Lost in Transformation

Composer as translator

by Benjamin O’Brien

Problems with Translation in Music

As a composer working in the medium of electroacoustic music, I am confronted with the double task of identifying sounds that interest me and developing tools to impose on them for some affective result. After selecting an intriguing sound, I favour applying procedures that underscore my attraction. In the process of rendering my target sound, peripheral sonic elements inherent to the source may be reduced or removed. One may argue that this practice is akin to a translator who determines the essential meaning of a source message, removes idioms or edits syntax, and communicates it in a target language. Others may adopt the position that a composer is merely a language creator who recognizes sound primed for treatment — nothing more. My goal is to instil in the reader a clear understanding of translation relative to music for the purpose of proposing it as a specialized, ontological branch of transformation.

To illustrate the principal problems of intention and reception as they relate to translation in music, I turn to select works by Ludwig van Beethoven and Olivier Messiaen. Beethoven’s Symphony No. 6 (1808), commonly known as the Pastoral Symphony, famously has the inscriptions “Nachtigall” (“nightingale”), “Wachtel” (“quail”) and “Kukuk” (“cuckoo”) in the second movement “Scene by the brook” (Andante molto mosso) in the flute (measure 124), oboe (125) and clarinet parts (126), respectfully. Traditionally these markings indicate the composer’s inspiration and musical direction for the performer to realize. Yet despite this overt symbolism, several questions emerge that concern the listener: does the listener actually hear the instruments performing birdcalls? Or, is Beethoven merely driven by his muse, a fact that may be lost on the listener? Alternatively, some scholars argue that the text is a commentary on the practice of tone painting (Will 1997). But this assertion challenges those who do hear the real or imagined — in the case of Schindler’s identification of the phantasmic Goldhammer (yellowhammer) birdcall in the work (Schindler and MacArdle 1966) — birdcalls as being accurate (Grove 1892). Whether Beethoven intended to make an accurate depiction of the birdcalls or merely offered a commentary on tone painting remains unclear. That being the case, if Beethoven did not intentionally map the birdcalls to instruments for the purposes of “accuracy”, his process cannot be considered translational.

Olivier Messiaen’s works Réveil des oiseaux (1953) and Oiseaux exotiques (1955–56), both composed for solo piano and orchestra, address how different processes of transcription alter our music consumption. For Réveil, Messiaen ventured into the woods, documented birdsongs, and implemented his transcriptions directly into the composition (Hill and Simeone 2005). While the aim of Réveil was to present the birdsong material through the medium of piano and orchestra with little to no deviation from the source, Oiseaux focused on the juxtaposition of different species of exotic birds (Hill and Simeone 2007). Not being burdened by having to experience the birds in their natural environment, Messiaen fabricated artificial interactions between transcontinental birds using his own transcriptions of an anthology of six 78-rpm recordings, American Bird Songs, released by Cornell University in 1942 (Ibid.). While the issue of authenticity may again be lost on listeners, there is a notable discrepancy between the works: Réveil is composed of real, localized bird interactions as experienced by the composer, whereas Oiseaux documents impossible avian encounters constructed from third-party recordings. In addition, we must also inquire about issues of fidelity. Many have compared Messiaen’s transcriptions to their source materials and have concluded them as modest (Hold 1971). Knowing that the composer cannot escape the nature of human error, I believe it is Messiaen’s inaccuracies or differences in capturing the birdsongs that define the work’s meaning. Thus, meaning is about difference. However, if translation is possible in music, at what threshold do these inaccuracies undermine a translation?

Music, Language and Translation

The issues of meaning in music, and fidelity in a translation, conjure up connections between language and music. Significant cognitive research has been conducted to prove links exist between the two, such as how linguistic and musical information is stored and accessed in the brain (Jackendoff 2009). While some maintain comparative studies with cognitive linguistics are a valuable method in finding commonalities, others argue this research superimposes principles and foundations from the study of linguistics — something inherently not musical and foreign to the totality of music — onto the practice of music (Feld and Fox 1994). The only way for a listener to retroactively identify sound as having characteristics intrinsic to linguistics is to simultaneously impose musical meaning. However, even if a priori musical meaning is constructed so as to map properties of linguistics, a problem still arises: there is no denotative meaning associated with the sonic form. Thus, we are left with Messiaen’s birdcall. But this categorization and implementation of vocabularies for the sake of expression are still common to language and music, creating a semblance between the two.

Philosopher Theodor Adorno believed that music “resembles” language, as particular musical codes such as cadence structures may be preferred over others (Adorno 1998). Such musical choices only exist in the nature of their creation and not outside. Both language and music relate to the issue of intention, however; their individual relationships with the “absolute”, or truth, are orthogonal. While words are limited in their attempt to capture truth, music is truth and is limited in its naming (Ibid.). Given the context of a music composition, Messiaen’s Réveil for example, where sonic properties are assigned particular values, the following question arises: is this value translatable or is it just subject to intention? If one adopts the latter position, then the composer is a language creator who chooses material based on subjectively assigned meaning and expresses it, be that algorithmically, experimentally or indeterminately. This intentionality is unexpected and without pattern. Thus, we can now deduce that translation is a process available for systemization.

A Cybernetic Explanation of Translation

Translation: Definition and Distinction

I define translation as the systematic mapping of a source input to a target output formed within a similar medium such that an individual separate from the system recognizes the preservation of source patterns despite changes to that which carries it. Through the compositional analysis and the observation of contextual discrepancies in the implementation of translation, we understand translation as systemic. Being able to understand the properties of said system assists in bettering our understanding of translation in music.

For philosophers Humberto Maturana and Francisco Varela, translation is not the method of reinterpreting a source message for a target, but rather the method of translation is the message (Maturana and Varela 1987). The success of the translation lays in the ability to define “what is meaningful” and “how does one search for that meaning.” How meaning is encoded and decoded is central to the process of communication. In other words, with respect to the issue of translation in music, just because a composer wants an audience to hear the mapping of birdsongs via an orchestra does not guarantee that listeners will hear birdsongs. As semiotician Jean-Jacques Nattiez points out, the poietic process and the esthesic process do not always correspond; a common connection between message producer, or sender, and receiver must exist, enabling the act of communication (Nattiez 1990).

This process of encoding and decoding is not without numerous technical, semantic and effective problems (Weaver 1949). Ignoring for the moment technical and semantic issues, one can certainly argue that music has the potential to affect us. But how does one gauge music’s effectiveness? To improve this quality, a transmission may have to be adjusted to optimize conditions concerning the sender and receiver, which of course may provoke various technical problems. Depending on the source material, further alterations may be necessary for the receiver to comprehend the decoded information. The key issue here is that alterations in encoding and decoding information yield more noise, which interrupts the signal flow and could potentially lead to miscommunication.

Noise is any interference that disturbs transmission signals (Ashby 1970). Relative to the process of translation, noise can result in a receiver’s inability to discern the exchange of information from sender to receiver, thus obfuscating a receiver’s understanding of the source meaning. Subsequently, the receiver’s misunderstanding may affect and alter the state of system. While noise threatens to destroy the structural integrity of a system, it also tests its technical strength — a good thing! The generation of noise forces systems to adapt, resulting in new meaning for the system and its products (Bateson 1967).

Music — A Problematic Source Material

Understanding music in terms of signals and noise within a system yields two unavoidable, interrelated problems. First, music is two entities: an open set that contains an infinite number of elements and the individual elements within the set. The set is infinite, as music includes all sounds known and unknown to a listener. This creates a metaphorical fissure between our epistemological and ontological understanding of music. Our absence of knowing a sound is something that must be acknowledged, however, it cannot be accommodated, categorized and systematized. If we are to understand music as a system, we must be able to account for all components within the system that distinguish it from its background. Since we can only retroactively identify an element as being musical after its existence is known, we have no indication as to whether it belongs in the system or its environment.

The second complication is that it is impossible to claim music has meaning or that musical truths exist. This complication stems from our first problem since our inability to apply value to an element in a musical set derives from our inability to name elements as being musical. It is important to note how this affects our cybernetic explanation of the musical process: if we have no value system for musical elements, then it is impossible to encode said values and translate their meaning. Music, of course, does not exist in a vacuum, but rather in the context of its actualization.

Music Devoid of Context

Functioning much like noise in a system, context reveals the structural qualities of environmental elements via the calculable differences that emerge as a result of time and situation. However, due to its temporal reliance, context cannot be described in terms of mathematics. Cyberneticist Gregory Bateson believed context and meaning to be irrevocably linked and equally difficult to define (Bateson and Brockman 1977). Thus, we are left with a Borromean knot between meaning, difference and context: meaning is connected to difference; difference is determined by context; and context is the site where circumstances and elements are assigned meaning. If context has the potential to change the structural meaning of environmental restraints, then it will be impossible to apply translational procedures since the musical values are subject to constant alteration. Thus, it is in our best interest to theorize a method of divorcing music from context.

To help us determine whether the process of translation is possible in music, the work of philosopher Slavoj Žižek serves as insight into understanding music devoid of context. Žižek’s paradox of self-identity illuminates how an entity’s meaning is not self-reflective (Žižek 2002). Continuing with his analysis of Hegel’s “a is a” and “a is not non-a”, Žižek demonstrates how the subject fails to competently complete the very goal established at the origin of its invention: it is without substance (Ibid.). Thus, the subject’s existence rests in a gap between our symbolic and imagined realities. This gap, or void, provides an opportunity for us to establish music as an ontology: the absence of a void is the presence of being.

This logic aligns with philosopher Alain Badiou’s theory of inæsthetics (Badiou and Toscano 2005), which provides a foundational logic for understanding music separate from its context. Badiou arrives at his definition of inæsthetics by first distinguishing it from the field of æsthetics. He argues that rather than the calculation of situational differences that affect a systemic structure, truth in a work of art is evaluated internally: meaning is intrinsic to the work itself and lies in its composition (Ibid.). Thus, the truth of a work is strictly individual and cannot be privileged over other works. If we claim music as an ontology, then we consider the instance of music composition to be what Badiou defines as an “event”: composed of situated elements and “itself”, it rests between the void and “itself” (Badiou 2005). To be clear, music composition is the self-referencing organization of known sound features.

Adopting the Badiouan idea of the event to explain music composition reveals certain truths, which stem from the site, or evental site, of its emergence. First, the existence of the event of music composition exposes the void that confirms music as an ontology: music either exists or it does not exist — there is no in-between. Second, in the moment of its realization, the event of music composition confirms the status of music being the set of all sounds known and not-yet-known. Since this set is subject to multiple readings, or Badiou’s term the “multiple”, we embrace the conceit of music as an ontology; it is an open set that includes all known sounds and the empty set of sounds not-yet-known. Thus, our unique associations with sound constantly revise our truths about the set of music. The status of music is updated, for example, when previously not-yet-heard sounds are discovered in the event of a music composition, thus transforming into known elements in the set of music. But we must now ask: is translation possible in this understanding of music?

In reaction to the music composition event, if an individual — composer or listener — claims to translate known sonic elements within the set of music, this event exists due to the transitive nature of sets: elements “belong” to the set and are “included” in their subsets (Ibid.). The evental site of translation yields links between subset elements, such as pitch or timbre. However, we can infer that the retroactive identification of elements within subsets available for translation is merely a hermeneutic approach to understanding the multiple. That we attempt to leverage our actions in favour of others asserts that another does not know, and as a result of their not knowing, we do not know. In addition, the temporal nature of music and our continual re-visitation of the music composition event further complicates this unknowing. However, an individual’s act of linking elemental groupings in music as actualized by the event of music composition in fact constitutes an event. If this is the case — that the act of translating is itself an event — then each evental site of translation is valid and available to the multiple. Relative to the musical practice, there are multiple modes of translation due to one’s individualistic criteria for categorization as subject to the multiple. Thus, a claim for a theory of translational modes is valid and worthy of further discourse.

A Theory of Translational Modes

Operation Definitions

My theory of translation modes proposes a theoretical framework for identifying translational operations. If we recall that an individual’s evental site of music is an open set composed of the empty set of all sounds not-yet-known and known sonic elements and subsets, translational modes define the former as signal elements and the latter as symbolic sets. To be more precise, a signal element is any identifiable sonic source material, such as a birdsong or a baby crying. Of course individual signal elements are composed of many different analysable component values relative to time, such as frequency — these components are symbolic sets. We can observe that symbolic sets combine to create the character of signal sound. Translational modes state six operations available to identified signal elements and symbolic sets: substitution, communicative, distribution, implication, exportation and absorption. The core operations of translational modes described here are based on foundations of propositional calculus.

Signal Operations

• Substitution — Replace signal A with an equal signal B, such that symbolic set a of signal A is maintained. To determine if signal B is equal to signal A, find the intersection of the symbolic set a of signal A and the symbolic set b of signal B and determine if its strength is suitable.
• Communicative — Given the sequence signal A follows signal B, which follows signal C, reorder the sequence of signals A, B, C.
• Distribution — If symbolic set c is common to signals A, B, then signals A, B may be sequenced (see Communicative), such that symbolic set c is distributed over the sequence group containing signals A, B.
• Implication — If signal A implies signal B, then the individual may choose signal not-A or signal B.
• Exportation — If the intersection of symbolic set a of signal A and symbolic set b of signal B implies signal C, then signal A implies the implication of signal B implies signal C.
• Absorption — If signal A implies signal B, then signal A implies the intersection of symbolic set a of signal A and symbolic set b of signal B.

Symbolic Operations

• Substitution — Replace symbolic set a of signal A with equal symbolic set b such that signal A is maintained. To determine if symbolic set b is equal to symbolic set a, find the intersection of the symbolic set a of signal A and symbolic set b, and determine if its strength is suitable.
• Communicative — Given a sequence of elements in the symbolic set a, reorder the sequence of elements in symbolic set a.
• Distribution — If signal C is common to symbolic sets a, b of signal C, then symbolic sets a, b may be sequenced (see Communicative) such that signal C is distributed over the sequence group.
• Implication — If symbolic set a of signal A implies symbolic set b of signal A, then the individual may choose symbolic set not-a of signal A or symbolic set b of signal A.
• Exportation — (see Signal Operations: Exportation)
• Absorption — (see Signal Operations: Absorption)

Of course this list is a concise abstraction of the translational modes operations. The reader will better understand how these operations are employed in the context of my fixed-media composition OSCines (2013).

Operations in “OSCines”

Before outlining the specific signal and symbolic operations utilized in OSCines, I offer some background into my pre-compositional practice. Inspired by the works of Messiaen, my interest was to compose a piece that utilizes the nightingale’s rich birdsong as source material (Audio 1). The nightingale belongs to the clade Passeri also commonly known as oscine, from the Latin root “oscen” meaning “a songbird”. The birdsong is comprised of a wide range of whistles, trills and gurgles. Reviewing the sonograms of many nightingale birdsongs revealed sinusoidal groupings, or oscillations, which also inspired the naming of the piece. While some of the birdsongs are presented unaltered, I identified and extracted specific characteristics of the birdsong, such as pitch, timbre and sequence, and translated their value within the context of the composition.

The operations of translational modes are applied in various manners in OSCines. The substitutive operation appears numerous times in both the signal and symbolic domains. More specifically, to give a detailed description of its use, I tracked the amplitude, frequency and spectral centroid of a specific signal nightingale birdsong, grouped them into separate symbolic sets and, given pre-determined heuristics, I designated different signal instrument samples of similar symbolic sets for replacement. For example, if I select a common symbolic set of frequency, while the product’s melodic contour matches the birdsong signal, the sonic result is different. The success of this approach lies in the identification of an equal substitute. An implementation of the following substitutive operation was used in the composition of the work: given a signal “birdsong” (Audio 2; Figure 1) and its symbolic sets of amplitude and frequency values over time (Figure 2), a directory of signal piccolo samples are selected for playback to replace the birdsong with the same symbolic sets (Audio 3; Figure 3).

The communicative operator creates variation from an existing, limited sonic palette. In the symbolic domain, this means re-ordering values, such as an amplitude envelope, to alter the sonification of a signal. In the signal domain, this procedure may have a more dramatic effect: By dividing a signal birdsong into smaller portions (less than a fifth of a second) via finding the zero-crossing point, I construct new birdsongs through re-juxtaposing these parts (Audio 4–7).

While identifying commonalities within and among signal elements and symbolic sets provides the basis for applying translational modes operations, it is the actualization of these processes that dictate a listener’s recognition. In other words, the event of music composition yields a site for creativity. Here, I return to Messiaen’s identification of birdsongs carrying qualities of music as an example of my inspiration. So the success of a translational procedure begins in the selection source and target materials. My research into understanding translation in music encourages my compositional interests in the alignment and collision of distinct sonic features.

Bibliography

Adorno, Theodor W. “Music and Language: A Fragment.” Quasi Una Fantasia: Essays on modern music. Trans. Rodney Livingstone. London: Verso, 1998.

Ashby, William Ross. An Introduction to Cybernetics. London: Chapman and Hall, 1970.

Badiou, Alain. Being and Event. London: Continuum, 2005.

_____. Handbook of Inæsthetics. Trans. Alberto Toscano. Stanford CA: Stanford University Press, 2005.

Bateson, Gregory. “Cybernetic Explanation.” American Behavioral Scientist 10/8 (April 1967), pp. 29–32.

Bateson, Gregory and John Brockman. About Bateson: Essays on Gregory Bateson. New York NY: Dutton, 1977.

Feld, Steven and Aaron A. Fox. “Music and Language.” Annual Review of Anthropology 23 (October 1994), pp. 25–53.

Grove, George. “The Birds in the Pastoral Symphony.” The Musical Times and Singing Class Circular 33 (1892), pp. 14–15.

Hill, Peter and Nigel Simeone. Messiaen. New Haven CT: Yale University Press, 2005.

_____. Olivier Messiaen: Oiseaux exotiques. Aldershot: Ashgate, 2007.

Hold, Trevor. “Messiaen’s Birds.” Music and Letters 52/2 (April 1971), pp. 113–122.

Jackendoff, Ray. “Parallels and Nonparallels between Language and Music.” Music Perception 26/3 (February 2009), pp. 195–204.

Maturana, Humberto R. and Francisco J. Varela. The Tree of Knowledge: The biological roots of human understanding. Boston MA: New Science Library, 1987.

Nattiez, Jean-Jacques. Music and Discourse: Toward a semiology of music. Trans. Carolyn Abbate. Princeton NJ: Princeton University Press, 1990.

Schindler, Anton. Beethoven As I Knew Him: A Biography. Edited by Donald W. MacArdle. Trans. Constance S. Jolly. Chapel Hill NC: University of North Carolina Press, 1966.

Weaver, Warren. “Recent Contributions to the Mathematical Theory of Communication.” 1949. http://ada.evergreen.edu/~arunc/texts/cybernetics/weaver.pdf [Last accessed 4 October 2013]

Will, Richard. “Time, Morality and Humanity in Beethoven’s ‘Pastoral’ Symphony.” Journal of the American Musicological Society 50/2/3 (Summer-Fall 1997), pp. 271-329.

Žižek, Slavoj. For They Know Not What They Do: Enjoyment as a political factor. London: Verso, 2002.