1. INTRODUCTION
Throughout the twentieth and twenty-first centuries technologies originally intended for the reproduction of recorded sound – such as vinyl and the compact disc – have been adopted by performers as a means for manipulating and generating sound. The practice of exploring playback media for its creative potential is widespread across a vast array of sound-based arts and this approach to playback technology has permeated popular culture. One example – the performance of vinyl, known as turntablism – reached near ubiquity in popular music towards the end of the twentieth century, spawning and defining genres of music and performance practices from hip hop to house DJ culture. Practitioners of turntablism – from DJ Kool Herc and Grandmaster Flash to Shiva Feshareki – have developed a rich tapestry of complex and highly idiomatic techniques for performing with vinyl. Because of this, discourse on the utilisation of vinyl and other playback media often refers to these technologies as a ‘musical instrument’ that can be utilised ‘in both … creation and performance’ (Smith Reference Smith2007: 86).
As practices exploring the creative potential of playback media have developed, improvisers have looked to test the boundaries of what is possible with these technologies within performance. Through doing this, practitioners have begun to actively engage with the sonic potential of pushing technology to the point of failure (Cascone Reference Cascone2000: 13). Approaches to instigating a state of failure in playback technology are extremely varied and the manifestation of this failure even more so. However, often improvisers look to push playback technology ‘to the edge of breaking’, stopping short of removing its ability to function entirely (Kelly Reference Kelly2009: 34). Many improvisers, such as Maria Chavez, Otomo Yoshihide and Yasunao Tone, aim to maintain the technology in a state of failure, setting up systems where the playback media exists on the cusp of complete technological breakdown. Within this system of failure, the playback technology often begins to behave erratically and throughout a performance the improviser engages and responds to this, altering elements of the system to produce an evolving sense of form. When working in this way, the relationship between the improviser and the playback media becomes increasingly complex and the idea that the playback media is just an instrument that the improviser is performing with begins to feel unsatisfactory, with practitioners describing the process of working with these systems as feeling ‘more like playing with another musician’ (Dunning Reference Dunning2019: 8).
This article explores the relationship between improvising human performers and playback media pushed to the point of failure. It looks to answer questions of agency and the role of failure in improvisation, which have arisen through the author’s practice-as-research as an improviser utilising systems of failure in modified CD players. The article will begin by examining the role of playback media as an instrument and will then interrogate how the utilisation of failing playback media within a performance extends beyond that of an instrument. It will examine agency in improvisation by drawing parallels between the agency of both human improvisers and computer-based improvisation systems and the agency exhibited by failing playback media. It will argue that when pushed into a state of failure, the agency exhibited by playback media over the improvisation exceeds that of a conventional instrument. Instead, the failing playback media begins to function as a non-human improviser, similar to computer-based non-human improvisers, with agency over the musical content and structure of the improvisation and the potential to engage in an evolving, dialogical back and forth with the human improviser.
2. PLAYBACK MEDIA AS AN INSTRUMENT
As is often the case when defining musical terms, there is no singular definition for what constitutes an instrument. However, in the context of this article, the need to define ‘instrument’ is perhaps a distraction. Regardless of whether there is an agreed definition, improvisers experimenting at the limits of technological boundaries continue to self-define what is or is not a musical instrument in their practice. As improviser and researcher Thor Magnusson states: ‘[p]ut simply: anything can be a musical instrument if framed as such’ (Magnusson Reference Magnusson2019: 17). Rather than attempting to identify a satisfactory and all-encompassing definition here, it is perhaps more pertinent to examine what it is about these tools that leads practitioners to consider them as musical instruments and what this brings to the process of improvised performance.
Descriptions of playback media as an instrument can be traced back to the first half of the twentieth century. In 1930, Paul Hindemith and Ernst Toch premiered works for Gramophone discs that contained pitch-shifted sound created by adjusting the speed of playback on a gramophone; Toch described this process as allowing him to create ‘a kind of instrumental music’ from the mechanical process (Katz Reference Katz2001: 162–4). By 1948, Pierre Schaeffer had begun experimenting with playback media as a means of manipulating sound to make music. Writing in April 1948 on one of his earliest experiments with record players, Schaeffer notes:
By arranging the discs on record players, I can, using the controls, play these notes as I wish, one after the other or simultaneously. Of course, the manipulation is unwieldy, unsuited to any virtuosity; but I have a musical instrument. A new instrument? I am doubtful. I am wary of new instruments. (Schaeffer Reference Schaeffer, North and Dack2012: 7)
The level of control this technology afforded Schaeffer led him to immediately link this experience of manipulating playback technology to that of playing an instrument. The technique described by Schaeffer is still used by turntablists today, and despite Schaeffer’s initial doubts about the credibility of his ‘musical instrument’, performers who continue to manipulate playback media in similar ways often consider the playback media to be a musical instrument (Schaeffer Reference Schaeffer, North and Dack2012: 7).
The instrumentality of playback media is apparent in the language used to describe this practice by practitioners and theorists alike. Hip hop DJ Gregory Keltgen, known as DJ Abilities, describes turntables and turntablism as ‘so complex, and rhythmically challenging. To me, the turntable is the last new instrument’ (McLeod Reference McLeod, Navas, Gallagher and Burrough2014: 86). On learning to perform with turntables, performer, composer and researcher Sophy Smith notes that ‘[a]s in many popular music genres, turntablists may begin by imitating favourite musicians and experimenting with techniques in order to learn to play their instrument’ (Smith Reference Smith2007: 81). What is interesting in these descriptions is not just the use of the word ‘instrument’ but also the way in which performers describe engaging with playback technology for music-making: a focus on technique and virtuosity of performance and the ability to learn and develop a nuanced practice. Performers using playback media as an instrument can learn how to control their instrument to produce repeatable results. Pioneer hip hop DJ Grandmaster Flash highlighted the importance of this repeatability in his seminal TV demonstration of turntablism on MTV’s The Cutting Edge when he described his process of mixing between two different records on the beat as ‘cutting them on time, all the time’ (Grandmaster Flash Reference Grandmaster1983). Performers utilising playback media as an instrument primarily view it as a tool for enabling their musical agency. Through study and practice performers can learn how to control the playback media with a high degree of accuracy and repeatability.
It is, however, important to note that instruments, including playback media, are not passive objects within a performance. While performers may view instruments as just a tool for enabling their musical agency, all instruments exhibit some agency over a performance. As Magnusson describes:
Against the logic that sees musical instruments as a direct channel for our musical thoughts, seamless media that materialise our imagination, the critical analytics of musical instruments reveals them as epistemic objects with agency; as technologies that are not supposed to merely channel but to also to converse, resist, challenge, surprise, and reject their performer. (Magnusson Reference Magnusson2019: 68)
The agency exhibited by instruments is perhaps more commonly referred to as the limitations of the instrument or, for more positive connotations, idiomaticism. Like all instruments, playback media exhibit their own sets of limitations; for example, without specially cut records it is incredibly difficult to use a turntable to ‘produce defined pitches … and playing sustained tones is almost impossible’ (Hansen Reference Hansen and Williams2015: 51). Conversely, playback media also have unique traits that enable them to produce music that would not be possible on many other instruments, the most notable of course being the ability to play back recorded sound. When performers are practising and learning an instrument, they are exploring and learning how the instrument exerts agency; by learning what an instrument can and cannot do, a performer is able to effectively utilise the instrument as a tool for expressing their musical agency.
Throughout much of the past century, performers have utilised playback media as an instrument, enabling them to express themselves musically in ways that would not be possible on other instruments. While playback media used in this way does exhibit some agency over the performance, the focus of the performance is primarily on the agency of the performer. However, as the practice of utilising playback media in performance progresses, performers have looked to explore the limits of the technology. Through doing this the distinction between the agency of the performer and the instrument becomes less clear.
3. FAILING PLAYBACK MEDIA
Improvisers exploring the limits of playback media often engage with what academic and sound art curator Caleb Kelly refers to as a ‘crack’ in the technology: ‘a point of rupture or a place of chance occurrence, where unique events take place that are ripe for exploitation toward new creative possibilities’ (Kelly Reference Kelly2009: 4). There are a multitude of approaches to creating and exploiting these ‘cracks’ in playback technology with varying levels of destructiveness.
At the extreme end is a destructive approach in which the mediating device or recording is broken, perhaps beyond repair. Toward the center is the less finally destructive crack, where impermanent or nonfatal damage is caused to the player or the recording media. The least extreme approach uses the simple manipulation of playback technology—sometimes an extremely subtle intervention. (Kelly Reference Kelly2009: 32)
While the means of producing a ‘crack’ vary, the goal is usually the same: to identify and exploit a point of failure in the technology that forces it to produce sonic results that differ from the designer’s original intention. Failure is often thought of as part of a binary – something has passed or failed, it is working or it has failed. However, rather than forming one side of a binary, failure within playback media can often manifest as a continuum between functioning correctly to complete technological breakdown. Improvisers exploring the cracks of playback media often seek to utilise or highlight this transitory state of failure. Rather than seeking to cause an instantaneous breakdown of the playback media into absolute failure, improvisers instead often look to set in motion systems of failure with drawn-out or extended gradual declines from functioning to not. Or, they may look to create systems that suspend the playback media in a perpetual state of failure somewhere between functioning and complete breakdown. How these states of failure manifest vary substantially between practitioners, performances and playback media; however, the ways in which practitioners choose to utilise and engage with the continuum of failure are often very similar.
3.1. Auto-destructive sound
Once a playback media is in a state of failure the performer has two main options available: let the system run its course, leaving it to work independently until it reaches the natural endpoint of complete breakdown, or interact with the system, altering its rate and methods of failure. The first approach shares many similarities with the practice of Auto-destructive Art developed by Gustav Metzger in the late 1950s (Metzger Reference Metzger and Spieker2017). Auto-destructive Art is a performative, usually visual, artform that focuses on destruction as process for creation. However, unlike other Destructivist art forms – such as the work of Rafael Ortiz, which places an emphasis on the artist as the destroyer – Auto-destructive art places an emphasis on the artwork destroying itself with minimal interaction from human agents; a system of failure is put in place and then left to run its natural course (Ortiz Reference Ortiz and Spieker2017: 75).
An example of Auto-destructive sound is William Basinski’s The Disintegration Loops. The Disintegration Loops are a suite of pieces composed by Basinski utilising loops of magnetic tape that had been recorded 20 years prior to their use in this suite. Owing to the age and poor storage conditions of the tape loops, the substrate on the tape had begun to deteriorate, meaning that as the tape loops were played, they began to flake apart (Doran Reference Doran2012). Each piece within the suite is a recording of a single tape loop running through a tape machine; on each loop the tape degrades slightly further, causing gradual dropouts in the playback and distortions to the recorded sound. Each loop is left to run until almost all semblance of the original material is lost. While The Disintegration Loops were originally recorded live, they now exist as fixed media compositions documenting the Auto-destructive process. Owing to the lack of human intervention, Auto-destructive systems of failure are more commonly utilised in fixed compositions than improvisations. One example of Auto-destructive sound that is not fixed is artist and experimental turntablist Christian Marclay’s Record Without a Cover. As the title of the piece suggests, Record Without a Cover is a vinyl record that was sold and should be stored without any protective packaging. The original audio content of the record is a fixed composition by Marclay; however, damage incurred to the record over time – exacerbated by its unconventional storage – results in the original material being gradually transformed, destroyed and lost in a way that is unique to each copy of the record. This Auto-destructive process is generally very slow and differences between each playback of the record are usually incredibly subtle. While improvisers can and do utilise Record Without a Cover and other records kept without protective sleeves as part of performances, due to its gradual nature this Auto-destructive process may be imperceptible in an individual performance. However, if the record is utilised across numerous performances the Auto-destructive process will likely be more apparent at this macro-scale.
3.2. Interacting with systems of failure
Rather than allowing the systems of failure to run autonomously, improvisers working with failing playback media often choose to engage with the system of failure, actively changing aspects of the system. Interacting with the system can have a myriad of results in different combinations; for example, when utilising a system of failure that is designed to end in full technological breakdown, altering the system may cause the rate of deterioration to speed up or slow down. Alternatively, by changing elements of the system of failure the performer may be able to reverse the rate of deterioration, pulling the media back from complete breakdown. If the system of failure is designed to suspend the playback media in an ongoing state of failure, then manipulating the playback media may change the sonic effect of the failure.
This engagement with and manipulation of systems of failure can be found in the work of improviser and composer Yasunao Tone. Tone has performed regularly with CD players since the 1990s, creating a system of failure by modifying or ‘wounding’ the CDs with pieces of scotch tape, causing the CD player to skip, outputting a variety of snippets of manipulated audio, harsh clicks and pops. During performances, Tone ‘knocks and jolts the CD tray … gently nudging it to continue reading the prepared CD’ (Kelly Reference Kelly2009: 238). By doing this, Tone influences the failing playback media, using his interaction to prolong the state of failure and to change what audio from the CD is being processed by the breakdown of the technology. Although Tone’s interaction with the system appears similar to that of an instrumentalist performing with their instrument, unlike an instrumentalist, Tone is unable to learn how the system of failure will respond to his actions as his manipulation of the CD causes the CD player to perform in unpredictable ways.
3.3. The Unpredictability of Failure
Owing to the nature of how playback media can fail, these systems of failure often cause the playback media to function erratically, producing unpredictable audio manipulations, textural and structural changes. Performers working with failing playback media often list this unpredictability as being a driving force behind their work. For example, Christian Marclay states:
When something goes wrong [in a performance], like when the needle skips, something unpredictable happens …. In that incident, something new and exciting happens. For me, it has creative potential. (Gross Reference Gross1998)
For Marclay, the unexpected moments in a performance caused by the failure of the playback media are generative opportunities, where the playback media can drive a change in the direction of the performance. Similarly, experimental turntablist and sound artist Maria Chavez describes:
By experiencing chance situations during performance, this created the basis of developing my vocabulary with the turntable. The more that ‘went wrong’ the more I learned about new sound possibilities, i.e. when a needle broke a certain way it began making interesting sounds on different records. The more they broke, the more sounds began to emerge that wouldn’t have without accidents and damage. (Neumann Reference Neumann2013)
Here, Chavez highlights the creative potential for the failing playback media to manipulate the source material in new and interesting ways, creating sonic manipulations that she would not have anticipated or sought out on her own. Both Marclay and Chavez view the unpredictability of failure as an intrinsic part of their practice. The shifts and changes caused by the instability of failure result in an unpredictable improvisatory experience where the agency of the failing playback media begins to determine key aspects of the performance.
4. PLAYBACK MEDIA AS AN IMPROVISER
As discussed, instruments exhibit a sense of agency over performance through their unique set of features and limitations (Magnusson Reference Magnusson2019: 68). However, with failing playback media, this far exceeds the level of influence commonly found in other improvised performances. When a system of failure is set into motion, the instrument gains influence and control over broad decisions of form and structure, disrupting the conventional relationship between performer and instrument. As sound artist and turntable performer Graham Dunning describes it, improvising with systems in this way can often ‘feel more like playing with another musician, another entity’ (Dunning Reference Dunning2019: 8). Dunning’s sentiment that failing playback media can feel like ‘another musician’ is mirrored in the language of a variety of other practitioners working in this way; Chavez personifies vinyl and turntables by describing failure as giving them ‘a personality’ (Uszerowicz Reference Uszerowicz2017) and composer and hardware hacker Nicholas Collins describes his use of skipping CD players as giving a performance ‘the tension and sense of uncertainty associated with improvised music’ (Collins Reference Collins2009: 3–4).
Playback media pushed into failure begins to exhibit a sense of agency over an improvisation that can mirror that of human improvisers. Rather than functioning as an instrument, failing playback media can be better understood as operating as a non-human improviser. The ways in which failing playback media manifest as a non-human improviser are many and ever evolving, varying between practitioners, playback media and performances. It is not the aim of this article to provide an all-encompassing study of the myriad ways failing playback media can function as an improviser or to suggest that the manifestations discussed are ubiquitous to all examples of improvisation using failing playback media, but instead to provide some examples that highlight this potential. However, before discussing how failing playback media feels ‘like playing with another musician’ (Dunning Reference Dunning2019: 8) it is first pertinent to briefly examine how developments in computer-based non-human improvisers can inform our understanding of what a non-human improviser is and does.
4.1. Non-human improvisers
Developments in computational processing power and approaches to artificial intelligence have led to an ever-expanding field of research into the creation of computer-based non-human improvisers (Herremans, Chuan and Chew Reference Herremans, Chuan and Chew2017). To what extent these systems are considered improvisers is a point of some contention (particularly among human improvisers) and one that is perhaps marred by an anthropocentric view of what improvisation is, as George Lewis observed:
it seemed to many that there was just something wrong with the notion of computers as improvisers. There was something special about improvisation—something essential, fundamental to the human spirit—that one just couldn’t, or shouldn’t, approach with machines. (Lewis Reference Lewis, Dean and McLean2018: 127)
However, even within ‘conventional’ improvisation there are already a multitude of non-human elements influencing the process and result of improvisation. Marcel Cobussen’s Field of Musical Improvisation Theory identifies and explores how a wide variety of human and non-human elements combine and influence each other during the process of improvisation; Cobussen describes improvisation as a ‘multiplicity made up of many heterogeneous actants of many different natures, a network of interconnected, human and non-human, nodes’ (Cobussen Reference Cobussen2017: 83).
If an improvised performance is the product of both human and non-human elements, it is hard to see why improvisation should be considered a purely human activity. However, the boundary between influencing an improvisation and improvising is unclear. This is likely due in part to our conception of what it means to ‘improvise’ being derived directly from our human experience of creating music in this way. Computer-based non-human improvisers predominantly seek to imitate, or at least appear to imitate, the behaviours of human improvisers. For computer-based non-human improvisers to be considered successful by their researchers, they usually undergo some variation on the improvisatory equivalent to a Turing Test: the system plays with human improvisers and the researchers see if the human improvisers describe the non-human system as seeming to improvise. Researchers seek to ensure ‘that contributions [from human and non-human improvisers] seem equally valuable’ and that the results of improvisations utilising computer-based non-human improvisers are ‘convincing [on their] own terms, with the same richness of sonic language, complexity of form, and effectiveness of expression that could be found in human-only music’ (Young and Blackwell Reference Young, Blackwell, Piekut and Lewis2016: 508). While improvisers do not need to be human, our understanding of what it means to improvise is still deeply rooted in the human experience. For this reason, when discussing the ways in which failing playback media can function as an improviser, we will primarily be relating this back to how it mirrors the behaviours of human improvisers. However, that is not to say that these systems can only operate with human improvisers; if and how failing playback media can function as an improviser without interaction with human improvisers is an interesting and rich avenue for future exploration.
4.2. Communicating with Failing Playback Media
Within group improvisation, communication between human improvisers is often considered to be a vital aspect of the process of improvising; in order to effectively improvise, non-human performers ‘must be able to react in a convincing way to stimuli, sonic or otherwise’ (Young and Blackwell Reference Young, Blackwell, Piekut and Lewis2016: 509). Human improvisers communicate their intention for the performance in a wide variety of differing ways, encompassing ‘verbal communication, non-verbal communication (e.g. eye contact, aural cues and body language) and musical communication’ (Seddon Reference Seddon2005: 47). Through this communication performers seek to share their ideas for the content of the improvisation in order to negotiate a collaborative performance where ‘the performance emerges out of the actions of everyone working together’ (Seddon Reference Seddon2005: 49). To what extent the communication between improvisers is successful is of some debate; studies have shown that improvisers often misinterpret the intentions of other performers in the group, yet despite this the performance is able to continue (MacDonald and Wilson Reference MacDonald and Wilson2020: 104–5). It appears that what is more important is the perception of clear communication, as music psychologists and improvisers Raymond MacDonald and Graeme Wilson state:
it did appear important to individuals that they felt they could make sense of what others were doing—that they could trust their own judgements, and could trust others to be making sense of them as well. Although improvisers may expect to share understanding, this may not be as important as believing that shared understanding exists. As long as each person thinks that they are working together, their interaction will show musical qualities. (MacDonald and Wilson Reference MacDonald and Wilson2020: 106)
For human improvisers, it is important that the performers they are working with appear to be able to interpret their communications, respond to it and communicate their own ideas. Therefore, for a non-human improviser to improvise it is important that to other improvisers it feels as though it is communicating, irrespective of whether there is an understanding or sentient intention behind the communication taking place.
During a performance, it is most common for human improvisers to communicate both through aural interpretations of musical intent and visually, through eye contact and body language. Due to its physical nature, playback media can often communicate visually to human improvisers, mitigating the issues surrounding physicality and embodiment that often arise with computer-based non-human improvisers (Frisk Reference Frisk2020: 38–9). For example, improviser and turntablist Otomo Yoshihide often replaces the needle in a turntable’s cartridge with a large spring and then overlaps and stacks various records in a pile on the platter, creating an uneven and unstable surface. As the turntable spins, the spring bounces over the records, lifting itself off the surface and causing the arm to swing in different directions and the records in the pile to shift and slide on top of each other. Visually this process aligns with the sonic result and by watching and following the physical motion of the spring, the turntable’s arm and the records beneath, Yoshihide can gain insight into what may happen next sonically. Just as an emphasised bowing motion on a violin may indicate to other improvisers an accent or downbeat, the physical motion of playback media can often offer some visual communication of the forthcoming sonic result. Similarly, just as human improvisers listen to and interpret the playing of other human improvisers in order to attempt to understand the intention and direction behind what they are playing, by listening and interpreting the sound of failing playback media human improvisers can gain insight into not just what it is playing in that moment but also where it may be going in future. The gradual build of feedback from a record cartridge, the looping of audio from a stuck CD player or the increase in dropouts from a worn tape loop can all communicate to human improvisers the future musical direction of the failing playback media.
While failing playback media can communicate both visually and aurally to human performers, there is no immediately apparent way for the human performer to communicate back. Computer-based non-human performers often utilise machine listening and vision to enable the system to engage with the aural and visual communication between human improvisers. It may seem as though failing playback media is incapable of communicating in dialogue with human improvisers, as it is generally not capable of listening and analysing. However, this is only true if we limit ourselves to a consideration of the specific ways in which human improvisers communicate.
There are means through which human improvisers can communicate their intention for the improvisation with the failing playback media. The physical nature of the playback media enables human improvisers to communicate with the system of failure directly, by altering aspects of the system by touching and manipulating the playback media. This method of communication between human and non-human performers is very similar to the mode of communication utilised in one of the earliest examples of computer-based non-human improvisers. Formed in the late 1970s, improvising ensemble the League of Automatic Music Composers consisted of a mix of human and non-human performers; during performances the human performers in the group ‘took hands-on improvisative roles with their machines, [doing so] from a collaborative rather than an instrumental standpoint, negotiating with their machines rather than fully controlling them’ (Lewis Reference Lewis, Born, Lewis and Straw2017: 96). This process of physical communication and negotiation is one that is commonly found in improvisations utilising failing playback media. Rather than producing a predictable outcome in the way in which pressing a key on a piano would, communicating with and influencing a system of failure through touch can produce a wide variety of responses, ranging from the expected, to the unexpected, to no change at all. For example, when Tone knocks the failing CD player, he is doing so in order to communicate to the CD player how he intends to progress the improvisation, similar to how an instrumentalist may try to direct human improvisers by changing tempo, rhythm or dynamic. Just like when playing with human improvisers, Tone has no clear conception of what the end result of his communication will be; the CD player could skip forwards or backwards as Tone intended, remain playing what it was playing or shut down completely (Kelly Reference Kelly2009: 238). The option for the system of failure to ignore the communication from the human improviser may at first seem like a flaw in its ability to communicate. However, this aligns very closely with the idea of ‘musical rudeness’ within improvisation; to ‘choose to not listen’ is for most improvisers considered a valid musical decision (Frisk Reference Frisk2020: 36).
While they may not communicate in the ways that are common in performances between human improvisers, the communication between human improvisers and failing playback media nevertheless manifests in a manner that is familiar to human improvisers. As MacDonald and Wilson have shown, it is more important to human improvisers that clear communication appears to exist irrespective of the actual effectiveness of the communication (MacDonald and Wilson Reference MacDonald and Wilson2020: 106). Familiarity with the means and outcome of communication between human improvisers and failing playback media gives the impression that communication is taking place and because of this, improvisations appear to emerge ‘out of the actions of everyone working together’ rather than the human improviser working as a leader (Seddon Reference Seddon2005: 49). Owing to the nature of failing playback media, responses from the playback media may be unpredictable, creating a dialogue between the human and the non-human performers that has the potential to be surprising and engaging for musical exploration.
4.3. Unexpected Music
One of the most commonly discussed improvisatory traits of failing playback media is its potential to produce ‘unexpected music’ (Knížák in Kelly Reference Kelly2009: 142). While commonly mentioned by practitioners, the implications of the unpredictability of failure are relatively underexplored: rather than being viewed as a single improvisatory trait, the potential for the unexpected underpins many of the ways in which failing playback media can function as an improviser.
Turntablist Graham Dunning performs what he has dubbed Mechanical Techno; by stacking multiple modified records on a single record platter separated by blocks of wood, Dunning utilises the spinning motion of the turntable to loop content from the records and to trigger, both electronically and mechanically, other percussive and synthesised sounds to create live dance music. The construction of Dunning’s spinning tower of records is often complex and simultaneously precarious. Because of this, small elements of failure often occur during a performance: ‘a small accidental pull on a cable might physically move the tone arm, resulting in a completely different loop. Leaving the machine running for too long whilst preparing the setup might lead to the groove wearing down and becoming audibly more noisy’ (Dunning Reference Dunning2015a). However, rather than trying to mitigate these unexpected elements, Dunning embraces them as these results from the failing playback media often force him to reconsider what he is playing: ‘[t]he chance elements and unpredictable aspects lead to [music] the artist would never think to deliberately make’ (Dunning Reference Dunning2015b).
Prolific improviser Mattin describes this process of reconsideration as being fundamental to the process of improvisation between human improvisers: ‘improvised music forces situations into play where musicians push each other into bringing different perspectives to their playing’ (Mattin 2005). As performers communicate, interpret and misinterpret the ideas of other performers, the improvisation can often move in directions that are unexpected to the members of the group; in order to incorporate and respond to these unexpected elements improvisers must often reconsider elements of their performance – for example, what they are currently playing, their intentions for the rest of the improvisation and their overall role within the performance. Just as human improvisers can influence failing playback media through touch during a performance, due to its ability to produce unexpected results failing playback media can function generatively within a performance, producing results that often cannot be fully anticipated by human performers. This gives the failing playback media agency over the performance as the unexpected outcomes of the system force the human improvisers to reconsider their intentions for the improvisation. The human improvisers and the failing playback media influence and shape each other’s performances, producing the musical output collaboratively. Rather than the playback media primarily enacting the agency of the human performer, the failing playback media works semi-autonomously with the human performer, each responding reactively to the other.
Failing playback media has the potential to produce unexpected results in a variety of different ways. For example, Chavez describes ‘creating unstable situations’ in order to structure her turntablism performances; by pushing playback media to failure, Chavez is able to utilise the unpredictable rate of breakdown as a structure for her improvisations (Neumann Reference Neumann2013). One method often utilised by Chavez to create ‘unstable situations’ is by layering fragments of snapped vinyl records on the platter of a turntable (Chavez Reference Chavez2012: 78; Neumann Reference Neumann2013). As the platter rotates, the needle drags across various shards of different records, creating a montage of different recordings, punctuated with percussive pops and scratches as the needle bumps up and down the edges of the fragments. As the needle bumps into the fragments, the tone arm of the turntable moves unpredictably, deciding for the human performer which sections of the record fragments are played. During this, the tone arm may bounce off the platter away from the records, or the cartridge may become detached from the tone arm, creating a period of silence or the end of the performance (Chavez Reference Chavez2012: 78). During a performance, Chavez nudges both the tone arm and the record fragments, influencing and altering the system of failure; by doing this Chavez works in collaboration with the failing playback media to create a structure for the improvisation.
Beyond creating structures for improvisation, failing playback media can also manipulate and affect the audio it is attempting to play back in unexpected ways. Tone’s work with ‘wounded’ CDs utilises this to create evolving and shifting manipulations of the source audio that are unique across performances. Owing to the unpredictable way the error correction built into the CD player attempts to interpret the ‘wounded’ CD, the ‘sounds produced by the affected discs [are] never quite the same’ (Stuart Reference Stuart2003: 48). Each time the disc is played, the error correction causes the audio on the CD to be distorted, warped and skipped in different and unexpected ways. Similarly, sound artist Lucas Abela’s High Powered Turntables utilised industrial motors to spin records nearly a thousand times faster than intended, which he then played with amplified skewers instead of record needles. Owing to the difficulty of holding the skewer on the spinning record and the damage this does to the record, Abela’s chaotic system manipulated the audio on the records unpredictably into imperceptible ‘loud bursts of sound’ punctuated by ‘occasionally recognizable high-speed cartoon-like tunes’ (Kelly Reference Kelly2009: 1). While failing playback media can produce unexpected results, they are not totally random, as they are still working within a system constructed and set into motion by the human improviser, with the source sounds on vinyl or CD pre-selected by the human improviser (Kelly Reference Kelly2009: 238). This means the musical results of the improvisation usually align with a fixed aesthetic, allowing the failing playback media to perform within the bounds of a specific style or improvisatory approach. In group improvisation settings, human improvisers will often share a ‘common purpose, set of skills’ and/or ‘musical vocabulary’ that can help them create music together with a sense of cohesion (Ashley Reference Ashley, Hallam, Cross and Thaut2008: 418). The ability to produce unexpected musical results that still adhere to a fixed aesthetic allows failing playback media to improvise in a manner that mirrors that of human improvisers while retaining a unique sense of identity as a non-human performer. This, combined with the communicative nature of performing with failing playback media, allows human improvisers to utilise failing playback media in a way that feels like performing with other human improvisers.
5. CONCLUSION
This article has explored the role of playback media – such as vinyl and CDs – in improvisation, looking to understand how the agency of the playback media over the performance changes when pushed into a state of failure.
Since the second half of the twentieth century, performers have adopted a variety of playback media and utilised them in different ways as instruments for live performance. Performers looking to explore the possibilities of playback technology have pushed at the operational boundaries of these technologies, actively engaging with the unique creative potential these media exhibit when in a state of failure. Improvisers working with failing playback media often personify the technology, describing it as feeling ‘more like playing with another musician’ (Dunning Reference Dunning2019: 8). By drawing parallels with how human improvisers perform and interact, this article has examined why performing with failing playback media can feel like playing with another musician, identifying that, when utilised in a state of failure, playback media can be understood as functioning as a non-human improviser. Two ways in which failing playback media can be understood to be improvising have been proposed: by exhibiting agency over the structure and musical content of the improvisation, and through the ability for the human improvisers and the failing playback media to communicate.
Understanding the potential for failing playback media to function as an improviser offers a reframing of the role of the playback media in performance. In this article we have primarily considered similarities between failing playback media and human improvisers; going forward it would be interesting to explore in more depth what failing playback media and other non-human improvisers bring to improvisation that goes beyond the human experience, considering how the utilisation of these devices may alter and de-anthropocentrise our understanding of what it means to improvise. It is hoped that this shift of perspective will encourage exploration through practice, further exploring the creative applications of failing playback media as a non-human improviser.
