banner image

A circuitous walk through the laboratory of psychopataphysics and ventriloquy at The Banff Centre

In a world which is characterized not by the autonomization of sensory channels and their corresponding media but by their remorseless interchange, the voice is kind of a sluggish impediment to the logic of sensory conversion and commutation.


Steven Connor

The Airless Chamber


In Joseph Derby’s painting, “An Experiment of a Bird in the Air Pump”, a little white bird, deprived of the spring of the air, expires in Robert Boyle’s now iconic vacuum chamber. Boyle’s demonstrations appealed at once to the veracity of eye-witness (although observers more empathic than curious are depicted as looking away), while emerging from a philosophical tradition that increasingly held the senses suspect of deception.


The macabre outcome of his performance was to suggest robust conclusions about the nature of air, conclusions uncorrupted by introspection or the narrating voice; it was the fulfillment of ‘Nullius in Verba’ (on nobody’s word), the motto of the Royal Society that Boyle was so influential in shaping.


It seems that in the ideal of this emerging society of scientists, the voice was disqualified from inquiry. Apposite then that the painter would depict Boyle as eliciting no outward signs of verbal accompaniment. His lips are closed. He seems to look out of the painting with portentous recognition. What can we assign to that look? Did the experimental criteria of replicability apply to the initial wonder as well, each demonstration as remarkable as the first?


Perhaps he is experiencing some vestige of the very introspection such empirical instruments were meant to replace. Or just maybe he is troubled by a glimpse into the future, his glass globe briefly exchanging data for clairvoyance where he spies a developed bow valley, and a pump that not only contradicts his law of soundless vacuums, but might even cajole him, berate him and, like Charlie Mcarthy [the now archetypal ventriloquy dummy from 1940's and 50's American radio and television], draw attention to the surreptitious throat movements of his interlocutor


Figure 1: Jospeh Wright of Derby, An Experiment of A Bird in the Air Pump

How it came to pass that a ventriloquial dummy was subjected to Boyle’s air pump


Excerpt from narrative written on the wall in the laboratory at The Banff Centre


I acted (I was acting anyway) as if the abduction of my dummy right from over my arm did not bother me. As they fled, the masked acousticians shot back armed looks to see, to their surprise, that I did not appear at all anxious to pursue them. The jaw of my puppet flapped loquaciously as they jostled it, and so I threw my voice into the fray in such a way as to reveal why I could only be thrilled with the way things were unfolding.


“Finally shucked your arm of me I see. I’ll curl my snickersnee (I would have said dagger here, but one of the men tripped, and the hinge I had carefully crafted many years hence bobbed tri-syllabically) around the throat of your thumb again, be sure of it.”


Having established in the abductors this belief that they had in fact delivered me from captivity, their vigilance subsided. It was at that point that I became a shadow, moving as they moved, pulled along through the alleyways and corridors, throwing my voice in supplication, and more importantly throwing it in concert with the terrain that rattled my dummies jaw, (the way the texture on a cylinder might vibrate a stylus), to sustain from the grottos of their circuitous wake the wonder that could only inspire such a theft.


Figure 2: An experiment of a ventriloquial dummy in an air pump

Vacuums, perception as action, and ventriloquial dummy-tones


Boyle’s pump has become a symbol for the scientific procedure of knowing the world through the process of removing elements of the world itself (Schapin & Schaffer, 1989). (…) If removing the air removes the aria, then sound must depend as a medium on what was removed. Yet applying such simple logic of causality to perceptual studies without considering the creative capabilities (vocal interpolation) of the participant can be the source of misleading conclusions. Take pitch perception for example: The limits and idiosyncrasies of the human auditory system as it resolves multifrequency sounds into singular pitched sensations have been studied by removing component parts of the waveform, and seeing how the brain lights up, or how people respond.


In 1841, through the use of mechanical sirens, August Seebeck found that to remove the frequency often associated with pitch perception (the fundamental frequency) did not preclude the perception of pitch. This effect has been called residual pitch, virtual pitch, or the missing fundamental illusion. Attempts to resolve how this kind of virtual pitch is perceived have led to competing models of pitch perception.


More recently, virtual pitch was used by Seither-Preisler et al.(2007) in hearing tests to measure whether musical training influences auditory perception. The observed variability in response patterns to these kind of manipulated tones — along with removing the fundamental, other manipulations were made to make the tones potentially ambiguous — was thought to suggest rudimentary differences in the perceptual systems of musicians and non-musicians.


Granzow and Vokey (2009) replicated this study. Through a series of subsequent experiments, they also showed that vocal mediation is a likely source of the observed performance gap; the experiments suggest that those that tracked the virtual pitch may do so by unconsciously voicing the missing and implied fundamental, thereby removing its synthesized ambiguity. They may in fact recreate the stimuli, performing a kind of covert ventriloquy to restore congruency to the auditory scene. These findings are aligned with current theories of embodied cognition where perception cannot be easily separated from performance or action.

The Rhythmicon


Raymond W Gibbs (2005) observes that in indo-european languages “to see” and “to know” often become synonymous. Many of our metaphors for describing the other sensory domains come from vision. Acoustic signals are often represented graphically as a set of discrete component parts or frequencies.


This spectral representation of sound emerged from Joseph Fourier’s insight that any wave form could be seen as the sum of various sine waves. These components are often represented as vermiform lines between two axes: Frequency and time.


This visualization of sonic components may contribute to the presumption that we can simply assign various auditory phenomena to sonic counterparts, like monsters to their makers. Yet we rarely experience sound in such a particulate way, and the perceptual interactions of component parts of a sound elude simple isomorphisms between physical properties and perceptual experience.


There is a tension then between our visual representations/metaphors and our more protean auditory percepts. In response to these considerations we built a version of Henry Cowell’s rhythmicon that was strung across a structure in the form of a three dimensional graph. The axes, like the jaws of a trap, were set to ensnare the spherical wave forms of their own sounding. But the sound got away


Figure 3: Dry wall screw plectrum of Distributed Rhythmicon


This tension between visual representations/understandings of sound and the sound itself can cycle in this kind of research, the phenomenal experience giving way to our visual deconstructions and back again to our renewed and perhaps altered listening. What has been the influence of this tension on the way we study sound, the way we listen to it, even the instruments we might build?

The Daxophone


In response to these questions we also made a set of Hans Reichel’s Daxophones that was constructed with a feed forward process that went from a visualization of sound energy to the use of those visualizations in the generation of unique timbres.


To make the daxophones we held a cherry plank at the 1st node, knocked it, recorded it and generated a spectrogram from the recorded sound. We then removed just enough of wood from the plank to make the first daxophone (figure 5). The first spectrogram from the knocked plank was used like a stencil to determine shape of the first daxophone tongue that in turn determined the timbre of the instrument.


Knocking the retreating plank again, a new tone at a different pitch and with a different spectrogram would emerge. This spectrogram was used to produced the next iteration of daxophone tongue and so on and so on.


Figure 4


Figure 5: knocked and retreating plank of Cherry with resultant spectrograms

Audio: Listen Up


In the laboratory, the voice-like sounds of the daxophones were amplified behind the dummy in the pump. At times the cranked pump coincided with a bow running across the daxophone producing the predicted leaps of surprise. With the help of Chris Chafe and Doug Van Nort, we are now looking into an analogue to speech recognition for the daxophone.


The dax listening agent will be able to recognize specific gestures from the instruments and produce an output response of video clips of the plank knocked at its various lengths. We look forward to improvisations between the daxophones and the audio/visual traces of their source wood.


These daxophones were performed this summer by Chris Chafe at the Subtle Technologies conference in Toronto, in the Walter Phillips gallery at The Banff Centre, the Sound and Music Computing Conference in Porto, Portugal, and telematically from Montana with musicians in Milan.











Connor, S. (2000). Dumbstruck: A cultural history of ventriloquism. Oxford University Press.

Gibbs, R. W. (2005). Embodied and cognitive science. Cambridge University Press.

Granzow, J. E., Vokey, J.R (2009) Musical Expertise: Judging the Pitch Direction of Implied fundamental frequencies (submitted)

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of ChicagoPress.

Seebeck, A. (1841). Beobachtungen uber einige bedingungen der entste- hung von tonen [Observations over some conditions of the emergence of tones]. Annals of Physics and Chemistry , 53 , 417– 436.

Seither-Preisler, A., Johnson, L., Krumbholz, K., Nobbe, A., Patterson, R., Seither, S., et al. (2007). Tone sequences with conflicting fundamnetal pitch and timbre changes are heard differently by musicians and nonmusicians. Journal of Experimental Psychology:
Human Perception and Performance, 33(3), 743-751.

Schapin, S., & Schaffer, S. (1989). Leviathan and the air-pump: Hobbes, boyle, and the experimental life. Princeton, NJ: Princeton University Press.

Video 1: An experiment of a ventriloquial dummy in an air pump

Video 2: Playing the Daxaphone

Image Gallery

Comments are closed.