Saussure and Hjelmslev spoke in fact of signs ruled by ratio dijficilis (Eco 1976, 3.4.9) where the expression maps, according to preestab-lished projection rules, some features of the corresponding content. In this sense one can call symbols those used by algebra and formal logic, at least insofar as their syntactic structure is concerned. They are such be-cause every transformation performed upon the syntactical arrangement of the expression mirrors a possible rearrangement in the structure of their content.
If, on a geographical map, one alters the borderline be-tween France and Germany, one can forecast what would happen if in a possible world (the new content corresponding to the manipulated ex-pression) the geopolitical definition of both countries were different. An algebraic formula and a map are diagrams. That is why in electro technics Seinmetz and Kennelly (following Helmholtz) called ‘symbolic’ the method postulating biunivocal correspondence between the emsemble of sinusoidal functions of the same frequency (which incidentally are expressed by mere conventional and by no means ‘analogical’ devices) and the ensemble of points upon an Arnauld-Gauss plane of rotating vectors. The rotation of a vector is a diagram that implies different sinusoidal functions.
It is, however, clear that there is a difference between diagrams and other phenomena labeled as symbols. Diagrams are based on precisely coded transformational and projective rules, in the same way in which in a musical score the ‘symbolic’ relation between rising points on the stave (spatial height) and frequency increments (phonic height) are ruled by a precise proportional criterion. On the contrary, many so-called symbols are characterized by the vagueness of their content and by the fact that the correlation is not precoded but invented at the same moment in which the expression is produced.
In Hjelmslev’s definition the category of symbols encompasses both phenomena, without acknowledging the radical difference between the way in which the Christ of Thorwaldsen is a symbol for compassion and the way in which a move on the chessboard has a symbolic nature. A different chess move would imply different interpretations of the further course of the game, whereas we do not know how many manipulations the Christ of Thorwaldsen should undertake in order to stand for something other than compassion.
Moreover, a diagram such as the map of a subway is certainly ruled by ratio dijficilis, but it is neither vague nor indirect: its meaning is a ‘literal’ one; one can extrapolate from one’s operations upon the map a precise oossible state of affairs. It could not be said that this possible state of affairs is a sort of ‘second’ sense that the map conveys. In the same way as one can interpret the word father by inferring that if there is a father there should be either a son or a daughter, thus, if one detects on the map that, for reaching the node С from the node A, one must pass through the node 5, one can infer that, if A and С were tied by a direct connection, the В would be avoided. In both cases the word and the map tell what they tell as soon as they are correctly interpreted according to given cultural criteria.
Rather different, on the contrary, is the image of the serpent biting its own tail. It is defined as a symbol because there is the strong feeling that it not only represents a snake in an unusual position but that it also aims at communicating something more.
4.3.2. Symbols as tropes
4.3.2.1. Oneiric symbols.
In The Interpretation of Dreams (1899), Freud speaks of oneiric symbols. Dreams convey images which stand for some-thing else, and Freud is interested in establishing how a “latent con-tent” is organized by the oneiric labor into the form of a “manifest con-tent.” The latent content is transformed by the dream distortion (ibid., 4), and the dream is the disguised fulfillment of repressed wishes. Freud does not interpret (as the ancient oneiromancy used to do) dreams as organic allegories.
Allegories do have a logic, whereas dreams do not. The psychoanalytic interpretation does not work upon organic oneiric discourses but upon fragments and their idiosyncratic mechanisms of substitution. Dreams work through condensation and displacement, and (even though Freud does not say it explicitly), since they do not have a Jogic, they have a rhetoric. Condensation and displacement are modal-ities of tropic substitution.
In the dream of the botanical monography (1899, 6a) the botanic sym-bol condenses Gartner, Flora, the forgotten flowers, the flowers loved by the author’s wife, a university exam: “Each of the elements of the dream’s content turns out to have been ‘overdetermined’ — to have been represented in the dream-thoughts many times over” (1899; Eng. tr., 4, P. 283).
Freud knows that the oneiric image is correlated to its content by a sort of ratio difficilis, since it displays certain features that in some way map equivalent features of the latent content. But, as happens in all cases of ratio difftcilis, the mapping relationship takes place between selected features of the expression and selected feature of the content.
To decide which properties have to be selected, that is, which properties are co-textually pertinent, is exactly the typical labor performed by dreams according to certain requirements of plasticity, immediacy, repre-sentability (1899, 6d).
Freud knows that oneiric symbols are not ‘stenographic’ signs endowed with a preestablished meaning; however, he tries to anchor these expressions to an interpretable content. To find such an anchor-age, Freud distinguishes between those oneiric symbols produced by idiosyncratic reasons, which must be interpreted by using the patient’s associations as their idiolectal encyclopedia, and those whose symbolism “is not peculiar to dreams, but is characteristic of unconscious ideation, in particular among the people, and it is to be found in folklore, and in popular myths, legends, linguistic idioms, proverbial wisdom, and cur-rent jokes” (1899; Eng. tr., 5, p. 351). It is true that every dreamer shows a remarkable plasticity in employing the most disparate images for symbolic purposes, but Freud tries repeatedly (see the various editions of the book, 1909, 1911, 1919) to find out a symbolic code so as to explain the intersubjective (or cultural) meaning of umbrellas, sticks, railway travels, staircases, and so on.
To look for an oneiric code means to touch on the hypothesis of a collective unconscious, as Jung will do; but Freud understands that in doing so one risks going backward, to the very sources of human mental activity, where there will no longer be a code. On the other hand, a code is indispensable in order to speak intersubjectively of a semantics of dreams beyond the idiosyncratic attitudes of the dreaming subjects. Thus Freud links the decoding of oneiric symbols to verbal puns, and in doing so he suggests that the knowledge of linguistic mechanisms can help one to understand the oneiric strategies of condensation and dis-placement. (The Lacanian decision to anchor the order of the imaginary to the order of the symbolic must be understood in this sense.) Freud suggests that the code can be reconstructed and that it is neither uni-versal nor innate, but is historical, semiotic, and depends on the cultural encyclopedia of the dreamer.
This assumption is not, however, so unambiguous. The dream must be interpreted according to a linguistic and cultural competence (that is, according to a competence which is external to the world of dreams); nevertheless, every oneiric image can be polysemous, as Freud explicitly says, and must be referred to the idiolect of the dreamer as well as to the whole dream as its co-text. Notwithstanding these perplexities and con-tradictions, Freud is undoubtedly looking for ‘correct’ interpretations of dreams, and in this sense his oneiric symbols are not constitutively vague.
Freud has thus elaborated on oneiric rhetoric, with its own rules for generating and for interpreting images.
4.3.2.2. Metaphors and other tropes.
Must we also exclude, from the rank of properly called symbols, metaphors, allegories, and other tropes? This is not to be taken for granted, because in many theories of literary crit-icism this distinction is not at all clear. However, even though they are ‘open’ to various interpretations, metaphors are always governed by rhetorical rules and controlled by their co-texts.
In any case, there is a clear-cut test for distinguishing a metaphor from a symbol: a trope cannot be taken ‘literally’ without violating a pragmatic maxim according to which a discourse is supposed to tell the truth; it must be interpreted as a figure of speech, since otherwise it would ap-pear senseless or blatantly false. On the contrary, the instances of the symbolic mode do suggest a second sense, but could also be taken liter-ally without jeopardizing the communicational intercourse. (I shall elabo-rate on this point in 4.5.)
More evident is the coded nature of allegories. They can be inter-preted according to complementary senses (see in 4.4.3 the medieval theory of the four senses of the Scriptures), but these senses are never vague or indefinite.
A radical difference between symbol and allegory has been definitely established by Romantic theorists, who have, however, dangerously identified the symbolic with the aesthetic.
4.3.3. The Romantic symbol as an aesthetic text
Originally, a symbol was produced by the mutual relationship of two pieces of a coin destined to acquire their full purport through their actual or potential rejoining. In other sorts of signs, the signans becomes irrele-vant at the moment at which its signatum is caught (the signans is thrown away, so to speak); instead, in the signs that Romantic philosophers and poets called symbols, the signatum acquires its full purport only insofar as it is continually compared to the physical presence of its signans.
This idea suggests that there should be some resemblances between symbolic activity and
