misleadingly, because the phrase suggests that there is a special problem of ‘transworld identification’, that we cannot trivially stipulate whom or what we are talking about when we imagine another possible world. The term ‘possible world’ may also mislead; perhaps it suggests the ‘foreign country’ picture. 1 have sometimes used ‘counter-factual situation’ in the text; Michael Slote has suggested that ‘possible state of the world’ might be less misleading than ‘possible world’. It is better still, to avoid confusion, not to say, ‘In some possible world, Humphrey would have won’ but rather, simply, ‘Humphrey might have won’. The apparatus of possible worlds has (I hope) been very useful as far as the set-theoretic model-theory of quantified modal logic is concerned, but has encouraged philosophical pseudo-problems and misleading pictures.
Of course I don’t imply that language contains a name for every object. Demonstratives can be used as rigid designators, and free variables can be used as rigid designators of unspecified objects. Of course when we specify a counterfactual situation, we do not describe the whole possible world, but only the portion which interests us.
See Lecture I, p. 273 (on Nixon), arid Lecture II, pp. 287–289.
There is some vagueness here. If a chip, or molecule, of a given table had been replaced by another one, we would be content to say that we have the same table. But if too many chips were different, we would seem to have a different one. The same problem can, of course, arise for identity over time. Where the identity relation is vague, it may seem intransitive; a claim of apparent identity may yield an apparent non-identity. Some sort of ‘counterpart’ notion (though not with Lewis’s philosophical underpinnings of resemblance, foreign country worlds, etc.), may have more utility here. One could say that strict identity applies only to the particulars (the molecules), and the counterpart relation to the particulars ‘composed’ of them, the tables. The counterpart relation can then be declared to be vague and intransitive. It seems, however, Utopian to suppose that we will ever reach a level of ultimate, basic particulars for which identity relations are never vague and the danger of intransitivity is eliminated. The danger usually does not arise in practice, so we ordinarily can speak simply of identity without worry. Logicians have not developed a logic of vagueness.
Philosophical Investigations, § 50.
Philosophers of science may see the key to the problem in a view that ‘one meter’ is a ‘cluster concept’. I am asking the reader hypothetically to suppose that the ‘definition’ given is the only standard used to determine the metric system. I think the problem would still arise.
Since the truth he knows is contingent, I choose not to call it ‘analytic’, stipulatively requiring analytic truths to be both necessary and a priori. See footnote 63.
Usually the Fregean sense is now interpreted as the meaning, which must be carefully distinguished from a ‘reference fixer’. We shall see below that for most speakers, unless they are the ones who initially give an object its name, the referent of the name is determined by a ‘causal’ chain of the communication rather than a description. Hartry Field has proposed that, for some of the purposes of Frege’s theory, his notion of sense should be replaced by the chain which determines the reference. In the formal semantics of modal logic, the sense of a term t is usually taken to be the (possibly partial) function which assigns to each possible world H the referent of t in H. For a rigid designator, such a function is constant. This notion of sense relates to that of ‘giving a meaning’, not that of fixing a reference. In this use of sense, ‘one meter’ has a constant function as its sense, though its reference is fixed by ‘the length of S’, which does not have a constant function as its sense. Some philosophers have thought that descriptions, in English, are ambiguous, that sometimes they non-rigidly designate, in each world, the object (if any) satisfying the description, while sometimes they rigidly designate the object actually satisfying the description. (Others, inspired by Donnellan, say the description sometimes rigidly designates the object thought or presupposed to satisfy the description.) I find any such alleged ambiguities dubious. I know of no clear evidence for it which cannot be handled either by Russell’s notion of scope or by the considerations alluded to in footnote 3 above. If the ambiguity does exist, then in the supposed rigid sense of ‘the length of S’, ‘one meter’ and ‘the length of S’ designate the same thing in all possible worlds and have the same (functional) sense. In the formal semantics of intensional logic, suppose we take a definite description to designate, in each world, the object satisfying the description. It is indeed useful to have an operator which transforms each description into a term which rigidly designates the object actually satisfying the description. David Kaplan has proposed such an operator and calls it ‘Dthat’.
P. F. Strawson, Individuals, Methuen, London, 1959, Ch. 6.
Searle, op. cit. in Caton, Philosophy and Ordinary Language, p. 160.
The facts that ‘the teacher of Alexander’ is capable of scope distinctions in modal contexts and that it is not a rigid designator are both illustrated when one observes that the teacher of Alexander might not have taught Alexander (and, in such circumstances, would not have been the teacher of Alexander). On the other hand, it is not true that Aristotle might not have been Aristotle, although Aristotle might not have been called ‘Aristotle’, just as 2 × 2 might not have been called ‘four’. (Sloppy, colloquial speech, which often confuses use and mention, may, of course, express the fact that someone might have been called, or not have been called, ‘Aristotle’ by saying that he might have been, or not have been, Aristotle. Occasionally, I have heard such loose usages adduced as counterexamples to the applicability of the present theory to ordinary language. Colloquialisms like these seem to me to create as little problem for my theses as the success of the ‘Impossible Missions Force’ creates for the modal law that the impossible does not happen.) Further, although under certain circumstances Aristotle would not have taught Alexander, these are not circumstances under which he would not have been Aristotle.
If someone fixes a meter as ‘the length of stick S at t 0, then in some sense he knows a priori that the length of stick S at to is one meter, even though he uses this statement to express a contingent truth. But, merely by fixing a system of measurement, has he thereby learned some (contingent) information about the world, some new fact that he did not know before? It seems plausible that in some sense he did not, even though it is undeniably a contingent fact that S is one meter long. So there may be a case for reformulating the thesis that everything a priori is necessary so as to save it from this type of counterexample. As I said, I don’t know how such a reformulation would go; the reformulation should not be such as to make the thesis trivial (e.g., by defining a priori as known to be necessary (instead of true) independently of experience); and the converse thesis would still be false. Since I will not attempt such a reformulation, I shall consistently use the term a priori in the text so as to make statements whose truth follows from a definition which fixes a reference a priori.
Strawson, op. cit. , pp. 191–192. Strawson actually considers the case of several speakers, pools their properties, and takes a democratic (equally weighted) vote. He requires only a sufficient plurality, not a majority.
See, for example, H. L. Ginsberg, The Five Megilloth and Jonah, The Jewish Publication Society of America, 1969, p. 114: “The ‘hero’ of this tale, the prophet Jonah the son of Amittai, is a historical personage… (but) this book is not history but fiction.” The scholarly consensus regards all details about Jonah in the book as legendary and not even based on a factual substratum, excepting the bare statement that he was a Hebrew prophet, which is hardly uniquely identifying. Nor need he have been called ‘ Jonah’ by the Hebrews; the ‘J’ sound does not exist in Hebrew, and Jonah’s historical existence is independent of whether we know his original Hebrew name or not. The fact that we call him Jonah cannot be used to single him out without circularity. The evidence for Jonah’s historicity comes from an independent reference to him in II Kings; but such evidence could have been available in the absence of any such other references — e.g., evidence that all Hebrew legends were about actual personages. Further, the statement that Jonah is a legend about a real person might have been true, even if there were no evidence for it. One may say, “The Jonah of the book never existed,” as one may say, “The Hitler of Nazi propaganda never existed.” As the quotation above shows, this usage need not coincide with the historians view of whether Jonah ever existed. Ginsberg is writing for the lay reader, who, he assumes, will find his statement intelligible.
In Ernest Nagel, Patrick Suppes, and Alfred Tarski, Logic, Methodology and the philosophy of Science: Proceedings of the 1960 International Congress, Stanford University Press, 1962, 622–633.
Pp. 629–630.
Searle, ‘Proper Names’, in Caton, p. 160.
D. Lewis, op. cit. , pp. 114–115.
An even better case of determining the reference of a name by description, as opposed to ostention, is the discovery of the planet Neptune. Neptune was hypothesized as the planet which caused such and such discrepancies in the orbits of certain other planets. If Leverrier indeed gave the name ‘Neptune’ to the planet before it was ever seen, then he fixed the reference of ‘Neptune’ by means of the description just mentioned. At that time he was unable to see the planet even through a telescope. At this stage, an a priori material equivalence held between the statements ‘Neptune exists’ and ‘some one planet perturbing the orbit of such and such other planets exists in such and such a position’, and also such statements as ‘if such and such perturbations are caused by a planet, they are caused by Neptune’ had the status of a priori truths. Nevertheless, they were not necessary truths, since ‘Neptune’ was introduced as a name rigidly designating a certain planet. Leverrier could well have believed that if Neptune had been been knocked off its course one million years earlier, it would cause no such perturbations and even that some other object might have caused the perturbations in its place.
Following Donnellan’s remarks on definite descriptions, we should add that in some cases, an object may be identified, and the reference of a name fixed, using a description which may turn out to be false of its object. The case where the reference of ‘Phosphorus’ is determined as the ‘morning star’, which later turns out not to be a star, is an obvious example. In such cases, the description which fixes the reference clearly is in no sense known a priori to hold of the object, though a more cautious substitute may be. If such a more cautious substitute is available, it is really the substitute which fixes the reference in the sense intended in the text.
Some of the theses are sloppily stated in respect of fussy matters like use of quotation marks and related details. (For example, Theses (5) and (6), as stated, presuppose that the speaker’s language is English.) Since the purport of the theses is clear, and they are false anyway, I have not bothered to set these things straight.
The cluster-of-descriptions theory of naming would make ‘Peano discovered the axioms for number theory’ express a trivial truth, not a misconception, and similarly for other misconceptions about the history of science. Some who have conceded such cases to me have argued that there are other uses of the same proper names satisfying the cluster theory. For example, it is argued, if we say, “Gödel proved the incompleteness of arithmetic,” we are, of course, referring to Gödel, not to Schmidt. But, if we say, “Gödel relied on a diagonal argument in this step of the proof,” don’t we here, perhaps, refer to whoever proved the theorem? Similarly, if someone asks, “What did Aristotle (or Shakespeare) have in mind here?”, isn’t he talking about the author of the passage in question, whoever he is? By analogy to Donnellan’s usage for descriptions, this might be called an ‘attributive’ use of proper names. If this is so, then assuming the Gödel-Schmidt story, the sentence ‘Gödel proved the incompleteness theorem’ is false, but ‘Gödel used a diagonal argument is the proof is (at least in some contexts) true, and the reference of the name, ‘Gödel’ is ambiguous. Since some counterexamples remain, the cluster-of-descriptions theory would still, in general, be false, which was my main point in the text; but it would be applicable in a wider class of cases than I thought. I think, however, that no such ambiguity need be postulated. It is, perhaps, true that sometimes when someone uses the name ‘Gödel’, his main interest is in whoever proved the theorem, and perhaps he refers to him. I do not think that this case is different from the case of Smith and Jones in footnote 3. If I mistake Jones for Smith, I may refer (in an appropriate sense) to Jones when I say that Smith is raking the leaves; nevertheless I do not use ‘Smith’ ambiguously, as a name sometimes of Smith and sometimes of Jones, but univocally as a name of Smith. Similarly, if I erroneously think that Aristotle wrote such-and-such passage, I may perhaps use ‘Aristotle’ to refer to the actual author of the passage, without supposing any ambiguity in my use of the name. In both cases, I will withdraw my original statement, and my original use of the name, if apprised of the facts. Recall that, in these lectures, ‘referent’ is used in the technical sense of the thing named by a name (or uniquely satisfying a description), and there should be no confusion.
It has been suggested to me that someone might argue that a name is associated with a ‘referential’ use of a description in Donnellan’s sense. For example, although we identify Gödel as the author of the incompleteness theorem, we are talking about him even if he turns out not to have proved the theorem. Theses (2)-(6) could then fail; but nevertheless each name would abbreviate a description, though the role of descriptions in naming would differ radically from that imagined by Frege and Russell. As I have said above, I am inclined to reject Donnellan’s formulation of the notion of referential definite description. Even if Donnellan’s analysis is accepted, however, it is clear that the present proposal should not be. For a referential definite description, such as ‘the man drinking champagne’, is typically withdrawn when the speaker realizes that it does not apply to its object. If a Gödelian fraud were exposed, Gödel would no longer be called the ‘author of the incompleteness theorem’ but he would still be called ‘Gödel’. The description, therefore, does not abbreviate the name.
As Robert Nozick pointed out to me, there is a sense in which a description theory must be trivially true if any theory of the reference of names, spelled out in terms independent of the notion of reference, is available. For if such a theory gives conditions under which an object is to be the referent of a name, then it of course uniquely satisfies these conditions. Since I am not pretending to give any theory which eliminates the notion of reference in this sense, I am not aware of any such trivial fulfillment of the description theory and doubt that one exists. (A description using the notion of the reference of a name is easily available but circular, as we saw in our discussion of Kneale.) If any such trivial fulfillment were available, however, the arguments I have given show that the description must be one of a completely different sort from that supposed by Frege, Russell, Searle, Strawson and other advocates of the description theory.
Princeton University.
Strawson, op. cit. , p. 181n.
The essential points of this example were suggested by Richard Miller.
A good example of a baptism whose reference was fixed by means of a description was that of naming Neptune in footnote 3. The case of a baptism by ostension can perhaps be subsumed under the description concept also. Thus the primary applicability of the description theory is that of initial baptism. Descriptions are also used to fix a reference is cases of designation which are similar to naming except that the terms introduced are not usually called names. The terms ‘one meter’, ‘100 degrees Centigrade’, have already been given as examples, and other examples will be given later in these lectures. Two things should be emphasized concerning the case of introducing a name via a description in an initial baptism. First, the description used is not synonymous with the name it introduces but rather fixes its reference. Here we differ from the usual description theorists. Second, most cases of initial baptism are far from those which originally inspired the description theory. Usually a baptizer is acquainted in some sense with the object he names and is able to name it ostensively. Now the inspiration of the description theory lay in the fact that we can often use names of famous figures of the past who are long dead and with whom no living person is acquainted; and it is precisely these cases which, on our view, cannot be correctly explained by a description theory.
I can transmit the name of the aardvark to other people. For each of these people, as for me, there will be a certain sort of causal or historical connection between my use of the name and the Emperor of the French, but not one of the required type.
Once we realize that the description used to fix the reference of a name is not synonymous with it, then the description theory can be regarded as presupposing the notion of naming or reference. The requirement I made that the description used not itself involve the notion of reference in a circular way is something else and is crucial if the description theory is to have any value at all. The reason is that the description theorist supposes that each speaker essentially uses the description he gives in an initial act of naming to determine his reference. Clearly, if he introduces the name ‘Cicero’ by the determination ‘by ‘Cicero’ I shall refer to the man I call ‘Cicero’,’ he has by this ceremony determined no reference at all. Not all description theorists thought that they were eliminating the notion of reference altogether. Perhaps most realized that some notion of ostension, or primitive reference, is required to back it up. Certainly Russell did.
Ruth Barcan Marcus, ‘Modalities and Intensional Languages’ (comments by W. V. Quine, plus discussion) Boston Studies in the Philosophy of Science, volume I, Reidel, Dordrecht, Holland, 1963, pp. 77 116.
p. 101.
p. 115.
There is a more elaborate discussion of this point in the third lecture, where its relation to a certain sort of counterpart theory is also mentioned.
New York, McGraw-Hill (1953), see Chapter VII, ‘Equality’.
Of course, the device will fail to convince a philosopher who wants to argue that an artificial language or concept of the supposed type is logically impossible. In the present case, some philosophers have thought that a relation, being essentially two-termed, cannot hold between a thing and itself. This position is plainly absurd. Someone can be his own worst enemy, his own severest critic and the like. Some relations are reflexive such as the relation ‘no richer than’. Identity or schmidentity is nothing but the smallest reflexive relation. I hope to elaborate on the utility of this device of imagining a hypothetical language elsewhere.
Recall that we describe the situation in our language, not the language that the people in that situation would have used. Hence we must use the terms Hesperus and Phosphorus with the same reference as in the actual world. The fact that people in that situation might or might not have used these names for different planets is irrelevant. So is the fact that they might have done so using the very same descriptions as we did to fix their references.
The same three options exist for ‘Hesperus is Hesperus’, and the answer must be the same as in the case of ‘Hesperus is Phosphorus’.
‘Internal and External Properties’, Mind 71(April, 1962), pp. 202–203.
The Prince and The Pauper.
Of course I was pointing to a wooden table in the room.
A principle suggested by these examples is: If a material object has its origin from a certain hunk of matter, it could not have had its origin in any other matter. Some qualifications might have to be stated (for example, the vagueness of the notion of hunk of matter leads to some problems), but in a large class of cases the principle is perhaps susceptible of something like proof, using the principle of the necessity of identity for particulars. Let ‘A’ be a name (rigid designator) of a table, let ‘B’ name the piece of wood from which it actually came. Let ‘C’ name another piece of wood. Then suppose A were made from B, as in the actual world, but also another table D were simultaneously made from C. (We assume that there is no relation between A and B which makes the possibility of making a table from one dependent on the possibility of making a table from the other.) Now in this situation B ≠ D; hence, even if D were made by itself, and no table were made from A, D would not be B. Strictly speaking, the ‘proof uses the necessity of distinctness, not of identity. The same types of considerations that can be used to establish the former can, however, be used to establish the latter. (Suppose A ≠ B; if A and B were both identical to some object C in another possible world, then A = C, B = C, hence A = B. )Alternatively, the principle follows from the necessity of identity plus the ‘Brouwersche’ axiom, or equivalently symmetry of the accesibility relation between possible worlds. In any event, the argument applies only if the making of D from C does not affect the possibility of making B from A, and vice-versa.
In addition to the principle that the origin of an object is essential to it, another principle suggested is that the substance of which it is made is essential. Several complications exist here. First, one should not confuse the type of essence involved in the question “What properties must an object retain if it is not to cease to exist, and what properties of the object can change while the object endures?”, which is a temporal question with the question “What (timeless) properties could the object not have failed to have and what properties could it have lacked while still (timelessly) existing?” which concerns necessity and not time and which is our topic here. Thus the question of whether the table could have changed into ice is irrelevant here. The question whether the table could originally have been made of anything other than wood is relevant. Obviously this question is related to the necessity of the origin of the table from a given block of wood and whether that block, too, is essentially wood (even wood of a particular kind). Thus it is ordinarily impossible to imagine the table made from any substance other than the one of which it is actually made without going back through the entire history of the universe, a mind-boggling feat. (Other possibilities of the table not having been wooden have been suggested to me, including an ingenious suggestion of Slote’s, but I find none of them really convincing. I cannot discuss them here.) A full discussion of the problems of essential properties of particulars is impossible here, but I will mention a few other points: (1) Ordinarily when we ask intuitively whether something might have happened to a given object, we ask whether the universe could have gone on as it actually did up to a certain time, but diverge in its history from that point forward so that the vicissitudes of that object would have been different from that time forth. Perhaps this feature should be erected into a general principle about essence. Note that the time in which the divergence from actual history occurs may be sometime before the object itself is actually created. For example, I might have been deformed if the fertilized egg from which I originated had been damaged in certain ways even though I presumably did not yet exist at that time. (2) I am not suggesting that only origin and substantial makeup are essential. For example, if the very block of wood from which the table was made had instead been made into a vase, the table never would have existed. So (roughly) being a table seems to be an essential property of the table. (3) Just as the question whether an object actually has a certain property (e.g. baldness), can be vague, so the question whether the object essentially has a certain property can be vague, even when the question whether it actually has the property is decided. (4) Certain counterexamples of the origin principle appear to exist in ordinary parlance. I am convinced that they are not genuine counter examples, but very exact analysis is difficult. I cannot discuss this here.
Peter Geach has advocated (in Mental Acts, London, Routledge and Kegan Paul, 1957, Section 16, and elsewhere) a notion of ‘nominal essence’ different from the type of essential property considered here. According to Geach, since any act of pointing is ambiguous, someone who baptizes an object by pointing to it must apply a sortal property to disambiguate his reference and to ensure correct criteria of identity over time — for example, someone who assigns a reference to ‘Nixon’ by pointing to him must say, ‘I use ‘Nixon’ as a name of that man’, thus removing his hearer’s temptations to take him to be pointing to a nose or a time-slice. The sortal is then in some sense part of the meaning of the name; names do have a (partial) sense after all, though their senses may not be complete enough to determine their references, as they are in description and cluster-of-descriptions theories. If I understand Geach correctly, his nominal essence should be understood in terms of a prioricity, not necessity, and thus is quite different from the kind of essence advocated here (perhaps this is part of what he means when he says he is dealing with ‘nominal’, not ‘real’, essences). So ‘Nixon is a man’, ‘Dobbin is a horse’, and the like would be a priori truths. I need not take a position on this view here. But I would briefly mention the following: (1) Even if a sortal is used to disambiguate an ostensive reference, surely it need not be held a priori to be true of that object. Couldn’t Dobbin turn out to belong to a species other than horses (though superficially he looked like a horse), Hesperus to be a planet, rather than a star, or Lot’s guests, even if he names them, to be angels rather than men? Perhaps Geach should stick to more cautious sortais. (2) Waiving the objection in (1), surely there is a substantial gap between premise and conclusion. Few speakers do in fact learn the reference of a given name by ostension; and, even if they picked up the name by a chain of communication leading back to an ostension, why should the sortal allegedly used in the ostension be, in any sense, part of the ‘sense’ of the name for them? No argument is offered here. (An extreme case: A mathematician’s wife overhears her husband muttering the name ‘Nancy’. She wonders whether Nancy, the thing to which her husband referred, is a woman or a Lie group. Why isn’t her use of ‘Nancy’ a case of naming? If it isn’t, the reason is not indefiniteness of her reference.)
I may have spoken too soon. That is what the financial pages said when these lectures were delivered, January, 1970.
Prolegomena to Any Future Metaphysics, Preamble Section 2.b. (Prussian Academy edition, p. 267). My impression of the passage was not changed by a subsequent cursory look at the German, though I can hardly lay claim to any real competence here.
Paul Ziff, Semantic Analysis, Ithaca, Cornell University Press, 1960, pp. 184–185.
Journal of Philosophy59, No. 22 (October 25, 1962), pp. 658–671. In subsequent work on natural kinds and physical properties, which I have not had a chance to see at the time of this writing, Putnam has done further work, which (I gather) has many points of contact with the viewpoint expressed here. As I mentioned in the text, there are some divergencies between Putnam’s approach and mine; Putnam does not base his considerations on the apparatus of necessary versus a priori truths which I invoke. In his entire paper, ‘The Analytic and the Synthetic’, Minnesota Studies in the Philosophy of Science, vol. HI, pp. 358–397, he seems closer to the ‘cluster concept’ theory in some respects, suggesting, for example, that it applies to proper names. I should emphasize again that it was an example of Rogers Albritton which called my attention to this complex of problems, though Albritton probably would not accept the theories I have developed on the basis of the example.
I am presupposing that an analytic truth is one which depends on meanings in the strict sense and therefore is necessary as well as a priori. If statements whose a priori truth is known via the fixing of a reference are counted as analytic, then some analytic truths are contingent; this possibility is excluded in the notion of analyticity adopted here. The ambiguity in the notion of analyticity of course arises from the ambiguity in the usual uses of such terms as ‘definition’ and ‘sense’. I have not attempted to deal with the delicate problems regarding analyticity in these lectures, but I will say that some (though not all) of the cases often adduced to discredit the analytic-synthetic distinction, especially those involving natural phenomena and natural kinds, should be handled in terms of the apparatus fixing a reference invoked here. Note that Kant’s example, ‘gold is a yellow metal’, is not even a priori, and whatever necessity it has is established by scientific investigation; it is thus far from analytic in any sense.
Even better pairs of ringers exist; for example, some pairs of elements of a single column in the periodic table which resemble each other closely but nevertheless are different elements.
Mill, op. cit.
I am not going to give any criterion for what I mean by a ‘pure property’, or Fregean intension. It is hard to find unquestionable examples of what is meant. Yellowness certainly expresses a manifest physical property of an object and, relative to the discussion of gold above, can be regarded as a property in the required sense. Actually, however, it is not without a certain referential element of its own, for on the present view yellowness is picked out and rigidly designated as that external physical property of the object which we sense by means of the visual impression of yellowness. It does in this respect resemble the natural kind terms. The phenomenological quality of the sensation itself, on the other hand, can be regarded as a quale in some pure sense. Perhaps I am rather vague about these questions, but further precision seems unnecessary here.
Of course, there is the question of the relation of the statistical mechanical notion of temperature to, for example, the thermodynamic notion. I wish to leave such questions aside in this discussion.
Some people have been inclined to argue that although certainly we cannot say that sound waves ‘would have been heat’ if they had been felt by the sensation which we feel when we feel heat, the situation is different with respect to a possible phenomenon, not present in the actual world, and distinct from molecular motion. Perhaps, it is suggested, there might be another form of heat other than ‘our heat’, which was not molecular motion; though no actual phenomenon other than molecular motion, such as sound, would qualify. Similar claims have been made for gold and for light. Although I am disinclined to accept these views, they would make relatively little difference to the substance of the present lectures. Someone who is inclined to hold these views can simply replace the terms ‘light’, ‘heat’, ‘pain’, etc., in the examples by ‘our light’, ‘our heat’, ‘our pain’ and the like. I therefore will not take the space to discuss this issue here.
Assuming, of course, that they are all gold, as I say below, some may be fool’s gold. We know in advance, a priori, that it is not the case that the items are typically fool’s gold; and all those items which are actually gold are, of course, essentially gold.
Obviously, there are also artificialities in this whole account. For example, it may be hard to say which items constitute the original sample. Gold may have been discovered independently by various people at various times. I do not feel that any such complications will radically alter the picture.
To understand this dispute, it is especially important to realize that yellowness is not a dispositional property, although it is related to a disposition. Philosophers have often, for want of any other theory of the meaning of the term ‘yellow’, been inclined to regard it as expressing a dispositional property. At the same time, they have been bothered by the ‘gut feeling’ that yellowness is a manifest property, just as much ‘right out there’ as hardness or spherical shape. The proper account, on the present conception is, of course, that the reference of ‘yellowness’ is fixed by the description ‘that (manifest) property of objects which causes them, under normal circumstances, to be seen as yellow (i.e., to be sensed by certain visual impressions)’ ‘yellow’, of course, does not mean ‘tends to produce such and such a sensation’; if we had had different neural structures, if atmospheric conditions had been different, if we had been blind, and so on, then yellow objects would have done no such thing. If one tries to revise the definition of ‘yellow’ to be, ‘tends to produce such and such visual impressions under circumstances C, then one will find that the specification of the circumstances C either circularly involves yellowness or plainly makes the alleged definition into a scientific discovery rather than a synonymy. If we take the ‘fixes a reference’ view, then it is up to the physical scientist to identify the property so marked out in any more fundamental physical terms that he wishes. Some philosophers have argued that such terms as ‘sensation of yellow’, ‘sensation of heat’, ‘sensation of pain’, and the like, could not be in the language unless they were identifiable in terms of external observable phenomena, such as heat, yellowness, and associated human behavior. I think that this question is independent of any view argued in the text.
Some of the statements I myself make above may be loose and inaccurate in this sense. If I say, “Gold might turn out not to be an element,” I speak correctly; ‘might’ here is epistemic and expresses the fact that the evidence does not justify a priori (Cartesian) certainty that gold is an element. I am also strictly correct when I say that the elementhood of gold was discovered a posteriori. If I say, “Gold might have turned out not to be an element,” I seem to mean this metaphysically and my statement is subject to the correction noted in the text.
Of course, the body does exist without the mind and presumably without the person, when the body is a corpse. This consideration, if accepted, would already show that a person and his body are distinct. (See David Wiggins, ‘On Being at the Same Place at the Same Time’, Philosophical Review, Vol. 77 (1968), pp. 90–95’. Similarly, it can be argued that a statue is not the hunk of matter of which it is composed. In the latter case, however, one might say instead that the former is ‘nothing over and above’ the latter; and the same device might be tried for the relation of the person and the body. The difficulties in the text would not then arise in the same form, but analogous difficulties would appear. A theory that a person is nothing over and above his body in the way that a statue is nothing over and above the matter of which it is composed, would have to hold that (necessarily) a person exists if and only if his body exists and has a certain additional physical organization. Such a thesis would be subject to modal difficulties similar to those besetting the ordinary identity thesis, and the same would apply to suggested analogues replacing the identification of mental states with physical states. A further discussion of this matter must be left for another place. Another view which I will not discuss, although I have little tendency to accept it and am not even certain that it has been set out with genuine clarity, is the so-called functional state view of psychological concepts.
Thomas Nagel and Donald Davidson are notable examples. Their views are very interesting, and I wish I could discuss them in further detail. It is doubtful that such philosophers wish to call themselves ‘materialists’. Davidson, in particular, bases his case for his version of the identity theory on the supposed impossibility of correlating psychological properties with physical ones.
For example, David Armstrong, A Materialist Theory of the Mind, London and New York, 1968, see the discussion review by Thomas Nagel, Philosophical Review 79 (1970), pp. 394–403; and David Lewis, ‘An Argument for the Identity Theory’, The Journal of Philosophy, pp. 17–25.
I have been surprised to find that at least one able listener took my use of such terms as ‘correlated with’, ‘corresponding to’, and the like as already begging the question against the identity thesis. The identity thesis, so he said, is not the thesis that pains and pain states are correlated, but rather that they are identical. Thus my entire discussion presupposes the anti-materialist points which I set out to prove. Although I was surprised to hear an objection which concedes so little intelligence to the argument, I have tried especially to avoid the term ‘correlated’ which seems to give rise to the objection. Nevertheless, to obviate misunderstanding, 1 shall explain my usage. Assuming, at least arguendo, that scientific discoveries have turned out so as not to refute materialism from the beginning, both the dualist and the identity theorist agree that there is a correlation or correspondence between mental states and physical states. The dualist holds that the ‘correlation’ relation in question is irreflexive; the identity theorist holds that it is simply a special case of the identity relation. Such terms as ‘correlation’ and ‘correspondence’ can be used neutrally without prejudging which side is correct.
Having expressed these doubts about the identity theory in the text, I should emphasize two things: first, identity theorists have presented positive arguments for their view, which I certainly have not answered here. Some of these arguments seem to me to be weak or based on ideological prejudices, but others strike me as highly compelling, arguments which I am at present unable to answer convincingly. Second, rejection of the identity thesis does not imply acceptance of Cartesian dualism. In fact, my view above that a person could not have come from a different sperm and egg from the ones from which he actually originated implicitly suggests a rejection of the Cartesian picture. If we had a clear idea of the soul or the mind as an independent, subsistent, spiritual entity, why should it have to have any necessary connection with particular material objects such as a particular sperm or a particular egg? A convinced dualist may think that my views on sperms and eggs beg the question against Descartes. I would tend to argue the other way; the fact that it is hard to imagine me coming from a sperm and egg different from my actual origins seems to me to indicate that we have no such clear conception of a soul or self. In any event, Descartes’ notion seems to have been rendered dubious ever since Hume’s critique of the notion of a Cartesian self. I regard the mind-body problem as wide open and extremely confusing.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1972 Saul A. Kripke
About this chapter
Cite this chapter
Kripke, S.A. (1972). Naming and Necessity. In: Davidson, D., Harman, G. (eds) Semantics of Natural Language. Synthese Library, vol 40. Springer, Dordrecht. https://doi.org/10.1007/978-94-010-2557-7_9
Download citation
DOI: https://doi.org/10.1007/978-94-010-2557-7_9
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-277-0310-1
Online ISBN: 978-94-010-2557-7
eBook Packages: Springer Book Archive