9. Conclusion

¶1 In concluding with the evolutionary value of music we’ve come a long way from the making of early recordings and the information stamped on discs. But once one starts to think about the evidence of recordings the rest easily follows. Recordings show us that musicianship has changed over the past hundred years more than we could ever have imagined. As soon as we start to ask how performances of the same piece in 1900 and 2000 could be so different we are already asking questions about music’s identity and about how it moves us and why. And those questions, because they are questions about perception and cognition, lead us inevitably into the brain, how it works with sound and why it evolved to work in that way. It’s now clear that to understand music we need to know how it is constructed in the mind (mind being what the brain does) and how mind allows music to be so variable and yet apparently to move humans now just as profoundly as it has in other times.

¶2 Now we have enough recordings to see how music changes, understanding it can no longer be a matter of a taking rule-based approach to studying scores and reading the results in a historical context. Scores still matter; their notes still function in relation to rules that change over time; they can still be placed usefully in their original historical context. But what they mean depends on how they sound to people, and how they sound to people depends on the configuration of their minds; and that is not just a historical issue. I don’t believe that we can sensibly talk about music any more without considering how it sounds and how it feels. How it sounds before our own time is something we cannot know in nearly enough detail without recordings. How it feels is something we cannot know in nearly enough detail without people to study as they listen. Music is becoming a subject concerned with the present and the recent past, and if we want that to change we must start collecting data now that will give the future enough detail about the past for it to understand what music meant to us.

¶3 What I’ve tried to do in this book is far more modest, needless to say, nothing more than a small step towards a view of how music might be studied if we were to take the evidence of performance more fully on board. I’ve aimed simply to show some of the things we need to know, and some of the things we need to think about, in order to study performances using recordings. At the same time I’ve tried to outline some of the areas that might usefully make up a field of study, areas in which more research effort might be focussed. Scholars interested in working on performance are still quite uncertain how to go about it, and I’ve tried to suggest some approaches that may be fruitful. But I’m just as aware of how undesirable it is to pin a subject down to a set of given methodologies. I’ve seen how limiting that is in other fields, and I’m concerned above all to set people thinking creatively rather than to give them a programme for study. It may be useful, though, to outline some of the areas in which work on music is to be found, and the kind of work being done there, if only because it helps us to rethink their categorisation. And that’s essential: categorisation is as much as anything the problem holding us back.

Archival research: documenting recordings, performers, organisations, technologies.
Discography: listing, dating, locating recordings
Sociology: how music in performance is understood and used; listening
Information and Media Studies: how recordings are disseminated and used
Music: what performers have done with scores
Linguistics: semiotics of performance; metaphor
Gesture studies: (especially) performer’s bodily gestures
Psychology: music cognition
Neuroscience: music perception; neural substrates of music; neuropathology of music
Healthcare: music therapy; music in medicine, palliative care, geriatrics, rehabilitation, mental health
Developmental studies: musical development in infancy and childhood; implications for adult music-making
Education: recordings as models of performance; learning; performance strategies
Evolution: music’s evolutionary value; music and natural selection; music and memes
Electronic engineering: approaches to sound analysis; visualisation of sound
Anthropology: musical artefacts and practices
Ethnomusicology: music in societies; musical practices
Computer science: artificial intelligence and performance; computational approaches to performance analysis
Physics: acoustics; modelling of instruments and voices
Mechanical engineering: surface mapping of recordings
Mathematics: techniques for modelling musical shapes
Zoology: vocal communication in animals

¶4 This is a very partial list of subject disciplines in which research on music is happening at the moment. There is a huge amount of it, and almost all of it is happening outside university music departments. As a statistic consider that the abstract book of the 2006 International Conference on Music Perception and Cognition, from the first third of which I compiled this list, ran to 600 pages. It conceals a great deal of overlap, similar work being done in quite different fields, much of it unnoticed by the overlapping disciplines because of lack of communication between experts in different areas. For instance, much of the work in neuroscience takes as its evidence the compromised performance of patients with brain damage, typically caused by stroke. By correlating the area of the brain that’s been damaged with the cognitive deficits that result it’s gradually been possible to build up both a map of what happens where and also a much clearer picture of the extent to which mental skills that seem unified are in fact distributed between separate components, each of which performs a simpler sub-task. What’s particularly interesting for musicians is that so many of these modules correspond to categories of music theory. We now know, for example, that the brain has distinct neural networks for each component of rhythm as music analysts understand it: tempo, duration, pattern, and meter; and that melodic perception deals separately with pitch, contour, interval, and tonality. 1 But do music theorists know about this? Very few, it would seem. Neuroscience finds out as much music theory as it thinks it needs; music theory on the whole doesn’t imagine that neuroscience might have anything to tell it about music.

¶5 To take another example, work by scholars in Medical Psychology and Behavioral Neurobiology, and in Cognitive Neuroscience, suggests that mental singing may involve more areas associated with emotional processing than actual singing, 2 which perhaps bears out the testimony of expert score readers who claim to prefer reading scores to listening. The sounds they imagine are more emotionally laden than the process of making those sounds in performance, or than decoding someone else’s performance would be. If borne out by further work this could be extraordinarily important for music departments, yet it seems unlikely to reach them thanks to the huge disciplinary gulf separating this kind of work from musicology.

¶6 Steven Mithen reports research in anaesthesia which finds that patients recuperating after minor operations recover more quickly through music, probably because they are more relaxed. Pain tolerance is greater when accompanied by soothing music, so much so that in some procedures it’s been found possible to reduce the dose of anaesthetic by a significant amount. 3 Mithen points out that this is relevant to anthropological studies of societies that use music in healing; yet anthropologists and music therapists seem largely unaware of each others’ existence: an example of the ways barriers between disciplines prevent the development of research with huge potential benefits. 4

¶7 All the subject areas I’ve listed above have vital knowledge and techniques to contribute to the study of how music works in performance. Yet for the most part, research groups active in each of these areas are working locally, talking to their own subject colleagues, using music as an exemplification of things their colleagues find professionally interesting. But for music research to fulfil its potential, both through its applications in our lives and though its ability to link disciplines together, we need exchange, especially between all these corners of science and the humanities in which research on music is being done. There’s huge potential. Music, considered as the sound and experience of musical performance, is an extraordinarily wide-ranging subject that transcends disciplinary boundaries just as it transcends modular boundaries in our brains and social and cultural boundaries in our lives.

¶8 I hope to have persuaded you, if of nothing more, that studying music as performance leads us towards a view of music that takes the subject beyond the scope of existing musicology, which at the moment has little idea of its potential. The first task, then, for anyone starting work in this field is to find out enough of what is going on elsewhere to see how knowledge drawn from other disciplines (most of it, incidentally, empirically acquired) can be used to inform work on music by those trained in the humanities. The consequence should be, over time, that the distinction between science and humanities work on music becomes ever more blurred. It’s worth considering that if there is any humanities subject that can potentially be substantially understood through the application of scientific method it is music. 5

¶9 It must also be clear that interaction and collaboration are key ideas that should become increasingly attractive to people who work on music as performance. We’ve seen again and again how performance gestures call on multiple associations at the same time as they interact with the musical score. There’s a close similarity between my view and Nicholas Cook’s interaction of autonomous agents. 6 My agents are slightly different—some physiological, some psychological, acoustic, perceptual, in the score, in the performance, in speech, emotions; some are universals, some deeply embedded in western musical culture, some more specific to academically trained musicians, some personal—but the processes we’re hypothesising are very similar. They’re negotiating new interrelationships in every performance and probably from moment to moment within a performance; so at any moment there’s a very complex mix being produced in the mind, determining how we feel about what we’re hearing (and how we hear it). It’s important that they’re not hierarchically arranged but interact, with leading roles changing according to the nature of the ingredients and what they seem to mean to us.

¶10 This process in turn requires a collaborative approach to study. Music is so powerful because it engages the mind on so many levels. No one person can be expert in all of them. So as part of the process of finding out what is going on elsewhere we need also to make contacts, to explore possibilities for collaborative research, and to get funding for truly interdisciplinary projects. That gives us a much better chance than we have at the moment of making real progress in understanding how music works.

¶11 To think of music in the way I’ve been suggesting is also to change the way we hear it. Music cognition is a physical process that generates signals in the brain which we teach ourselves to perceive in ways that suit us. For a long time musicology has preferred that perception to be predominantly pattern-recognition, since that has proved to make the richest use of the information provided in scores. But it’s also possible for us to perceive those electro-chemical signals in other ways, not necessarily to the exclusion of pattern-recognition. Musical meaning can be generated, too, by attending to the quality of sound, to details of timing, frequency and amplitude that are not indicated by the composer’s notation and that may vary widely between instances of the piece. And the meanings generated by listening that way—by listening the way performers listen (when they’re really concentrating)—are potentially just as rich, perhaps richer, than the meanings we apply through interpretation of the score.

¶12 To experience music more fully, I’d suggest, we—and academics above all—need to make more contact with the emotional side of musical perception. Non-academic music lovers seem to have far less difficulty with this. We have laboured long and hard to make our knowledge of music different from theirs, and to train our students to be like us. We’ve learned many fascinating things about music as a result. But we’ve overlooked a lot too. A musicology of performance can help us to recover some of that and add it to what we already know. By attending to music as performance we can relearn what it is like to hear music as sound. By becoming closer listeners to sound and more responsive listeners to the quality of sound we can not just recover that experience of music but can enrich it by learning to hear more and to respond more deeply.


For a good introduction see Peretz (2003). For a schematic map of these networks, based on Peretz's work, see Mithen (2005), 63. Back to context...
Boris Kleber, Niels Birbaumer, Ralf Veit & Martin Lotze, 'The brain in concert-activation during actual and imagined singing in professionals', in ed. Mario Baroni, Anna Rita Addessi, Roberto Caterina & Marco Costa, 9th International Conference on Music Perception and Cognition, 6th Triennial Conference of the European Society for the Cognitive Sciences of Music: Abstracts (University of Bologna, August 22-26, 2006), 247-8. Back to context...
Ulrica K. Nilsson, Narinder Rawal & Mitra Unosson, ‘A comparison of intra-operative or postoperative exposure to music – a controlled trial of the effects on postoperative pain’, Anaesthesia 58:7 (2003), 699-703; Caroline Lepage, Pierre Drolet, Michel Girard, Yvan Grenier, & Richard DeGagne, ‘Music decreases sedative requirements during spinal anesthesia’, Anesthesia and Analgesia, 93 (2001), 912–16. Back to context...
Mithen (2005), 96. Back to context...
It may even be some dim realisation of this that lies behind the relative lack of attention given music by cultural studies, as lamented by Sterne (2003), 3-4. Back to context...
‘Theorizing Musical Meaning’, Music Theory Spectrum, 23 (2001), 170-95. Back to context...