20 Here it is not enough to object that this is a matter of representation and not of referring. Apart from the fact that this would come to contradict the fairly established opinion that we can use an image of something to refer to something (think of a news photo that to all effects and purposes constitutes news), the paradoxes of inconceivability—craftily excluded from the phenomenology of reference—would resurface in the phenomenology of representation, and we would have gained nothing.
21 For a good compendium of the various arguments, see Salmon (1981, Appendix 1).
22 On the other hand, the same argument could also be applied to countries with a stabler history, France included, and it is not easy to say what the expression United States of America refers to if one asks oneself whether it was said before or after the purchase of Louisiana or Alaska.
23 The most recent is that put forward by Santambrogio 1992, for whom fictional characters are similar to «general objects.»
24 While properties of type (iv) can seem of little importance for an encyclopedic definition of the fictional character, consider this curious news item I found in La Repubblica of 1 September 1985 (since the same piece of news appeared in what was only a slightly different version in the Corriere della Sera of the same day, it can be supposed that it came from a news agency): «The false death notice published by The Times was the revenge of a jealous woman and not a message in code for some spy, reveals The Sun. The death notice announced the sudden death in Cornwall of Mark, Timothy, and James, ‘favourite sons’ of a German countess. Rita Colman, a London magistrate, has admitted to having had the text published at the request of countess Margareth von Hessen, the mother of the three boys. The paper has now discovered that Rita Colman’s present husband divorced from countess von Hessen five years ago, and is the father of Mark, Timothy, and James. The three boys are alive and kicking and one of them, Mark, is in fact holidaying in Cornwall, where he was contacted by The Sun. ‘Whoever is behind this macabre joke is the same person who attempted to blacken my mother’s name two years ago,’ commented the young man. In 1983 an English newspaper had published a false news item according to which an Anglican minister, Robert Parker, was on the point of abandoning wife and career to run off with the countess von Hessen. Since yesterday Rita Colman has been untraceable in London: she has left with her husband for a holiday in Devon.» Note that this text specifically names those involved in the affair, it links them reciprocally through S-necessary relationships, and attributes to them both «Registry Office» properties and fairly precise actions. However, if we thought this was a story, we would fall prey to much puzzlement. There is certainly a Rita Colman who has admitted to having had the false news item p published, but why did she have it published? On the request of the countess von Hessen, her husband’s ex-wife, we are told. But since the countess knows that the three boys are alive, why did she induce Colman to have the story published? To terrify her ex-husband? And why did Colman accept, given that the ex-husband is now hers, and that she clearly wants him in good spirits if they are to holiday together in Devon? To please the countess? But why, if, according to the insinuations of one of the sons, Colman bears no love for the countess and on the contrary had spread a false story about her alleged affair with an Anglican vicar, in order to blacken her name? If this news item were a story, we would not be able to paraphrase it in a sensible way, precisely because it confuses our ideas about properties of the type (iv). Or we might take it as the beginning of an affair whose mysteries will have to be cleared up later. Naturally the text is confusing as a news item too, but at this point it suffices to think that the agency editor was an incompetent bungler or that the Italian press had mistranslated an English text, and there’s an end to it.
25 Semprini (1997) devotes a paragraph to the conditions for the recognition of legendary characters from the history of comics, and shows they are identifiable by name, by pronounced physiognomic features, by unmistakable clothing, by civil status, by a series of specific skills, and by various other details (typical phrases, sounds that constantly accompany some canonical gestures, etc.). Are there many real people for whom we have such detailed instructions for recognition?
26 An indexical or deictic term has a meaning independent of context and circumstances. But it is amid circumstances that the way it is to be used for referring must be negotiated. Ducrot (1995: 309) gives an example that reminds us of the uncertainties of Quine’s explorer faced with the native’s gavagai. «This or that, even taking the gesture of designation into account, are not enough to delimit an object. How can I know that what someone shows me on the table is the book in its entirety, or its dust jacket, or its colour, or the contrast between its colour and that of the table, or the particular impression it gives me in this moment? A noun, possibly implicit, is necessary to complete the act of reference.»
27 See in Eco (1979, 1.3) the analysis of two contextual uses of instead.
Chapter Six / Iconism and Hypoicon
1 For a review of the debate, see Calabrese 1985; Fabbrichesi 1983; Bettetini 1996 (1.3 and II.1.1.)
2 Fabbrichesi (1983) puts forward the suggestion that that debate did not die a natural death, because semiotics refused to reflect «philosophically» on the concept of likeness, and this likeness to be explained was not the correspondence between two objects (let’s say a drawing and its original) but Peircean Firstness, as an internal difference, which does not distinguish concrete objects but prepares their individuation and constitution (1983: 109).
3 This distinction is not homologous with that, to use the words of Dennet (1978, III, 10), between iconophiles and iconophobes in the cognitive sciences. I would say that having distinguished (i) the iconic value of knowledge and (ii) the nature of the hypoicon, Dennet’s objection is internal to point (i). In any case, according to the various registers already mentioned, the number of the iconoclasts usually includes Goodman, Gombrich, most of the followers of Greimas, the Liège Group, and even psychologists such as Gregory, while among the iconists one might mention the early Barthes and the early Metz, Gibson, the early Wittgenstein, and Maldonado.
4 I am justly accused of this by Leo Fabbrichesi (1983: 3), even though he perhaps underestimates how the problem resurfaced in A Theory of Semiotics apropos of the «inventions,» and does not consider (owing to the publication date) the way in which I partly tried to retable it in the essay on mirrors (1985).
5 Sonesson, for example, is one who reproves me for having treated only of visual iconism: but, in those very years, I published in VS two essays by Osmond Smith (1972, 1973) on musical iconism, and I mentioned for example the experience of syntactical iconism. However, it is also true that on at least two occasions I wrote that to talk of iconism for Peirce’s existential graphs was pure metaphor, because they do not reproduce morphological and spatial relations. A sign that in that cultural climate, he who uttered the word «icon» was already naturally anchored to the pictorial universe.
6 There is an extremely wide-ranging debate on this subject. On the one hand, there are experiments that show how even animals recognize images (starting with the legend of Zeuxis); on the other, there are ethnographic reports that show us a «primitive» (in any case, someone who has no experience of photographic images or even portraits) toying with a photograph of a known person and manifesting perplexity, anxiety, or even an absolute lack of interest. These are almost always insufficiently tested experiments: in certain cases what strikes the primitive is the offer of a piece of paper, an object unknown to him, whereas if the image is printed on a piece of cloth, he will approach it with greater confidence. In other cases, the problem is the poor quality of the image. In others, the fact that the primitive is still puzzled means not that he has not recognized the subject portrayed but that he simply cannot understand how the features of a known person can appear as if by magic on a piece of paper. In others again, it is evidently a matter of badly formulated questions, which recall the misunderstandings in Quine’s radical translation.
7 Otherwise
