All too often it seems as if for a composer the text is only a pretext for an already predetermined compositional conceit, and not a challenge to question his or her own means by actually working with the text. In most cases the question is never raised whether singing is at all necessary; however, between opera and drama we could conceive of scenic works with forms of composed language which are still little explored. Many composers have no real interest in the spoken text, and the sung text often degenerates into merely creating an atmospheric mood. Yet, literary text not only serves as food for the voices of singer and actor. For me, the material quality of the text, and the demand it poses to make this quality transparent through musical means, is of much greater importance. This may sound more academic than the actual compositional result proves to be: I am equally interested in the paper quality of literature. To be clear from the outset: in general, in my 'audio plays', scenic concerts and music theatre pieces I work with non-dramatic texts, texts that were not explicitly written for the (acoustic) stage. On the one hand, I find it almost impossible to read dramatic literature (theatre plays), yet, on the other, I can hardly imagine a drama as a written text when heard on stage, totally dissolved into the plausibility of character discourse. I imagine something different: I do not want to disembody the text (I know too many dusty performances of 'New Music' where this happens as often as its opposite, the disappearance of the writing). I imagine adding another dimension to the physical dimension of the text (i.e. the sound and volume of a voice), one that is dedicated to the pleasure of reading. I want to draw attention to the physical shape in which a text appears on the page, for example, preferably in its original form and not in one of the cleaned-up editions of many publishers. (The distorting interventions into original texts, to which the editors of more recent editions of Hoelderlin, Kleist and Kafka have borne witness, are truly shocking.) Is it prose? Can the punctuation marks tell us something? Or particular spellings? Capital letters? Breaking of lines? Where are the paragraphs situated in the original, where did the author want to write through, where does the typeface allow a glimpse of the architecture of a text (behind the scenes, so to speak)? These are methods which go beyond the semantic reading of a text in order to reveal clues about a deeper structural stratum of literature. A musical composition can attempt to incorporate this written quality of a text in order to prevent reducing the wealth of experience that the text allows to a single level of acoustic performance. Quite consciously therefore, I read a text not primarily in order to reveal its contents, semantics or interpretation. One could be tempted to say that I read it from a purely formal standpoint, were not the contents always the basis on which I make my preliminary selection of the texts I want to work with. What does 'formal' mean in this context? During my work on 'The Man in the Elevator' by Heiner Mueller,' for instance, it was only when I looked at the English translation of the text lying upside-down on my desk and noticed the many capital 'I's standing out amongst the small English typeface, that I became aware of the 'I's at the beginning of each sentence. From this I composed the accents on the first person at the start of each line which became characteristic of the opening sequence of this audio play and scenic concert:

I am standing among men who are strangers to me ... I am dressed like an office clerk or a worker on a Sunday. I have even put on a tie, my collar rubs against my neck. I am sweating. When I move my head, the collar constricts my throat. I have been summoned to the boss....
(Heiner Mueller 1984: 93)

In the German original, my view of this obvious sequence was obscured by the story it told. A little later in the text I noticed the conjunctions 'and', 'or', 'but' which structure the fantasies of the clerk on his way to the boss. I had them shouted, i.e. emphasised in dynamics and pitch, and, later on, even added to by the German voice of the clerk, completely detached from the flow of the narration, so as to hint at the command character of its structure:

Schnell ueberdenke ich meine Lage: ich kann beim naechsten moeglichen Halt aussteigen und die Treppe hinunterlaufen, drei Stufen auf einmal, bis zur vierten Etage. Wenn es die falsche Etage ist, bedeutet das natuerlich einen vielleicht uneinholbaren Zeitverlust. Ich kann bis zur zwanzigsten Etage weiterfahren und, wenn sich das Buero des Chefs dort nicht befindet, zurueck in die vierte Etage, vorausgesetzt der Fahrstuhl faellt nicht aus, oder die Treppe hinunterlaufen (drei Stufen auf einmal), wobei ich mir die Beine brechen kann oder den Hals, gerade weil ich es eilig habe. Ich sehe mich schon auf einer Bahre ausgestreckt, die auf meinen Wunsch in das Buero des Chefs getragen und vor seinem Schreibtisch aufgestellt wird, immer noch dienstbereit und, oder, aber nicht mehr tauglich.

'Formal' means that in my audio play The Liberation of Prometheus (again based on a text by Heiner Mueller)' I took the longest sentence of the text - the one which most clearly describes the constantly frustrated efforts of Herakles to approach Prometheus whom he must liberate - and I emphasised the conjunctions (so dass, als nach, zwar schon, aber, immer wieder, so dass) in my composition to make them transparent:

Der Kot war seine Nahrung. Er gab ihn, verwandelt in eigenen Kot, an den Stein unter sich weiter, so dass als nach dreitausend Jahren Herakles sein Befreier das menschenleere Gebirge erstieg, er den Gefesselten zwar schon aus grosser Entfernung ausmachen konnte, weissschimmernd von Vogelkot, aber, zurueckgeworfen immer wieder von der Mauer aus Gestank, weitere dreitausend Jahre das Massiv umkreiste, wahrend der Hundskoepfige weiter die Leber des Gefesselten ass und ihn mit seinem Kot ernaehrte, so dass der Gestank zunahm in dem gleichen Mass wie der Befreier sich an ihn gewoehnte.

'Formal' means that I composed the culmination of the drama in Mueller's text 'Herakles 2 oder Die Hydra' with the shortness of breath that the rhythm of the syllables suggests: this is the moment when Herakles discovers that the forest he is passing through to fight the Hydra and which has begun to hinder and attack him is itself the Hydra:

Il eut
du mal
à ne pas
crier.
Il se
lanca
dans une
rapide course
en avant,
hors de l'encerclement.

Il savait
qu'il n'avait
jamais couru
aussi vite.

Il ne s'éloigna
pas d'un pas,
la foret
fut aussi
rapide,

il resta
dans la pince
qui maintenant
se refermait
sur lui,
comprimant
ses viscères,
aplatissant
ses os,
combien de temps
pourrait-il
endurer
la pression,
et, dans
sa panique
montante,
il comprit:
la foret
était la bête,
depuis longtemps
déjà
la foret
qu'il avait cru
traverser
était la bête...

When the architecture of a text, its style, refers to and hints at its content, or contents, which is always the case with Heiner Mueller's texts, this procedure proves to be not merely a formal one. It reflects rhythmical, structural, architectural references. Beyond interpretational illustration, the compositional work enables these layers to become audible, the writerly strategy to become transparent and the 'text to be experienced. To treat the text us a landscape means not to pass through it superficially in the manner of a tourist or, to remain in the picture, to grab hold of it from inside a moving car, but to travel through it like an expedition. Or to look at the text, in the words of Walter Benjamin, as a 'forest in which the reader is the hunter'; or, following Soeren Kierkegaard, as an 'impenetrable virgin forest' - a forest where, with a side-swipe, the critics are like 'swarms of wild animals, which one must keep at bay with all sorts. of noise-makers... The very best, perhaps, would be if one could do with critics as is done with rats: train one to bite the other' (Kierkegaard 1967: 51).
An expedition does not come up with results in the case of all authors. It is a procedure that needs the luxury of brief texts because it practically reads and composes with a magnifying glass. You cannot do this kind of thing with a five-act drama of four hours. I usually take about forty-five minutes for each page. It works with authors like Buechner, Kleist, Kafka, Mueller - other attempts were successful with Ponge. But almost never with poetry: there the whole sound is already pre-composed by the poet and leaves nothing for the musician to do, unless you want to reiterate. The quality of a text as libretto is decided at the point where it makes musical offerings on a rhythmical, tonal, structural level, yet remains strong enough to sustain the application of musical devices. The 'view' of a text can be taken quite literally. I look at the text, sentence by sentence, searching for its characteristics, starting with simple questions about its length, its complexity, its tone colour, its accumulation of words or letters. I remember a rehearsal with director Ruth Berghaus. We got stuck with the heckling of a 'rose maiden' in Kleist's play Penthesilea.' Suddenly it sprang to my eye, beyond the content, that the short text of the girl contained seven syllables that started with the letter 'z'. From this Ruth Berghaus created the energy of the entire scene:

Zukuenftig, wenn, beim Zimbelnschlag, von neuem
Das Amazonenheer ins Schlachtfeld rueckt,
Ziehn wir zwar mit, doch
nicht mehr, das versprichst du,
Durch Rosenpfluecken bloss und Kraenzewinden
Den Sieg der Muetter zu verherrlichen.
Sieh, dieser Arm, er schwingt den Wurfspiess schon,
Und sausend trifft die Schleuder mir das Ziel.

I found a similar accumulation of letters in a text by Poe (The Narrative of A. Gordon Pym) which Heiner Mueller adapted. I asked actor David Bennent to elaborate them vocally and musically:

Wir treiben mit
zunehmender
Geschwindigkeit auf die
Nebelwand zu manchmal
reisst die Nebelwand und
wir blicken in einen
Wirbel aus flackernden
Bildern wie Fetzen von
Fotografien im Feuer.

My involvement as a composer in another production consisted mainly in exploring the entire text of Georg Buechner's Dantons Tod (Danton's Death) to reveal its poetic structures and visualise them with the help of a computer. This was done by simply dividing lines at their punctuation marks, emphasising sequences, etc. Following Buechner who once said that 'every comma is a stroke of a sabre and every full stop is a head cut off', I later attempted to make the punctuation audible.
All these examples are of an intentionally exaggerated, almost anecdotal nature, and are meant to illustrate my attempts to take the formal offerings which a text makes seriously. This involves not approaching texts with a predetermined concept ('I am still looking for something for voice and piano') but deciding on the instrumentation and the musical form after an analysis of the text.
The expedition into the landscape of text can be of a sociological, literary, or archaeological nature. This depends not only on the antiquity of the author. For me, a complex experience of this kind was the preliminary work on the audio play Shadow/Landscape with Argonauts after Poe and Mueller.' I asked about a hundred people on the streets of Boston to read the text aloud. In the sound of their pronunciation, the passers-by, all totally unprepared for this, gave me not only an idea about biographical information, such as class and education, but also an impression of the ethnic fabric of Boston with its Irish, Russian, Italian, Asian immigrants. Not only did I get to know the text in all its possible meanings, but with the help of the recordings I can convey these to the listener as well. Quite often the understanding of the text is assisted by an error made by the reader which is corrected either by himself or herself, or by the listener. I believe listening to be a process that makes something of the written quality of the text transparent and that is therefore comparable with the process of reading - going forwards and backwards, misreading and noticing the errors, discovering a word in other words. When I stated at the beginning that texts don't interest me as mere food for singers and actors, I actually meant the opposite: on stage I try to blur or even break the identity between speech and speaker in order to make the 'speaker' disappear. There are two reasons for this: first, to rescue the language, to develop the hearing of language independently, and second, to acquire an actor who can not only physically elaborate what he or she has already said but who can present himself or herself as an independent body - to arrive finally at having two bodies: the text as a body and the body of the actor. This can be achieved by using radio-mikes on stage, but also by strict stage directions such as: do not reiterate the movement of the text with body movements or with gestures of the hand, do not echo the contents with the enunciation by saying, for instance, 'bbbloodd' and 'sssworddd' instead of 'blood' and 'sword' to make it sound more violent. It is a common mistake amongst actors not to trust an understanding of words but to believe that through over-illustrating the sounds the narrative will be somewhat enriched. Heiner Mueller once stated in a discussion with Robert Wilson:

'[I]n German theatre, text is not accepted as a reality, it is only employed to make statements about reality. This is a degradation of the text.... The theatre is seen as a mere surrogate and not recognised in its vital function as a reality, as a part of life.'

In another discussion he added, alluding to a performance by Joseph Beuys in New York:

'This for me is the ideal metaphor for the way in which the actor handles the text - the text is the coyote. And you never know how he/it is going to behave. But how do I tell this to an actor who is used to handling the text like an office clerk, at best administering the text.'

I need to add from my perspective that there have been quite a few occasions where I would have preferred to have seen a more indifferent ,administration' of text, its distribution to the audience from behind a 'counter', rather than the enthusiastic way in which an actor often takes possession of the text and monopolises its experience, thereby preventing its distribution to the audience.

(Translated by Heike Roms)