downloaded from the web on dec 28 1995 David A. Jaffe's Home Page copyright 3039w

Impossible Animals: Notes on Birds and Musical Style, Perspectives of New Music (Winter 1995)



Copyright (c) D. A. Jaffe, 295 Purdue Ave., Kensington, CA. 94708

The visual arts would be greatly impoverished without birds and all that birds represent. Music, except for a few notable exceptions, has left bird song a largely untapped resource. I have been watching and studying birds for twenty years, nearly as long as I have been composing. During that time, I have been repeatedly drawn to the avian world as a source of musical inspiration.

This brief essay chronicles some of my shuttlings between the domains of music and nature, and looks closely at examples from three different genres: instrumental music, vocal music, and computer music. More generally, it illustrates the compositional strategy of starting from the known, in this case bird song and behavior, and abstracting to the unknown. The familiar, whether it be a musical style, a bird song or a computer-simulation of the human singing voice, is already something of great richness and character. The process of abstraction then involves combining several familiar elements in an unconventional manner, or stretching the familiar in strange directions or to unnatural proportions. The result can be something quite alien, but with a strong hauntingly-familiar identity, as if viewing a face from a long-forgotten childhood dream.

Use of bird song can be viewed as one example of "external" artistic inspiration that, far from the popular view of a passive experience that overcomes an artist, involves active molding and sculpting of raw material into something new. In taking such an inclusive approach, the artist has, quite literally, the world to gain. Paradoxically, when he or she allows private personal experience to find its way into music, the result can be more broadly relevant, compelling work.

Bird Song in Instrumental Music

But how to begin translating the twitters, trills and squawks of bird song into musical material? Clearly, a strict transcription would be musically irrelevant. A violinist's rendition of a Wood Thrush song could never match the original, and what would be the point? It seems that the secret to an effective use of bird song lies in the particular details of the translation, as the material travels through the filter of personal sensibility and emerges as human musical expression.

My own first clear use of bird song was in a work for two violins entitled Territory (1977). At the time, I was engrossed in a detailed breeding bird study within a quadrant of southern Vermont. While going out every morning at 4 AM to chronicle breeding behavior, searching for evidence of birds staking out and defending their territory, I became fascinated by the song of the Grey Catbird, a common breeder in that area. It was not specifically the song itself, but its quality, variety and rhythm that fired my imagination, suggesting a violin music with just such a jerky, abrupt, wheezy character. I attempted to convey this sense in the language of the violin, using a rich vocabulary of pizzicato, double stops, tremolo, ponticello and a combination of trills, shakes and sliding pitches.

The avian influence in Territory extended further. The large-scale formal structure for the piece was provided by the behavior of song birds such as the Catbird; the relationship of the two violinists' parts follows a hypothetical Catbird encounter, with one violin playing the role of a male, the other a female. The work begins competitively, just as a female who newly arrives on the territory of a male may first be taken for another male, as both violinists play similar material in contrasting counterpoint. Later, the tempo and energy of the music increases, symbolizing the increased intensity of the birds' competition. Then, just as a female Catbird will change her behavior to reveal her gender, thus transforming the male's aggression into courtship, one of the violinists begins a sweeter, more lyrical style of playing, provoking a gradual calming in the playing of the other violinist. Eventually, the two play a "love duet" in a harmonic texture. Near the end of the piece, the agitated music of the opening reasserts itself, symbolizing post-nesting return to non-nuptial behavior patterns. This leads to a final showdown, in which one violinist takes over entirely, playing a long cadenza, as if having won the song contest and driven off all competition.

This structure may be criticized as contrived, a kind of reverse anthropomorphism in which bird behavior is used as human musical structure. But notice that this musical structure has a dramatic logic that stands on its own, quite independent of any reference to birds. A traditional view of the musical form would outline a large-scale three-part form, with the center section contrasting in character. The piece would be described as a contrapuntal invention, a harmonic development of this material, and a recapitulation with elaboration. Still, if the bird associations are not necessary, why include them at all? The answer is that they provide both the composer and the listener with additional associations that serve as points of reference, act as guides through the work and help make a generic structure unique.

As an example of bird material in symphonic music, in Whoop For Your Life! (1987), the sound and plight of the endangered Whooping Crane form the basis of both the sonic and the affective quality of the work. Not only is the bird's sound rendered (albeit in a stylized manner) by French horns, tubas and trombones, but the entire emotional thrust of the piece came from a visit to the Aransas Wildlife Refuge on the Gulf Coast of Texas, where I saw Whooping Cranes going about their business as if they were nothing special, not teetering on the edge of extinction. To further symbolize this contradiction, the piece invokes the folk music of Brazil---vibrant, life-affirming music from a country where the rain forests are being destroyed at a catastrophic rate. This technique of using multiple, yet linked or contrasting associations, can lead to extraordinarily rich results. Music is uniquely suited to such expression, since several distinct points of view may be simultaneously presented with no loss in comprehensibility, thanks to the techniques of counterpoint (in the broadest sense of the term.)

Bird Song in Vocal Music

Vocal music with a text allows for a more literal influence of bird-derived material. Such an approach appears in Bird Seasons (1984), scored for four voices (soprano, alto, tenor and bass) and based on my own text. The four movements of the piece represent the seasons from a bird's point of view. The first movement, Winter Lullaby is a musical rendition of Great Horned Owl vocalizations (the familiar "hoot owl"), in a dark setting that equates winter, the time of little sun, with night. The owls appear singly, in harmony and in antiphonal relationships, eventually leading to a tongue-in-cheek English rendition of the Barred Owl call, "Who cooks for you, who cooks for y'all?," drawled in a Southern accent. Next comes Love Song for Spring, a courtship process described in a sexually-energized, excited manner. The text of this movement ("In all birds it is essential that male and female come together at some time for copulation...") was assembled from various ornithology texts. In Circle Dance for Summer, a great deal of activity goes on, with each singer having his or her own text and musical style. The soprano swoops and sails, like a swallow in flight, as she describes various flight characteristics ("...flies buoyantly, often reeling, with legs trailing behind...hovers, then drops on prey like a shrike...can capture in air and swallow whole an adult Dovkie...") Meanwhile, the alto rattles off a rapid-fire series of adjectives used in bird species names, mixing in some hunters' nicknames for birds, in the matter of an auctioneer ("...white-fronted, white-crowned, white-winged, white-throated, black and white, black-throated, black-chinned... spoon-bill, speckle-belly, sea coot...") After a particularly rambunctious solo of this kind she whispers "Sold!." In parallel, the tenor yodels latin names for bird families, while the bass tells stories about different nesting locations (" crotch of shrub, football-shaped....") This heterogeneous movement is structured around solos alternating with ensemble passages, just as an observer's attention in the field may be drawn to one bird, then a group, then another individual bird. The fourth movement, Autumn Meditation is concerned with migration and uses a text from the Bible: "Doth the crane fly by Thy wisdom and stretch her wings to the south?" Toward the end of this peaceful nostalgic movement, the owl calls of winter return, completing the cycle of the seasons. In this piece, bird life provides at once the text, the material and the form.

Bird Song in Computer Music

Particularly effective musical uses of bird song are made possible by computer technology. In 1986, I was commissioned by the Hamilton College Chorus to compose a piece for chorus and synthesized voices. With the notion of creating a fanciful exploration of the boundary between human and animal behavior, as well as between Nature and imagination, I wrote a text (Impossible Animals after a reference from Carl Sandburg's poem "Sky Talk") about imaginary animals that I saw in the clouds, one windy autumn afternoon. The text describes several such animals ("a llama with a llama belly...his neck is made of an opossum carrying another opossum in his mouth..."), interspersed with proverbs about the weather ("blue sky in morning, sailor take warning; red sky at night, sailor's delight...mackerel sky and a mare's tail make lofty ships carry a low sail..."), concluding with a description of a more familiar, though no less unlikely, beast ("Has an upright posture, has an opposable thumb, has the ability to communicate by means of organized speech and a variety of symbolic systems. The most impossible animal...") To complement this text, I searched for a computer technique that could create a hybrid between human and animal sounds. It seemed natural to begin with a bird song and see in what musical direction it would lead.

I began with a recording of a song of the Winter Wren, chosen for its considerable length and variety. Fortunately, Julius Smith, a brilliant engineer at CCRMA (Stanford University), had just developed software to analyze a recorded sound, break it up into basic sinusoidal components (which need not be harmonically related), and track these components through time, deriving extremely finely resolved contours of pitch and amplitude. The technique (Smith and Serra, 1987), which had been originally designed for analyzing violin tones, turned out to work beautifully on the Winter Wren song. One of the software's novel features is that the user may specify the number of components used to model the sound. I knew I wanted more than one component, since birds can produce multiple simultaneous sounds, due to the unique construction of the syrinx (a bird's equivalent of our larynx.) As a guess, I chose three components, and this proved to do well in representing the wren. In fact, I was later able to simplify the data to a single sinusoidal component, since fast frequency and amplitude fluctuation ("modulation") of a single component can produce multiple components when necessary.

Armed with the huge amount of analysis data produced by the program, I began the task of shaping it to the musical context. The bird-derived data seemed a perfect starting point for creating the hybrid between human vocalizing and animal utterances for which I was searching. I wasn't sure how to "get there from here," but I headed in what seemed a promising direction.

First I slowed the wren song down by a large factor, but without changing its pitch. This was possible because of the parametric representation provided by the analysis program. Normally when a recording is slowed down, its pitch changes. But if the sound is represented in terms of pitch and amplitude functions of time, it can be slowed down by simply moving more slowly through the pitch data. I then went ahead and lowered the pitch by a different factor, to bring it into the range of the human voice. I was shocked to hear the Winter Wren turn into a Northern Mockingbird! On reflection, this doesn't seem so strange, since the trills of the Winter Wren become repeated phrases when slowed down, with an incessant quality, repetition rate and variety strongly reminiscent of a mockingbird or one of the other mimics.

My goal was to create a synthesized vocal solo that would appear to "sing" over the live chorus. To do this, I needed to tune the pitch of the bird solo to that of the chorus, but I soon ran up against the problem of how to tune material that is frequently trilling or sliding in pitch. To address this, I wrote a series of computer programs. The first searches through the analyzed data and looks for quiet spots between phrases. When it finds such a location, it marks it as a new "segment" of the data. The result is a set of separate mini-phrases or "chirps." The second program takes this data and shapes the range dynamically, according to the musical score, providing a kind of meta-control over the dramatic contour of the bird song. The third program looks for places where the pitch is relatively constant. When it finds such a pitch plateau, it re-tunes the segment to be in harmony with a series of jazz-style chords that the live chorus sings. As for the sliding pitches, they are not perceived to have a particular pitch, so the program leaves them alone.

The final step was the most revealing, and was the result of a lucky coincidence. Vocal synthesis expert Xavier Rodet had just spent several months visiting from the IRCAM computer music facility in Paris, and implementing a novel vocal synthesis system on Stanford's special-purpose music computer. This technique, called "Chant" (Rodet, 1984), represents human singing by a series of resonated pulses (vocal tract impulse responses), a model derived from human anatomy, since our vocal mechanism is composed of a vocal cords that produces a series of pulse-like waveforms, coupled with a vocal tract (throat, mouth and nose), through which these waveforms resonate. Part of the power of this approach is that it can synthesize transitions between notes of a musical phrase by simply interpolating (cross-fading) between pulse resonance shapes. Thus it can synthesize not just a series of separate isolated notes, as is too often the vocabulary provided by MIDI instruments, but fluid phrases with vowels that change gradually and pitches that slide from one note to another.

Using yet another home-brew program, I transformed the wren-derived pitch contour so that it not only specified pitch, but also the transitions between a series of human vowels. For example the lowest pitch might correspond to "Ah," the highest pitch to "Ee" and the pitches between might correspond to "Oo," "Uh" and "I." As the pitch glides up, the Chant vocal synthesis gradually shifts from vowel to vowel, producing "Ah-Oo-Uh-I-Ee," as well as the transitions between each adjacent pair of vowels. This pitch-to-vowel mapping is always in effect, even during extremely rapid trills, giving trills a natural, un-mechanistic sound. For additional variety, I changed the set of vowels as the music progressed, as if the imaginary animal were pronouncing different words in some unknown language. Once the evolving map of pitches to vowels was set up, I re-synthesized the data into a new and greatly transformed rendition of the original wren's song. The result is a true hybrid, as if the brain of a Winter Wren had been transplanted inside a wildly-gifted human singer. When this solo is combined with the live chorus, the effect is weirdly evocative, a solo vocalise of utterances full of emotion and expression, but from a species and in a language that has never existed.

The disconcerting combination of human and bird vocalizations is another case of combining diverse elements into a single musical context, manipulating the material in such a way as to bring out and resolve (or not) its inherent contrasts and contradictions. The result is a music that is both radically challenging on the one hand, and strangely reminiscent of past experience on the other. As in a cubist painting, a nose may be sideways, sticking out from the wrong side of the head, but we know it is a nose, and that gives it an expressive power that an abstract shape would not have, while simultaneously setting up a rich network of associations with everyday life.

Final Thoughts

These are only a few of the possibilities that presented themselves to me when I allowed bird vocalization and behavior to enter my musical world. Musical "inspiration" might better be called "transformation," a two-way street where we receive ideas from the world, filter them through our personal experience and point of view, and give them back to the world in a new form. And, by combining several seemingly mutually-exclusive aspects of experience, we have an opportunity to approach the realm of the mystical, to begin to reconcile the irreconcilable.


Andrew Schloss and Susan Williams provided valuable feedback on this manuscript. Special thanks to Henry Brant and Paul Lansky for their support.


1. D. Jaffe, Territory, for two violins (unpublished manuscript, available from the author, 1977)---12'.

2. D. Jaffe, Whoop for Your Life!, for orchestra (unpublished manuscript, available from the author, 1987)---15'.

3. D. Jaffe, Bird Seasons, for four voices (unpublished manuscript, available from the author, 1984)---10'.

4. D. Jaffe, Impossible Animals, for chorus and synthesized voices. Also versions for solo voices and synthesized voices, violin and synthesized voices, oboe and synthesized voices, and five winds and synthesized voices (unpublished manuscript, available from the author, 1986, 1989, 1990, 1991, 1994, 1995)---10'.

5. J. O. Smith and X. Serra, "PARSHL: A Program for the Analysis/Synthesis of Inharmonic Sounds Based on a Sinusoidal Representation", Proceedings of the International Computer Music Conference Urbana, Illinois (1987).

6. X. O. Rodet, "Time Domain Formant Wave Function Synthesis", Computer Music Journal, Vol. 8, No. 3, pp. 9--14 (1984).