I intend to exclude a “broad” use of the term,1 and I think it would be appropriate (also in the light of the previous essays) to limit the notion of referring to what is perhaps more properly describable as cases of designation, that is, to utterances that mention particular individuals, groups of individuals, specific facts or sequences of facts, in specific times and places. From now on I shall also be using the generic notion of “individual” for identifiable spatiotemporal segments, such as 25 April 1945, and I shall hold to the golden decision by which nominantur singularia sed universalia significantur.
I refer the reader to Eco (1989) for the extremely tortuous story of terms such as denotatio and designatio, which have taken on different senses in the course of the centuries, but I think we can accept what has now become the established custom whereby general terms “denote” properties of class or genera, while singular terms or expressions that circumscribe precise portions of space-time “designate” individuals (see, e.g., Quine 1955: 32–33).
I maintain therefore that we perform acts of reference by using designative sentences like Look at that platypus, Go fetch me the stuffed platypus I left on the table, The platypus in Sydney Zoo is dead, while I maintain that sentences like Platypuses are mammals or Platypuses lay eggs do not refer to individuals but assert some properties that are attributed to genera, species, or classes of individuals. To return to the computer example I gave in 4.2,1 am talking not so much about platypuses as about the way in which our directory tree (or that of the zoologist) is organized. We are not referring to any individual or group of individuals but reasserting a cultural rule, making a semiotic and nonfactual judgment,2 reiterating the way in which our culture has defined a concept. Defining a concept means elaborating a unit of content, which corresponds in fact to the meaning, or to part of the meaning, of the corresponding term. Saying that “one refers” to meanings is at best a bizarre way of using the word referring.
If instead I say In 1884, Caldwell saw a platypus while it was laying eggs, I am referring to an individual x (Caldwell), who at the time y (1884) examined an individual platypus (which one I don’t know, but he did, and it was certainly that platypus and not another, and I imagine it was a female) to discover that it laid ovoidal objects S1, s2 … sn (I don’t know how many, but he certainly knew, and the assertion refers to those objects and not to others).
While some authors hold that there are cases of reference to essences, which I shall call quidditates, here I should like to deal only with the designation of haecceitates. Naturally I mean quidditas in its Scholastic sense, as the essence itself seen as knowable and definable. To quote Aquinas, who however was referring in his turn to the words of Averroes (De ente et essentia III), “Socrates nihil aliud est quam animalitas et rationalitas, quae sunt quidditas ejus.” In this context I am insisting on the fact that one can designate Socrates but not his quiddity, and I harbor doubts regarding the legitimacy of saying that we refer to the quiddity of Socrates.
By bringing into play the concept of haecceitas (Scotist and not Thomist), I am calling into question the notion that Socrates is nihil aliud than his quidditas. And as a matter of fact Aquinas was well aware that, to talk of Socrates as an individual, it was necessary to appeal to a principium individuationis, which was the matter signata quantitate. Since my purpose here is neither to teach the history of medieval philosophy nor to profess neo-Thomism or neo-Scotism, I shall make free use of the notion of haecceitas as an unrepeatable characteristic of individuals (whether it depends on the matter signata quantitate or any other principle of individuation—as, for example, a genetic inheritance, or registry office records).
I assume the notion of individual in its most intuitive sense, the way we use it in everyday speech. Usually we think not only that there are unrepeatable objects of which no replica or double is conceivable (such as my daughter or the city of Grenoble) but that even in the case of groups of objects in which each is the double of the other (such as the sheets in a ream of paper) it is always possible to choose one of those sheets and decide that, although it has all the properties of the others, it is nonetheless that sheet, even though the only mark of individuality I can allow it is that it is the sheet I am holding in that moment. But that sheet is so individual that, if I burn it, I have burned that one and not another.
It seems to me that the medieval notion of materia signata quantitate is no different from the idea of the principle of individuation expressed, for example, by Kripke (1972: 350): “If a material object has its origin in a certain hunk of matter, then it could not have its origin in any other matter.” This idea that the individual possesses a haecceitas still has nothing to do with the idea that man or water (in general) has an essence, even though in current causal theories of reference these two problems often appear together. Which is a good reason, in itself, for distinguishing between designation (of individuals) and denotation (of genera).
However, I did specify that I intend to use referring not only for the designation of individuals (in the broadest sense of the term, and so even 25 April 1945 is an individuable segment of space-time, and the assassination of Julius Caesar is an individually punctual fact) but also for groups of individuals. By “groups of individuals to which we can refer” (also including generic spatiotemporal segments, such as the thirties) we must understand a set of individuals that has either been counted, or was countable, or might one day be countable (so that every single individual could be individuated).
References to the first victim of the Second World War or to the first men to settle in Australia are certainly very vague: but in using them, we nevertheless presume that it may be theoretically possible one day (or that it may have been possible in the past) to ascertain who the individuals in question were, if nothing else because of the fact that they certainly existed.
Deciding whether a sentence designates individuals or classes depends not on its grammatical form (on the basis of which one can construct an infinite number of bold examples and counterexamples without ever solving the problem definitively) but on the intention of the senders and on the assumptions of the receivers. Therefore a first contract is necessary to decide whether the sentence has a referential function or not.
Sometimes discriminating is very simple: This stick is one meter long certainly designates a certain individual stick, while One meter equals 3.2802 feet expresses a law or a convention. But other cases require more thought. If Herod, before the birth of Jesus, had said to Herodias that he hated all babies, she probably would have agreed about the fact that Herod was not referring to some particular babies but expressing his dislike of babies in general. But when Herod ordered his cutthroats to kill all the babies in Galilee, by his order he intended to designate all the babies born that year in a precise place, one by one (apart from anything else, they were identifiable thanks precisely to the census that had just been made).3
But there is a point that needs to be made clear, even though it should have been clear since the days of Plato and Aristotle. Isolated terms assert nothing (at best, they have a meaning): what is true and false is said only in the sentence, or in the corresponding proposition. Now, I am not saying that referring is the same thing as saying what is true and false (we shall see that acts of reference can be made even when we have not decided if what we are referring to is in fact the case or not), but without a doubt, if we always refer only to individuals, we refer to states of a world (any world).
And to do this we need to articulate a sentence. If I say cat, I am not referring to anything. I can refer to one cat only, or to some cats localized or localizable in time and space. On the other hand, when people say we can refer to generalia, they are suggesting that referring is something we do with isolated terms. I often happen to hear otherwise entirely respectable persons stating that the word cat refers to cats, or to the essence of cats. For the reasons given previously, this strikes me as misleading, so I shall refrain from putting the problem that way.
The word cat always means or denotes, if you will, the essence of cat (or the NC, or the corresponding MC) in all circumstances, outside all contexts, and therefore its signifying or denotative power belongs to the lexical type. The same word designates a given cat only in the context of a sentence that has been uttered and that contains specifications of time and place, and therefore the function of designation is performed by the token. The sentence type Cats are mammals expresses a thought, in whatever context it appears, even if it
