1The scientific works of the Prague linguistic circle on linguistics and literature are usually found under the heading 'structuralism'. The rival term 'functionalism' is frequently used, but more in the U.S.A. and the Soviet Union than in Western Europe. For American linguistics it is the functional approach which is the truly distinctive feature of Prague linguistics, that which sets it apart both from post-)Bloomfield structuralism, which limited itself to the formal classification of linguistic utterances, and from the transformational grammar of the Chomskyans, which dismisses the functional perspective as inessential for the explanation of language and of linguistic competence. In early accounts of their own position members of the Prague school habitually used simultaneously both the expression 'functional' and the expression 'structural' to characterise their works.
2In fact, functional analysis was the point of departure for the two main concerns of Prague linguistics, phonology and poetics. What distinguishes a linguistic sound, a phoneme, from a mere noise is the function it has of discriminating between meanings. What marks the difference between a poetic text and an ordinary linguistic text is the fact that the communicative function - through which language comes into being and which shows itself in the concentration on what is said - is overlaid by the aesthetic function - which shows itself in a turn of attention (or set) towards the structure of the means of communication, i.e. of the linguistic utterance as such.
3Functional analysis led directly to structural analysis. Once function had been discerned the question arose what effect the function of discriminating between meanings has in the sound-system of a language and how the aesthetic function, the focus on the linguistic medium as such, is brought about. In phonology questions like these led to the discovery of a hierarchically ordered system of binary distinctive features. A system of binary oppositions can be conceived as the most rational procedure for decoding information. In poetics the attempt to determine the underlying structural base of the aesthetic function issued in the thesis that the principle of equivalence, which in normal speech is constitutive of the paradigmatic axis of selection (for the selection of a linguistic unity as a member of a linguistic chain from an arsenal of unities which are meaningful in a similar way), also determines the axis of combination in a poetic text (Jakobson, 1960). Poetic texts are marked by a complex and varied network of relations of similarity and contrast between their components which gives them the character of a dense unity (Dichtung).
4Within the theory of science the teleological approach has been conceded a heuristic value ever since Kant. It serves as a guiding thread for research which aims to set out in a systematic way the mechanisms which are the basis of a process oriented about some goal. One thinks of the way Harvey systematically set out the mechanism of the circulation of the blood in discovering the functions of the heart. In the linguistics of the last sixty years the heuristic fruitfulness of the teleological approach has shown itself in the way teleological explanation has complemented not so much causal analysis but rather structural analysis. When teleology and functional analysis is discussed in contemporary literature in the philosophy of science any reference to the extremely successful linguistic research in this direction is missing.1 Examples are limited to material taken from anthropology, psychology, sociology and biology, although only in the case of biology are recent examples dealt with, while in the other sciences the old examples, to be found already in the classic works by Hempel and Nagel of the '50s, continue to be traded back and forth. The absence of linguistics is all the more regrettable in view of the fact that, quite apart from the combination of teleological and structural analysis, this discipline has brought to light a second original feature, the phenomenon of plurifunctionality.
5Literature in the philosophy of science proceeds as though an object or a process could have only one function. Even in the case of the single venture in the direction of a plurifunctional perspective in the anthropological distinction between manifest and latent function, the former is characteristically dismissed from the start as a pseudofunction. On this point analytic philosophy of science con curs with the functionalism of the '20s in the areas of art and architecture, which indulged in a pronouncedly extreme and, as can be shown, literally infantile monofunctionalism. The phenomenon of functional equivalence (one end, many means) has indeed been dealt with at length in the theory of science. A sentence can be emphasised not only by being written in capitals, but also by means of italics, by letter-spacing, or by underlining. The four techniques are functionally equivalent. The other alternative (one means, many ends) remains however unnoticed and unevaluated Yet the (potential) plurifunctionality of an object is a factor which should not be underestimated if we want to talk about its survival, its future existence, something which is as important for a pragmatically oriented philosophy of science as is the explanation of its present existence. An explicit thematisation of the phenomenon of pluri functionality is a contribution of the Cercle Linguistique de Prague to the general philosophy of science which has still scarcely been acknowledged. It thereby anticipated by forty years in linguistics and, what should attract just as much attention, in the science of folklore, a movement which has only gained prominence in recent years in architecture. After functionalism had stripped away so many functions people are today trying to give back to buildings streets and whole cities ('multi-purpose user-facilities') something of their lost wealth of functions.
6Before turning to some problems posed by plurifunctionality it would seem to be a good idea to provide the begin>ning of an explanation of the word-field 'function'. Especially in German-language literature, talk of an 'aesthetic' or 'poetic' function of language may be bound up with scientific and technological connotations, from which one would like to see both they and one's self dissociated.
7Talk about a goal is much less likely than talk about function to lead to the suspicion that linguistic phenomena are being distorted: yet function and goal are two correlative concepts. Goals are normally ascribed to independent entities, functions to non-independent entities. What counts as independent and non-independent is to a certain extent a matter of one's perspective. If one considers a speaker as independent, one asks about the goal he has in mind when speaking. If on the other hand he is considered as nonindependent, as in the case of the spokesman for a corporate body such as a government, one asks about his function. Assigning the concept of function to a non-independent part of a more inclusive whole has also established itself in the metonymic use of the expression 'part' for function. Cultural entities, such as language, and actions, such as speaking, are per se non-independent. With reference to languages and speech acts one therefore asks not about their goal, except in the case where these are made independent through personification, but about their function. One only speaks about the function of an object or process when these are either used by a subject in the service of a goal, or when they are integrated in a system which is goal-directed in the cybernetic sense. Functional analysis (of a non-independent part) and a final or teleological analysis (of the independent whole whose part it is) therefore presuppose each other. It is this intrinsic connection of functional analysis with the relation between a part and its wholes which yields the common denominator of functionalism and structuralism. In coining the slogan 'structuralism' in 1929 Jakobson had put forward the exposition of the inherent lawful relations of an arbitrary group of phenomena, which were to be dealt with not as though they were a mechanical heap, but as a structural whole, as the specific interest of a science which could properly call itself structuralist.
8While the concept of a goal is only used when something is a means for a subject, 'function' can be used both for something which refers to a subject and for something which is integrated in a system. If not (only) the goal-oriented entity, but (also) the means is a subject, it is best to speak about its role. Talk about its function is also usual, while talk about its goal is used only in special cases where objectualisation is involved.
From Bühler's three-function schema to Jakobson's six
9A characteristic trait of Prague structuralism in general and of Roman Jakobson's theory in particular is the fact that poetics is treated as a component of linguistics which cannot be anything other than integral. One can go still further and put forward the view that the most important principles of structural linguistics have their starting point in the study of poetical language and are marked by this (Holenstein, 1975). This holds not least of functional analysis. This was neither an abstractly formulated functional principle nor a schema of functions which was first of all discovered and tested in ordinary language and then subsequently extended to deal with poetic texts. It was, rather, precisely the other way round: the search for the essence of poetical language, poeticity, that which makes a linguistic utterance poetic, had led to the distinction between different linguistic functions and to the discovery of changing hierarchies of functions.
10The earliest moves in this direction come from the period of Russian formalism. Already in 1916, L.P.Jakubinskij, like Jakobson a founder member of the Petersburg "Society for the Investigation of Poetic Language" (Opoyaz), had described the difference between ordinary and poetical language in a teleological perspective:
Linguistic phenomena should be classified by taking account in each particular case of the intention of the speaking subject. If they are employed for the purely practical aim of communication then what is involved is the system of everyday language (of verbal thought) in which the linguistic components (sounds, morphological elements, etc.) have no autonomous value and are only a means of communication. But one can imagine other linguistic systems (and they do in fact exist) in which the practical aim takes second place (although without diappearing completely) and where the linguistic components then take on a value of their own (Eikhenbaum, 1927, p.39).
11Three years later in his famous lecture on "modern Russian poetry" to the Moscow Circle of Linguistics Jakobson distinguished poetical language additionally from emotional language in the process of arguing against a psychologistic interpretation of poetry. Whereas practical everyday language and emotional language are defined by their communicative function, poetry is determined as 'language in its aesthetic function'. Poetry is 'simply an utterance with a set (Einstellung) towards expression' (1921, p.31). Here words have, in Chlebnikov's formulation 'a power and value of their own' whereas they remain indifferent to any relation to the object of discourse (the reference) towards which one is normally turned (eingestellt) in practical uses of language. Poetic language has this in common with emotional language, that there is in both a close connection between the phonic moment and the meaning.
12In the programmatic "Theses of the Linguistic Circle of Prague" on the occasion of the first Slavists' conference in 1928 these approaches were taken over from Russian formalism. The section on the functions of language which, in accordance with the interests of the Prague audience, culminated in a paragraph about 'the poetic function', stresses throughout the variety of different functions. These were indeed listed, and again - as was typical for the Prague group - in binary oppositions: inner and manifest, intellectual and emotional (including the speaker-centred expressive and the hearer-oriented appellative function), practical and theoretical, as well as language with a communicative and language with a poetical function. The prominent position of inner language, which has such an interior position in analytic philosophy of language, may likewise be due to Russian influence. At the IXth International Congress of Psychologists the two Russian psychologists of language, Vygotskij and Luria had put forward the specific function of internal speech, and this, together with its specific structure, were dealt with by Vygotskij in 1935 in his book Thought and Language which has since become a classic of the psychology of language: its specific function consists in the solving of problems and in the organisation of the pursuant actions. It is important for the next stages that the "theses" (1929, pp.14, 21) assigned the two central functions, the practical communicative and the poetic function to the constitutive components of the speech event. This is completely in line with Jakobson's Moscow lecture mentioned above. In the communicative function the intention of the speaker is directed towards what is designated (le signifié), in the poetic function towards the sign itself (le signe lui-même), the linguistic expression as such.
13In the early '30s Karl Bühler, building on an earlier sketch (1918), published his famous scheme (1934, pp.24ff.) of linguistic functions (Fig.l):
14It was natural for the Prague Circle to take up this so- called Organon model. There was first of all a historical reason for this: Bühler was well-known to them through lectures he gave to their circle and through his Viennese colleague Trubetskoy. And also two substantive reasons: on the one hand there was a gap in Bühler's approach, with its claim to being systematic, namely the poetic function - which for the Prague Circle was a cornerstone of linguistic functional analysis. Bühler connected poetic language with the function of expression, as was then quite common, and in any case not unexpected in the case of a psychologist. On the other hand the approaches of the Prague group and of Bühler had in common the attempt to assign functions to the constitutive components of the speech event, and to anchor the former in the latter. Bühler limited himself to three extra-linguistic bases (Fundamenta) of the speech situation, the speaker, the hearer and the object of discourse. The linguistic components proper on the other hand he did not take into account, although his famous schema should have brought him to realise the importance of these. He does not make any use in his functional analysis of the central factor in his schema, the sign Z, although at the same time it is noticed and discussed by him in a differentiated way. Rendering the linguistic sign by means of a triangle overlaid by a circle and the incongruence of the two figures is intended to indicate the apperceptive complementation which the sensible sound-material acquires in linguistic use. The model and the accompanying text directly invite one to pick out the code as an independent component - the code of the linguistic system, which is decisive for the apperceptive Gestalt of the sound-material; the apperceptive Gestalt being represented by the triangle, the sound-material by the circle. The physical channel which connects speaker and hearer is clearly distinguished from the merely intentional or symbolic reference to the objects of discourse by the continuous lines. In other words it was to be expected that it would only be a question of time before there arose an immanent linguistic complement of Bühler's model (and not merely an external ad hoc accumulation as is the fate of most scientific models). The poetic function of language corresponds to the sign as a whole, the meta-linguistic function to the code, and the phatic function to the channel.
15Bühler's Organon model was filled out in two stages. The first step consisted in bringing together the two approaches, the Bühlerian trichotomy and the Russian formalist/ Prague structuralist dichotomy between the practical and the poetic function. It was first set out explicitly and in print in Mukařovský's contribution to the Copenhagen Congress of Linguists in 1938:
a fourth function, unmentioned in Bühlers scheme, emerges. This function stands in opposition to all the others. It renders the structure of the linguistic sign the center of attention, whereas the first three functions are oriented toward extralinguistic instances and goals exceeding the linguistic sign. By means of the first three functions the use of language achieves practical significance. The fourth function, however, severs language from an immediate connection with practice. It is called the aesthetic function, and all the others in relation to it can be called collectively the practical functions. The orientation of the aesthetic function toward the sign itself is the direct result of the autonomy peculiar to aesthetic phenomena (Eng.trans.,p.68).2
16The metalinguistic function was added in 1952 as a fifth function. (See Jakobson, 1953, pp.556ff.) It was the heyday of communication theory and of the logical analysis of language. Communication theory led to a strengthened and more precise attitude towards the linguistic code as an independent factor of the speech event and the replacement of the ambiguous Saussurian opposition of langue-parole through the more concise opposition code-message. The systematic distinction between object- and meta-language was urged from the side of logic (Tarski, 1944)3 as a means of avoiding the antinomies (of the liar, of heterological terms, etc.,) and in order to make possible a definition of truth. In the case of Jakobson (cf. his 1962) a different priority stands in the foreground: the indispensability of metalinguistic utterances for the acquisition and understanding of language. New and strange linguistic unities are acquired and made understandable through differentiation from and paraphrased by means of units which are already familiar.
17Finally in 1956 came the integration of what was now a six-member schema of functions of the phatic function, which had first been described by the functionalist Malinowski.4 Certain linguistic utterances have as their function only to bring about, confirm or extend contact with someone without conveying information with any content. (Jakobson, 1960, pp.355f.) The text of the 1956 Presidential Address to the Linguistic Society of America (first published in 1976) combined into a single diagram (Fig.2) the six components of the speech event and the functions assigned to these (which four years later in "Linguistics and Poetics" were to be displayed in two figures: 1960, pp.353,357):
From formalism to structuralism - from the separation of the work of art to the autonomy of the aesthetic function
18In the early phases of Russian futurism and formalism there was proclaimed and indeed there began the creation of a poésie pure and, as an extreme form, a poésie concrète. The poetic text was celebrated as a pure sound-formation which, when it pointed to anything outside itself at all, pointed not to objects which it designated clearly and about which it made propositions with a truth-value, but rather pointed - synaesthetically - to neighbouring sensations such as odours and colours. It was in this way that Ossip Brik analyses parallelism - an eminently grammatical procedure - as a mere repetition of sounds (Pomorska, 1968, pp.28ff). Those formalists who were acquainted with the philosophy of language were nevertheless soon making use of the Frege-Husserl distinction between the reference and the meaning of linguistic signs, particularly in order to specify the nature of poetic language. As a particularity of poetic language an inner connection between sound and meaning was described, one indifferent to the object and to the truth-value of the proposition (and to its emotional potential). Jakobson's pronouncement at that period has often been quoted (1921, p.41):
Let others superimpose upon the poet the thoughts expressed in his works! To incriminate the poet with ideas and emotions is as absurd as the behavior of the medieval audiences that beat the actor who played Judas and just as foolish as to blame Pushkin for the death of Lensky.
19Later the thesis to the effect that the objectual reference retreats into the background, that there is in the extreme case a complete lack of objects for (also fictionality of) poetic propositions was corrected. It is not the lack of an extra linguistic reference which is characteristic of poetical texts, but the way in which meaning and/or reference is made ambiguous (Jakobson, 1960, p.370). The move towards greater differentiation went hand in hand with the discovery of phonology, particularly with the discovery that the expressive material is inseparable in its Gestalt from its sign-function. Sound associations were experimented with, not only the associations with other sounds and with colours and odours and with varieties of sense- data but also with meanings. Sounds turned out to be phonemes, 'i.e. accoustic presentations suited to associate themselves with meaning-presentations' (Jakobson, 1960, p.94). Certain sounds become associated at all only because they have meaning in common. The association owes its poetic attraction to the (semantic) unity in or underlying the (sensual) contrast between the two sound-formations: 'He is naked and bare'. Other sounds have a poetical effect, because they are ambiguous: 'The kosa ("braid" or "scythe") sometimes adorns the head, hanging down to the shoulders; sometimes cuts the grass.' In short:'the contents are part of the structure of the poem.'5
20Language has a multi-dimensional structure and is plurifunctional. The poeticity of the text consists not in the exclusion of one of the relations which are constitutive of language, but only in a re-orientation of these relations. Having defined language with a communicative function in their "theses" (Thèses..., 1929, p.14) as 'dirigé vers le signifié' the Prague linguists, with their keenness for binary oppositions, described language with a poetic function with good reason not as 'dirigé vers le signifiant' but as 'dirigé vers le signe lui-même'. Signs have, on the Saussurian view, two sides. They consist of a signans and a signatum. Both components surface in poetry.
21Nor, however, does the poeticity of a text consist in the exclusion of all practical function in favour of an isolated poetic function, but only in the presence and - in the optimal case - in the dominance of the poetic function. In opposition to tendencies in early formalism Prague structuralism did not concentrate on isolating poetic language, the pure poetical text, but on the isolation of the poetic function. 'What we emphasise is not the separatism of art, but the autonomy of the aesthetic function' (Jakobson, 1934, p.412). There is no opposition to the view that a poetic text can also have functions other than the purely aesthetic: it can be propagandistic, therapeutic, documentary, informative, for example. What is opposed is simply, on the one hand the reduction of the poetic function to other functions of language, the psychologistic reduction to its emotive function, the spiritualistic reduction to the idea of disclosure of some sublime theme, the revelation of otherwise inaccessible truths; and on the other hand the reduction of procedures which owe their development primarily to the poetic function, to procedures with a merely utilitarian function such as the reduction of parallelism, an almost paradigmatic procedure, to humdrum mnemotechnical procedure (Jakobson, 1966, p.423). 'The poeticity ... is an element sui generis, an element that cannot be reduced mechanically to other elements' (1934, p.412). With the conception of an autonomous yet isolated poetical function Prague linguistics, - for which is characteristic the dialectical unification of the opposites which Saussure and Russian formalism had split apart: synchrony- diachrony, form-content, langue-parole , etc. - has gone beyond the false antagonism of those who would play poésie pure and poésie impure (engaged literature) off against each other. Simultaneously they have provided their most original contribution to functional analysis, the thematisation of pluri- or multi-functionality and of variable hierarchies of functions. This was decades ahead of discussions in the theories of science and of art (in particular of architecture).
Multi- versus monofunctionalism
22What makes the works of the Prague Circle so extraordinarily fruitful is their concrete approach. Their programmatic theses were immediately followed by analyses of prototypical examples. After their insight into the necessity to semiotics and aesthetics of perspectives which take into account the theory of action, they began with a sure instinct the analysis of those works of art in which action appears most directly, that is to say drama and in particular that sort of drama in which the 'recipients' are most actively involved, people's theatre. They developed an insight of theirs of great importance for functional analysis - that a thing is rarely given with only one function, but more normally with a whole bundle of functions - for the case of two illustrative multi-functional phenomena: dress and architecture, of which they gave detailed descriptions. As soon as multi-functionality as such comes into the centre of attention the question arises of the relation between the functions. Is it a matter of affinity, of implication, of subordination, of incompatibility, of aversion? And what is the sort of possible change of function? This change does not consist in one function giving way to another which has nothing in common with the first, as is still assumed in a somewhat unreflective way in the actual German work on aesthetic reception (i.e. the Rezeptionsästhetik). A more likely form is a change in the ranking (dominance) of functions and in the varying degrees of manifestation or latency of the individual functions (Jakobson, 1935b).
23A first function of objects of use consists in the fact that - given their relation to a cognitive subject - they do not merely exercise their practical function but also - semiotically - express it. In the technical terminology: 'The object not only performs, it also signifies its function' (Mukařovský, 1937, p.236). A 'natural device' in architecture is the way in which building-elements, rooms, etc., are given a form which indicates their specific function. Aesthetic architecture is distinguished from merely utilitarian architecture through the way the practical function of furniture and rooms is secondary, and serves primarily to give a semiotic form to space by directing the attention and the movements of the visitor. Of course the practical and the semiotic function may also part company. There are entire stylistic movements which set their heart on concealing practical functions behind architectonic constructions. Such a split is artistically attractive only when it is experienceable as something intentional or when the distracting form stands in some other connection with the building, such as for example the case where, with its monumentality, it signals its importance or signals aesthetically its outstanding qualities. Only in bad architecture do function and form stand in a relation to one another which says nothing, neither in a positive nor in a contrastive way, nor so as to present consciousness with additional aspects through amplification.
24Certain functions are regularly supported by specific other functions. Thus the magical function of verbal as well as non-verbal rites is bound up in a striking way with the aesthetic function. Aesthetic effects are frequently the same as those intended by magic. Aesthetic works' are able to enchant. Their characteristic attractiveness predestines them regularly to accompany the erotic function, which is likewise out to capture the senses. According to Bogatyrev (1937, p.75) the aesthetic function of clothes frequently has the subsidiary function of concealing the erotic function, precisely because of this affinity of effect. Today, where the erotic function no longer has to be hidden but at most justified, it is possible that the aesthetic function is merely an alibi. Bogatyrev's analysis of national costumes shows that there are indeed combinations of latent and manifest functions in which the manifest function is by no means a pseudo-function, as is supposed to be the case in the examples of the Hopis' rain dance which is so popular in analytic philosophy of science. Here the manifest magical function is dismissed as a pseudo-function and the true function which is all that needs to be taken seriously by science - the strengthening of the group identity of the Indian tribe - is said to be latent. This function is claimed to be something the Indians are not aware of, but is noticed only by anthropologists (Stegmüller, 1969, p.564).
25Apart from explanation, analytic functional analysis aims at the prediction of the existence of an object on the basis of a function which the object possesses in some system. It ignores the fact that frequently secondary functions guarantee the survival of an institution which has since lost the primary function to which it perhaps owes its origin. Magical rites, particularly, survive a rationalising period of enlightenment in virtue of a strong secondary aesthetic function, often too in virtue of a religious, regionalist-nationalist or even merely commercial transformation, which is often very easy to bring about. An example which is closer to home than the rain dances of the Hopi Indians is our own Christmas tree. Here a cycle of functions can be observed which is also to be found in poetry. 'When, that is, products of literary poetry, themselves based on folk poetry, in turn penetrate back through to the people' (Bogatyrev, 1932/33, p.257). The Christmas tree owes its wide popularity in Eastern Europe to an aesthetic and/or status function, which caused it to be introduced by the mobile intellectual upper classes into the villages in which, then, its original magical function again gained the upper hand.
26It is however by no means always the case that when a primary function disappears, secondary functions continue to ensure the existence of a phenomenon:
It happens that the loss of one function also involves the loss of other functions. Thus when national costumes disappear old clothes simultaneously lose their aesthetic value and are felt by the peasants to be less comfortable, less practical, etc. (Bogatyrev, 1930, p.335).
27Certain functions can dominate so strongly that they are able to conceal functional injury or breakdown in secondary respects. The dysfunctionality, the imperfections in the latter come to be realised only when the dominating function falls away. The aesthetic function in particular seems to be such that it induces secondary justifications of additional functions. Mukařovský (1936, p.34) also sees in the aesthetic function a guarantee of the further development of a cultural phenomenon which has lost its practical function. Its aesthetic character is retained for future periods in which there occur possibilities of its being used in another practical function.
28Secondary functions can also serve for the classification of a domain of phenomena defined by the dominance of another function. Thus literary genres can be characterised to a certain extent on the basis of a function which for them is of only secondary importance. At least in the case of classical examples the referential function is of great importance for epic poetry, the emotive function for lyric poetry (Jakobson, 1960, p.357) and, one might add, the appellative function for dramatic poetry with its moralising and didactic tendency.
29The fate of whole artistic genres and stylistic tendencies, their coming and going, seems often to be bound up less with their aesthetic quality as with secondary functions. A work of art which is emotionally no longer moving loses its topicality. It is in any case striking that aestheticians who see delectare et movere as an essential property of works of art (cf. Jauss, 1977) frequently belong to the proponents of a relativistic historical point of view. For Jakobson, who defines the aestheticity or poet- icity of a text in structural terms, being moved, which shows itself in pleasure, is secondary, something which varies according to the historical and biographical situation. One can appreciate a poem by Goethe as a work of art without having any pleasure in it.6
30Not only can different functions be compatible with one another, complement each other reciprocally, they can likewise stand in conflict with one another. The aesthetic function suggests itself once again as a prototypical example of such conflict. For in a certain sense the aesthetic function turns out to be the 'dialectical negation of functionality' (Mukařovský, 1937, p.244), in so far as the subject's attention turns from causes and effects, and hence from the practical use of a work, to its structure in all its proud autonomy. The fact that a piece of furniture is outstandingly beautiful can be an obstacle to its use. One experiences the tendency to put it on show instead of using it. In the same way the artistic form of a poem can distract from its content and its communication - unless it is a perfect poem in which content and concrete realisation (in audible recitation) are themselves structural elements of the artistic whole. An object of use can also be constructed in like manner, i.e. such that its beauty is only fully displayed when it is used. In the case of a car the streamlined design stands out only when the car is driven. There is no aspect of reality and hence no function of language which cannot be employed as an aesthetical structural element. This is a first answer to Jauss's criticism (1977, p.168) of Mukařovský that he leaves unexplained 'how the aesthetic function can be a dialectical negation of the communicative as well as the emotive function and is simultaneously supposed to be communicative'. A second answer involves referring to the attractiveness of the aesthetic structure, which directs the attitude or set of the subject to the work of art, certainly no longer only to its content - as is the case where the communicative function dominates - but also the interplay of formal and material factors which make of a text a whole valuable in itself ('selbstwertiges' Ganzes). Thirdly Mukařovský's semiotic principle,'an object not only performs but also signifies its function', holds not only for objects with a practical function but also for those with an aesthetic function. A work of art performs not only the aesthetic function it is intended to perform, it simultaneously signals it also. Fourth, the truly dialectical possibility that precisely because in poetry 'communication is not really important' (Jakobson, 1934, p.406) and the truth content of a poem is secondary, the censorship exerted by one's own unconscious and by authority - and how many poets have not used this opportunity, often unadmitted but highly consequential - is weaker, so that (prima facie fictive) poetry is often more faithful to the truth than a diary, more informative than a newspaper report, and more provocative than a handout of a professional agitator.
31With a little less pathos and a little more sarcasm one might quote here an example which is today more than well known: the disclosure of 'what we always wanted to know about (or see of) sexuality' occurred first on film and then on stage, step by step, in that exceedingly aesthetically sensitive directors, first and foremost Ingmar Bergmann, proclaimed each successive stage in this stripping away as an imminent aesthetic necessity. This somewhat sarcastic excursus should not be taken to deny the close affinity between the erotic and the aesthetic. What leads to the mutual involvement of art and the erotic is their common sensual attractivity (see discussion of Bogatyrev, 1937, above). While the sexual-erotic function in many contemporary films is nothing more than a substitute for the lack of any aesthetic function, the reverse is often true for dress. The aesthetic function may be a substitute for the weak or absent erotic function of clothes which have nothing attractive to reveal.
32So much concerning Jauss's problem of how an aesthetic work can be both communicative and emotional in spite of its dialectical negation of and its distraction from communication and emotion. The relation between functions is particularly complicated in the case of linguistic sounds (cf. Jakobson and Waugh, 1979). The phonic materiality of sounds is such that they function immediately both synaesthet- ically and associatively. There is a correlation between the sound-opposition i and u and the visual opposition light and dark. The strikingly discrete structure of linguistic sounds, however, also makes them mediately suitable for the construction of an exceptionally rational system of discriminations. The thirty phonemes of an average language permit the formation of for example a repertoire of signs of 304 (810,000) combinations of four sounds. In natural language this bi-functionality of sounds leads to different constellations. The immediate latent function can be used in a subsidiary fashion, in that for certain semantic fields only particular phonemes, which directly symbolise these fields phonetically, are admissible. It can also censor selectively: it is statistically demonstrable that lexemes, the meaning of which coincides with the immediate symbolism of the sounds, have greater chances of survival in the course of linguistic development. But the immediate meaning (that which is natural, latent and directly pertinent to sound-symbolism) and the mediate (artificial, manifest, arbitrarily assigned) meaning can also come into conflict. In everyday language, where attention is normally directed to what is said, this conflict usually goes unnoticed. In poetry on the other hand, with its set towards (Einstellung auf) the sign as a whole and its tendency to make manifest all the relations which contribute to the latent make-up of this sign (cf. Jakobson, 1960, p.373), this conflict is more than noticeable. The poet can exploit the conflict as a means of estrangement, but he can also try to balance it out by paying particular attention to the context. Thus French poets, faced with the relation between the dark sound of jour and the light sound of nuit, compensate for this by means of a phonological and/or semantic context which fits more neatly the sense-content of day and night (Jakobson, 1960, p.373). Finally, the conflict can also be heuristically effective by directing the attention of sensitive language-users to other less dominating components of the signatum, which agree more readily with the latent sound-symbolism.
Instead of perceiving a contradiction between the signata and the phonetic peculiarities of the corresponding signantia, I unconsciously assign to the various signata differential natures. Thus jour, in conformity with its dark vowel, presents the durative aspect of continuing action, nuit, with its light vowel, the perfective aspect of completed action. For me the day is something which endures, night something which sets in, or comes to pass (night falls, as we say). The former connotes a condition, the latter an event (Lévi-Strauss, 1976, p.17).
33The Prague theorists of aesthetic reception of the '30s are ahead of their descendants in Konstanz in explicitly thema- tising the complexity of changes of function. In particular Bogatyrev's semiotic analyses of folk-costume and customs (1936 and 1937) are a mine of information and insight about the manifold aspects of functional analysis. The dynamic form of the phenomenon of plurifunctionality, which involves a continuous change in the hierarchy of functions and a dominant function which determines the mode of appearance of all the other functions, is here clearly presented in what is one of the earliest phases of the investigation of these matters. Bogatyrev (1937, p.43) sets out the following order of ranking for everyday dress (ranked in order of power):
- practical function (best suited to protection from cold and heat, dirt, injury);
- indication of social status or class;
- aesthetic function;
- indication of regional background.
34Sunday and holiday costume on the other hand shows the following re-distribution of values:
- holiday or ceremonial function;
- aesthetic function;
- ritual function;
- nationalistic or regionalistic identification;
- indication of social status or class;
- practical function.
35In the exchange of the dominant function, individual subordinate functions are strengthened, others recede into the background or are eliminated, new ones are added, certain forms become obligatory, others optional. For example, if clothing associated with work or with a profession advances to the status of holiday costume, it becomes binding on a whole age or gender group (think, for example, of the growth in popularity of naval attire as Sunday clothes for children). Functions change with the transition from one bearer to another. Galoshes, whose dominant function in the Russian cities in which they first came into use was the practical one of protection from dust and dirt, acquire a largely aesthetic function in the countryside, where they are worn by young people sauntering through the streets on holidays in fine weather. Songs and costumes, where at first the aesthetic function dominated, come to lose this in favour of indication of age when they become reserved for children.
36The reality of the function of the identification of region and religion so heavily emphasised by Bogatyrev can be illustrated by a moralising gloss of the Freiherr von Knigge:
The calvinist merchants of Emden set out their gardens in the Dutch style; now I once heard one of them say of another merchant of the same confession who had however taken it upon himself to include in his garden certain alterations shocking to the reformed community, that this garden had a distinctly Lutheran stripe.
37Characteristically, Knigge objects to this hypertrophy of the identification function because it is detrimental to what is for him the more important function of supraregional communication in society.
I am of the opinion that these differences in the customs of the German states make it very difficult, in foreign provinces outside one's own native area, to start up friendships, to please in society, to have an effect on or win over other people.
38In view of this multifunctionality which one soon comes up against when one pays a little attention to the phenomena themselves, the question arises how both on the one hand analytic philosophy of science and on the other hand the architectural current which as 'functionalism' has entered into the history of art were able almost simultaneously to arrive at their naive unproblematic monofunctionalism. Monofunctionalism is typical of machines, of highly developed biological organs, of formalised languages and - strikingly - of an early stage in the development of children's intelligence. The architectural monofunctionalism of Le Corbusier and of the Bauhaus may in part go back to the fascination with machines: the house as machine à habiter, the chair as machine à s'asseoir, etc.7 The unreflected monofunctionalism of analytic philosophy of science may well be conditioned by too one-sided a concentration on biological examples, on the highly developed individual and collective studied in physiology and ethology (cf. the specialised organs of 'higher' animals and the division of labour in communities of bees and ants; on this Mukařovský, 1966, p.121), as well as by the logical ideal of univocal - and in their own way also unifunctional - terms of scientific language. One forgets too easily that the higher development consists not necessarily only in the formation of specialised organs but also conversely in a plurifunctional adaptation of what was a monofunctional organ. Think of the role assigned to the human hand in the process of hominisation, its polyfunctionality in contrast to the hands of the other primates which merely scratch and seize. In the realm of sociology the experience of plurifunctionality, at least of that of members of human societies, is formulated most impressively by Max Frisch in what he says about the problems which have arisen in connection with foreign works (Gastarbeiter) in Switzerland: 'We called for labour power, but what we got was human beings'. In the third area of examples, the language of science, it has likewise become common knowledge that the survival value and the chances of survival of natural languages vis à vis formalised languages consists precisely in their plurifunctionality (including context-sensitive ambiguity).
39Monofunctionalism is, above all, as was remarked at the beginning of this paper, infantile. The child in the process of acquiring language is for a long time unable to assign two different meanings to one linguistic form. Thus initially a French child uses the article les exclusively (as a pluraliser) to signalise a plurality. In order to signalise a totality (as a totaliser) as this article can simultaneously be employed in adult language, the child resorts at this early stage ostentatiously to an additional morpheme: toutes les voitures. Occasionally it will even take refuge in ungrammatical forms in order to keep apart two functions, such as indefinite reference and numerical indication which can simultaneously be expressed by the indefinite article in the adult language. Thus in order to indicate that what is in question is a single cow, it creates the ungrammatical expression une de vache (Karmiloff- Smith, 1976, pp.300ff.). The child is unable simultaneously to express two functions such as a wish and a piece of factual information, as this is normal in adult language by means of constative forms such as it is cold, when what is principally intended is the demand to close a window (cf. Halliday, 1977, pp.42, 71).8
On the plurifunctionality of the poetic text
40A consequence of the plurifunctionality of language is that the linguistic analysis of a poem cannot restrict itself to the poetic function (see Jakobson, 1960, p.357). It has to take into account the other functions of the text, not in an additive fashion but in an integrative way in relation to the dominating poetic function which transforms and unifies them into a structural and functional whole. Conversely, every study of linguistic phenomena has to analyse the role played in these phenomena by the poetic function. It plays a not inconsiderable role in propaganda texts, in rhetoric generally and, as Jakobson has pointed out, in the acquisition of language.
41The obstinate prejudice to the effect that structural poetics limits itself to a text-immanent, if not merely phonological-grammatical analysis of poems, is a complete misunderstanding of its intention and scope. It is a misinterpretation of the dominance of the poetic function (with its primarily phonological and grammatical means) to take this function to be exclusive. The failure to recognise the central role played from the beginning in Prague linguistics by semantics, as well as pragmatics - in the form of functional analysis - in contrast to contemporary tendencies in American linguistics and analytic philosophy of language, is probably due to a fixation on the revolutionary one-sidedness of Russian formalism and on positi- vist/anti-mentalist one-sided emphases within currents of thought which likewise claim for themselves the term 'structuralism'. Against the tendency to reduce linguistics to mere combinations, Jakobson (1949, p.38) coined the phrase: 'The primary function of the sign is to signify and not to figure in certain given constellations.'
42The same prejudice passes over the programmatic theses and theoretic reflections about just this theme (Jakobson and Tynjanov, 1928; Jakobson, 1935b). It passes over the exemplary studies in which interpretations are given of the poetry of the Hussites in its historical and sociological context (Jakobson, 1936) and of the poems of Pushkin and Mayakovsky in their biographical and psychological context (Jakobson, 1930, 1937). Finally it overlooks Jakobson's interdisciplinary cooperation with a specialist on myths, Lévi-Strauss, in one of his best known poetic analyses, that of Baudelaire's Les Chats (1962). As a linguist, and in view of the dominance of the linguistic medium as such Jakobson does indeed limit himself, in most of his analyses, to displaying immanent linguistic structures. The occasional moves in the direction of the extra-linguistic context may, when considered independently of his theoretical reflections and studies of examples, make an anecdotal impression, as when Yeats' geometrical symbolism in which the number 3 and its exponents (32 ,33) plays a central role, is discovered not only in the number of word-categories of his poems but also in a confession in his 27th (33) year (1977, pp.42,51).
43Considerations of the extra-linguistic context are frequent and consistent enough to enable one to recognise the relation which, according to Jakobson, characterises the relationship between a poetical text, its author, and its period of genesis. The relation is a double one. It is poetic and/or functional. In the first case the historical situation is a context for the relation of which to the poetical text as such the same principle of equivalence is decisive as is constitutive of the poetical text in the narrower sense, that is, an equivalence principle in the form of multiple relations of similarity and contrast (cf. 1937, p.162). This continued relevance of the principle of equivalence is valid primarily for the verbal context of a poem, consisting of contemporary poetry, of which an individual text is a part, and of the traditional poetry which precedes it.
Certainly we do not want simply to deduce the work from its context. On the other hand we are not allowed in the analysis of a poetical work to ignore disguised and repeatedly occurring correspondences between the work and its context, especially regular connections between certain traits common to many works of a single author and some specific place, time or biographical detail which in their origin they all share. Context is one of the components of speech: the poetic function transforms it as it transforms every other component (Jakobson, 1937, p.154).
44The poetic attraction of Pasternak's metonymic style is doubled when it is contrasted with the metaphorical style of Mayakovsky. Both styles also have their correlates in the authors' respective biographies, and in the artistic and scientific Weltanschauung of the time, in which relations as such are more important than the atomistically considered objects they connect (Jakobson, 1935a).
45Beyond this relations of equivalence play a role not only in poems but also in the interpretation of poems. Jakobson's analysis of Brecht's party poem (1965) is dedicated to the East German linguist and folklorist Wolfgang Steinitz, a dedicated adherent of the communist party. Jakobson had got to know him in 1940 during their common exile in Sweden. While they were bidding each other farewell at the station in Stockholm, just before Jakobson was to start his hazardous voyage to New York, Jakobson remembers Steinitz replying to his offer to help him settle in America as follows (1975, p.xii): 'I am a communist, my place is here...and after the war I intend to continue my work in a revolutionary Berlin.' Brecht's poem contains a repeated demand to those who are 'wise' not to leave the party:
>Gehe nicht ohne uns den richtigen Weg
Ohne uns ist er
Sei bei uns weise!
Trenne dich nicht von uns!
46One fails to grasp the intention of structural poetics and the essence of the aesthetic as such if one (cf. Jauss, 1975, p.148) fails to see more than the classical theory of reflection in the setting out of positive or negative (or contrastive) relations of equivalence. The theory of reflection fails to take into account the specific hierarchy of linguistic functions in a poetic text. For the idea that the poetic text functions as a depiction - whether typified, alienated or whatever - of reality accords far too much importance to the cognitive function. Reflections, i.e. equivalences between literary text and surrounding context have no (or only a secondary) cognitive function, but rather a (dominating) aesthetic function in that they turn the attention of the subject to the artistic medium as such. Reflection is, like any content, part of the structure of the work of art and not its cognitive aim. The relation between the work of art and reality is pointed to in a direct indeed crude way (ad usum delphini) by wall paintings of the kind where there is a continuous transition from >painting to its surroundings and where one barely notices when the illusion of the picture stops and the palpable reality of the surroundings begins, or which objects (flower pots, windows, flights of steps, etc.) are real and which are merely painted. Thanks to such continual equivalence relations, reality becomes part of the work of art.
47What both structuralism and functionalism are opposed to is an exclusively causal-historical explanation of cultural phenomena, a one-dimensional, uni-directional relation between a work and its time. The correlation between the literary and historical series has its own structural laws which need uncovering (Jakobson and Tynjanov, 1928). More important than the causal relation which points back to the past is the teleological and functional relation which anticipates the future. A work has to be understood both on the basis of its genesis and on the basis of its character as a project. The futurists had declared the future to be the tense appropriate to poetry. Only weak poets wait on events, as Mayakovsky put it, in order to reflect them or, in the extreme case, to reflect about them; strong poets run ahead of their time, pulling it after them. In his obituary article for Mayakovsky Jakobson (1930) inquires into the literary anticipation of suicide by its most famous representative which, during the lifetime of the poet, was taken for a literary stunt and whose realisation was then evaluated not as a 'literary fact' but as an absurdity. Before the effect upon the reader, which is placed in the foreground by the historians of the effects and of the reception of literature, there comes the (reflexive) effect upon the author himself.
The poet lives in a myth which he himself creates and consequently directs his life - more or less subliminally - towards the pattern created. In short, according to the results of Jakobson's analysis, not only is the life situation active in the process of literary creation, but the product created is likewise active and often decisive in the poet's actual biography. Thus one cannot draw a sharp borderline between the two, nor can one generate one from the other unequivocally. Both sequences are in a relation of complementarity and form a kind of feed-back system (Pomorska, 1977, p.373).
48In the words of the poet Yeats (quoted by Jakobson and Rudy, 1977, p.8):
The friends that have it I do wrong
Whenever I remake a song
Should know what issue is at stake:
It is myself that I remake.
On the plurifunctionality of the poetic device
49The analysis can be taken further. Not only is the poetic text plurifunctional but so also is the poetic texture, the poetic device involved in lending shape to a text. The device to which a text owes its aesthetic charac ter - the projection of the principle of equivalence, of relations of similarity and contrast belonging to the axis of selection, onto the axis of combination, has many other consequences. The first of these is the much remarked plurivocity of poetic texts.
50First however a word about the 'structuralist' definition of poeticity in the terms of Jakobson's notion of projection. One can get a grip on what has value - beauty, goodness - or on what is dynamic - force, causation, intention - only by setting out invariant presuppositions and/or entailments of a structural sort. It is important to give metonymic definitions (to use Jakobson's terminology) e.g. of force as the product of mass and acceleration, of causality as the invariant sequence of two events given additional boundary conditions to be more closely specified, of intentionality as the dependence - in the form of a feedback system - of a process on a plan. It is to metonymic definitions such as this that modern natural sciences owe their striking success. Attempts to provide metaphorical definitions, on the other hand, lead to open or hidden tautologies, as when causation is defined as production, or beauty as brightness, elegance, etc.
51The same structural property (a network of relations of equivalence in the syntagmatic axis of combination) which makes a text 'beautiful' is, we can say - modifying words of Kant9,- conducive to its plurivocity. For every linguistic unity a distinction can be made between an invariant core-meaning and variable context-meanings. 'Coach' has quite different connotations when uttered in a station and when uttered in a garage, although the two referents have many common properties, a fact that is decisive for the use of a term. The context determines which are the connotations which the core-meaning acquires. If the context to which a term is related is manifold then the term, too, will be ambiguous. The context is decisive for the over-determination of the term.10 In poetry, because the linguistic sequence is submerged under relations of similarity and contrast (at all linguistic levels), the context in which a term is experienced is at one and the same time made manifold and acquires its own distinct profile. Relations of similarity and contrast have an associative effect, that is, they form contexts, and hence determine meaning. 'In poetry, any conspicuous similarity in sound is evaluated in respect to similarity and/or dissimilarity in meaning' (Jakobson, 1960, p.372). The most transparent examples of plurivocity from poetically formed contexts are to be found in examples of parallelism. In parallelism individual sentence-parts are made to approximate to one another through metrical and grammatical parallelisation and in their way their meaning is modified either metaphorically or, in a suitable context, antonymically (cf. Holenstein, 1976b, p.11f.). Particularly illustrative is a triplet from the Song of Judith (16.9) in which the beheading of Holophernes is described in song:
Her sandal ravished his eyes,
Her beauty captured his mind,
The sword severed his neck.
52Through the immediate context, their object (eye and mind), the verbs of the first two lines acquire a metaphorical meaning, whereas the additional context which is the parallel relation to the third line works to dissolve this metaphorical meaning. It becomes mixed with the original concrete meaning. At the same time to ravish and to capture take on a metonymic connotation. They appear as a preparation for the beheading (cf. R.A.P.Roberts, 1977, p.988).
53The plurivocity of poetic texts is not something which is or should be resistant to inquiry, something for a romantic, anti-scientific 'symbolist' or 'hermeneuticist' to accept devoutly as a gift of grace or fate. Plurivocity is an automatic consequence of the mechanisms of language which the poet brings to emphasis (cf. Jakobson, 1960, p.370f.).
54The poetic device which makes a text beautiful functions both cognitively and emotively. Psychobiological investigations of aesthetic experience (Berlyne, 1971; quoted by Bock, 1978, p.192), yielded the result that linguistic and non-linguistic material of medium complexity is felt to be pleasant whereas a degree of complexity which is too high or too low is experienced as unpleasant. An optimal degree of complexity for some material is present when the recipient succeeds in grasping a unifying principle in the variety of what is offered. Invariance in what is variable and hence unity in multiplicity is just what the projection of the principle of similarity onto a linguistic sequence achieves.
55The same poetic device can, finally, function in yet another way - namely as a device of disautomatisation and as a mnemotechnic device. These are two functions which are cultivated by pragmatists among theoreticians of literature, the first by those with a sociological interest, the second by those whose orientation is psychological. The relations of similarity brought into being by the poet do indeed create not only unusual but also what, for the laws of everyday speech, are anomalous connections. The simplest example is the metaphor 'a laughing meadow'. Laughing and meadow are instances of incompatible word-categories according to the laws of linguistic combination. Only an animal, or even perhaps only a human being, can laugh. The unusual connection hinders understanding and calls forth the attention of the reader until he discovers the property which gives the connection a sense (cf. Hörmann, 1973, p. 320ff).
56Jakobson (1965; 1966, p.423) has objected in quite striking terms to the two theses of disautomatisation and mnemotechnics. His opposition to the disautomatisation thesis in particular, which followed his own early work on deformation as a poetic device (1921, pp.90, 92, 103) to which his Prague colleague Mukařovský had paid so much attention, may well surprise. But all that Jakobson is really objecting to is the elevation of disautomatisation of experience of the world and social life to the essence of poetic language, to the primary function of poetic language. The same holds of the function of making memorable (the mnemotechnical function) which people have tried to employ to give a genetic account of grammatical parallelism, the more so as parallelism is the prototype of the poetic principle of the projection of relations of equivalence onto the syntagmatic axis. The easy memorisability of poetic texts can be explained as much in terms of the psychology of association (based on relations of similarity and contrast) as in terms of the cognitive psychology which is based on meaningful organisation. Poetic texture may, after all, serve memorising and disautomatising functions. But the pragmatic-utilitarian view that the primary function lies in such use is perverse. For the primary function consists in what the tradition called beauty and what nowadays is called aestheticity, a term no less rich in problems of its own.
- 1 An exception is the biologist E. Mayr (1974, p.114) who makes reference to Jakobson's analyses.
- 2 Already in his 1928 (pp.l09f.) H. Dempe had extended Bühler's three basic functions (after the version of 1918) with 'secondary functions', and therewith also, referring to Croce and Vossler, mentioned too the aesthetic function, though without any further development.
- 3 "The first of these languages is that 'about which we speak'.... The second is the language in which we 'speak about' the first language, and in terms of which we wish, in particular, to construct the definition of truth for the first language. We shall refer to the first language as the object language, and to the second as the metalanguage." (Tarski, 1944, p.21f.)
- 4 "Phatic communion - a type of speech in which ties of union are created by a mere exchange of words... They fulfil a social function and that is their principal aim, but they are neither the result of intellectual reflection, nor do they necessarily arouse reflection in the listener... Phatic communion...serves to establish bonds of personal union between people brought together by the mere need of companionship and does not serve any purpose of communicating ideas." The stock example of phatic language is provided by gossip: "...this atmosphere of sociability...achieved by...the specific feelings which form convivial gregariousness, by the give and take of utterances which make up ordinary gossip." - "Always the same emphasis of affirmation and consent, mixed perhaps with an incidental disagreement which creates the bonds of antipathy." (Malinowski, 1923, p.314ff.)
- 5 Jakobson in a lecture given in Louvain in 1972; cited by Holenstein, 1975, p.34.
- 6 The acceptability of sentences like 'That is beautiful, but it doesn't please me' is a favourite topic of linguistic analytic aesthetics: cf. the discussion in Strube, 1980. Incompatible with Jauss' aesthetic experience are utterances such as, 'Oui, cela est parfaitement beau, mais il me fait bâllir' (Madame de Longueville); or, 'I admit that Raphael is a great painter but I do not like his work; it does not move me' (M. Macdonald). - Precisely such a running together of the pleasurable and the unpleasurable is characteristic of 'new art': 'The impression which is made by living art is...rather that of something hard and strict...than of something elegant and smooth - it distinguishes itself sharply from the conventional and from the agreeable comfortableness of the artistic expression with which one is familiar.' (F.X.Šalda, cited by Mukařovský, 1936, p.46). A running together of this kind is especially typical of those cases where beauty is sought for dialectically in something that is prima facie ugly.
- 7 "Partout on voit des machines qui servent à produire quelque chose et qui le produisent admirablement, avec pureté" (Le Corbusier, 1928, p.233). "Architectural functionalism proceeds from the premise that a building has a single, precisely delimited function given by the purpose for which it is built. Hence Corbusier's well-known comparison of a building to a machine, a typically unambiguous product from a functional point of view." (Mukařovský, 1942, p.37).
- 8 In industry too monofunctionalism is something infantile, characteristic of an early phase of industrialisation. Machines are, it is true, monofunctional (and with advantage); the energy with which they are driven however, and the raw materials which they transform, are both distinctively polyfunctional. The failure to take account of this fact leads to uneconomic wear and tear which does justice neither to the material of the economy nor to its needs.
- 9 "...For this reason the same property which makes a building beautiful also contributes to its being good (to its bonitas), and a face would also have to have for its purpose no other form than for its beauty." This sentence, from Kant's Nachlass (1923, p.628), may serve as a guideline for this section and especially as a historical support.
- 10 Bock (1978, pp.186ff.) distinguishes a threefold semantic effect of context. It works in an integrative way - the context clarifies or reveals from the first the connection between individual parts of the text which, without this additional information, remain unintelligible. It works in a selective way - the context determines which of the possible meanings of a linguistic unity become actualised. It operates in such a way as to expand the meaning, activating potential connotations which without it would have remained unremarked. An example of the integrative effect: 'Franz, enraged, is tearing the mirror to pieces.' A mirror is not torn but rather broken into pieces. The sensefullness of the connection between 'tearing to pieces' and 'mirror' is revealed only with the addition of the sentence: 'The shreds of paper flew about wildly in the wind', which tells us that we have to deal with a London newspaper and not with an article of bathroom furniture. The not wholly adequate example, which I use for the sake of its brevity, derives from me, and not from Bock. In itself it serves as a more suitable example for the selective effect. The context, the verb and the second sentence, indicate which of the possible meanings of 'mirror/Mirror' is to be selected. Of quite special importance for poetics is the third effect, that of extension of meaning. Bock and his co-workers discovered that plurivocity of textual passages is most often realised where the context - the relevant experiments deal with visual contexts - is also plurivocal. Thus the ambiguity of ’Dorfkapelle' (village chapel/village band) was most frequently noticed where the accompanying image showed a man by the side of the group of musicians and in front of a small country church. The different meanings of a term are not, as a rule, equal in value. The less common remain latent in the absence of an appropriately stimulating context. The sense-modificatory effects of the changing context surrounding a work is the subject matter of the theory of (literary) reception (Rezeptionsästhetik).