In general, however, when we refer to fictional characters, we do it on the basis of the properties most commonly registered by the encyclopedias, and the encyclopedias usually register properties of the type (ii) and (iii), because those are the ones made explicit in the texts and not the ones they presuppose or lead one to conjecture. To talk of properties made explicit means thinking of a fictional text as a musical score: just as a score prescribes the pitch, duration, and often the timbre of sounds, so a story establishes the S-properties and the essential properties of the characters. The fact that a story also supplies accidental properties (the majority of which may be deleted without losing the identity of the character) could be similar to the fact that in order to identify a musical composition, it is not strictly essential for, let’s say, certain differences between forte and fortissimo to be respected, and a given melody can be recognized even though it is not executed con brio, as called for in the score.
I have returned to the analogy with the musical score because I intend to refer to the discussion (in 3.7.7) on formal individuals. On that occasion both a musical composition and a painting or a novel were considered as formal individuals. Now I intend to suggest that fictional characters (inasmuch as they are intersubjectively and encyclopedically identifiable through the S-necessary and essential properties that a text has attributed to them) may be referred to just as one refers to Bach’s Second Suite for Solo Cello. We have said that (above and beyond the practical and theoretical difficulties involved in its recognition on the basis of two or three notes) whoever talks of SC2 intends to refer to that formal individual that, in the impossibility of ascertaining the musical thought in Bach’s head when he composed it, is represented by its score or an execution of it held to be correct and faithful.
In this sense fictional characters are formal individuals to whom we can refer correctly as long as all the properties made explicit by the original text are attributed to them, and on such a basis we can establish that anyone who asserts that Hamlet married Ophelia or that Sherlock Holmes was German is stating a falsehood (or is referring to some other individual who by chance bears the same name).
However, what I have said may be applied to fictional characters insofar as they are recounted by a specific work, which constitutes their score. What can we say about mythical or legendary characters who migrate across various works, occasionally performing different actions, or who simply survive in the mythic imagination without being bound to any one work? One typical example is Little Red Riding Hood, where the variations between the popular tradition and the literary versions are extremely numerous and also involve marginal details (see Pisanty 1993: 4). Let us limit ourselves to dealing with a fundamental difference between Perrault’s version and that of the Brothers Grimm: in the first, the story ends when the wolf, after having devoured the grandmother, devours the little girl too, and the tale ends with a moralizing warning for rash and imprudent young ladies; in the second, the hunter comes on the scene and opens up the beast’s belly to release both the little girl and her grandmother. To whom are we referring when we speak of Little Red Riding Hood? To a little girl who dies or to a little girl who emerges from the belly of a wolf?
I would say that the cases here are two. If someone talks of the resurrection of Little Red Riding Hood (reference to the Grimm score) and the interlocutor has in mind the Perrault score instead, then the interlocutor will ask for extra information; negotiations will continue until an agreement is reached regarding which score is being referred to. Or else the interlocutors are thinking of the popular score, the one that showed itself the stronger in the end, which is less complex than that of the various written versions and which circulates in a given culture as a basic fabula. Thisj fabula is substantially the Grimms’ version, and it is to this popular score that we usually refer (namely, the girl goes into the woods, encounters the wolf, the wolf devours the grandmother, takes on her appearance, devours the child, the hunter frees them both), while details that are important in the cultivated versions (e.g., whether the little girl undresses and gets into bed with her grandmother) or are only marginal (e.g., whether the girl took her grandmother cakes and wine or cakes and butter) are dropped. On this popular basis therefore we refer to Little Red Riding Hood in a contractual way that is defined independently of the detail of whether she took her grandmother wine or butter.
It likewise happens that certain characters from novels, once they are famous, come to be a part of—as they say—the collective imagination, and in terms of the basic fabula they become known even to those who have never read the work in the first place. I think that The Three Musketeers is a typical case. Only those who have read Dumas can take part in trivia games knowing that the nag on which d’Artagnan appears in the first chapter is from Beam, is thirteen to fourteen years old, and was a present from his father. Most of the time, the three musketeers are referred to in terms of a basic fabula (they are daring, they duel with Richelieu’s guards, they perform swashbuckling deeds in order to recover the Queen’s diamonds, etc.). In this basi c fabula not much distinction is usually made between their actions in The Three Musketeers and the actions they perform in Twenty Years After (while I would say that the popular fabula pays no attention to what happens in the less famous The Vicomte de Bragelonne—proof of this being that the infinite series of film versions ignore it). Thus we recognize d’Artagnan or Porthos even in film versions where events occur that do not happen in the novels of Dumas, and we are not disturbed by this, as we would be if someone were to tell us that Madame Bovary cheerfully divorced Charles and lived happily ever after.
In all these cases it is a matter of negotiating the score to be referred to (a specific work or a fabula deposited in the collective imagination), and, afterward, the reference occurs without ambiguity. So much so that, in the case of trivia games, one may hear protests of the type: «Look here, the daughter of Milady that you are referring to appears in the film! In Twenty Years After she is Milady’s son!»
Ultimately, in such cases, the possible world in question is negotiated. And that agreement is not always reached may be because of the number of possible worlds at issue, not because it is impossible to fix the reference in a possible world negotiated with precision.25
Ich Liebe Dich
Anyone who maintains that a pronoun in the first person singular is identified with the person speaking—without the mediation of an agreement regarding its own content—ought to explain what happens when a foreigner, whose language is unknown, says Ich liebe Dich. The objection that this is not a case of failed reference but simply of linguistic incompetence, is self-defeating: in point of fact I am saying that in order to understand the reference I must know not only the meaning of a verb such as liebe but also the meaning of the two pronouns—otherwise that declaration of love will end up as a case of infelicitous reference (and never was adjective more appropriate).
We had begun by taking as implicit and almost obvious the fact that, in order to use terms in acts of reference, it was first necessary to know their meaning. As we went on, we realized that, at least in part, we can understand acts of reference even without knowing the meaning of the term. Then we had to conclude that there are no «white boxes» without at least a label, that meaning creeps back in everywhere, and that finally we cannot have a reference crowned by success unless we first agree on the meaning of the terms, and only at that point can we proceed to negotiate with regard to the individual we intend to refer to. Let us conclude with some observations regarding the importance of a NC and of a successive negotiation, even for those terms that seem, so to speak, to acquire life, to make sense only when directly attached to an individual—and, when detached from them, to float in a fog of nonsense.
I am always puzzled by the fact that some people maintain that the indexical terms (the ones usually accompanied by a gesture, such as this or that), the deictics (relative in the context to the speaker and his spatiotemporal position, such as yesterday, now, soon, not far from here), not to mention the personal pronouns, designate directly without any mediation on the part of any possible meaning they might have. I tried to show in Eco (1975, 2.11.5) how even these types of signs must be contained in their meaning if they are to be applied in acts of reference; but I always find someone who denies this because of the simple fact that the instructions for understanding how one can use cat to refer to cats are different from the instructions for understanding how one can use I or this to refer to either the person who emitted the
