However, in Aristotelian semantics, unlike standard semantics, there is no fundamental distinction between names and predicables. Both proper and common names lie in the same sort of relation to an object, a relation which the medieval semanticists called suppositio. But a proper name is proper to just one object, so proper name propositions are universal as well as existential. “Socrates is a philosopher” states that at least one person is (identical with) Socrates, and that every such person is a philosopher. Hence, just as a common name like “planet” can be empty if there are no planets, so a proper name can be empty yet function perfectly well in a proposition. “Bilbo Baggins is a hobbit” states that at least one person is (identical with) Bilbo, and that every such person is a philosopher and so, while it states something perfectly coherent, is false. On the Aristotelian theory of the proposition,21 a subject-predicate sentence containing a proper name asserts rather than presupposes the existence of the proper name’s bearer, and asserts (or denies) that the predicate applies to it. Thus, there are two causes of the sentence being false: (i) the predicate does not apply to the bearer (or does apply, if the sentence is negative), (ii) there is no bearer at all. There are also two forms of negative: wide scope, where the negation applies across the whole sentence, and narrow scope where the negation applies to the predicate only. Thus, in Aristotelian semantics, sentences with empty names are false, rather than lacking a truth value.
In summary, a singular term signifies by telling us which individual satisfies the predicate of a proposition. In the proposition,22 “Moses is a prophet,” the proper name “Moses” tells us which object is said to be a prophet, or which individual satisfies the predicate “is a prophet.” Any theory of reference needs to explain how this is possible. According to the standard theory, this is achieved by means of a semantic relation between the term and the object, that is, a semantic relation between a linguistic item, a word such as “Moses,” and an extralinguistic item, namely Moses himself. The standard theory is the target of this book.
Problems with the Standard Theory
There are a number of well-known problems with the standard theory, and an extensive literature has been devoted to engaging with them. I summarize the main difficulties as follows.
(1) The theory leads to the absurdity that objects, including large planetary masses, are actually a part of our thought. We use language to signify our thoughts, with the aim that others can understand or grasp what we have said. What is signified is what is understood, as the medieval philosophers put it,23 and what is understood is the thought the speaker has expressed. But if a proper name signifies its bearer, the bearer must somehow be a part of the thought expressed. This is absurd. I can express the thought that Jupiter is a planet, but how can Jupiter, with its massive gravitational field and poisonous atmosphere, be literally a part of my thought?24
(2) The theory provides no coherent explanation of how we establish a connection between names and their bearers. Mill says (A System of Logic, I. ii. 5, see also I. v. 2.) that they are simply marks for objects, giving the example of a chalk mark upon a door, but, perhaps seeing how this fails to explain how a proper name can be a mark of something that is not in front of us, or which has long since ceased to exist, he says that this is by analogy only, and that the mark is upon our idea of the object. “A proper name is but an unmeaning mark which we connect in our minds with the idea of the object, in order that whenever the mark meets our eyes or occurs to our thoughts, we may think of that object” (my emphasis). He does not explain how a mark can be meaningless, yet be connected in our mind with the idea of an object, nor does he explain what the idea of an individual object is. He says elsewhere that our concept of Caesar is “the presentation in imagination of the individual Caesar as such,”25 but this does not help much.
(3) The theory has difficulty in explaining how we frequently use names that we know to be empty, for example in fiction, and how it is possible that many names that we believe to have a bearer may possibly not have a bearer, for example “Moses.” According to the theory, the meaning of a proper name is the bearer itself, so a sentence containing an empty name cannot have a meaning. But the Torah appears to be meaningful, whether or not Moses existed, as is a work of acknowledged fiction such as The Lord of the Rings. Furthermore, it requires that a name must have a bearer, which seems absurd. “Moses is a prophet” says of someone (Moses) that he is a prophet, “Moses is not a prophet” says of the same person that he is not a prophet. One or the other must be true, so on the standard theory, both propositions require us to say something of Moses, which we can’t do unless he exists. How then do we deal with the possibility that Moses does not exist? Indeed, how do we deal with the possibility that God does not exist? When the atheist denies “God exists,” the name “God” must signify precisely what the fideist asserts by the same sentence. On the standard theory, this seems impossible.
(4) The standard theory suggests that different proper names for the same bearer could be substituted without changing the meaning of a sentence. Thus “Cicero is Tully” has the same meaning as “Cicero is Cicero,” given that “Cicero” and “Tully” have the same bearer. Yet no one would disbelieve “Cicero is Cicero,” for it expresses a logical truth, while someone might not believe “Cicero is Tully.” This suggests the two names have a different meaning, yet the standard theory says they have the same meaning.
Though these are not the only problems, they are recognized as the main ones.
The Intralinguistic Thesis
The intralinguistic thesis defended here consists of three connected claims. The first claim, the co-reference thesis, is that there is a phenomenon I call signified or guaranteed co-reference, where it is clear simply from the meaning of two statements that if one statement asserts (or denies) that some thing is such and such, then the other statement asserts (or denies) that the same thing is so and so. The paradigm is pronominal back-reference. It is part of our understanding of pronoun use that if “Herod realized that he had been outwitted, and he was furious” is true, then the first part of the sentence says that someone had been outwitted, and the second part says that the same person was furious. When I talk about “co-reference,” I shall always mean this form of anaphoric co-reference. Pronouns are a paradigm, but clearly different tokens of the same proper name can co-refer in the same way, for example, the first and second occurrences of “God” at the beginning of the book of Genesis.
In the beginning God (‘ĕ-lō-hîm, ὁ θεὸς, Deus) created the heavens and the earth. Now the earth was formless and empty, darkness was over the surface of the deep, and the Spirit of God (‘ĕ-lō-hîm, θεοῦ, Dei) was hovering over the waters.
I claim that there is nothing philosophically difficult about explaining co-reference. That is not to say that the explanation, which is a scientific and technical matter, is not complex and difficult, but rather that it is a task for computational linguistics, not philosophy. It is non-philosophically difficult to explain the exact rules by which we determine co-reference in all cases, which I will discuss in the next chapter, but it seems clear that in the natural and obvious reading of the passage given earlier, the two tokens of “God” have a common referent, if they have a referent at all, and so there is no philosophical difficulty. The philosophical difficulties are whether co-reference implies reference (I argue that it does), whether reference implies reference to something (I argue that it does), and whether, if so, reference to something implies that there is (or there exists) something such that it is referred to (I argue that it does not, but this raises some difficult questions that I defer until chapter 7).
The second claim, the reference thesis,26 is that the semantic value of a proper name consists solely in its anaphoric co-reference with its antecedents in a chain of co-referring terms, and that the truth of a reference statement depends upon such co-reference, even if no referent exists.
As stated earlier, a reference statement contains a mentioning term, such as “second word of the Quran,” a mentioned term, a token available in some antecedent text or utterance, and a used term. For example:
• In the beginning God created the heavens and the earth.
•