By Marcus Skjold Pedersen, Junior Researcher
This research project has been an exploration of the social complexity of how music is reproduced and played from digital files. Specifically, I have focused on the role of the DJ, because I found a tension between how the DJ is seen to be a performer of music, yet is also understood to reproduce the performance of others. I started out asking basic questions, like: What even is a piece of recorded music? What does it mean to DJ? And what is the difference between listening to music and performing it? I challenged my thinking through interviews, play activities and practical involvement in event organizing. This eventually led me to discover conflicting yet coexisting ways that music is considered a social and economic phenomenon. I suggest that these ways of considering music each expands and limits our imagination for how to study and intervene in contemporary music culture in different ways.
This blog post will be a reflection piece, where I will interleave my thoughts on doing exploratory research with my current thinking about the subject of this research. I am currently at a point where some of those thoughts feel like they overlap, and I thought it would be exciting to see if they could have some greater value if juxtaposed with each other. In fact, it is a similar question of value that I have arrived at on both fronts as this research project winds down.
Mixing it up
To start with, I’ll mention the context for my project and my motivations, to get us all on the same page. I used to study anthropology, am currently studying software design and development, and have been in love with electronic music for many years. This project was meant to be an opportunity for me to experiment with the intersection of these interests. The goal for me was to figure out how to bring my existing expertise and passion into this (to me) new world of computer science. I have interpreted this as a challenge to see if I can use anthropological methods of inquiry to generate interesting ideas and approaches for designing software for a specific domain – in this case music. I don’t have the space to present my ideas in this blog post, instead, I will reflect on the value of generating them in this way.
I think this exercise is important because it is a way of promoting interdisciplinary collaboration and cross-pollinating ideas, which in my experience at least is something desired, yet seems quite difficult to achieve. A core insight of anthropology is that the researcher herself is the instrument of science, and so her interests, her position, and her assumptions are constituent parts of her field of study. And so I also think that going about research with the intention of using the findings in a different field of study, may in turn lead to different insights than what would come from a strictly anthropological interest. And so, my theory went, this project should not only be an experiment in researching a new empirical field but also be an experiment in muddying and mixing disciplinary concerns.
My first insight is this: Interdisciplinary ambitions are really hard to fulfil, mainly because by intentionally straying from a discipline, the value of my research became hard to judge. In one way, I didn’t feel like I was doing “interdisciplinarity” enough, whatever that means; It felt like I was still interviewing, still reading ethnographies and histories, still going about my research in general in the same way I usually have done, just more confused and less systematically. In another way, I felt like I had lost all relevancy to any discipline; by focusing on digital objects and technical systems rather than on people and social theory, I felt like a child trying to force a square peg into a round hole.
What should we make of these frustrations? Is the difficulty of valuing this experimental research a reason for devaluing it? It would certainly seem so, going by how I felt. At times, I felt like I wasted time, asked stupid questions, tried reinventing banal things, in general just like I was wasting resources trying to force something cool to happen that just wasn’t going to.
Yet, in other ways, this was the point: I had set out to experiment, to challenge myself and to see what would happen. It should not be expected that my experimental activities bring much value, by the standards of valuation that I knew beforehand. The corollary of this is that the research should bring value in some other way, perhaps even by helping me discover new metrics for valuing research. And indeed, I feel like it has done that, though I am perhaps still too in the weeds to clearly articulate that value.
In combination, I would suggest that there is a duality to experimental research, where it at once fails to deliver value by established standards, while it creates new forms of value, that can only be understood in hindsight, as new standards develop. This observation is paralleled in how I came to think about the subject of my research – digitally mediated music.
There are many ways of valuating music
Throughout my project, I have come back to issues of economics. When considering the role of software for DJ’s and music lovers, it always figured as a site of economic concern. For example, people would bring up streaming platforms as causing a general devaluation of music, or people would have nuanced opinions on the use of sampling and automatic content detection software. This general interest in economics was reflected in my first blog post for ETHOS lab. Related to the point above, we may instead recast the issues of economics as issues of value and valuation. In the first blog post, I argued that when music files are played to an audience, there are different economic logics that are in conflict with each other. One of these logics values rarity, and another values fame. I now propose that these are not universal or timeless logics, but two examples of systems of valuation that grow and develop as performers and audiences experiment and create new styles and scenes. Perhaps this is a banal insight: economic issues in music are entangled with the development of musical culture, and the concept of valuation takes both into account. Much like experimental research mentioned above, new developments in music then simultaneously frustrate earlier systems of valuation, while generating new ones that are only understood in hindsight.
This thinking prods at something that still confuses me about music. It seems like music is simultaneously always in an economic crisis, while constantly generating great sources of value. Music is one of the most powerful generators of economic activity, with clubs, stadium concerts, and advertisements standing as the most obvious examples. Music is also constantly generating new styles and fads, inventing new ways of getting people excited. Yet each development also seems threatening, and the livelihood of people in the music industry precarious and constantly on the verge of being ruined by the latest cultural or technological innovation.
Why do we come up with the same solutions?
While I can’t pretend to understand this dynamic in any depth, I will try to connect it to the goal I started this research with: To find interesting approaches for designing software for digitally mediated music. My theory is that software design in relation to music is currently wound pretty tightly to certain kinds of valuation systems, while it is assumed to be relevant to all of them. Take for example the question brought up in my first blog post of what to do about the fact that certain kinds of music may not be paid fairly by established systems of royalty distribution. Many directions could be taken here, but when asked about specifically software-based solutions, every person I have spoken with defaults to suggesting some sort of system for more efficiently registering each song played and automatically distributing royalties.
This is weird to me, as people seem able to imagine such a grand scale (to me quite utopian) software system, even after having themselves often just moments earlier described issues that might not be solved by such a system. Maybe it is because most people are stuck judging software interventions by certain kinds of valuation systems. This makes sense, as the people I have spoken to are not software developers themselves, and as such are not in a place to experiment with or challenge the valuation systems of software design.
When programming or designing software, one is usually thinking about solving problems, or satisfying requirements. But the goals, problems and requirements of software change and morph, just like styles of music. Currently, the influence of grand streaming catalogues and other data-heavy genres of software dominate many peoples’ imagination of what good and effective software looks like. But getting stuck on some value criteria for software limits what kinds of problems they can solve. For example, a fully automatic system for registering what songs DJs play is not as relevant if instead of valuing fame you value rarity. So certain values of software design have an affinity for certain values in music.
I hope this is where the value of messy experimentation across disciplines can show itself: By becoming aware of the different ways of valuing music, I can challenge the implicit values I hold for software design. By intentionally messing with the differences of how music, research, and software are valued, I may come up with new kinds of problems to solve and prioritize otherwise overlooked requirements.