Music Composition for Dance in the Twenty-First Century:

Questions about the Dance/Music Relationship

presented at the Stockholm 1998 IGMID conference

Allen Fogelsanger




               This conference has been organized under the title of "Music and Dance in Unity."  This slogan is certainly one to which I subscribe in my daily work as a dance class accompanist and a composer for dance.  But what I would like to do in this presentation is explore the ways in which, at least with regard to much American modern dance of the last several decades and to some extant with regard to other dance, music and dance are not in unity at all.

               First of all we have to establish some sense of what is meant by the phrase "music and dance in unity."  For that I would like to refer to a 1992 book by Paul Hodgins called Relationships between Score and Choreography in Twentieth-Century Dance.[1]  Some of you may recall his talk on "Music, Movement and Metaphor" at the 1992 IGMID Conference at SUNY-Brockport where he pointed out choreomusical relationships in Balanchine and Stravinsky's Agon.  In his book Hodgins develops what he calls a "paradigm" for choreomusical analysis, which he applies to six acknowledged masterworks of twentieth-century dance.  In his analyses, he covers the gamut of potential relationships between dance and music that choreographers and/or composers use in creating a rich and unified work.  For this audience I don't think it necessary to go into much detail about what these relationships are--very briefly, they include the following: relationships between rhythms in the dance and rhythms in the music, between sound volume and the size of choreographic gestures, between musical textures such as polyphony or homophony with their particular organizations of instrumental voices and analogous choreographic organizations of dancers, between the timbre of instruments or sounds and the characteristics of the groups of dancers appearing on stage, and many, many more relationships.  Hodgins' systematic and quite likely exhaustive categorization of these relationships allows for thorough investigations of the dances about which he chooses to write: The Green Table (1932) choreographed by Kurt Jooss with music by Fritz Cohen; Billy the Kid (1938) choreographed by Eugene Loring with music by Aaron Copland; Errand into the Maze (1947) choreographed by Martha Graham with music by Gian-Carlo Menotti; and three works choreographed by George Balanchine with music by Igor Stravinsky, Apollon Musagte (1928), Orpheus (1948), and Agon (1957).

               In explanation of how he picked these six works, Hodgins writes:

               In choosing six masterworks for analysis, I decided first of all to examine only those works in which complete and thorough collaboration between choreographer and composer was achieved....  I chose only those works in which I felt music and dance seemed to play equal interpretive and expressive roles; I focussed on choreographies of the last seventy years, the era since the iconoclastic productions of the Ballets Russes in which the two disciplines have struggled toward a more or less equitable co-existence.  This selection of analyses is regrettably small; more recent works are, unfortunately, omitted.[2]

Hodgins' choices are works that, for me, exemplify the phrase "music and dance in unity," so I propose using his words to define what the phrase means: works exhibiting music and dance in unity are those "in which complete and thorough collaboration [is] achieved," and in which "music and dance [seem] to play equal interpretive and expressive roles" and are in "more or less equitable co-existence."  Thus I am choosing to understand our conference banner in a relatively narrow sense, a sense which is best exemplified by Agon.

               Consider two descriptions of the relationship between music and dance in Agon.  Dance and music theorist Stephanie Jordan writes:

               It is as if choreography and music mesh together in Agon like interlocking parts in a sophisticated piece of machinery, a rare example of deep interpenetration disciplined by a shared motoric drive.  And yet the machine metaphor must be applied sparingly.  The excitement of Agon is that shifting and volatile musical/choreographic relationships continually enliven our visual/aural awareness.  Our perceptions constantly challenged, in Agon, the dance virtually begins to sound and music to move.[3]

And dance critic Marcia Siegel writes:

               Stravinsky's rhythms are irregular but, as Balanchine has pointed out, his pulse is steady, and the reliable underlying metre supplies all the support that's necessary to open the field for a great rhythmic interplay among the instruments of the orchestra, the solo dancers and the ensemble, and between the dancers and the orchestra.  So the ballet really has two lines of imagery to follow: the varieties of rhythmic invention--syncopation, suspension, canonic devices, explorations of the dynamic range from percussive to legato--and the constantly shifting arrangements of dancers in the space and of the dancers' body shapes.[4]

I would like to stress the words "interlocking," "interpenetration," and "interplay" in the particular ideal of "music and dance in unity" that I am establishing here.  This ideal is actually that of Gesamtkunstwerk, or the combined and integrated work of art.  I think it is no surprise that Balanchine has perhaps come closest to creating such integrated works of art, given that, as pointed out by critic Roger Copeland, "one could argue that for Balanchine the medium of dance is by definition 'bodies moving in response to musical stimuli.'"[5]

               In many dances, if not most, music and dance are not in unity in the way I've described.  There are numerous works in which integration between music and dance is not the goal of the choreographer and composer.  It is these sorts of works, and how music and dance function together in such works, that I would like to talk about today.




               Although Hodgins "focussed on choreographies of the last seventy years," all of his chosen works come from the period 1928-1957, and none from the subsequent thirty-five years.  This later period is coincidentally one in which dance's traditional relationship with music, the idea that dancers dance to music, was being sharply criticized with a number of alternative relationships being proposed and explored.            Much of the changed attitude towards the relationship between music and dance is, of course, responsible to Merce Cunningham and John Cage.

               Their basic collaborative philosophy is well known: in Roger Copeland's words, "movement and sound existed independently of one another; choreography and music were both performed in the same space and time, but without affecting (or even acknowledging) one another."[6]  I think that a look at the consideration of Cage's music by music theorists can illuminate something more about how Cunningham uses music.

               According to music historian Robert Morgan, in the case of Cage,

               each musical unit existed more or less for itself, essentially independent of any relationship it might have with other units.  A musical sound was not derived from the sounds that preceded it, nor did it imply those that followed.  It simply "was."  According to this conception, music is "purposeless."  Its components have no meaning--that is, no discernible connections with one another.[7]

I am going to argue that, essentially, Cage has taken the traditional musicality out of music (narrowly defined); on the other hand, I also think that Cage has opened up new areas of music by teaching us not to insist on traditional musicality from the sonic works presented to us.  Cunningham, analagously, has opened up new areas of dance by teaching us not to insist on a traditional idea of dancing, that of dancers dancing to music; but while Cage has removed musicality from music, Cunningham has revealed the inherent musicality of dance.  Of course, I should more carefully define what I mean by musicality before I make such statements, and I will.  But first I'd like us to take a look at an excerpt from a Cunningham dance, keeping in mind this explanation of Cunningham's work by Dance Magazine editor Nancy Dalva:

               In performance, Cunningham's dances usually are accompanied by live music...but it is not the music of the dance, merely the music that happens at the same time.

               The dance's music can be seen but not heard, except in the footfalls of the dancers and their breathing.  The dance's music is its rhythm.  Perhaps the easiest place to see Cunningham rhythm is his unison sections, and the easiest place to find such sections is in the early video works choreographed first for camera, then transferred to stage.  Here one finds the dancers disposed in squads.  To see one squad opposed against another is to see two unisons at once: basic Cunningham counterpoint.

               Always--in stage, on film, in videos, and in rehearsal--the dancers seem to be dancing to something--keeping up with it, slowing down to it--their phrasing exquisite and driven.  By what?  By the sound--or the memory--of Cunningham's own snapping and clapping.  Merce Cunningham works with a stopwatch.  He his own way the most musical of all [choreographers].  One could see this when he revived Septet, made in 1953 to Erik Satie's Trois Morceaux en Forme de Poire.  To see Septet is to realize that, for Cunningham, working to music must be like turning on the radio when the record player is already on.  It interferes with music he already hears.[8]

This three-minute video excerpt is from Event for Television (1976) and shows part of the dance Westbeth, which was "remade and angled for the camera;"[9] the music comes from Branches by John Cage.




               In this example I think there is a certain way in which the dance has musicality and the music does not.  The music has sounds, but no melody.  We do not hear sounds connecting into musical lines.  This makes it difficult to hear phrases.  It seems that the sounds are not organized, at least in short time-spans, in any recognizable way.

               On the other hand, the choreographic events are very organized.  Movement is occasionally repeated, dancers interact, there may even be unison.  There may be no pulse, but the sense of rhythm is much more apparent than in the sound.  Sections are defined by entrances and exits and by the number of dancers on stage.  Inevitably the audience sees phrases of movement bound by stillnesses.

               For Cage, the musical unit is the sound; for Cunningham, it is the phrase.  Cage "set about discovering means to let sounds be themselves;"[10] Cunningham says, "dancing, for me, is movement in time and space."[11]  Cage's music is essentially still, like a painting--it is a soundscape; dance cannot quite become still, and so retains its musical qualities.  Furthermore, each movement unit cannot exist more or less for itself, essentially independent of any relationship it might have with other units, because we viewers automatically connect a dancer's movements into a phrase.  Unlike a musical line, the movement line always exist, continuously connecting a succession of movements through the inescapable connective tissue of the dancer's body.  Unlike the sounds in Cage's music, a movement cannot help but be derived from the movements that precede it, and imply those that follow.

               In the work of Cunningham and others who choreograph to soundscapes, the dance provides the propulsion once provided by music.  The dance develops according to its own needs, not to the music's.  Music has become a backdrop, a dŽcor, scenery in front of which the dance holds our attention.  Music, like the other accessory elements, helps to paint a space within which the dance occurs.  Cunningham freed dance from music, and ever since choreographers have been developing choreography that depends only on itself, on the movement, and not on music.  Choreography is inspired not by music, but by movement, and dance can look within itself for its own tools for organizing material over time.

               Considering that in the nineteenth century the idea of "absolute music" developed in reaction to Wagner's Gesamtkunstwerk, it is no surprise that in the twentieth century an idea of "absolute dance" not even dependent on music, has developed in reaction to Balanchine, Graham, and the entire history of modern dance and ballet.

               Agon is the epitome of the ideal of an integration of "traditionally musical" music and dance.  We need to expand Paul Hodgins' paradigm for choreomusical analysis and extend its application to recent developments in collaborative work.




               Since Cunningham's revolution, there has been more than a generation of choreographers for whom the function of music is to create a static or slowly changing emotional space--the movement, the motion, the essential rhythm, the dynamics of the dance are the province solely of the choreography.  Many have used music like Cage's--music of sounds, sonic designs.  Another favorite accompaniment is minimalist music.  To make a connection between these two types of music, I'd like to bring up the question of how we listen to music and sounds.

               The model of listening I'm going to outline is fairly standard.  I, personally, was first introduced to it through the works of Leonard Meyer, particularly through his books Emotion and Meaning in Music[12] and Music, the Arts, and Ideas.[13]  When we listen to traditional music, we are not just listening to sounds, we are evaluating how they are related and building a model in our minds of what the music is.  The more we know about music, and the more we know about the style of the music to which we are listening, the more we are able to evaluate.  One important factor in the music that affects our evaluation is the redundancy in the music.  Redundancy occurs when something is presented again--for example, the repetition or varied repetition of a pitch, interval, melody, duration, rhythm, or chord structure.  When we recognize a redundancy we draw a connection to something we've heard before in the piece, and we add this information to our mental model and adjust it to reflect our new understanding of the music.  So as we listen to the music, we're always thinking about the music, though we may not be conscious of that thinking.  The frequency with which we are presented with new redundancies and new information, make new connections, and adjust our mental model, affects how we are mentally engaged by the music.

               Although this model of listening has been extended by Meyer in the direction which I would now like to go, I prefer to use another theorist's terminology.  In a fascinating book called The Time of Music,[14] music theorist Jonathan Kramer explores how the way we are engaged by music affects our experience of time.  I am going to borrow some of his terms linear music and vertical music.

               Linear music is music where one event implies another.  The simplest example of this is the way in which the leading tone implies the tonic.  Another example would be the way in which, in the second time through a repeated section of music, once the listener recognizes that a section is being repeated, there is an implication that the section may continue to be repeated to its conclusion.  In both example, composers may play with the implications, but the very fact the implications are present is an indication of "linear" music.  Most Western music has been "linear".

               Dance, as least as far as it has movement, seems to be inherently "linear".  As movements are not static, they imply their own continuation: someone in the air must come down, someone off-balance must eventually balance.

               Vertical music is music where "linearity" has broken down, allowing listeners to abandon their constant evaluation of relationships and lose themselves in listening.  This can happen in two ways.  On the one hand, "linearity" can break down because we are unable to hear many relationships between musical events, as in Cage's music.  On the other hand, "linearity" can break down because the relationships are so predictible that we relax and take them for granted.  In either case we are left only with the sensuousness of sound.  In the first case, we have minimal redundancy, randomness, aleatoric music; in the second case, we have maximal redundancy, predictability, minimalist music.

               When choreographers use "vertical" music, they interfere with the music's capacity to lull people out of their normal "linear" listening habits.  The audience may be engaged by the dance, with little competition from the music.

               We are going to look at two examples of dance choreographed to highly redundant, minimalist music.  The first example is a three-minute excerpt near the beginning of Trisha Brown's Set and Reset with music by Laurie Anderson titled Long Time No See.  This music is a good example of "vertical" music.  It has an ever-present energizing pulse that creates a static musical space in which to dance.  Anderson adds various sounds to the music's ever-changing palette, but there is little sense of progression.  The music simply creates a background for Brown's dance, which is of an entirely different order of complexity.  The choreography is full of unisons and near-unisons, movement phrases that move in and out of each other.  Sections are created by different groupings of dancers on stage, material is developed and varied over time.  In short, the choreography has many of the characteristics of traditional music.  It is "linear" dance.  The music supplies a beat and an emotional space.

               The second example is the five-minute penultimate movement of Twyla Tharp's In the Upper Room with music by Philip Glass.  Glass's music is not as "vertical" as Anderson's is.  Earlier I referred to redundancy as being a measure of whether music is "vertical" or "linear", but it is not a case of either/or.  As Jonathan Kramer explains, there is a continuum from non-redundant music through somewhat redundant music to totally redundant music.  The extremes of this continuum are "vertical" music, the middle is "linear" music, but most music has both "linear" and "vertical" qualities.  "Much vertical music retains vestiges of linearity."[15]  Glass's music is such, primarily "vertical" but with important "linear" landmarks.

               Even so, it is the "vertical" quality of the music that strikes dance critic Arlene Croce as she writes about the music/dance relationship:

               Arranged in nine sections...In the Upper Room is unambiguously a dance suite....  Glass's music, with its tootling ostinatos and keening strings, contributes exactly the supercharged atmosphere that Tharp wants here, although it's easy to see why she has not collaborated with Glass before.  He sets a properly frenetic pace but builds no momentum; each dance is pinned in its own gridlike cage of sound....  Glass makes almost no rhythmic or textural demands on Tharp.  And she doesn't seem to want them.[16]

Through the many notes, the repeated pulsations, Glass's music fills the dance with underlying energy, but at the same time leaves the choreography practically free of any obligation to interact with the music.  Thus Tharp can attend to her material, develop it at its own pace.  But in this section at least, she recognizes and uses certain aspects of the music.

               The music divides the section into three subsections, the second two of which repeat the harmonic structure of the first.  Within each subsection the harmony is repetitive, except that near the end it leads to a dominant seventh chord to take us back to the tonic.  Furthermore, each latter subsection has one more layer of musical material than the preceding section: first a high legato string line is added, then a slow staccato bass line.  The whole effect is of a very, very slow increase of tension.

               Tharp uses and sharpens this skeleton of "linearity" with her choreography.  The section begins with a lone female dancer.  Three male dancers appear.  At first all four dance alone, but it is not long before the woman briefly dances with one of the men, then another, then the third.  Tharp gradually increases the complexity of the interactions until the dominant seventh comes around in the music.  Then a second woman enters as the repetition begins.  While the first woman now has close partnering with the men, including considerable lifting, the second woman seems to closely follow the movement sequence danced by the first woman in the first section; in other words, Tharp has created a long drawn-out canon, the period of which matches the length of Glass's subsections of music.  Two more women enter at the beginning of the last repetition, dancing in unison, and apparently creating a three-part canon.  The action intensifies until the last dominant seventh is reached, when the dancers relax.  The potential cadence is interrupted by silence, and all exit save a lone woman.

               Let's now look at these examples.




               There are a few more comments I would like to make about these examples.

               To begin with, although in Set and Reset it is fairly easy to recognize relationships between different parts of the choreographic material in the three-minute excerpt, it is difficult to see how these relationships will give way to a satisfying structural form over the entire twenty-minute duration of the dance.  Because of this, Anderson's music may appear to have far more coherence than Brown's choreography, at least within this excerpt.

               Another aspect of the Set and Reset example worth pointing out is the camera work.  This gives me an opportunity to remark that in all my video examples, the camera obviously affects how we perceive the dancing.  A more thorough analysis will have to wait for another time.

               Furthermore, In the Upper Room practically begs for interpretive analysis, especially with respect to gender and the use of classical and popular dance vocabularies.  But that, too, is beyond the scope of this talk, which is concentrated on comparing formalistic elements in contemporary choreography and music.

               Finally, I'd like to note that In the Upper Room is, in fact, a work where many of the choreomusical relationships written about by Paul Hodgins can be found.  But I think it is not incorrect to think of Glass providing the chords and Tharp providing the melody.  Although In the Upper Room  may be a work "in which complete and thorough collaboration is achieved," I'm not sure that the "music and dance play equal interpretive and expressive roles" and are in "more or less equitable co-existence."  There is little "rhythmic interplay" between them.  The "vertical" music establishes a strong emotional base while the "linear" dance engages most of our attention.




               One of the choreographers with whom I've collaborated is Jumay Chu, a dancer whose training derives from Cunningham through Viola Farber.  For much of her choreography, I find I prefer to interfere as little as possible with the rhythms and phrases of her movement.  A natural response would be to write "vertical" music, but my interest is in writing "linear" music.  So, what options are open to me that will still allow that Cunningham musicality to be seen in the dance?

               One long-used solution would be to write very slow music.  Take, for instance, Pina Bausch's use of arias by Henry Purcell in Cafe MŸller.  The movement, consisting of people wandering around a stage full of chairs, sometimes running into many of the chairs at once, has very little rhythmic relationship to the music.  The music has so little forward motion that it takes on the quality of "vertical" music--it is changing so slowly that if we were listening to it we would lose ourselves in the sound of the singer.  But we're watching movement, we see it's rhythms, and the music has only our peripheral attention.  That is enough for it to profoundly affect our experience of the Tanztheater.

               In this excerpt of Jumay's and my collaboration, we have partially adopted the "slow-music" solution.  I have tried to leave plenty of "space" in the music for the dance to occupy.  But there are also "spaces" in the dance for the music to occupy.  To some extent the dance and the music alternate in engaging the audience.

               The name of the dance is Movements.  The name of the music is Logues, Ludes, and Improvisations.  The excerpt consists of the first six minutes of a twenty-minute work.




               Considerably more could be said about the impact of "vertical" music on dance, and on "vertical" dance, but I would like to move on and mention what I think is the other important development in the contemporary relationship between music and dance, one which may have great repercussions in the next century: the increasing ability of dancers to produce their own sounds, create their own music, and create their own soundscapes.  For centuries dancers have been using their voices and producing percussion with their bodies.  But as we head into the twenty-first century, fascinating additions to the dancer's sound-producing repertory are being brought into reach by the advent of computer-driven technology.

               Not surprisingly, one of the earliest works with enhanced dancer-produced sound was by Merce Cunningham, Variations V (1965).  In this piece, there were two sound sources that could be triggered by dancers.  According to Cunningham,

               The first was a serires of poles, twelve in all, like antennae, placed all over the stage--each to have a sound radius, sphere-shaped, of four feet.  When a dancer came into this radius, sound could be triggered.  Each of the twelve antennae had different sound possibilities....

               The second sound source was a series of photoelectric cells which were to be positioned on the floor along the sides of the stage.  The stage lights would be focussed in such a way as to hit them, and when a dancer passed between the cell and the light, more sound possibilities were triggered.  This did not work out precisely....  So at the last minute the cells were put at the base of the twelve poles....[17]

               That there is an interest in following this line of exploration was seen at the 1997 IGMID conference in Tempe, where John Mitchell and Robb Lovell showed a piece in which the dancer could cause more complicated sound events, as well as changes in lighting and video, that were activated through a video camera feeding information to a computer about the dancer's position, velocity, and shape.[18]

               In these examples, the dancer had very little control over "playing" with the sound as an instrumentalist does.  Not surprisingly, one piece that does give the dancer more control over sound was devised by a composer, Karl-Heinz Stockhausen.  This piece, Harlequin (1975), has also been presented at an IGMID conference, in Miami in 1993 by Michael Seaver[19].  In Harlequin, a clarinetist moves about the stage, in effect dancing, in ways related to the music he is playing.  Of course, musical training is at least as important as movement training for this piece, but that should not be surprising for dancers who want to make music.

               But what dancer/musicians should be most excited about is the recent work of composer Tod Machover.  Machover has been interested in developing what he calls hyperinstruments--acoustic instruments with electronic sensors that send information to a computer about instrument and performer parameters, whcih then enhances the sound.  Machover has written a solo piece for hypercello, and concertos for hyperviolin and hyperviola.

               In an article in The Sciences, Thomas Levenson describes how the computer interfaced with the musician playing the hypercello.  "[There] are [magnetic] sensors that detect the cellist's physical gestures.  In a performance, hypercello and cellist are wired with the devices that measure bow-hand wrist angle, finger pressure on the bow, bow position, left-hand finger position on the neck, and the pitch and loudness drawn from each string."[20]  In other words, the computer senses, among other things, human movement and responds by modifying the sound of the cello.  Without much work it could respond to movement by producing and modifying sound.  This technology is eventually going to give dancers another avenue for producing sound directly from their movement, an avenue for virtuosos to combine dance and music more complexly and more together than ever before.

               I have some small experience in this area, via a piece developed at my institution, Cornell University, under the direction of my colleague Byron Suber, using some simple sensing devices designed by another colleague, Warren Cross.  These devices, fitted to a dancer's elbow and/or knees, are more mechanical than electronic--each one can send a MIDI signal determined by the angle made by the two parts of the limb to which it is attached, for example by the angle between a forearm and the connected upper arm.  The signal is activiated by a button in the palm of the dancer's hand.  With this equipment the dancer can play a synathesizer by bending the elbow and pressing the button.  Or he can affect other MIDI parameters of an ongoing soundscape.

               This equipment was used in a dance developed during the fall and winter of 1996-97, culminating in a performance in March.  Although the dancer using the equipment was controlling a significant part of the sound environment, the performance raised questions for me about how that control was perceived, questions also provoked by the performance at the 1997 IGMID conference in Tempe.

               Central to these questions is my observation that I had great difficulty seeing that the dancer on stage was actually producing the sounds I heard through the speaker system.  From my close association with the Cornell project, I knew exactly how the dancer was affecting the sound, but as an audience member I had trouble making the connection between the movement and its sound system-mediated result.  We designers could have saved ourselves a lot of trouble and just created a sound score without movement input.

               For the technology-aided sound-producing dancer of the future, this question of establishing a clear connection between the sound producer and the produced sound is of utmost importance, because it plays a pivotal role in determining the reception of the sound/movement relationship.  Choreography "mickey mousing" the music is generally frowned upon, with the glaring exception of sound-producing movement like tap dancing, which, of course, must "mickey-mouse" the music because it is the music.

               The reception of dancer-produced music is also addressed by Michael Seaver in his article for the 1992 IGMID journal describing his experience working on Harlequin.  He writes:

               It seems the effectiveness of Harlequin relies to a large extent on the fact that one person is the source of the movement and the music and the impulse is never perceived as being from one or the other but from the combination of the two.  In the case of different people being the sources for the movement and music there is a perceived tension between the dancer and the musician.  This is not a bad thing, of course....

               But when experiencing a performance by a musician and a dancer there is a part of us that wants them to remain separate, we don't want one to do the same as the other....  We want to experience the interaction between the two disciplines to heighten the overall experience--the synergy principle.  Yet we think this synergy only possible when there is a lack of dependency between the two.  If this is the norm then how do we access a performance that, by it's nature, hasn't got this tension and portrays an almost total dependency between the two?  Is it a lesser experience than had the one performer tried to keep the two elements as separate and disparate as possible?[21]




               In summary, I think that Balanchine and Stravinsky on the one hand, and Cunningham and Cage on the other hand, have defined the choreomusical relationships of the twentieth-century.  As we enter the twenty-first, creative activity seems to occupy all the space between these two extremes.  But if I were to make a prediction of what developments are in store for us, I would say that the next revolution of choreomusical activity will be triggered by advancing technology, namely the increasing possibilities for dancers to make music themselves.  The questions raised by Seaver are a mirror image of the choreomusical questions raised by the twentieth-century as a whole, and there is no way to tell how they will play out in the new arena created by technology, nor how they will affect the ongoing work in traditional collaborations.

               I see choreographers experimenting with various ways of controlling music.  For centuries they danced to and with the music.  In the present, using "vertical" music, they have great liberty to dance over the music.  In the future, instead of dancers moving to an extraneous sound source, they may be the sound source, and a wonderfully rich one at that.

               Thank you.


[1]Paul Hodgins, Relationships between Score and Choreography in Twentieth-Century Dance: Music, Movement and Metaphor (Lewiston, New York: Edwin Mellen Press, 1992).

[2]Hodgins 30.

[3]Stephanie Jordan, "Agon: A Musical/Choreographic Analysis," Dance Research Journal 25/2 (Fall 1993): 11.

[4]Marcia B. Siegel, The Shapes of Change: Images of American Dance (Berkeley, California: University of California Press, 1979) 228-229.

[5]Roger Copeland, "Backlash Against Balanchine," Choreography and Dance 3/3 (1993): 6.

[6]Roger Copeland, "Merce Cunningham and the Politics of Perception" (1979), What Is Dance: Readings in Theory and Criticism, ed. Roger Copeland and Marshall Cohen (Oxford: Oxford University Press, 1983) 310.

[7]Robert P. Morgan, Twentieth-Century Music (New York: W. W. Norton & Company, 1991) 362.

[8]Nancy Dalva, "The Way of Merce" (1992), Merce Cunningham: Dancing in Space and Time, ed. Richard Kostelanetz (Chicago: A Cappella Books, 1992) 181-182.

[9]Merce Cunningham, "A Collaborative Process Between Music and Dance" (1982), Kostelanetz 148.

[10]John Cage, "Experimental Music" (1957), Silence: Lectures and Writings (Middletown, Connecticut: The Wesleyan University Press, 1961) 10.

[11]Merce Cunningham, Event for Television, WNET/"Dance in America" (1977).

[12]Leonard B. Meyer, Emotion and Meaning in Music (Chicago: University of Chicago Press, 1956).

[13]Leonard B. Meyer, Music, the Arts, and Ideas: Patterns and Predictions in Twentieth-Century Culture, with a new postlude (Chicago: University of Chicago Press, 1994, 1967).

[14]Jonathan D. Kramer, The Time of Music: New Meanings, New Temporalities, New Listening Strategies (New York: Schirmer Books, 1988).

[15]Kramer, 389.

[16]Arlene Croce, "Postmodern Ballets" (1987), Sight Lines (New York: Alfred A. Knopf, 1987) 321.

[17]Merce Cunningham, "A Collaborative Process Between Music and Dance" (1982), Kostelanetz 145.

[18]Aixe' Djelal, "Sixth Annual Guild Conference Takes Place in Tempe, Arizona," International Guild of Musicians in Dance Newsletter 6/2 (Spring, 1997) 1.

[19]See Michael Seaver, "A Singular Impulse: Musician and Dancer as One Performer," International Guild of Musicians in Dance Journal 2 (1992) 19-22.

[20]Thomas Levenson, "Taming the Hypercello," The Sciences 34/4 (July, 1994) 15.

[21]Seaver, 20-21.