Back in the early 1980s, the synth pop guru Thomas Dolby was asked on British television to describe his ideal synthesizer. Although I can’t find any evidence of this on YouTube, I have a vague recollection that his reply was something like: "I sit at the synthesizer, I imagine any sound, the synthesizer makes the sound and then I play it."

According to Dolby’s model, the sound begins its life in his head, the technology then converts that imagined sound, as accurately as possible, into a tangible form. This method sounds quite appealing, and I know of at least one university that set up a research programme to do just that. It is, however, entirely unlike any synthesizer I have encountered. Furthermore, it’s an ideal I find very problematic.

Let’s skip forward a few years to 1987, to the arrival of Acid House, and another interview on British TV, which tells a very different story. Here, Earl Smith Junior (aka Spanky) and Nathaniel Pierre Jones (aka DJ Pierre), collectively known as Phuture, describe the making of "Acid Tracks", widely regarded as the first Acid House record. The story goes that neither of them knew how to use the Roland TB303, which was in those days a more or less ignored little synthesizer known for its astonishingly bad imitation of the bass guitar. Pierre explains how he couldn’t figure out how to work the 303 – it didn’t come with a manual – so he just started to turn the knobs.

The result became the sonic signature of Acid House – not just the familiar squelchy Acid sound (which often steals the limelight in the Acid House story) but also to the repeating musical sequence, the use of accents, portamento and varying note lengths. When Pierre talks about "not being able to figure out the thing", I think he’s referring primarily to the 303’s convoluted step time sequencer, which is much less familiar than the filter and envelope controls common to many synths of that era.

If Phuture had rented a studio containing Dolby’s synthesizer, we wouldn’t have got the Acid House we are all so familiar with. And this is precisely because they did something with the 303 that they had not previously imagined. The music was not an expression of an idea starting in their heads.

In his 2008 paper "Putting A Glitch In The Field", sociologist Nick Prior suggests, "The history of music bulges with cases that point to the unpredictable, productive and unstable", referring to a "slippage" between "prescriptions" encoded into the machine and “the unforeseen uses that these technologies end up affording through breakdown, error and misuse". He cites the making of "Acid Tracks" as a case in point, describing "monophonic bassline generators such as the 303 misprogrammed to beget Acid House".

Although I can agree that the history of music production offers many examples of technology used in unforeseen and unpredictable ways, there are parts of his account that I feel uncomfortable with. Firstly, what is it about this particular recording session, or the technology itself, that strikes Prior as unstable? Is it because there is no identifiable sonic forethought driving the process in a specific direction? I’m not sure how this makes an activity unstable. For example, I often go to the supermarket without knowing exactly what I’m going to cook that evening. Is my shopping methodology unstable?

Secondly, what is it about this particular recording session that demonstrates misprogramming? What is misprogramming? What test could we apply to ascertain the incidence of misprogramming? Could we say "if Spanky asked Pierre to make sound x and he instead made sound y, then Pierre misprogrammed the machine"? But if Spanky asks Pierre to make any sound, the only way the 303 could feasibly be misprogrammed is if it made no sound at all. Why, then, categorise this as a case of misprogramming?

Finally, how did Phuture invoke error, breakdown or misuse to transgress the prescriptions encoded within the machine? Where is the error, breakdown or misuse in the recording of "Acid Tracks"?

The machine was not malfunctioning; the group did not misuse it. Although both Spanky and Pierre had limited technical understanding of the 303, their exploration of it, and the resultant recording, could equally constitute discovery rather than error. And if we agree that there are prescriptions encoded in the 303, how were these actually transgressed?

"We can redefine technology, not as a tool subservient to creativity or an obstacle to it, but as part of a wider context within which creative activity happens."

Did Phuture manage to turn a dial further than it was intended to go? Was there a message on the front of the machine saying, "If you have the filter’s resonance turned up to maximum, please do not wiggle its frequency control, as you might inadvertently discover a new musical vocabulary"? Was this caution somehow hardwired into the construction of the machine at a physical rather than symbolic level – for example a bit of very strong plastic inside the machine designed to prevent rapid manually induced changes in a filter’s cut-off frequency when the resonance was set beyond a certain point, as this was deemed non-representative of real bass playing?

Although the music made by Phuture that day was undeniably remarkable, I do not see anything remarkable about the role of technology here. Their hands-on exploration is a very common way of working, and I suspect if we could travel back in time and observe Thomas Dolby in his studio, he might have behaved in much the same way, and only occasionally used his hypothetical synth.

The German philosopher Martin Heidegger thought that this absorbed, non-theoretical mode of activity offers a way of understanding the world that is more fundamental than detached and theoretical analysis. Heidegger argues that the privileging of detached theoretical reflection over absorbed activity is a fundamental error at the origin of Western thought, one that casts a shadow over Western society and culture.

The comparison of Thomas Dolby’s hypothetical synthesizer with Phuture’s use of the 303 demonstrates this hierarchy. The difference between the two is this: the function of Dolby’s system is to more or less accurately express a predefined musical proposition; Phuture, by contrast, enacted a previously undefined musical proposition.

This difference – between technology as a means of construction and as a means of expression – is important when considering the relationship between musicians, technical systems and music. It means we can redefine technology, not as a tool subservient to creativity or an obstacle to it, but as part of a wider context within which creative activity happens. Recently, the artist Ernest Edmonds brought together several pioneers of computer art for an event at Sheffield’s Site Gallery at which Manfred Mohr described his creative process as "a dialogue between me and the programming language" – not merely a one-way journey from imagination to implementation. I would go one step further. Recent studies in cognitive science refer to this dialogue as "coupling", where the human agent and the technological environment become an integrated cognitive mechanism.

I suspect Prior would approve of this description. In his paper, he considers the work of French theorist Bruno Latour in the context of glitch musics (a subject uncomfortably close to my heart). Latour promotes the idea that technologies play an active role within networks containing both humans and non-humans. Decisions are constructed within these networks, and not imposed on them by an isolated human agent. If we accept Latour’s position, and in the light of Heidegger’s standpoint, we can see Phuture’s encounter with the 303 not as one driven by error, confusion or breakdown, but as an absorbed exploration, and a series of 'what if?' questions that lead to a non-theoretical understanding of the system. Here, decisions are not made in resistance to what is encountered, but in response to it.

For this column I was originally asked to consider how musicians alter the technologies they work with – principally thinking about programming environments and circuit bending.

I like to believe that all uses of a technology alter and define that technology. Any tool is subject to redefinition through its uses, and dependent on its placement within wider social and cultural contexts; for example, my Dad’s use of a screwdriver to open a tin of paint, or a friend’s use of a shoelace to commit suicide. Some musicians deliberately alter and define technologies to produce unexpected sound or music, such as Matthew Herbert’s use of a perpetually boiling kettle as a sound source, or Yasunao Tone’s foregrounding of a specific CD player’s error correction system. Both are alterations and redefinitions of technical systems.

In his paper "The Folk Music Of Chance Electronics", Qubais Reed Ghazala, a founder of the circuit bending movement, describes his working method: after opening the equipment’s shell, he places a length of wire between different points on the circuit board – if the results are pleasing, these connections can be hardwired later. He calls a circuit-bent device an "out-of-theory" instrument that is "chance wired". The circuit board is transformed into an "immediate canvas". Ghazala’s account suggests processes of intuitive, in-the-moment discovery, principally directed towards the non-expert.

"...despite the rhetoric of openendedness, the first thing people do when they encounter these allegedly open environments is to develop variations on extremely limited systems."
I think the circuit bending movement shows an emphasis on absorbed activity over detached theoretical reflection, and a preference for active, perhaps unpredictable systems, not the subservient machines dreamt of by Thomas Dolby.

Let’s consider what might happen if Ghazala got hold of one of Dolby’s synthesizers. Imagine he added an extra dial to the front with nothing on it but a large question mark. So Dolby turns up at his studio, goes to his synth, imagines a sound and starts to play. After a few moments, he notices the question mark. We tell Dolby that the question mark dial induces a different, unpredictable and unplanned transformation to the sound each time he turns it. Do you think Dolby would feel upset or cheated if we told him he could not touch this dial? Do you think he would want to twiddle it? I’m pretty sure he would, even though it directly contradicts the ideal he described all those years ago on TV.

Software environments such as Max/MSP, Pure Data and Supercollider share some commonalities with Ghazala’s description of circuit bending. Although the background knowledge necessary to make the first steps in Max/MSP is still quite considerable, it allows the user to engage in what Ghazala would call immediate and intuitive non-theory-driven exploration, functioning in real time without the need to stop the process, compile and rerun.

It is often suggested that environments like Max/MSP are open systems. These are unlike closed systems such as audio editors or software synthesizers, which have narrowly defined functions. Open systems, by contrast, offer the user a sort of blank canvas.

But for me, this distinction shares a conceptual kinship with Dolby’s ‘only limited by your imagination’ synthesis model: all technical limitations, internal characteristics and boundaries removed. And despite the rhetoric of openendedness, the first thing people do when they encounter these allegedly open environments is to develop variations on extremely limited systems. I did this myself in Max/MSP: first using a few simple objects, then, as I got more proficient, attempting to emulate machines like the TR808. Presented with hypothetically infinite openness, we start to construct systems with an inbuilt closedness. Users of these software packages may say they are drawn to openness, yet the same users demonstrate a much more significant interest in the narrower systems that can be built with them.

If we follow the paradigm promoted by Dolby’s example, a system’s structural logic would only cramp our creative activity. But if we follow a paradigm like Phuture’s encounter with the 303, or theorised by Latour, we see those structures actually facilitating such activity.

Imagine, for example, that we could change the rules of football midway through a match. Would this lead to a better game? Would fans cheer as much if a player randomly decided that stuffing the ball up his shirt and walking into the net constituted a goal? No. In football, the laws of physics, the rules of the game, the technologies, their size, shape, weight, etc, combine to keep the system in a state of equilibrium and give it significance.

I met a sports scientist who was working on tennis balls that travelled more slowly – responding to the concern that, as people get better and better at serving, tennis could become reduced to a series of unreturnable serves. I wish I could visualise the musical equivalent of an unreturnable serve, but if we want to carry on believing in a thing called creativity, let’s not assume that technical limits equate to creative limits.

Mark Fell’s collected Sensate Focus releases, Sentielle Objectif Actualité, is out now on Editions Mego.