Music for Solo Performer by Alvin Lucier in an Investigation of Current Trends in Brainwave Sonification Professor Michael Gardner, Analysis University of Pittsburgh, 30th of April, 2011
The sonification of brainwaves produced by analog electrical and acoustical sounds in Music for Solo Performer by Alvin Lucier, was both an attempt to bypass the acculturated frame of sound (“I think that acoustical phenomena are universal, not cultural, so I decided…to explore the natural characteristics of sound waves”1) and also a door through which we might hear the sound of the mind in process (and incidentally to experience our own musicreactions to it,) to the extent to which it is possible through the amplification of alpha waves. Music for Solo Performer poses an audacious challenge to conventional musical analysis; while normally a main goal of musical analysis is to address the mind of the composer, performer and/or audience, Lucier makes us listen to the rhythm of waves formed in the performer s mind itself, the minute ʼ
specifics of which are, by nature, involuntary. How do we analyze the sonic projection of a natural waveform occurring in the electrical pattern of the brain? Although the piece would seem to be angled specifically to transcend normal “musical meaning” and address state of mind itself, Alvin Lucier says, “I found ʻ
1
ʼ
Matthew Rogalsky, “’Nature’ as an Organising Principle: Approaches to chance and the natural in the work of John Cage, David Tudor and Alvin Lucier,” in Organised Sound Vol. 15 No. 2, 136.
the alpha s quiet thunder extremely beautiful.” bea utiful.”2 We can explicate that beauty ʼ
through sonic and musical analysis. By looking at the sound of this natural phenomenon as projected onto electrically triggered percussion ensembles, Lucier has provided us with a way to evaluate aesthetic judgment in its most raw state.
In many performance and design aspects, Music for Solo Performer is not free of musical intention however. There is a hidden performer who is capable of changing essential parameters of the piece both in the form of the person who decides on, and builds the percussion set-up and the person who attenuates them live. In most cases this person also changes the instrumentation as the piece progresses. There is the performer s ability to open and close the eyes, ʼ
which trigger major changes in alpha wave amplitude. In addition, there is the possibility of biofeedback, wherein the performer is able to control the strength of their alpha waves. And there are the theatrical aspects of the act itself and it s ʼ
situation. From the structural and surface differences between various versions of MFSP and it s relation to other brainwave music we can ask the question, which ʼ
ʻ
ʼ
criteria can we use to assemble the most meaningful “translation” of brainwaves? First I will examine and compare the available recordings of MFSP to each other. Thereafter I will discuss recent neurological research in the realm of sonification,
2
Alvin Lucier, “Music for Solo Performer” recorded on December 8 and 9, 2007, Lovely Music LCD 5013, Liner notes.
both medical and aesthetic and see how a close reading of Music for Solo Performer enlivens and informs the current discussion.
Various other composers, some of whom have gathered experience in the field of neurology, (most notably David Rosenboom) have been engaged with making music with interactive brainwaves for some time. But there has been a definite revival of interest in the attempt to make brainwaves into sound since around the turn of the 21st century. One reason for this is the utilization of the new, more powerful digital means at our disposal. Another is the continuing rise of cross-disciplinary study, including the acknowledgement on the part of the scientific community, as well as in the liberal arts, of the limitations of positivistic methods in scholarship. Neurologists regularly work with and consult experts in other fields in order to aim their experiments for maximum relevancy. The sciences and the arts have not been closer together for centuries than they are now.
Nearly 50 years after Alvin Lucier composed and performed Music for Solo Performer, the new focus on sonifying brainwaves is being brought to bear by
neurologists (not just composers) not only for pathology and even potential treatment purposes, but also for artistic experimentation, in order to open possible doors to new ways of thinking, and perceiving, that could provide neurological insights. On both sides, utilitarian and aesthetic, they are faced with
the same questions which composers face in their work. In the case of sound for diagnostic and treatment purposes, a host of specific practical criteria come into play for the utilization of the sonification of brainwaves that place limits on creativity. How do these developments change the perspective on Music for Solo Performer ? How can the different sound settings of various performances of
MFSP inform the renewed engagement with brainwave sound today? A majority of the articles dealt with in this paper cite MFSP, but none of them go any farther than merely mentioning it, as if by just knowing about the piece their scholarship is significantly enriched. Lucier and other s various extant recordings from 1965, ʼ
1976, 1982, 2007 and other realizations not by Oliveros or Lucier in the last 2 years all have valuable software under the hood into which researchers could profitably delve.
The local rhythms in MFSP, as portrayed through percussion surfaces excited by amplification are, due to their source, similar in all the recordings. In all of them we hear dense clusters of attacks. Rolls are separated by short asymmetrical sporadic silences. The rolls accomplished by speakers are superhumanly fast, and therefore often have a distorted quality. As the hyper-rolls increase or decrease in intensity, various pitches are elicited from the drumheads. (We see the most exaggerated example of this effect in the version from 1982.) The pause s rhythms are reminiscent of Morse code signal. The ʼ
rhythms are tantalizingly speech-like. We form phrasal closure in connection with
the pauses between bursts. But the constant, fast-paced, (mostly) undifferentiated string of the thousands of closures presented overloads our cognitive system. The unrelenting stream of such rhythms without clear larger boundaries forces our imagination to create arbitrary larger formal downbeats in ʻ
ʼ
their stead. These arbitrary downbeats are often the markers of our own state of ʻ
ʼ
consciousness going in and out of focused attention as we deal with the overload. Music for Solo Performer creates a performative, interpretive role for the listener by leaving the local rhythmic structures up to the indeterminate actions of the performer s alpha waves. Figures Lucier1 and Lucier 2 detail ʼ
ʻ
ʼ
ʻ
ʼ
some of the local alpha rhythms, taken from the 2007 version of Music for Solo Performer minute 29:56 to 30:16. Lucier1 is at speed and Lucier 2 is more ʻ
ʼ
ʻ
ʼ
than twice as slow.
The overall form of the piece is to some degree dependent on flux in the performer s “alpha state” from one minute to the next, but rests even more on the ʼ
“hidden” performer who decides how sensitive the amplifiers triggering the drums should be set initially, and who runs the mixer in the performance, improvising how the amplified alpha waves are dispersed through the ensemble over the course of the work. This person, the unnamed duo partner in Music for Solo Performer , is actually responsible for improvising the form of the piece. By
looking at different performances of the piece we can trace very different priorities for generating large-scale form in each and also similarities. Especially
significant are the comparatively very low level of intensity at the beginning and the end of all of the versions of Music for Solo Performer.
We know that on the premier of Music for Solo Performer on May 5th, 1965, Lucier s amplified alpha wave output was routed to “16 loudspeakerʼ
percussion pairs deployed around the museum.” Lucier further explains that, “During the course of the 40-minute performance Cage randomly raised and lowered the stereo amplifiers volume controls, channeling the alpha signal to ʼ
various instruments around the room.” (This is, by the way, the first time where I have ever read that Cage did something “randomly.”) This instrumental distribution would seem to imply that the sound was spatialized and also sectionalized into parts where various instruments came to the fore or receded. It would also seem to imply that various different volume levels were employed.3 The diagram for the piece of 1965 lists a piano as one of the instruments, a feature excised from future versions of the piece.
It is this version from 1965 that the video of 1976 seems to closely emulate. The sparseness of the reduced activity in the beginning of this version is the most dramatic of all the versions, taking up the about whole first 4 minutes. The large-scale form is periodic, with 5 main areas of high activity separated by moments of very reduced
3
Ibid.
amplitude in which the low rumbling of the timpani stops and smaller less impactful single instrument solos come out (see figure 1.) Some of these quieter moments reach near-stillness. Over the course of the piece we visit various solos and groupings of all the available instruments. The third and forth periods of high activity are marked by having parts where the high frequency cymbals and garbage can drop out. Instead the bass drum, timpani and cardboard box are featured. The fourth period of high activity is much quieter than the others, recalling the sparse first 4 minutes. The last 15 seconds trail off and are reduced down to being only a quieter solo for the snare with chain off. This version possesses a silent ending: in the video even once the sound is turned off, Alvin Lucier keeps his eyes closed for a good 10 seconds before finally opening them to aknowledge that the piece is over.
The surface placement and angle of the amplifiers on the surface of the drums would have allowed for a range of different roll speeds and sounds. It sounds in the 1976 video as if they aimed for a maximum diversity of these various electroacoustic excitation techniques. An example of this is the presence of one snare with the chain off and one with it on. Various instruments are panned 75% left and right to preserve the sound of the spatialization of the instruments throughout the room. Occasionally there is an ethereal constant midrange electronic ambient sound: this is the “tape” part of the composition –
recordings of alpha waves sped up in to the audible range. Here their deployment is very sparse.
In the vinyl version from 1982 we see a “ramp” form where it gains in intensity roughly from the beginning through to the end. It is vastly different from the video recording of 1976. The recording of 1982 is the most unrelenting of all the recordings. In the beginning the tape of the alpha rhythms continues over the sporadic silences in the low bass drum strata. It is apparent that the bass drum articulates particular spikes of the alpha waves while the alpha wave tape is a more continual sound with micro fluctuations. The alpha wave signals coming from the performer seem to get higher however from about the middle of the piece on. The pauses stop and instead we get only a steady surging and receding of the loud rolls in the low drums. About 30 seconds after the pauses in the rolls disappear, the tape sound re-enters but at a higher amplitude with more mid-range to the sound, continuing until almost the end. About a minute after the tape starts, the snare drum entrance begins to articulate high points in the signal more audibly. About 4 minutes before the end, cymbals start to be triggered. The “ramp” form escalates at a higher rate at this point just under discussion, shortly before the midpoint of the piece and continuing until the end, where there are no more spaces between low drum attacks. This point of escalation heightens the “ramp” impression that a listener receives of the piece overall. (See figure 2.) At the very end there is a major dip in intensity however, which is foretold by a slight
decrescendo of the higher frequency sounds, that returns to the texture of the beginning where pauses once again return between clusters of low drum attacks.
The “thunder” of the alpha waves is in the 1982 Lucier setting is the most striking aspect of the piece. Although there are layers of higher sounds above, it seems like there was a production decision to make the low (bass) drums be, far and away, the most predominant sound. It sounds as if there were even contact mics placed on the drumheads to get the “hottest” most overdriven sound possible. Since the cymbals and snare are so far back in the mix, the overall impression is that the low drums are so loud that the high sounds can hardly get through. The unrelenting rumble of the timpani and bass drum in the 1982 version represents a total re-vamping of the concept of the piece. The tape of the alpha rhythms also figure far more prominently than in the 1976 version. This new approach may have been spurred by the fact that a stereo audio recording, lacking the ability to convey the sound and the drama inherent in a spatial placement of the various instruments might have stimulated a different, more frontal and aggressive confrontation with the alpha rhythms by and for themselves. This approach is also evidenced and supported by the “ramp” form.
By looking at the spectrogram of the whole of Pauline Oliveros version ʼ
(figure 3) of MFSP on the B-side of the vinyl disc from 1982 we can see that the form is very different, where the densest most energetic clusters of events occur
in the first fifth of the whole piece. It then generally gradually thins out, becoming less and less dense until a short reoccurrence of heightened density towards the very tail end of the structure occurs. A lapse in density that begins in the last third of the piece confirms the impression of dissipation. Here the attack points are more sporadic than ever. The reoccurrence of heightened density at the tail end, at about minute 13:25 – 14:25 is split into two parts: one strong and one weaker, echoing the overall thinning out structure of the whole recording. All the more surprising, coming as it does, after the most sporadic section, the first thirty seconds of the high-density reoccurrence also has a high degree of regularity in respect to other sections of the piece. The second half is more diffuse and is marked by a slight decrescendo of the higher gong sounds in relation to the rest of the piece. From 14:26 – 14:50 we have a return to the texture of the beginning of the piece; the attacks here are the sparsest besides the opening, and the low drums drop out entirely. This occurrence at the end is the first time where there is any significant change in the instruments that are playing; otherwise there are no shifts or layering of instrumental groupings (as in the second half of the Lucier on the A-side described previously.)
The timbres of Pauline Oliveros version are more weighted on the high ʼ
frequency spectrum, dominated as it is by a shaker, cymbals and gongs, using a “world music percussion orchestra: Indian, West African, Chinese, Japanese and
Korean.”4 The low drum sounds do not inhabit such a wide resonant spectrum and are quite dry, with a quick decay; it is, rather, a muffled thumping. This, the high metallic almost wind-chime like sounds and the lack of a snare drum characterize the sound of Oliveros version. ʼ
The spectrogram from the 2007 version of Music for Solo Performer demonstrates the most complex layering and overlapping through various distinct frequency areas of all the versions. Aside from the opening, which is also quiet like in the other versions, there are 4 “breaks” - clear periods where the activity is markedly decreased, two of which are only very short, but complete, silences (See figure 4.) There is a build-up progression of sound in the beginning: first we have a timpani, then another timpani enters a fourth below it. Soon a bass drum is added and finally after a short time a tam-tam comes in. After the short introduction-buildup of this combination of instruments we encounter the first silence, which lasts for 4 or 5 seconds. Then those same instruments start up again and after about 10 seconds a snare drum is added. After about minute 8:30 we get the second period of reduced activity, wherein the low drums drop out leaving only tambourine and triangle. At about 10:30 the triangle is elided with the start of another build-up, which is essentially the same as the beginning – 2 timpani solo, which is then doubled by bass drum, followed by tam-tam entrance and again with the snare drum added last. Soon after this quasi-recapitulation, about a third of the way into the piece, we get the first instance of the tape part. 4
Alvin Lucier, Music for Solo Performer, Lovely Music VR 1014. Liner notes.
The third “break” is preceded by a fairly quiet section, which includes tam-tam and a very dry tapping surface only. This time the silence lasts a strikingly long 12 seconds. This is even a sort of “fake ending,” which, coming at the eighteenth minute denotes the average length of the other versions; the Music for Solo Performer of 2007 is twice as long, 39 minutes. The fourth “break” is much more
extended, about 4 and a half minutes long. Here there is a solo for both timpani, which is then reduced further to a solo for only one of the timpani. The tape part comes in and the other timpani joins in again. It sounds very much like the version from 1982 here but not overdriven. The dying away at the end is very short in this version, only about 20 seconds long. Here the timpani continue but now quieter, with the tambourine above played sparsely, sputtering out very quietly. Overall there are 4 high points, each time with the triangle and/or tambourine forming the apex of the buildup and triangle or tambourine also overlapping with the following quiet section.
The 2007 version includes some new instruments, like the tambourine and triangle, which redress the balance of the domination of low rumbling sounds of timpani and bass drum found in older Lucier versions. Tambourine and triangle also further highlight well the rapid-fire capabilities of the electric stimulation of the acoustic percussion. The unison attacks of bass drum, timpani and snare drum are also somehow sharper than in other versions, making the alpha rhythms generally stronger.
The 3 versions from the different years and the Oliveros version all follow an implied score of performance aesthetic: the alpha waves are introduced in a quiet section, which is restated at the very end. A build-up and withdrawal from the “alpha state” is dramatized by this framing mechanism in all of the versions. The performer in each case starts the piece with the eyes open and when he/she closes them, the ocular linked boosting of the alpha waves are demonstrated to the audience and the piece begins in earnest – as in the video from 1976. The two versions from 1982 stay mostly in a steady state, while the other performances from 1965, 1976 and 2007 go through various ensemble groupings.
All spikes in brainwave activity are the observation of synchronized neuron movement in the cortex. One theory is that an increase in activity in the central (command center) part of the brain causes the ceasing of activity on the surface (cortex area.) Hence, the increase of waves signals a resting state of certain command processes. The reduction of Alpha waves have been linked to ocular activity. There is also speculation in line with the aforementioned theory of cortical activity that alpha waves are boosted during moments of “alert immobility” like the moment just before the archer lets go of the arrow. 5 Many experiments
5
Reference is to Dr. Tom Mulholland’s theory of “behavioral stillness.” See: Shaw, John Crosley, “Chapter 14: Mulholland’s Alpha Feedback Paradigm and ‘behavioral stillness’ model,” in The Brain’s Alpha Rhythms and the Mind , (Amsterdam: Elsevier Science B.V., 2003)
conclude that it is possible to control alpha rhythms to varying degrees, but the significance of this fact and of the waves themselves, apropos related tasks beyond certain observed phenomenon and a specific group of mental processes is the subject of ongoing inquiry. Therefore the conclusion that boosted alpha waves (or any other waves for that matter) yield a “meditative state” of wellbeing is premature. In 1965, however, the matter would have been a subject for greater debate and speculation than it is now after several decades more of experimentation. The fact remains, however, that still no one has conclusively disproven that biofeedback, meditation and hypnosis lead to heightened alpha waves.6
A compelling and widespread finding is that EEG can predict a subject s ʼ
listening states in terms of active vs. passive forms. Eduardo Reck Miranda goes further, claiming that mental focusing on left vs. right audio channels can also be predicted.7 David Rosenboom has been making use of this ability of EEG since before 1990 to enable a performer to structure computer generated musical forms.8 Especially recent scholarship in musicology that take passive forms of listening as an equally qualitative musical activity in society could possibly shed light on possible further avenues of research and experimentation in this realm.
6
Ibid., 214-215. Eduardo Reck Miranda, “On Harnessing the Electroencephalogram for the Musical Braincap,” in Computer Music Journal, Vol. 27, No. 2, 2003. 8 David Rosenboom, “Propositional Music from Extended Musical Interface with the Human Nervous System,” in Annual New York Academy of the Sciences Vol. 999, 2003, 263-271. 7
In any case the fact that this detection has been proven to be fairly reliable would tend to lend confirmation to Oliveros instinct to use biofeedback in her version of ʼ
MFSP, as it says in the liner notes, “the performers listened through headphones to their previously recorded performances, which reinforced and influenced their alpha.” If the excessive phrasal closures characteristic of the alpha rhythms do succeed in overloading and therefore shutting down the closure mechanism of ʻ
ʼ
active listening reflexes, it would hypothetically provide an event related potential (neurology: ERP) for strengthening brainwaves associated with passive, i.e. “meditative” (or in any case not expectation heightened) listening, alpha among them.
An agreed-upon basis for work in EEG sonification among a segment of neurologists is the superior ability of the auditory, as opposed to the visual apparatus, to pick out and interpret simultaneous stimuli (as in counterpoint, or in discerning variable layers of speech vs. noise etc.) Gerald Baier, et al. have demonstrated audio models to identify pathologies more distinctly than with graphic representations, for instance.9 Since this is irrefutably the case, it seems that it would be an even greater aid in perception to have a spatialized and/or acoustic array of instruments to allow the diagnostician even greater ease of discernment of various layers. One most pronounced feature of our experience of
9
Gerold Baier, Thomas Hermann and Ulrich Stephani, “Even-based Sonification of EEG Rhythms in Real Time,” in Clinical Neurophysiology, Vol. 118, 2007, 1377-1386.
listening to MFSP is, as previously observed, our simultaneous awareness and discernment of multiple layers of instrumental strata and their differences.
Several recent articles attempt a sonification by mapping EEG wave amplitude to pitch. All of them use standard MIDI notes and tunings. Tuning is a one of the most highly sensitive mechanisms of human hearing and it would be worth applying microtonal parameters to EEG for the purposes of complex data assessment. Also the rich world of affect associations and musical imagery that microtonal inflection carry could open whole new vistas in the realm of biofeedback, cognitive emotional processes and other areas of research if harnessed in EEG sound experimentation. Aside from this, microtonality is a more fitting correlate in the domain of pitch to the complex resultant brainwave rhythms.
MFSP is exceptional for Lucier s work in that it mostly focuses on rhythmic ʼ
activity rather than the fine sine waves, spectral resonances and discreet beating found in the great majority of his output. Another major piece of Lucier s dealing ʼ
with the body and its nervous system, Clocker , is also atypically preoccupied with rhythm. Baier, Hermann and Stephani, enrich and complicate rhythmic projection of the brainwaves by linking EEG with MIDI velocity rather than pitch, unlike most of their colleagues experimenting in EEG sound. Their resultant sonification is reminiscent of MFSP and has a highly satisfying aesthetic and utilitarian design.
Thilo Hinterberger and Gerald Baier make an astute observation about the role of rhythm in cognition:
“researchers generally believe that brains work as pattern-recognition devices that first encode all sensory information into multiple streams of neural rhythms and then use rhythmic reorganization of their ongoing activity to process the input.” 10
This says something directly relevant to the whole attempt of brain rhythm mapping and specifically to the Morse code rhythm sound of MFSP. Rhythms ʻ
ʼ
are the most germane factor in neurological patterning and are thus probably the paramount aspect for aesthetic and medical contemplation and exploration of brainwaves turned into sound. The raw rhythms heard in Lucier s MFSP could be ʼ
used as a starting point for various types of rhythmic transformation, perhaps by modeling the same type of flows represented in larger scale tendencies of brainwave potentials. The types of multiple streams of encoding that Hinterberger and Baier refer to provide excellent inspiration for the modeling of analogous rhythmic processes and transformations in the acoustic realm.
One main difference between Lucier s MFSP and current ʼ
sonification attempts by the neurology community is that by allowing the aforementioned hidden performer of the piece free reign over aspects of ʻ
10
ʼ
Thilo Hinterberger and Gerold Baier, “Parametric Orchestral Sonification of EEG in Real Time,” in IEEE Computer Society, April-June, 2005
amplitude and instrumentation, the surface is in most recorded versions of MFSP highly variegated on the large scale. Most new attempts by neurologists however, seem to try to extract rules that are meant to stay in place for the entirety of that particular sonification attempt. In trying to find the right formula for optimal auditory presentation, the possibility to modulate the rules of sonification over time is ignored in all cases in the medical sources used in this paper. Most composers who attempt sonification of brainwaves today, who aim to add to the neurological literature, are mostly as inflexible in their approach, only with more sophisticated musical goals. This attitude represents a lost opportunity for the scientific as well as for the artistic community. In Reflexionen, Lucier expounds on the significance of hiding the performer who changes the instrumental groupings and amplifier attenuation:
“You see, one of the inaccuracies of the title is that it s not really for solo ʼ
performer. You need someone to run the amplifiers, to pan the sounds around, to turn on one loudspeaker and then turn on another, and I ve always…done it with ʼ
another player, an assistant. In the score that I wrote, I stipulated that someday, when electronics became what it s now become, you could have an automatic ʼ
switching arrangement, such that so many bursts of alpha would be a code to a switching device, and the alpha could control itself without an assistant.” 11
11
Alvin Lucier, Reflexionen: Interviews Notation Texte (Cologne: MusikTexte, 1995)
To this day no one, musician or neurologist has implemented such a selfswitching system, and it is time that they did.
Another significant lost opportunity is the hemming in of all recent attempts at sonic brainwave transcription into the digital realm. Acoustic sounds have an ʻ
ʼ
organic richness that yields more sonic information than speakers alone do. The diffraction of sound into various organic materials, (metal, wood, skins etc.) gives our ears various and differentiated points of contact and resonance on which to test, and modulate our focus. Perhaps it would be possible to build physical acoustic structures that mimic and map various regions of the brain along with an acoustic analog to their differences in structure to test certain hypotheses.
Figure 1 – Music for Solo Performer , 1976
Figure 2 – Music for Solo Performer , 1982 (Lucier version)
Figure 3 – Music for Solo Performer , 1982 (Oliveros version)
Figure 4 – Music for Solo Performer , 2007
5
7
3
3
7
5
7
5
3
7
7
9
5
5
6
5
Lucier 1 (local alpha wave rhythms at speed)
5
3
7
w
7
3
5
7
9
5
5
6
3
3
7
5
7
7
6
3
6
7
3
2
11
13
3
5
3
3
3
7
15
17
19
3
5
3
3
5
Lucier 2 (local alpha rhythms slowed down)
5
7
