Every language has 'movements' and 'rests', giving rhythm to the discourse. The 'pace' of a language is given by its conceptual nature. The organization of thought can occur along two semiotic lines that we might define - to borrow from the terminology of music - as 'melodic' and 'harmonic'. The former is horizontal, and seeks continuity, polished intonation, fluidity of expression. In the melodic, 'thought' and 'word' come together. It is structured centrifugally. Punctuation serves to imitate the strategy of spoken discourse. In 'melodic' languages, the conclusive definition of the expressive act is gestural mimicry. English, French and Italian all belong to this first group. Not one ancient language belongs. Latin has a centripetally organized structure of words that is entirely different, and in this light the so-called 'neo-latin' languages can be defined as neo-latin, in reality, only allowing considerable abstractions. Saxon languages are typified by a vertical, centripetal structure that creates a whole series of differences, giving them many codes with an incomparable semantic aftertaste. The effectiveness of those compound words so beloved of the German language rests on the fact that, within their compass, they remove the distinction between subject, predicate and complement, so that in these words it is as if the quality of a real or imagined being - its 'character' - takes on a non-representational aspect, reflecting not the consequence of an action, but the nature of the process by which the action is brought about. In Saxon languages things are not brought about: they come about. The semanteme Gesängegives the notion of something that was sung, at some earlier time, and has therefore become 'song': there is the idea of an epic, a tradition that sustains the song of the single individual. Similarly, inUnsichtbaren, the idea of the 'immeasurable' evokes the prospect of a being unable to gather itself, much less all else, within the limits of its own consciousness.

We know thatUrteil is 'judgement'; but what is impossible to express is the sensation, even before the idea, of a fragment turned up from the past, of an almost biological archetype, that this word triggers in the psyche of a German. The vaguely sinister aftertaste of unreliability, of both logical and ethical ambivalence, of this 'judgement', makesUrteilsomething almost opposed to the solid immanentist origin that the word assumes in neo-latin languages. In this context, elision takes on the character of a cross between denotation and connotation. The beginning of the Goethe's poem The King of Thule, "Es war ein König in Thule", has the fixity of madness: that initial Es is a monolith at the limit of conscious reason. By no accident, Es will be the step-sister not invited to the ball of the Ego and the Superego, the wedding of psychoanalysis and linguistics. An Italian translation regarded by now as a 'classic' commits the error of attributing to this passage the flavour of a fairy tale: "C'era un re / Un re di Thule"(There was once a king / A king of Thule), which, refuting the hermetic power of elision, looses the chain of the 'side effects' and opts for a quite definite field: "Once upon a time, in Thule", to paraphrase the beginning of Carlo Gozzi'sTurandot. But, if Goethe had wanted to allude to the end of time? If this king were the one, solitary man remaining on Earth and his power therefore redundant, a practical joke of nature which that elliptical incipit seeks to disguise with the appearance of sense? Thule rhymes withBühle; the death of the loved one has robbed the king of his only connection with sense. Thule is madness; and that lonely madman on the stumbling block of senselessness believes himself to be a king. This is why, in Saxon culture, fairy-tales are the realm of nightmare: the archetypal fixity of symbols, typifying the 'wild' fancies of childhood, enables the language of hard fact, the elliptic monumentalism of Saxon languages, to release all of its power.

Faced with this complex mesh of problems, one needs the courage to observe and retain the monolithic character of compound words: that process whereby, in German, matter thinks, and each reconstruction of terms in a new semanteme has the attraction of novelty. Thus, "ich blickte nach Osten, befreien" implies the Eastern horizon as a perception, but also as an existential perspective; at the same time however, it alludes to something that, in the West, prevents me from looking at the horizon. If I go East from this wall, I am 'liberated', 'befreien': but this liberation comes only 'nach', as the consequence of a limitation endured and perhaps accepted. So, the most apt translation would be: "Seguo con lo sguardo il paesaggio d'Oriente, i suoi liberi profili" (My gaze follows the Eastern landscape, its free outlines), attributing to the landscape that which would appear to be the character of the subject; almost a destiny, of that subject whose contemplation of the horizon becomes introspection. This simple and terribly elliptical phrase is the opening of a recently published novel: the choice of field made by the translator will influence our perception of the protagonist, his exact topography in the setting: the relationship between space, time and consciousness in the entire interior vision the reader will have of the ensuing narrative. In cases like these, the problem is not one of language but of imagination. It is in the mind of the translator that ambiguities must be analyzed, resolved and made to assume accessible dimensions, whereviluppobecomessviluppo, complexities are unravelled, and the aporiae of these same ambiguities are then recombined in a further complex expression ('semanteme') able to generate explosive implications of opposites to the same degree: a linguistic process based on a culturalUrteilultimately producing an estrangement, a polarization of meaning and significance, in the mind of the reader. One can appreciate to what extent the consciousness of the translator, in this game of mirors, where that which is convex for the writer becomes concave for the reader, should feel like the face of Escher, encapsulated in those glass spheres onto which the Dutch artist projected the perspectives of his recurring worlds, where only Achilles and the Tortoise can feel at home, racing toward their infinitely unreachable finishing line.

The greater problem, then, is that of retaining semantic density, the effect and at once the expression of an exact dramaturgical design. Another distinction of merit that might be useful to this end is that which distinguishes 'exclusive' languages from 'explicative' languages. In ancient Greek, the propositional particles with which a phrase begins - 'now', 'in fact', 'because' - might appear explicative, at the beginning of a discourse of which the elements have not yet been expounded. Likewise in Latin, 'nam' often has this type of function. In reality, these are allusions to an argumentative dimension, which, insofar as it proceeds from the analytical consciousness of the writer, is considered as an antecedent to the declaration proper. Old English also offers many examples of this strategy, represented in particular by the 'for' with which certain of Shakespeare's 'moral' statements begin ('und' has the same role in the Lutheran Bible). In this case the elision is obtained, paradoxically, through a redundance of expressive media employed. In 'explicative' languages, on the other hand, articulation of the discourse always proceeds from an exact spatio- temporal perception. The cause of this distinction is found in the greater or lesser influence that the conception of verbs in ancient languages has had on modern languages. Verbs do not have tenses as such in ancient languages, but simply 'aspects': they indicate the impact of a past action on present events. The Greek Aorist and the Latin Perfect signify that one event underlies a succession of events proceeding from it, logically and unfailingly, in the present. It is therefore inevitable that languages like German, and English to some extent, are typified by a higher incidence of elision than the so-called neo-latin languages which, in effect, have gone against this attribute of their mother tongues.

Just so, in French, the subject that performs an action is inserted as an epexegetic particle within the action described. Subtending the discourse, in French, there is always a question as to who performs that given action, described in its endlessly temporal immanence. This is why Verlaine could say: "de la musique avant tout chose" (music above all); because in French the subject of the factual situation is deduced, and not imposed apodictically, as is the case, by contrast, in English. When Baudelaire writes: "Les sons et le parfums tournent dans l'air du soir" (sounds and scents swirl, spin, etc., in the evening air), the translation of this phrase into more 'vertical', more 'exclusive' languages must concentrate on the notion of whirling and eddying, seen as a perceptive totality, rather than on the description of an event. Here in fact, Baudelaire wants to eliminate as far as possible the point of view, the perception, that one single onlooker may have of a phenomenon seen as an alternative to human consciousness. The French language is the language of hazard. Translating Mallarmé, in this sense, becomes a war with empty space; a building of elision strong in its reasoning preciously by virtue of the absence of any point of view within the discourse. Rimbaud's affirmation: "Je est un autre", concisely indicates the degree to which any poetic discourse becomes untranslatable when expressed in a self-regarding language like French. A term coined by Alfred Jarry sums up this entire argument: "pataphysics": this is a discipline running 'alongside' Physics, not 'beyond' it, as metaphysics would be, etymologically speaking. In pataphysics language dislocates itself; signifies only itself. Jarry's doctor Faustroll is the elliptical translator par excellence, according to a lesson which Queneau, in hisExercices de Style, has erected systematically. Elision, in this superb little opus, celebrates its own triumph. The ninety-nine different ways of telling about a meeting, trivial enough in itself, become a philosophical investigation of how our presumption of meaning, as readers, may produce the meaning of what we are reading. Only the 'variation' in mathematical language, variation in interjections, and variation in algebraic 'style' remain universal. For a literary translator wanting to experience the problem of elision at first hand before seeking to understand it, an attempt at translation of theExercicesis an absolute must. Alfred Jarry, in his Exploits & Opinions of Doctor Faustroll, Pataphysician, wittily condenses the entire question thus: "An epiphenomenon is that which is superinduced upon a phenomenon". I do not know any better statement of respect for the power of elision, in a literary translation.

In effect, if we define an epiphenomenon as the process of consciousness which, like a ground bass, precedes and accompanies the description of the event, we will understand that every literary work departs from a presumption of meaning that involves, a priori, an elliptical process. So, like Jarry's pataphysics, translation is the "science of the exact detail in an imaginary context". And this is what happens as a result of 'degeneration', from the 'centripetal' world of ancient languages to the 'centrifugal' world of modern languages. Whilst the Latins said "tenet res, verba sequentur" (if you know what you want to say, the words will come naturally), the very opportunity to create poetry in modern languages comes from the opposite notion: the res are connoted by theverba. Which is why modern poetry is first and foremost the poetry of ideas, and therefore highly elliptical. Heidegger refers to this whole process asHolzweg: the path closed off, brought to a dead end, lost in the woods. An elliptically effective expression quite impossible to translate into any other language, more's the pity.