and Romantic poets in this sense, see L. Anceschi, Autonomia ed eteronomia dell’arte, 2nd ed. (Florence: Valleech’, 1959).

See W. Y. Tindall, The Literary Symbol (New York: Columbia University Press, 1955). For an analysis of the aesthetic importance of the notion of ambiguity, see the useful observations and bibliographical references in Gillo Dorfles, 11 divenire delle arti (Turin: Einaudi, 1959), pp. 51ff.
5.Edmund Wilson, Axel’s Castle (London: Collins, Fontana Library, 1961),p.178.

Pousseur, «La nuova sensibility musicale,» p. 25.
7.J. Scherer, Le «Livre» de Mallarrni: Premieres recherches sur des documents inédits (Paris: Gallimard, 1957). See in particular the third chapter, «Physique du livre.»

Werner Heisenberg, Physics and Philosophy (London: Allen and Unwin, 1959), ch. 3.

Niels Bohr, in his epistemological debate with Einstein; see P. A. Schlipp, ed., Albert Einstein: PhilosopherScientist (Evanston, Ill.: Library of Living Philosophers, 1949). Epistemological thinkers connected with quantum methodology have rightly warned against an ingenuous transposition of physical categories into the fields of ethics and psychology (for example. the identification of indeterminacy with moral freedom; see P. Frank, Present Role of Science, Opening Address to the Seventh International Congress of Philosophy. Venice. September 1958). Hence, it would not be justified to understand my formulation as making an analogy between the structures of the work of art and the supposed structures of the world. Indeterminacy, complementarity, noncausality are not modes of being in the physical world, but systems for describing it in a convenient way. The relationship which concerns my exposition is not the supposed nexus between an «ontological situation and a morphological feature in the work of art, but the relation between an operative procedure for explaining physical processes and an operative procedure for explaining the processes of artistic production and reception. In other words, the relationship between a scientific methodology and a poetics.

Edmund Husserl. Meditations cartisiennes, Med. 2, par. 19 (Paris: Vrin, 1953), p. 39. The translation of this passage is by Anne FabreLuce. 1. JeanPaul Sartre, L’etre et le niant (Paris: Gallimard, 1943), ch. 1.

M. MerleauPonty, Phenominologie de la perception (Paris: Gallimard. 1945), pp 381383.

Ibid..p.384.

On this «eclatement multidirectionnel des structures,» see A. Boncourechliev, «Problemes de la musique moderne. Nouvelle revue franfaise (DecemberJanuary, 196o61).

Luigi Pareyson, Estetica: Teoria della formativit a, 2nd ed. (Bologna: Zanichelli, 196o), pp. 194ff, and in general the whole of chapter 8, «Lettura, interpretazione e critics.»

Analysis of Poetic Language

Benedetto Croce. Breviario di estetica, 9th ed. (Bari: Laterza, 1947), p 134

Ibid.,p.137.

John Dewey, Art as Experience (New York: Minton Balch, 1934), ch.
9, pp 194195.

Dewey has even been accused of idealism. See S. C. Pepper, «Some Questions on Dewey’s Aesthetics,» in The Philosophy ofJ. Dewey (Evanston and Chicago: Northwestern University Press, 1939), esp. p. 371 et passim. According to Pepper, Dewey’s aesthetics brings together two incompatible tendencies: organicism and pragmatism.

Dewey, Art as Experience, ch. 9, p. 195. 6. Ibid., ch. 4. p. 75.
8.Ibid., ch. 5, p. 98.
9.Ibid., ch. 5, p. 103. Whereby it follows that «the scope of a work of art is measured by the number and variety of elements coming from past experiences that are organically absorbed into the perception had here and now» (ch. 6, p. 123).

Ibid., ch. 6, p. 109. Thus, one can say that «the Parthenon, or whatever, is universal because it can continuously inspire new personal realizations in experience» (ch. 6, pp. 108109).

See F. P. Kilpatrick, Explorations in Transactional Psychology (New York: Ncw York University Press, 1961).
Nicolas Ruwet, «Preface» to Roman Jakobson, Essais de linguistique generale (Paris: Editions de Minuit, 1959), p. 21. See also Roman Jakobson, Selected Writings (The Hague: Mouton, 1981).
12.Jakobson, Selected Writings, vol. 2, p. 556.

In the following analysis, I shall often rely on the notions of the referential (or symbolic) and the emotive uses of language; see C. K. Ogden and I. A. Richards, The Meaning of Meaning (London: Kegan Paul, Trench, Trubner. 1923; rpt. New York: Harcourt, Brace, 1946), esp. ch. 7. The referential or symbolic use of language implies: (1) that there is a corresponding reality; (2) that the correspondence between the linguistic symbol and reality is indirect—that is to say, mediated by a reference to a concept, a mental image of the real thing. The emotive use of language, instead, relies, on the symbol’s power to evoke feelings, emotions, intentions. This, of course, does not mean that we make an equation between the emotive and the aesthetic uses of language, or that we make a drastic distinction between its referential and its emotive uses; quite the contrary, as the following pages will clearly show. Occasionally, I shall also use the terms sign and denotaturn proposed by Charles Morris to designate, respectively, the symbol and the referent. See Morris, «Foundations of the Theory of Signs,» in International Encyclopedia of Unified Science, vols. t and 2 (Chicago: Chicago University Press, 1938); also Signs, Language, and Behavior (New York: Prentice Hall, 1946), ch. 2. The following analysis will also take for granted the subdivision of the speech act into four distinct parts: the addresser, the addressee, the message, and the code (which, as we have seen, are not only abstract logical categories but also encompass, and account for, emotive attitudes, tastes, and cultural habits).

Jakobson, Selected Writings, vol. 3, pp. 18ff. («Linguistics and Poetics»).

See Charles Morris, Signs, Language, and Behavior, ch. 8. The meaning of a word can be determined by the psychological reaction of the listener: this is what we call its pragmatic aspect. Its semantic aspect concerns the relationship between sign and denotation, and its syntactical aspect the organization of words within a given discourse.

See Jakobson, Selected Writings, p. 218. «The set (Einstellung) toward the MESSAGE as such, focus on the message for its own sake, is the Poetic function of language.» (See also «Linguistics and Poetics,» in Se. leered Writings, vol. 3, p. =S.)

We can attenuate the rigor of Ogden and Richards’s distinctions with Charles Stevenson’s conclusions, according to which «the growth of emotive and descriptive referential dispositions in language does not represent two isolated processes.» In a metaphoric expression, the cognitive aspects of the total discourse affect its emotive aspects. As a result, the descriptive and the emotive meaning of an expression are «distinguishable aspects of a total situation, not ‘parts’ of it that can be studied in isolation.» Then, after examining a third type of meaning (neither descriptive nor emotive but rather the result of a grammatical incoherence that produces a sort of «philosophical perplexity»), which he terms «confused meaning» (and here we cannot help thinking of Joyce’s ambiguous, open words), Stevenson concludes by saying that «there will be emotive meaning dependent on descriptive meaning…but there will also be emotive meaning dependent on confused meaning.» See Charles Stevenson, Ethics and Language (New Haven: Yale University Press, 1944), ch. 3, pp. 71, 76, 78. The studies of the Russian Formalists have yielded analogous results. In the twenties, ShIclovsky and Jakubinsky had classified poetry as an example of the emotive function of language. Their point of view was eventually changed thanks to the increasing formalization of poetic expression. In 1925. Tomashevsky had relegated the communicative function of language to the background in order to stress the absolute autonomy of the verbal structures and the laws of immanence that govern poetry. In the thirties. the Prague Structuralists tried to distinguish a multidimensional structure of poetry that included the semantic level. «While a ‘pure’ Formalist brashly denied the existence of ideas and feelings in a work of poetry, or declared dogmatically that it is impossible to draw any conclusion from a work of literature,' the Structuralist would emphasize the inevitable ambiguity of the poetic statement, poised precariously between various semantic planes." Victor Erlich, Russian Formalism (The Hague: Mouton, 1965), p. 200. 18.According to Charles Morris, "a sign is iconic to the extent to which it itself has the properties of its denotata" (Signs, Language, and Behavior, p. 23). This seemingly vague definition is, in fact, quite restrictive: as Morris goes on to explain, a portrait cannot really be iconic, "since the painted canvas does not have the texture of the skin, or the capacities for speech and motion, which the person portrayed has." In fact, Morris sub sequently reamplifies his definition by admitting that iconicity is a matter of degree: onomatopoeia may well be considered an excellent example of linguistic iconicity (p. 191), just as one can find iconic characteristics in those poetic tnanifestations where style and content, matter and form arc perfectly in accord (pp. 195196). In cases such as those, iconicity becomes synonymous with the organic fusion of the elements of a work of art that I have discussed throughout this chapter. In a later work, Morris defines the iconicity of art by stating that "the aesthetic sign is an iconic sign that designates value" ("Science, Art and Technology," in Kenyon Review 1939), since what the addressee seeks in an aesthetic sign is its sensible form, the way in which it reveals itself. Rene Wellek and Austin Warren characterize the aesthetic sign in a similar way: "Poetry organizes a unique and unrepeatable scheme of words, each of which is at once object and sign and each of which is used in a fashion that no other system could predict" (Theory of Literature New York: Harcourt, Brace, 19421). Similarly, Philip Wheelwright defines the aesthetic sign as a plurisign, the opposite of the referential monosign, and insists on the fact that the plurisign is "semantically reflexive, insofar as it is a part of what it means" ("The Semantics of Poetry," Kenyon Review 2 19401). See also Galvano della Volpe, Critica del gusto (Milan: Feltrinelli, 1960); according to della Volpe, the poetic discourse is plurisenso (polyvocal, whereas the scientific discourse is univocal), by virtue of its organic and contextual nature. 19.Charles Stevenson