Written by Ollie Austin
In a possible world…
A: I want to know, how do names succeed in referring?
A: Well, only yesterday was I speaking with Bertrand when unexpectedly he said to me, and unashamedly, I might add, ‘I believe that I have solved four puzzles about referring.’[i]
A: Yes! For I believe that he is very wrong, though my incredulity disquiets me and I feel quite uncertain about whether I can refer at all.
B: If it is any consolation, I maintain that you did just indeed refer. And successfully so.
A: I doubt that. How so?
B: It is quite simple really. I believe that I understood you perfectly.
A: But you could not have possibly had in mind exactly whom I had in mind.
B: Perhaps not. Is that important?
A: Indeed it is. Imagine one day that I confide to you, frustrated. I say, ‘Bertrand Russell is really a very punctilious man,’ but you retort, offended, ‘He is most certainly not. I find him rather absent-minded.’ It is very clear to me that these cannot be examples of our both referring to the same man, certainly not on his account.
B: I cannot confess that I know him very well. What does he have to say about all of this?
A: Well, according to him there are four puzzles: (1) the problem of apparent reference to non-existents, (2) the problem of negative existentials, (3) Frege’s puzzle about identity, and (4) the problem of substitutivity. I will explain them each first by reference to definite descriptions. The problems of apparent reference to non-existents and negative existentials, we might claim, can be envisaged by way of the baldness and nonexistence of the present King of France. First, that there is no referent ‘the present King of France’ to whom we can refer and of whom we can predicate baldness. This is the problem of apparent reference to non-existents. Second, that because there is no such referent to whom we can refer and of whom we can predicate baldness, a privative claim about such a person cannot be made about ‘the present King of France.’ This is the problem of negative existentials.
B: And so both claims are infelicitous, though they each sound natural, albeit a little strange?
A: I suppose so. Now, Frege identified the third puzzle[ii]; he discovered that two singular terms, denoting the same person, might yield new information not contained in the one or the other. Suppose I say to you ‘Homer Simpson,’ and you respond ‘The husband of my sister, Marge.’ Do we refer to one and the same man?
B: I should hope so.
A: And so an identity statement of the sort, ‘Homer Simpson is the husband of my sister, Marge’ ought to be trivial?
A: And yet it is not. Neither term expresses the same information. This brings us then to the last puzzle. Now, imagine someone else comes along. He says of your brother-in-law, ‘Homer Simpson? Yes, I know him very well. He is the co-author of the article Horizons of a Nuclear Future, published last year in ‘Nuclear World,’ a thoroughly decent man.’
B: You flatter him!
A: But he believes it. He is unaware of course that your brother-in-law, though he does indeed work in nuclear power, and though he has written on occasion, is, more often than not, a ‘drooling moron,[iii] hardly capable of writing a single word, and whose irascibility compels him to violence.’ We cannot now substitute ‘drooling moron, hardly capable of producing a single word,’ and so on, for ‘the co-author of the article Horizons of a Nuclear Future, published last year in ‘Nuclear World’ without also changing its truth-value, since this person believes your brother-in-law to be a thoroughly decent man.[iv]
B: Well I must say: I am thoroughly offended.
A: I apologise if I have struck a nerve.
B: No, not at all. I think your claims about my brother-in-law are quite justified. Rather, I am offended that you believe these four puzzles are problems at all. For what it is worth, I think they are each of them perfectly soluble.
A: I cannot believe that. It would bring our conversation to a very abrupt conclusion. How do you propose to solve these puzzles?
B: Again, it is all really very easy. Now-
A: Oh look. Here comes John.
A: Yes, Mr Searle. He teaches at Berkeley.
John Searle: Hello.
A and B: Good afternoon.
John Searle: I overheard your discussing Horizons of a Nuclear Future, in ‘Nuclear World.’
A: Well, we had really been discussing the four puzzles about referring. In fact, I was just about to hear how these puzzles are actually soluble.
B: I think that can wait. I am interested to hear what John has to say. Are you also familiar with Bertrand’s response? I myself am not.
John Searle: Yes. Though I think that it is quite unsatisfactory, and I will endeavour to rebuke it.
A: Then we are both in agreement. Before you do give us your account, would you kindly explain for my friend here the purport of Bertrand’s response?
B: Please do.
John Searle: Certainly. Bertrand has for a long time been interested in definite descriptions, in the logic of the word, ‘the.’ He believes that it is responsible for each of your four puzzles.
B: I cannot imagine why.
John Searle: Ah, it is really very interesting. Consider a simple subject-predicate sentence, a definite description, ‘The author of Waverly was Scotch.’[v] It is his belief that this sentence abbreviates a series of generalisations, that: (1) ‘At least one person authored Waverly,’ (2) ‘at most one person authored Waverly, and (3) ‘whoever authored Waverly was Scotch.’[vi]
B: Yes, we discussed something similar earlier. It seems to me very opaque, perhaps even esoteric.
John Searle: Perhaps, though I think its bearing will become clear when we apply it to your four puzzles. Suppose then that when we say ‘the present king of France is bald,’ we really mean to say that: (1) ‘at least one person presently kings France,’ (2) ‘at most one person presently kings France,’ and (3) ‘whoever presently kings France is bald.’ Here we deny that our sentence is a subject-predicate sentence since ‘the present king of France’ is not a singular term, and so we circumvent the problem of apparent reference by denying (1), that ‘at least one person presently kings France.’[vii] To clarify, he said to me: ‘If I say ‘Scott was a man’, that is a statement of the form ‘x was a man’ […] But if I say ‘the author of Waverly was a man’, that is not a statement of the form ‘x was a man.’[viii] Nothing here precludes a definite description’s denoting an individual, but the puzzle arises when we analyse definite descriptions as singular terms.
B: That is all well and good, but I cannot imagine what it has to do with our question about names. You see, this discussion came about because my friend here came to me troubled, and said: ‘I want to know, how do names succeed in referring?’ I should like to answer him.
John Searle: Well, what you will understand is that this theory provides Bertrand with the conceptual skeleton on which to build his next claim: ‘The Name Claim.’ He contends that proper names abbreviate descriptions; just as definite descriptions abbreviate various quantificational relations, for Bertrand observed that his four puzzles about definite descriptions recur just as frequently for proper names. Consider another subject-predicate sentence, ‘James Moriarty is bald.’[ix] Here, just as with subject-predicate sentence, ‘the present King of France is bald,’ there is no referent ‘James Moriarty’ to whom we can refer and of whom we can predicate baldness. But by reimagining names as abbreviations of descriptions, the force of the problem of apparent reference to non-existents dissipates. This solution follows a similar trajectory for the remainder of our four puzzles. For Bertrand, then, proper names are semantically equivalent to definite descriptions.[x]
B: I am afraid that I do not follow.
John Searle: To clarify, imagine that I am searching for someone. Bumping into you, I ask, ‘Where might I find the reprehensible Joe Quimby?’ Bemused, you respond, ‘Who?’ I tell you, ‘He is the mayor of Springfield, a horribly obnoxious man.’ Acquainted, you now remark, ‘Oh! Yes. I know him. When last I heard of him, he was being carried away by Chief Wiggum.’
For, here, the proper name ‘Joe Quimby’ does not denote any particular individual, but abbreviates a series of definite descriptions each identifying the man ‘Joe Quimby.’
B: That can hardly do. ‘Joe Quimby’ might stand for any number of innumerable descriptions, none of which I may have had in mind on any particular occasion. If the proper name ‘Joe Quimby’ is identical with the description ‘the mayor of Springfield,’ then it must also be identical with the description ‘the corrupt politician who takes bribes from Fat Tony,’ for both descriptions purport to describe him. Yet this cannot be true, because neither description entails the other. Indeed I may not have had in mind any description.[xi] What then is a proper name’s domain of quantification?
A: Perhaps now you understand my anxiety. I fear that this challenge is insurmountable, and that I will never again refer.
B: You are far too dramatic. I still maintain that these puzzles are each quite soluble; that Bertrand’s response seems to me to miss the mark entirely. For-
John Searle: If I might interject? I think that I may be able to both console your friend, and offer an answer to your question about a proper name’s domain of quantification.
A: I too would be surprised.
John Searle: Since I last spoke with Bertrand, his response and what it professes to demonstrate have perplexed me. I have accordingly devised my own approach, a remedy to the challenge you have presented. I call it my ‘Cluster Theory.’[xii] I have also noticed that for each proper name, we require some corresponding description with which it is equivalent.[xiii] But its consequences are terribly unpalatable. I said to Bertrand: ‘If it is asserted that a proper name is a kind of shorthand description then we ought to be able to present the description in place of the proper name.’[xiv] We must ask ourselves therefore, ‘What are the necessary and sufficient conditions for applying a particular name to a particular object?’[xv] I propose that names do not purport to specify those characteristics of the objects to which they refer, but, as I told Bertrand, that ‘their referring uses nonetheless presuppose that the object to which they purport to refer has certain characteristics.’[xvi]
A: And in what does this presupposition consist, and how would we differentiate an account of this sort from that which Bertrand proposes?
John Searle: It consists in a sufficient but vague and unspecified number of descriptions. Suppose that we asked the citizens of Springfield to describe what they believed to be certain essential characteristics of the name ‘Joe Quimby.’ They would each give uniquely referring descriptions, crucially of a sufficient but vague and unspecified number. To use a proper name then is to presuppose that the object to which it refers has certain characteristics. Importantly, it is neither to assert these descriptions nor individuate them.[xvii] You will see then how we have solved your problem of co-denoting but non-equivalent descriptions. On my ‘Cluster Theory,’ we dispose of the requirement that with each name there must be associated a particular description with which it is identified.
A: I cannot say that I am wholly convinced, though your theory does indeed console me. Do you still maintain that proper names abbreviate descriptions, even if they are many?
John Searle: I do not. I hold instead that proper names signify certain characteristics with which the referent is associated.
A: Well, that is that then. It seems to me that you have done away with the biggest obstacle to Bertrand’s theory, since we need no longer assert that a proper name is associated with a particular description. Perhaps then you did indeed have in mind that same Bertrand, though our descriptions were different. I will therefore happily give my assent!
B: I loathe to be judged churlish, but I am afraid that I cannot accept these amendments. I cannot imagine how you have any more solved these four puzzles than Russell. For these puzzles are only intractable when proper names are held to be rigid designators. Imagined instead as flaccid designators, these puzzles about referring disappear entirely. I hold that all proper names have flaccid uses across possible worlds.[xviii]
A: Possible worlds? That is absurd.
John Searle: This is, I think, a conversation for another time. You will at least concede that we are now more confidently placed to talk about referring than when we began.
A: I have already esteemed your methods, John.
John Searle: Excellent.
B: I hardly think that we have addressed the issue at all, though your account is certainly more amenable than Russell’s. I was never in any doubt that we might succeed in referring.
[i] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, pp.13-15.
[ii] Frege, G., ‘On Sense and Nominatum’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, pp.186-187.
[iii] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.15.
[iv] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.15.
[v] Russell, B., ‘Descriptions’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.-211.
[vi] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.16.
[vii] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.18.
[viii] Russell, B., ‘On Denoting’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.204.
[ix] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.37.
[x] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.37.
[xi] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.41.
[xii] Searle, J., ‘Proper Names’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.249.
[xiii] Lycan, W.G., Philosophy of Language: A Contemporary Introduction. London: Routledge, 2000, p.40.
[xiv] Searle, J., ‘Proper Names’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.251.
[xv] Searle, J., ‘Proper Names’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.251.
[xvi] Searle, J., ‘Proper Names’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.252.
[xvii] Searle, J., ‘Proper Names’. Ed., Martinich, A.P., The Philosophy of Language Third Edition. Oxford: Oxford University Press, 1996, p.252.
[xviii] Heller, D., Wolter, L, On identification and transworld identity in natural language: the case of -ever free relatives, ‘Linguistics and Philosophy’, Vol. 34, No. 2 (2011), pp. 169-199.