Reference and referents¹

1 In discussing the topics of definite descriptions, referring expressions, and proper names, mistakes are made due to a failure to distinguish referring and making a reference, in the ordinary meanings of these terms, from what philosophers call ‘ denoting ’ and ‘ referring Of first importance here is the consideration that it is the users of language who refer and make references and not, except in a derivative sense, the expressions which they use in so doing. Ryle, for example, says, ‘ A descriptive phrase is not a proper name, and the way in which the subject of attributes which it denotes is denoted by it is not in that subject’s being called “the so-and-so”, but in its possessing and being ipso facto the sole possessor of the idiosyncratic attribute which is what the descriptive phrase signifies.’² I do not wish to deny that what Ryle says here is true, in his technical sense of ‘ denote ’. The example is chosen only to bring out how different this sense is from what we ordinarily understand by ‘ referring ’. I might, for example, refer to someone as ‘ the old man with grey hair ’. Still, the phrase ‘ the old man with grey hair ’ does not ‘ signify ’ an ‘ idiosyncratic attribute’, if what is meant by this is an attribute belonging to just one person. It is equally obvious that I might refer to a person as ‘ the so-and-so ’, even though that person did not possess the attribute (idiosyncratic or not) ‘ signified ’ by that phrase. I might, for example, refer to someone as ‘ the old man with grey hair ’, even though that person was not old but prematurely grey. In both cases I would be referring to someone not ‘ denoted ’ (in Ryle’s sense) by the expression used in so doing. But these expressions do not refer to that person, I do. The question ‘To whom does the phrase “ the so-and-so ” refer? ’ is, in general, an odd question. What might be asked is, e.g., ‘ Who is the President of the United States? ’ or ‘ To whom are you referring? ’; not ‘To whom does the phrase “the President of the United States” refer?’

The question ‘to whom (what) does the phrase “the so-and-so” refer?’ is generally odd. It is not always odd. Certainly it sounds odder in some cases than in others. I think one might ask, ‘To what does the phrase “the morning star” refer?’ Or, pointing to a written text, I might ask, ‘To whom is the author referring with the phrase “the most influential man in Lincoln’s cabinet”?’ But, in speaking about referring, philosophers have written as though one might sensibly ask such questions in an unlimited number of cases. What else could have caused Russell to say in ‘ On Denoting’, ‘A phrase may denote ambiguously; e.g., “a man” denotes not many men, but an ambiguous man’?³

It is, of course, perfectly true that one can ask, ‘ To whom does the pronoun “ he ” refer?’, if one is oneself referring to a particular passage in a text, or to something which has just been said. But it does not follow that one can ask this question apart from such a context. Clearly, the question ‘ to whom does “ he ” refer? ’ is a senseless question unless such a context is indicated. The same is true of Russell’s example, ‘ a man’. It is senseless to ask ‘To whom does “a man” refer?’; or (using Russell’s term), ‘Whom does “a man” denote?’ But even when the context is clearly indicated, this question does not always make sense. If, for example, I tell you that I need a wife, you can hardly ask me, ‘To whom are you referring?’

Failure clearly to mark these distinctions leads to confusions about uniqueness of reference. Russell says that a definite description ‘ will only have an application in the event of there being one so-and-so and no more’.⁴ But can I not refer to someone as ‘ the old madman ’, even though he is not mad and more than one man is ? Does my phrase not have ‘application’ to the one to whom I am referring? Certainly, I was speaking of him. What is usually said here is that uniqueness of reference is secured by making the description more determinate, e.g., by saying, ‘the old man who lives next door’. But this attempt to secure uniqueness of reference through in¬ creased determination of the ‘ referring expression ’ is otiose, for what secures uniqueness is the user of the expression and the context in which it is used together with the expression.

We may now notice Ryle’s futile attempt to get uniqueness of reference somehow guaranteed by the words themselves. ‘Tommy Jones is not the same person as the king of England’ means, Ryle says, what is meant by: ‘(i) Somebody, and - of an unspecified circle - one person only is called Tommy Jones; (2) Somebody, and one person only has royal power in England; and (3) No one both is called Tommy Jones and is king of England.’ But surely when I say, ‘Tommy Jones is not king of England ’ I am not claiming that exactly one person of any circle is named ‘Tommy Jones ’. What is indeed necessary, if I am to make a definite assertion, is not that one person only be named ‘Tommy Jones’; but that I be referring to just one person, however many others there may be with the same name as his. It is a mistake to think that the ‘ referring expression ’ itself can secure and guarantee this uniqueness. This is obvious in the case of proper names, for here we cannot appeal to meaning. ‘Tommy Jones’ does not have a meaning, and many people share it. Proper names are usually (rather) common names.

Ryle’s account makes it appear that it is an intrinsic characteristic of certain groups of words that they denote something or other. They possess this characteristic in virtue of their ‘ signifying an idiosyncratic attribute ’. Perhaps he is thinking of such an expression as ‘the oldest American university’. It is a matter of fact that the oldest American university is Harvard. But nothing prevents one from referring to another school (by mistake, or in jest) with these words.

Perhaps Ryle has confused referring to something with referring to it correctly as this or that. I might, for example, refer to L. W. in saying, ‘ He is the president of the bank ’. Still I would have referred to him incorrectly as the president of the bank, because he is not the president of the bank, but the vice president. Some of what Ryle says will be correct if we interpret his comments about denoting as giving an account of what it is to refer to something correctly as such and such. But it is, after all, possible to refer to something incorrectly as such and such, and that is still to refer to it. Furthermore, for one to refer correctly to something as ‘ the such and such ’ it is not necessary that the thing referred to be the sole possessor of the ‘property signified’ by the phrase, though it must certainly have that property. Looked at conversely, we can say that it is not necessary that the property ‘ signified ’ by a phrase of the form ‘ the such and such ’ be ‘ idiosyncratic ’ if one is to refer to something correctly as ‘ the such and such’.

2 The question, ‘To whom (what) does the phrase “the so-and-so” refer?’ is generally odd. But it is not always odd. I am arguing that the sense in which expressions (as opposed to speakers) can be said to refer to things is derivative. I mean by this that the question ‘ To whom (what) does the phrase “the so-and-so ” refer? ’ means the same as the question with regard to some person, ‘To whom (what) is that person referring with the phrase “the so-and-so”?’ Where the question cannot be so rephrased, it cannot be asked at all, e.g., ‘ To whom does the pronoun “ he ” refer ? ’, ‘To whom does the phrase “the old man” refer?’

Much of the philosophical discussion of this topic has assumed that this was not so. Russell says that a denoting phrase is such ‘ solely in virtue of its form ’. Thus we should be able to ask ‘To whom does the phrase “the tallest man in the prison” refer? ’, for the denoting phrase here is of the same form as ‘ The Sultan of Swat ’, and this phrase can be said to refer to someone, viz., Babe Ruth. But the first question cannot be asked. The second question, ‘To whom does the phrase “The Sultan of Swat” refer? ’ does not require a special context and is not the same question as the one which asks with regard to some person, ‘To whom was he referring with that phrase?’ For clearly this last question might receive a different answer than the first. This would occur if the speaker in question had erroneously been referring to Mickey Mantle. So the question ‘To whom does “the so-and-so” refer?’ seems not always to be the same question as the one with regard to some person, ‘To whom was he referring with the phrase “the so-and-so”?’

I am claiming that the counter-examples are only apparent and that the general thesis is still true. There is a class of expressions which (using Strawson’s happy description) have grown capital letters. Some examples are ‘ The Sultan of Swat ’, ‘ The Morning Star ’, ‘ The City of the Angels ’. One can ask, ‘To what city does the phrase “The City of the Angels” refer?’ The answer is ‘Los Angeles’. Such expressions are on their way to becoming names, e.g. ‘ The Beast of Belsen ’. They are what a thing or person is called often and repeatedly, and that is why one can ask to what they refer. Philosophers were perhaps concentrating on such examples as these when they said or implied that the question ‘To whom (what) does “the so-and-so” refer?’ can always be asked. But it cannot.

Perhaps another source of this mistake derives from a confusion between meaning and referring. One can ask both ‘What does this phrase mean?’ and ‘Whom do you mean? ’ Also ‘ I referred to so-and-so ’ and ‘ I meant so-and-so ’ seem very close indeed. But these verbs are radically different, as can be seen from the following considerations. One can ask, ‘ Why did you refer to him? ’, but not, ‘ Why did you mean him ? ’ One can say, ‘ Don’t refer to him! ’, but not ‘ Don’t mean him! ’ ‘ How often did you refer to him?’ is a sensible question, but ‘How often did you mean him?’ is not. One can ask, ‘Why do you refer to him as the such and such?’, but not ‘Why do you mean him as the such and such?’ I can ask why you refer to him at all; but not why you mean him at all. The verb ‘ to mean ’ has non-continuous present tense forms, e.g., ‘I mean you’; but it lacks the present progressive tense form, ‘ I am meaning you ’. The verb ‘ to refer ’ has a present progressive form, ‘ I am referring to you as well as a non-continuous present form, ‘ I refer to Adlai Stevenson’.

What these grammatical considerations show is that referring to someone is an action; meaning someone is not an action. As an action it can be right or wrong for one to perform. Thus it can be wrong of you to refer to someone; but not wrong of you to mean someone. It can be important or necessary that you refer to someone, but not important or necessary that you mean someone. One can intend to refer to someone, but not intend to mean him.

3 In discussions of statements such as ‘Edward VII is the king of England’ it is sometimes said that in making them one is referring to the same person twice. Frege says that the person is referred to in different ways each time. This way of looking at them leads to their interpretation as identities. But consider the following conversation to see how odd it is to talk of referring twice to the same person in such contexts:

A He is the king of England.

B To whom are you referring?

A That man behind the flag.

B How many times did you refer to him?

Referring to some one several times during the course of a speech would be a rather different sort of thing. If I mention a man’s name I would not ordinarily be said to have referred to him in so doing. Using a man’s name is in some ways opposed to referring to him rather than an instance of it.

If we assume that whenever in an assertion something is mentioned by name by a speaker he is referring to that thing certain very paradoxical conclusions can be deduced. It would follow that when I write in my paper ‘ I am not, of course, referring to Ludwig Wittgenstein’ I would be referring to Ludwig Wittgenstein. But if someone were asked to show where in my paper I had referred to Ludwig Wittgenstein it would be absurd for him to point to the statement in which I say, ‘ I am not referring to Ludwig Wittgenstein.’ The same would be true of the statement in which I say, ‘ I am referring to Ludwig Wittgenstein’. In both cases I would have used Wittgenstein’s name. Therefore, to mention someone by name is not necessarily to refer to him. And consider this example. Suppose the porter at Magdalen College asks me whom I am looking for. I answer, ‘ Gilbert Ryle’. Would anyone say I had referred to Gilbert Ryle? But if I say, in the course of a talk, ‘ I am not referring to the most important of present-day philosophers ’, I would then and there be referring to Ludwig Wittgenstein; though in saying as I just did, ‘ I would then and there be referring to Ludwig Wittgenstein ’, I could not be said to have referred to Ludwig Wittgenstein. And this is so notwithstanding the fact that Ludwig Wittgenstein is the most important of present-day philosophers. This, then, is the paradox of reference. In saying ‘ I am referring to Ludwig Wittgenstein ’ I am not referring to Ludwig Wittgenstein.⁵

Strawson has argued correctly that if I claim, e.g., ‘ The king of France is Charles de Gaulle’, what I have said is neither true nor false. The reason for this is not, as he says, that I have failed to refer in saying, ‘ The king of France. . . ’ The reason is that France is not a monarchy and there is no king of France. Just so, and said of a spinster, ‘ Her husband is kind to her ’ is neither true nor false. But a speaker might very well be referring to someone in using these words, for he may think that someone is the husband of the lady (who in fact is a spinster). Still, the statement is neither true nor false, for it presupposes that the lady has a husband, which she has not. This last refutes Strawson’s thesis that if the presupposition of existence is not satisfied the speaker has failed to refer. For here that presupposition is false, but still the speaker has referred to someone, namely, the man mistakenly taken to be her husband.

Of course a man may ‘ fail to refer ’, but not as Strawson uses this expression. For example, in your article you may fail to refer to my article.

4 Referring does not have the omnipresence accorded to it in the philosophical literature. It sounds odd to say that when I say, ‘ Santa Claus lives at the North Pole’ I am referring to Santa Claus, or that when I say ‘The round square does not exist’ I am referring to the round square. Must I be referring to something? Philosophers ask, ‘ How is it possible to refer to something which does not exist? ’ But often the examples produced in which we are supposed to do this (‘ Hamlet was a prince of Denmark ’, ‘ Pegasus was captured by Bellerophon ’, ‘ The golden mountain does not exist’) are such that the question ‘To whom (what) are you referring?’ simply cannot sensibly arise in connection with them. In these cases, anyway, there is nothing to be explained.

How is it possible to make a true statement about a non-existent object? For if a statement is to be about something that thing must exist, otherwise how could the statement mention it, or refer to it ? One cannot refer to, or mention nothing; and if a statement cannot be about nothing it must always be about something. Hence, this ancient line of reasoning concludes, it is not possible to say anything true or false about a non-existent object. It is not even possible to say that it does not exist.

It is this hoary line of argument which, beginning with Plato, has made the topic of referring a problem for philosophers. Still, ancient or not, the reasoning is outrageously bad. Surely here is a case where philosophers really have been seduced and led astray by misleading analogies. I cannot hang a non-existent man. I can only hang a man. To hang a non-existent man is not to do any hanging at all. So by parity of reasoning, to refer to a non-existent man is not to refer at all. Hence, I cannot say anything about a non-existent man. One might as well argue that I cannot hunt for deer in a forest where there are no deer, for that would be to hunt for nothing.

It must have been philosophical reflections of this genre which prompted Wittgenstein to say in his Remarks,

We pay attention to the expressions we use concerning these things; we do not understand them, however, but misinterpret them. When we do philosophy we are like savages, primitive people, who hear the expressions of civilized men, put a false interpretation on them, and then draw queer conclusions from it.⁶

Let us look a bit closer at what it is to talk about things which do not exist. Of course there are a variety of different cases here. If we stick to the kind of case which has figured prominently in philosophy, however, this variety can be reduced. What we now have to consider are characters in fiction such as Mr Pickwick; mythological figures such as Pegasus; legendary figures such as Paul Bunyan, make-believe figures like Santa Claus and fairy-tale figures like Snow White. (And why not add comicstrip figures like Pogo?) And do not these characters really exist? Mr Pickwick really is a character in fiction, Professor Ryle is not. There really is a figure in Greek mythology whose name is ‘ Pegasus but none whose name is ‘ Socrates ’; and there really is a comic-strip character named ‘ Pogo In talking about these characters I may say things which are true and I may also say things which are not. If I say, for example, that Pogo is a talking elephant that is just not true. Neither is Pegasus a duck. In talking about these things there is this matter of getting the facts straight. This is a problem for me; it is not a problem for Dickens or for Walt Kelly. What Dickens says about Mr Pickwick in The Pickwick Papers cannot be false, though it can be not true to character; and in the comic strip Walt Kelly does not say anything about his possum Pogo, for Pogo talks for himself. Still, Pogo could say something about Walt Kelly (or Charles de Gaulle), and that might not be true.

There is, however, another group of cases, and this group has the important characteristic that in talking about its members there is no such thing as getting the facts straight. Here we find Russell’s famous example, the present king of France; and Meinong’s equally famous example of the golden mountain. What are they supposed to be examples of? Well, just things that do not exist. But in saying this we must keep in mind how different they are from Mr Pickwick, Santa Claus, Snow White, etc. Keeping this difference in mind, we can see that though it makes perfectly good sense to ask whether Mr Pickwick ran a bookstore or whether Santa Claus lives at the North Pole; it makes no sense whatever to ask whether the golden mountain is in California. Similarly, though we can ask whether Mr Pickwick was married or not, we cannot sensibly ask whether the present king of France is bald or not.

If the question is ‘ How can we talk about objects which do not exist?’, then it is wrong to use the examples of the golden mountain and the present king of France. These famous philosophical examples, the round square, the golden mountain, are just things we do not talk about (except in telling a story or a fairy tale or something of the kind). Meinong, Russell, and Ryle all puzzle over sentences such as ‘ The gold mountain is in California ’, as though one just had to make up one’s mind whether to put it in the box with all the other true propositions or into the box with the other false propositions. They fail to see that one would only utter it in the course of telling a story or the like. It does not occur in isolation from some such larger context. If it did so occur, if someone were just to come up to us and say, ‘The gold mountain is in California ’, we would not concern ourselves with truth or falsity, but with this man. What is wrong with him ? When the sentence occurs in a fairy tale it would never occur to us to raise the question of its truth. And when we are asked to consider whether it is true or false outside of such a context we can only say that it does not so occur, we just do not say it.

Of course, we may sometimes in error, or by mistake, talk about non-existent things, e.g., Hemingway’s autobiography. So here is one way in which it can occur that we speak of non-existent objects. As a result of a mistake!

5 It is difficult to read the philosophical literature on these topics (the existence of chimeras, the round square, imaginary objects) without arriving at the conclusion ; that much of the difficulty surrounding them stems from an insistence that complex questions be given simple, unqualified, and categorical answers. One asks, ‘ Does : Santa Claus live at the North Pole?’ The philosopher asks it with the assumption ) that there must be just one correct answer, for by the law of the excluded middle, either Santa Claus does or he does not live at the North Pole. Russell says that the correct answer to the question is that it is false that Santa Claus lives at the North Pole, for since he does not exist he does not live anywhere. Meinong would say, I suppose, that it is true that Santa Claus lives at the North Pole, for his so-being (So-sein) is independent of his being (Sein).

But it seems to me obvious that there is no single, categorical, and unqualified answer to our question. Why must there be one? Why are we not allowed first to consider what the person who asked the question wants to know? Surely it is obvious that there are different things which may be at issue. Maybe we are dealing with a child who has just heard the Santa Claus story. Perhaps the child believes in Santa Claus and wants more details about him. Then if we say, ‘ Yes, he lives at the North Pole ’ we encourage the child in his mistaken belief, so from this point of view it is the wrong thing to say. But if the child knows already that Santa Claus is just makebelieve, and wants more details about the Santa Claus story, surely it is perfectly correct to say, ‘Yes, he lives at the North Pole’, for that is the way the story goes. A third case is one in which we do not know whether the child believes in Santa Claus and we do not want to encourage him in his belief, if he does believe in Santa Claus. Here we might say, ‘ No, he does not really live at the North Pole. It is just make-believe.’

Now in view of the obvious complexity of the situation, why should philosophers insist that there be single correct answers here? Not only are there different things one may want to know in asking ‘ Does Santa Claus live at the North Pole? ’, there are different things that may be at issue when one asks, ‘ Does Santa Claus exist? ’, ‘ Is Mr Pickwick an imaginary man? ’ Here are some of the things that one may want to know in asking this last question. One may want to know if Pickwick was a real person or only a character in one of Dickens’ novels. One may know that Pickwick was not a real person, and one may know that he is a character in a novel, but wonder wdiether he is a real person in the novel or (perhaps) the imaginary friend of a child in the novel. Clearly what is correct to say (or better, what is the least misleading thing to say) depends upon the background in which our question arises. Further, there is nothing, in logic, which requires us to give our answer in one short sentence. Why should we not say something as complicated as this: ‘ Pickwick is a real person in the novel, he is not the creation of the imagination of anyone in the novel. Of course he is a creation of the imagination of Dickens. He is not a real person at all.’ We may have to say at least this much in order not to mislead our audience.

In speaking about movies, plays, novels, dreams, legends, superstitions, make-believe, etc., our words may be thought of as occurring within the scope of special ‘ operators ’. Let me explain. Watching the western, I say, ‘ I thought the sheriff would hang the hero, but he didn’t’. The context in which these words are said makes it clear that they are occurring under the ‘ in-the-movie ’ operator. I am telling you of my expectations concerning the course of the movie. I thought that a hanging would take place in the movie, not in the cinema. Similarly, I might say, ‘ Leopold Bloom lived in Dublin’. Is this true or false? Obviously it depends upon whether my words are or are not within the scope of the ‘ in-the-novel ’ operator. It depends whether or not I am talking about (say) James Joyce’s Ulysses. If I am talking about the chief character in Joyce’s Ulysses and if I mean to tell you that according to Joyce this character lived in Dublin, what I say is true. But if I mean to say that Joyce’s fictional character is modelled on a real Dubliner it may not be true. I do not know.

But surely what is being said, and what is being asked, is generally apparent in context, and if it is not a few questions can usually make it clear.

There is no doubt that behind these discussions of Pegasus, chimeras, the round square, there is a kind of ontological anxiety produced by certain pictures. One comes across talk of realms of being, of supersensible worlds occupied by shadowy entities not occupying space and time but nevertheless real. Certain images come to mind. Images of things with shapes and sizes, yet not tangible. We think of them as being ‘ out there ’ and ‘ up there ’, yet nowhere. (It is curious that the picture requires numbers, chimeras, and Santa Claus to be ‘ up there ’, not ‘ down ’ or even ‘ level ’ with us. We picture the numbers standing next to each other like clothes pins on a line. But why should they be ‘up there’ rather than below our feet?)

It is the influence of such pictures, I believe, which Russell felt when he argued that talk of Hamlet is ‘ really ’ talk about Shakespeare. It is difficult even dimly to comprehend what the dispute about whether Pegasus has being, in some sense, can amount to unless one sees the operation of these pictures. One can almost feel the anxiety in this passage from Russell. ‘It is argued, e.g., by Meinong, that we can speak about “the golden mountain”, “the round square”, and so on; we can make true propositions of which these are the subjects; hence they must have some kind of logical being, since otherwise the propositions in which they occur would be meaningless. In such theories, it seems to me, there is a failure of that feeling for reality which ought to be preserved even in the most abstract studies. Logic, I should maintain, must no more admit a unicorn than zoology can; for logic is concerned with the real world just as truly as zoology, though with its more abstract and general features. To say that unicorns have an existence in heraldry, or in literature, or in imagination, is a most pitiful and paltry evasion.’⁷

One wonders just what the issue is between Russell and those against whom these remarks are directed, for surely no one is arguing that Santa Claus is not just make-believe and that Hamlet is not just the creature of Shakespeare’s imagination. How, then, does one’s ‘feeling for reality’ enter here at all? I am unable to explain what is at issue except in terms of the ‘pictures’.

6 It is said to be an astronomical fact of some importance that

(1) The morning star = the evening star.

This was not always known, but the identification was early made by the Greeks. Frege said that it was because the two expressions ‘the morning star’ and ‘the evening star’ had the same reference that (1) was true, and because these two had different senses that (1) was not a trivial thing to say.

Frege’s way of putting the matter seems to invite the objection that the two expressions ‘ the morning star ’ and ‘ the evening star ’ do not refer to the same thing. For the first refers to the planet Venus when seen in the morning before sunrise. The second phrase refers to the same planet when it appears in the heavens after sunset. Do they refer, then, to the same ‘ thing ’ ? Is it, as Carnap says,⁸ a matter of ‘ astronomical fact ’ that they do? One wants to protest that it is a matter of ‘ linguistic fact ’ that they do not.

Perhaps Frege’s view is better put if we think of the two expressions as names, that is, ‘The Morning Star’ and ‘The Evening Star’. Thus Quine,⁹ in repeating Frege’s example but adding capital letters, speaks of the expressions ‘ Evening Star ’ and ‘Morning Star’ as names. Quine would say that what the astronomers had discovered was that

(2)The Morning Star — The Evening Star.

This is better, for (1) implies (or presupposes) what (2) does not, that there is only one star in the sky both in the morning and in the evening. Also, a purist might object that it cannot be taken as ground for (1) that Venus is both the morning star and the evening star. Venus is not a star but a planet. It would be wrong to say that what the astronomers discovered was that the morning planet is the evening planet.

(2) is free from these criticisms, but still the same protest is in order as was made against (1). The name ‘The Morning Star’ does not refer simpliciter to the planet Venus. It does not refer to the planet in the way in which the demonstrative ‘that’ might be used to refer to the planet on some occasion. The names ‘ The Morning Star ’ and ‘ The Evening Star ’ are not that sort of ‘ referring expression ’.

It would be incorrect for me to say to my son as he awakens, ‘ Look to the place where the sun is rising and you will see The Evening Star ’, for that is not what the star is called when seen in the east before sunrise. Again, the proposal that we stay up until we see The Evening Star is quite a different proposal from the proposal that we stay up until we see The Morning Star. In dealing with failures of substitutivity in some ways like these, Frege developed his concept of ‘oblique’ (ungerade) discourse, and Quine has talked about ‘ referential opacity ’. Names in oblique contexts, according to Frege, do not have their ‘ordinary’ referents but an oblique referent which is the same as their ordinary sense. But it would be absurd to suggest that when I tell my boy that if he looks to the East on arising he will see The Evening Star I am not referring to a planet but to a ‘sense’, whatever that might be. Using Quine’s notion of referential opacity, one might suggest that the reason why the proposal to wait up until we see The Evening Star is a different proposal from the proposal to wait up until we see The Morning Star is that here the context is referentially opaque, so that the two names in these contexts do not refer to anything at all. But surely this result is too paradoxical to be taken seriously, and in any case no one has yet told us how to understand the view that a proposal can be referentially opaque.

Under the entry for ‘Venus’ in the Encyclopaedia Britannica we are given the following information: ‘When seen in the western sky in the evenings, i.e., at its eastern elongations, it was called by the ancients “Hesperus”, and when visible in the mornings, i.e., at its western elongations, “Phosphorus”.’ Did the astronomers then discover that

(3) Hesperus = Phosphorus?

In the entry under ‘Hesperus’ in Smith’s Smaller Classical Dictionary we read, ‘ Hesperus, the evening star, son of Astraeus and Eos, of Cephalus and Eos, or of Atlas’. From this, together with (3), we are able to get by Leibniz’s Law

(4) Phosphorus is the evening star.

And avoiding unnecessary complications, let us Interpret this as meaning

(5) Phosphorus is The Evening Star.

Any competent classicist knows that this is not true.

Under the entry on ‘ Phosphorus ’ in Smith’s Smaller Classical Dictionary we find: ‘ Lucifer or Phosphorus (“ bringer of light ”), is the name of the planet Venus, when seen in the morning before sunrise. The same planet was called Hesperus, Vespergo, Vesper, Noctifer, or Nocturnus, when it appeared in the heavens after sunset. Lucifer as a personification is called a son ofAstraeus and Aurora or Eos; of Cephalus and Aurora, or of Atlas.’ So the stars were personified, and it seems to be a matter of mythology that

(6) Hesperus is not Phosphorus.

Then did the astronomers discover that the mythologists were wrong?

Of course (3) is false, and no astronomical research could have established it. What could we make of the contention that the Greeks mistakenly believed that Hesperus was not Phosphorus? According to the Encyclopaedia Britannica (s.v., ‘ Hesperus ’): ‘ the two stars were early identified by the Greeks.’ But once the identification was made, what was left to be mistaken about here?

Could not one mistake The Evening Star for The Morning Star? Certainly one could. This would involve mistaking evening for morning. One could do this. In the morning it is just getting light, and in the evening it is just getting dark. Imagine someone awaking from a sleep induced by a soporific. ‘ But there aren’t two stars, so how could one be mistaken for the other?’

Hence, though it is sometimes made to look as though the Greeks were victims of a mistaken astronomical belief, this is not so. And Quine suggests that the true situation was ‘probably first established by some observant Babylonian’. If that is the case, a knowing Greek would not have said

(7) The Morning Star is not The Evening Star.

unless, of course, he were in the process of teaching his child the use of these words. And, drawing on his unwillingness to say (7) (except in special circumstances when he might want to say just that), we might push him into saying that The Morning Star is The Evening Star, and even that Hesperus is Phosphorus, though now he would begin to feel that these sayings were queer.

The moral is that if we allow ourselves no more apparatus than the apparatus of proper names and descriptions, sense and reference, and the expression ‘ x — y ’ we just cannot give an undistorted account of what the astronomers discovered, or about Hesperus and Phosphorus. Only the logician’s interest in formulas of the kind ‘ x = y ’ could lead him to construct such sentences as ‘ The Morning Star = The Evening Star ’ or ‘ Hesperus = Phosphorus ’. Astronomers and mythologists don’t put it that way.

¹ This paper is reprinted from L. Linsky, Referring, New York, 1967, pp. 116-31.

² ‘ Systematically Misleading Expressions’, reprinted in Essays on Logic and Language, ed.

A. Flew, New York, 1951, p. 23.

³ Logic and Language, p. 41.

⁴ Principia Mathematica, 1, p. 30.

⁵ Philosophical tradition sanctions the production of such paradoxes. I am thinking of Meinong’s paradox about objects, of which it is true to say that no such objects exist; and Frege’s paradox that the concept horse is not a concept.

⁶ Remarks on the Foundations of Mathematics, Oxford, 1956, p. 39

⁷ Introduction to Mathematical Philosophy, Allen and Unwin, London, 1920, p. 169.

⁸ Meaning and Necessity, Chicago, 1947, p. 119.

⁹ From a Logical Point of View, Cambridge, Mass., 1953, p. 21.