Paper presented at the 3^rd Society for Music Production Research conference at the University of Huddersfield on 9^th September 2026

The original proposal was accepted from the following abstract:

Practice-based PhD research, both generally and in record production, faces a persistent epistemological question: what kinds of knowledge are genuinely new, transferable, and useful — and what kinds merely describe what practitioners already know in language designed for outsiders? Or, worse still, in language designed for academic gatekeeping.

This paper examines three common knowledge-generation strategies in production-based doctoral work and interrogates their respective claims to significance.

When an ethnographic researcher is already an industry participant, ethnographic study of professional practice risks producing detailed descriptions of knowledge that experienced practitioners hold tacitly. Textbooks increasingly codify industry-standard techniques; if ethnography largely confirms or elaborates what those sources contain, what is its added value? The insider/outsider distinction — central to ethnographic method — becomes problematic when insiders are studying other insiders within a relatively well-documented professional culture. Rather than research-informed practice and education, are we producing practice-informed research?

Do theoretical frameworks such as Actor-Network Theory, the systems approach to creativity, sensemaking, and semiotics generate genuinely new understanding of production practice, or do they primarily translate insider knowledge into academic discourse, making it legible to those outside the field without materially helping either group?

Does analytical knowledge about creative practice demonstrably improve that practice, inform policy about it, or contribute upstream to cognitive and creativity theory? Or does it circulate within academic systems, adding nuance to existing theoretical frameworks without creating meaningful feedback loops into either practice or foundational theory?

The paper argues that production-based PhDs are most valuable when they generate knowledge that practitioners couldn’t have produced themselves, or that they can only produce in a research context. Perhaps more importantly, how can the subjectivity and singular case studies of ethnography, autoethnography and practice research into commercial and artistic activity extend beyond its originating context?

As can be seen from the video on:

And in the following transcript, the three common knowledge-generation strategies were replaced by a model more in line with my current thinking (and, I think, more coherent as well).

Transcript of the Paper:

Hi, so this paper is called what counts as new knowledge in production-based research, and I’m going to start with just a little fragment from my Abstract.

I’ll just read this out: “Practice-based PhD research, both generally and in record production faces a persistent epistemological question: what kinds of knowledge are genuinely new, transferrable and useful?

And what kinds merely describe what practitioners already know in language designed for outsiders, or worse still, in language designed for academic gatekeeping?”

So, I then went on in the abstract to identify 3 common knowledge generation strategies in production-based doctoral work and research – but I’ve actually changed my mind about how best to present this since the abstract – We’ll keep the idea of 3 common knowledge generation strategies and we’re still talking about Production Based Practice Research but…

I’m going to take a leaf out of Allan Moore’s writing on authenticity and think about the issue in terms of 1st, 2nd and 3rd person forms of knowledge generation and new knowledge. So – just to go through the table: 1st person knowledge generation is based on self-reported evidence, is interpreted through direct phenomenology and can only really be corroborated by repeating the process. 2nd person is based on the testimony of others, reported by those participants but can be corroborated by repeated similar results if there are multiple participants. And 3rd person involves a 3rd party observer who interprets their evidence using a theory that is different to the participants’ interpretation and is corroborated by how well the evidence fits the theoretical prediction.

Some quick examples from my own research: My 2023 video album Failing Upwards is an example of 1st person research – an exploration of the experimental process of self-production that led to the creation of the album. It involved other musicians as performers but the research component was a reflection on my own practice rather than on their participation. In 2016, I worked with pianist Dr Emily Capulet on my AHRC Classical Music Hyper Production project where we worked on collaborative versions of Debussy piano prelude recordings, sharing and comparing our interpretations of the process – a 2nd person example. In 2015, for the same AHRC project, Andrew Bourbon and I, with a group of masters students, recorded this string quartet in a variety of unconventional spaces on the estate of the Bishop of Wroclaw’s summer palace in Piotrowice Nyskie, Poland. In this instance, I’d prepared a recording plan for a fragment of a Mozart String Quartet using different spaces for different sections and instruments based on a musical analysis of the piece – a 3rd person theoretical model.

I’ll just park that here and we’ll come back to it. Before we talk more about production based practice research, I want to talk a little bit about the more general subject of vocational subject practice research. And I’m going to do that by talking about the differences and the overlap between ethnography and practice research.

The 1st thing I want to point out is the difference between methodology and methods in this area. I’m lumping together ethnography and autoethnography and distinguishing them from practice research. Let’s start by talking about common research methods. In ethnography and autoethnography a foundational method is describing practice, and in practice research is doing that practice. Obviously, there are examples of people in ethnography who are participant researchers, and people in practice research who write descriptions of what they do as well as doing it – these are the primary methods in these areas. Whereas the difference in the methodology, the overarching principles of the mode of investigation, are that ethnography and autoethnography are about exploring practice, as it is already or normally being done in the world.

And that practice research is about using theory and experimentation to find new ways of doing practice. So it’s a difference between looking at what already happens and trying to find new ways of making things happen based on experimentation and theory.

So for practice research, we’re talking about using theory and experimentation to find new ways of doing practice – so lets park that as well. But what kind of new knowledge are researchers producing with these methodologies?

I’ve been writing for a long time about the ecological approach to perception and cognition and the idea of embodied cognition and 4E cognition in music practice and it’s a common way of thinking now so I’m not going to explain it. My theoretical work proposes there are 3 ways that the brain learns or understands new things: catgorisation, causality and value. So categorisation – through saying that 2 things are similar to each other in some way – from colours on the retina up to complex time-based phenomena – they have one or more shared attributes and therefore we categorise them according to that shared attribute.

Or we recognise that some aspect of experience causes another, that there is some level of causality – from things that are fundamental laws of experience like gravity to more vague connections like “I always get drunker than I should when I go out with that group of people”. And then finally, there is the question of what value that knowledge has, that in the brain we’re always looking for things that are useful, that are going to feed us when we’re hungry. But also just as importantly, solve problems for us when we face them – and that includes aesthetics: that snare drum EQ sounds better in this track than my previous setting.

We’re still talking about vocational subject areas where value is, very often an empirical topic. Does this infection control practice in hospitals save lives? Saving lives is good – has value. But if we go back to production-based practice research, And other areas of arts focused practice research, The value question isn’t that simple.

I mentioned aesthetics, but there are many value-based decisions which are subjective in other ways – participants have different priorities and goals. What is interesting to me about practice research, well, about art-based research in general, is that subjectivity in general and aesthetics in particular are very rarely discussed as part of a research project. For example: stating that one of my musical goals was to identify moments when important priorities or aesthetic choices were made and based on which criteria? They’re often accepted or justified as industry standard, or popular, catchy – it stood the test of time as part of the canon. These are all value judgments and we need to find ways to theorise them.

Our slide’s getting a bit complicated now but in addition to these 3 categories. I also want to think about practice research as having two modes of creativity, of creating knowledge.

One is much better established than the other, and that one is the output focused project: the idea of making a work of art, performing, composing a piece of music, making a record, whatever it might be, that is output focused. Whereas in process focused practice research the record or the piece of music or the artwork, whatever it might be, is irrelevant to the knowledge that I’ve learnt through doing it. And in fact, very often, conducting experiments that fail tells me as much about how to do my practice in a constructive way as doing something that is successful.

That’s much less well established. It’s one of the things that we’re working on in the 21st century music practice Research centre.

And we have 2 mechanisms for exploring that process-focused approach. One is what we call the open research project – here’s a little screen recording from the website – ORP’s are being led by our research fellows from around the world. We’ve got about half a dozen of them at the moment and there are more of them coming online all the time. And in addition to that, we’re running what we call research challenges, where we invite early career researchers and PhD students mostly, but it’s open to anyone to address a particular research challenge through process rather than through the creation of a result.

So for example, one of the 1st ones that we did was about the creative use of guitar tuning, and the idea was, choose a particular guitar tuning that you want to experiment with, experiment with it for a while, but not for too long because you’ve got to make just a single 10 minute video that demonstrates something about what you learned through that process of playing around with it, about what its potential is, how it makes you feel about playing the guitar in a way that’s different. Whatever it might be that it that the process of experimenting with it reveals to you. And then we pair people up with another person who’s experimented in a different way with guitar tuning and they’ve made their 10 minute video. They watch each other’s video and then they use what they’ve learnt from the other person’s video to do another set of experiments, produce another 10 minute video each. And then once they’ve done that, so we’ve got 4 10 minute videos, 2 from each round, – then they conduct a 20 minute discussion between them about what they learned from each other and what they learned more generally about playing guitar with alternative tuning. And this idea that there is new knowledge that could be created from experimentation without a result is what we might call the practice research version of Blue Sky research.

Earlier I mentioned the AHRC Classical Music Hyper Production project with Andrew Bourbon and Amy Blier-Carruthers and various other musicians and sound engineers. The idea was to explore how contemporary production techniques from popular music could be used to record classical repertoire in new and interesting ways. And for the most part, we knew that we weren’t going to be producing any outputs that were going to be useable. We did however finish the project with a live performance at King’s Place in London – doing live experimental performed mixes of classical repertoire for piano and strings. So clearly – all these types and approaches to new knowledge through practice research are not mutually exclusive – while the overall project had a third person goal of exploring the creative possibilities of producing classical repertoire with popular music production techniques, several of the researchers had 1st and 2nd person goals at different moments in the project and we were sometimes using ethnographic descriptive documentation of practices that we used already and sometimes experimented with unusual techniques. Some were output focused and some were process focused.

And several times different phases of the projects would move from output-focused to process-focused – so working with Louise Cournerie between 2020 and 2022 on replacing performed piano ornaments with vibrato, tremolo, distortion or other audio processing started as a series of experiments and only when we had a set of working practices and production techniques did we become output focused and produce what I call a Du(ed)it – a version of Rameau’s Enharmonique study for solo performer and editor/mixer.

[MUSIC]

And I’m currently working with Jane Chapman and Tom Armstrong on a composition of his for harpsichord and synthesisers that involves experimental production – and there’s very much a ‘we don’t know what it will be until it’s finished’ approach.

[MUSIC]

I also worked with Congolese guitarist Kiamfu Kasongo on another set of Du(ed)it pieces – many of which were discarded failures – a different approach to Louise Cournerie’s piece.

[MUSIC]

Finally as part of my Failing Upwards album fro 2023 I recorded Jas Kayser and Winston Clifford – both amazing drummers – doing guided improvisations that I then edited into tracks for the final album and the subsequent dub mixes.

[MUSIC]

So let’s just park that diagram as well and start another new one. And this is about exploring how new knowledge emerges out of this iterative and varied process. And my point is that rather than deduction or induction – the top down or bottom-up processes of traditional methodologies, we are using a version of what is known as abduction. In conventional abuctive methodology, we know where we are and where we want to be and we hypothesise the process that will get us there. In this practice research model, however, we mostly don’t know what the end result is going to be – but we do have a temporary goal, a sub-goal or a value-based direction of travel. And the process is to hypothesise some potential causal mecahnism that will get us there and then test that hypothesis (a deductive process) or experiment (an inductive process) until we feel we are closer or have achieved some aspect of that goal. And then we revise the goal or direction and try again – a continuing iterative process until we decide that either the artwork is finished or we have learned what we wanted to about our process.

You can see from the examples that my practice research projects have all had various common threads – and you could say that they have all been building to where I am now – and hopefully further into future projects. But the point I’d like to make is that this looks coherent and guided only in retrospect – the actual process was much more haphazard. At each new point in my journey, my sense of “value-based direction” at each stage was itself shaped by second-person testimony (colleagues, conference conversations) and third-person patterns I’d absorbed by watching the field of both research and practice — but the selection and weighting of all of that, at the moment of abductive synthesis, is irreducibly first-person – a unique subjective decision.

And what I’m doing with this horrendously complicated set of diagrams is to try and map the rather chaotic state of current practice research methodology without privileging any single way of working but in a way that facilitates clearer methodological decisions and promotes rigour and clarity.