What can’t music do?

Abstract

Background

In this article we consider how the question of what music can or cannot do is linked to the kind of light cast upon musical engagement and its outcomes.

Methods

We describe how a concern with flourishing, as opposed to a concern with more conventional understandings of ‘health’ versus ‘illness’, can help to illuminate some of otherwise invisible processes by which music ‘helps’.

Results

We show how the processes by which music ‘helps’ can slip past modes of enquiry associated with more ‘scientific’ and ‘rigorous’ investigate modes.

Conclusions

A focus on flourishing challenges more conventional imageries of what comes to count as ‘evidence’ of music’s role in relation to health and well-being.

Background

Over the last decade, research on music, health and wellbeing has burgeoned. The scholarly literature now includes systematic reviews (Evans [2002]; Clift et al. [2010]; Nilsson [2008]; Gold et al. [2009]; Daykin et al. [2012]) and at least two large handbooks devoted to music, health and well-being (MacDonald et al. [2012]; Koen et al. [2008]). In addition, the economic arguments in favour of musical ‘interventions’ for health and well-being are increasingly clear. Not to mince words, music costs less than pharmaceutical and medical procedures and – in some cases – may offer alternative forms of ‘help’ in areas as diverse as respiratory therapy, pain management, physical therapy and psychiatry (DeNora [2013a]; Ansdell [2014]; Stuckey Heather and Jeremy [2010]). Why then does music remain on the margins of ‘mainstream’ healthcare provision, more of a ‘nice to have’ add-on for hospital or care-home ‘ambience’ rather than a partner in healthcare and health promotion?

The answer to this question is by no means simple. It includes reference to many factors: organisational, economic, and occupational interests, ‘do-ability’ (i.e., easily integrated into existing institutional structures and work practise [Fujimura [1987]) and the lag between research and policy implementation. But there is another factor, one that is at once more subtle and more insidious. At the heart of discussions about the role of music in healthcare, is, we suggest, a deeply-rooted and often unrecognised confusion. This confusion arises from further, and mostly unvoiced questions about the nature of music, how it ‘helps’ and what it can do. These questions are in turn related to further questions about the aetiology and phenomenology of health and illness and thus to the question of what it actually takes to move away from the things associated with illness (identities, technologies, places, practices, communities) and toward the things associated with well-being and health (identities, technologies, places, practices, communities).

We think that research and practice in the arts and health domain has much to gain by exploring these mostly ignored, and more philosophical, questions about the nature of health and well-being. Doing so, we think, can enhance our understanding of illness and health because it draw a arts/health practitioners closer to the core of healthcare practice. Such enquiry can also help to clarify just what sort of thing music ‘itself’ actually is.

Thus, in this brief piece, we address our list of questions in two parts. First we consider music in terms of what it cannot do for health and well-being. We describe how music is not like a pill or a medical procedure, how it does not, in itself, ‘make anything happen’, to paraphrase (Auden [2014]) (‘poetry makes nothing happen’ but was rather, a ‘way of happening, a mouth’). Following this thought, we suggest that claims in favour of what music can do have been grossly exaggerated and that in considering more realistically what music can’t (and can) do, and how, we learn a great deal about how both music and health are more complex, less stable and more emergent than the RCT discourse might allow. Then, in part two, we describe how, once health and illness are understood as emergent and relational matters, as linked to communities of practice and as ultimately connected to notions of flourishing rather than health per se, it is possible to return to the question of what music can and cannot do in the sphere of health and well-being in ways that illuminate the manifold ways in which music can be health promoting. For this task we will draw on our on-going longitudinal research project known as BRIGHT.

Before getting started, however, a few caveats. First, this is not an ‘empirical’ paper. We are not attempting to back up an argument, test a hypothesis or otherwise analyse collected data. Rather, we aim to offer a reflective essay that considers the terms and conceptual boundaries of health-music as a field still in development. Second, in the contrast we draw below between quantitatively focused RCTs and the more ‘delicate’ empiricism associated with ethnographic research, we do not mean to dismiss RCTs (we realise they play an important role in evidence-based discussions and the evidence culture of policy/practice and we respect the aim of serious attempts to control, measure, and compare). Rather, we are trying to remind readers, through a consideration of ‘what can’t music do’ of ‘what can’t RCTs do’. Our argument, then, shall run as follows:

We shall suggest that a paradigm drawn from medical and, perhaps more specifically, pharmaceutical science, applied to the question of music’s ‘effectiveness’ as a health technology is only occasionally appropriate. Moreover, attempting to apply such a paradigm to all forms of music and health enquiry may actually impede our understanding not only of music and what it can or cannot do, but also, and equally importantly, our understanding of ‘health’ as a phenomenological experience. To begin our discussion we turn to the rather narrow terms of evidence-based medicine and its so-called ‘gold standard’ - the RCT (randomised controlled trial). Just what kind of light can be shed on music’s workings and how does the kind of light we shed affect what we can and cannot see?

Which methodology? Bright lights, stark contrasts

As one of us has described in previous work, randomized trials may show (or purport to show) what music can do (e.g., decrease anxiety, increase sense of well-being, decrease need for medication), but they rarely (if ever) manage to describe the actual processes by which music effects change (DeNora [2006]; DeNora [2013a]; [2013b]). Too often, music’s role as a health technology is assessed through techniques that seek to measure ‘how much’, ‘how often’ music can make a difference. The Randomized Controlled Trial, for example, proposes that music is a form of variable, which, once it is introduced, effects a change in health status.

More specifically, the investigative gaze of the RCT is directed to two temporal time periods, before and after the introduction of an independent variable. So, for example, people are [fill in the blank here: depressed, psychotic, in pain, suffering from respiratory, circulatory, dietary, rheumatic conditions] and with a probability of [fill in the blank here] percentage, music will allay some/all of the symptoms. Meanwhile, the middle period, in other words the time in which music is active, is left in shadow, which means that the processes by which music might be ‘having an effect’ are left in darkness, made mysterious.

It is worth considering more closely the discourse here: cause and effect, before and after, independent and dependent variables, treatment and control. So too, the implicit assumption that both music and illness are somehow stable, identifiable ‘things’, one of which is independent, the other dependent. This discourse is aligned with a mechanistic conception of illness, its causes and cures; one that is congruent with the classical image of causality likened to billiard balls on a table, with one ball hitting and thus transforming the state of another (Hill [2003]). On the one hand ‘illness’ is understood as a condition capable of being changed, on the other hand, music is understood as an objective medium, capable of making change. A treatment group (music) and a control group (no music) are compared and the difference between the two is measured (for example, via questionnaires administered to patients, or physical assessments such as assays or scans) so as to determine whether music ‘did’ something to make difference and how much difference it made.

In principle, then, the RCT is neat and clean: conducted under the bright light of, as it were, controlled, antiseptic, conditions. But theory and practice are not identical and the result is that, in different contexts, music’s and associated musical ‘therapies’ may be (too hastily) deemed ‘inconclusive’ for methodological reasons (lack of control; sample size; procedure) or due to actual findings (MacDonald et al. [2012]: 129; 464; Gattino Gustavo et al. [2011]; Sen et al. [2011]). The problem is compounded because, typically, when results are deemed ‘inconclusive’ more research - and more ‘rigorous’, RCT style research - is called for in perpetuam.

Moreover, because a music intervention is always a complex intervention, even in studies where a clear change in health and illness indicators is found in the ‘after music’ time period (e.g., different score on a well-being questionnaire), the cause of this change may be deemed due to what experts in experimental design speak of as confounding or intervening variables such as extra attention to the subjects (Hawthorne Effect), belief that the procedure will help (Placebo or Expectancy Effect), to the mere fact that an activity has been offered, or to factors associated with music (for example, the fact that it may involve bodily movement). Why music, then, one might ask, and not, say reading poetry aloud, reminiscence therapy, watching a film, woodworking, horticulture, or bingo?) These forms of uncertainty in turn afford those who might wish to do so to claim that there is still no clear and compelling reason to allocate funding for music alongside mainstream medical interventions (or for bingo or poetry read aloud).

The usual response to objections of this kind is that it is impossible to isolate music from other things for the simple reason that any musical intervention is inevitably complex, namely, that music cannot be neatly severed from other potentially ‘intervening variables’ (that in fact it is the complex that is being tested) and that increasingly medical tests compare the independent variable to ‘treatment as usual’ (i.e., no other activity). But then this ‘complexity’ is, one might argue, where music-in-action takes place: what is it, then, about what happens in this complicated situation? On that, the RCT is tacit; the lights go out. The investigative gaze of the RCT is, in other words, reserved for the ‘before’ and ‘after’ period and the lights are cut precisely when the really interesting stuff takes place, the ‘during’ period when engagement with music happens. And because of this power cut, the really critical matter of ‘how’ music works is left in shadow leaving unanswered the arguably most important question: in the words of Maratos et al. ([2011]), ‘music seems to work but how? It is here we come to the key point of what we have to say.

We think there is an important, unarticulated reason why music has remained mostly on the margins of healthcare to date. That reason is that the terms in which music is expected to prove itself are incommensurate with what music is, with what it can do, and how it helps. We think that until we are more able to address this question, opponents of music and health can insist that we have yet to find the ‘smoking gun’ (apologies for using a munitions metaphor). One could imagine a future in which the music, health and well-being ‘moment’ had passed….

But there are other ways of investigating whatever it is that music does, or can do, ways of shedding a different kind of light on music as it ‘acts’? We suggest that the very bright, hygienic light of the experimental situation (and the implicit ontology of music and of health/illness associated with this situation) is probably the wrong kind of light for seeing what it is that music does and what it is that music is. We believe a different, softer (dimmer!) form of light is needed in order to perceive the subtle things that music does, to see it in its natural workings and in ecologically valid circumstances. And that a slower form of dwelling with music in situ can help us to see the variegated processes by which music helps. (Think about how some things cannot be seen in overly bright light, the screen of your computer when in brilliant sunlight, for example. Then think about how the shutter speed of the camera will determine what it is able, and unable, to record on film.)

Discussion: blinded by the light?

In our opinion, it is all too easy to be, as it were, blinded by the overly bright light of the experimental gaze. We think that too much important information is lost when music is forced into the ‘before/after’, ‘yes/no’ discursive grid of variables and outcome assessments because this grid limits the ways we can see health, music and the processes by which things get ‘better’. We believe it is time to reintroduce ambiguity, the un-hygienic, and mess. Perhaps if we dim the lights it is easier to see illness and health as multivariate conditions; as not easily reducible to a binary of better/worse. It is time to reflect on how conditions of being in illness or health are instable and temporally variable, and therefore also to reflect on how it is unrealistic to attempt to measure health-states through any simple set of indicators (which themselves often privilege certain imageries of ‘health’ over others in culturally biased ways and in ways that do not match well with everyday experience [see DeNora [2013b]).

While the RCT and its associated notion of musical-medical ‘intervention’ may well be appropriate for some types of problems and questions (e.g., if one is attempting to measure and modify unilateral changes in some physical or physiological state such as hormone levels or vital signs [Fancourt et al. [2013]), the investigative gaze of the RCT is arguably not appropriate for more complex, psych-social-physical-cultural-temporally malleable phenomena and for assessing how music contributes to flourishing. Indeed, the RCT discourse and associated imagery promote overly mechanistic understandings of health/illness aetiologies (cause and effect). They lead away from attempts to explore the actual processes in time and space through which music may indeed be linked to complex forms of change in health/illness status.

In short, we believe that too much pressure is placed upon music in this paradigm, as if music were a force or power to be deployed and too little attempt is made to more slowly, more gently explore (Ansdell and DeNora [2012]; DeNora [2014]) the actual processes of how music helps (if it does) from within the situations where it is made, encountered and deployed. From within this paradigm is it any wonder that music’s effectiveness remains vulnerable to being labelled, ‘inconclusive’? It would seem that music can’t do enough to please: and yet the question remains, what can’t music do enough of?

Unanswered questions or in praise of partial shadow

In our opinion, the really rich and nuanced ‘big’ question of how music works (Maratos et al. [2011]) is perhaps not well addressed from the vantage point of medicalised discourses. We think in part this is because a great deal of the research that seeks to ‘test’ what music does begins with an overly medicalised understanding of health and an overly individualised understanding of well-being and that neither of these understandings leave sufficient space for understanding health/illness and well-being as emergent, temporally and situationally varying, phenomenological and relational entities.

By contrast, we suggest that health and well-being be subsumed under the concept of flourishing (Ansdell and DeNora [2012]). By flourishing we mean something that eludes neat categorical divisions of health versus illness or ‘better’ health versus ‘worsened’ health, whether physiologically, psychologically, socially, or culturally conceived. At the same time the concept of flourishing draws into the foreground the phenomenological experience of health/illness situations. So, for example, it is possible to ‘be flourishing’ while nearing death, mentally deranged, or in pain (for example, one may remain in, or even gain, excellent spiritual health, still love and be loved, be communicatively at one’s best, feel life’s sweetness as never before, or take comfort in the ways that family and friends have bonded because of one’s condition). Thus, as a working definition, we understand flourishing as something that goes beyond individuals and discreet physical indicators to encompass the interconnections between people and their environments, understood as a reciprocal, ecological relation:

in which both illness and health are reconfigured within a more spacious social and cultural landscape. Well-being involves our flourishing together, within our socio-cultural community (Ansdell and DeNora [2012]:)

It is here that the qualitative, indeed the ethnographic – so often shunned as ‘not rigorous’ - comes into its own and within this come concerns with ecological validity and with participants’ experiences as they participate in building worlds and shaping themselves and others through reference to these worlds. And that in turn shifts the focus from what music can’t do (i.e., is it incapable of making a change) to what can’t music do (i.e., all the things that music can do to make a change). It also shifts the focus from experimental manipulation to grounded, qualitative and interpretive enquiry (the ‘gentle empiricism’ to which we referred above [Ansdell and DeNora [2012]; Ansdell and Pavlicevic [2010]) which instead of becoming perturbed by the messy ways in which music is linked to so many ‘confounding’ factors, instead explores precisely how music comes to be inextricably attached to many things besides itself.

Music-with – the idea of ‘para’ music

As various scholars have described (Stige et al. [2012]), music is always music-plus-something else, or music with. That ‘plus what/with what’ is added locally and always during specific circumstances of music engagement. Indeed, many attempts to theorise music’s ontological status recognise the musical object as always in flux because it is always music allied with other things – practices, people, situations, technologies (DeNora [2000]). It is this melding of music and other things that Small had in mind when he coined the term ‘musicking’ (Small [1998]). By this logic, any ‘musical intervention’ is inimical to the RCT protocol since it is not merely a ‘complex intervention’ but involves – by requirement – confounded variables! So, for example, specific tunes or works can be understood to have careers (i.e., different connotations for different people at different times and under different circumstances. So too, musical experience rarely if ever takes place in a vacuum but is always linked to other circumstances, to things on-going. The question of what music does, in other words, is misleading since it is always a question of how music-plus-what can do what.

Spending time with people who are doing things with music so that music can do things with them illustrates this point and in ways that are richer and more surprising than any randomised controlled trial could ever imagine. In the next section we describe a case in point – a seven year longitudinal study of musical activity in and around a mental health domain. Our aim here is not to describe the study – we do that elsewhere (DeNora [2013a]; Ansdell [2014]). Instead we simply want to describe the varied and often unpredictable ways that music is mobilised in ways that ‘help’ on a day-to-day basis within the circumscribed space of a hospital and community centre in a large British city.

Shedding a little BRIGHT light on what music can do

The longitudinal project we will now describe is known, pseudonymously as the BRIGHT project (Borough Centre for Rehabilitation, Integration, Group Activity, Healing and Training). The BRIGHT data includes field notes, interview transcripts, Gary’s music therapy log, and audio-visual material. One of our key findings was that the furnishing of musical spaces and what those spaces in turn afford are two sides of the same coin: making the ecology for action within which one can then act, and act in ways that are seen – by self, other – to be approximating wellness go hand in hand, each helping to develop the other: manus manum lavat (‘one hand washes the other’).

On the one hand, we have seen how the musical/cultural space at BRIGHT has come to be furnished. This furnishing of space in turn distributes materials, resources, within that space in ways that afford opportunities and possibilities for action, experience and relation to others. On the other hand, those opportunities are expressed in and through the furnishing activities in the space; that furnishing consists of a wide range of heterogeneous practical engagements with music-plus. And it in turn furnish that space with their by-products, such as new identities, skills, culture, capacities and much more. So for example, consider this admixture of musical and para-musical furnishing (on the one hand) and changed health/illness status on the other, couched in the literal metaphor of flourishing:

…it’s a bit like a plant…it’s under the earth…you’ve sown a seed in the ground…and it’s germinating…but you have no idea what’s going on there, because it’s invisible. But once that shoot comes up through the earth, you can begin to see things visibly growing…so if you take that year…in the last year…and you consider that I’m a little shoot…[both laugh]…and I’ve actually appeared…visibly… and I’m developing a bit here and there…with a little bit of water and a little bit of sunshine…you know… it’s part of a whole context […] [S]o it’s this [participating in the music project] … that gives me the platform for me to come out of myself… to emerge…that’s the best way I can put it…and, um…. and this is actually hugely important to me… and it’s not just through music… it’s through personal confidence… being able to go up to a microphone and sing… to get myself back into mainstream life… and it’s all helped… it’s probably been the single most therapeutic catalyst…for my recovery….that I can actually name… and I really mean that! (quoted in Ansdell and DeNora [2012]: 108)

Conclusion

We would be hard-pressed to isolate any particular musical ‘factor’ here that ‘made’ a change; rather music and its own changing role within this field site can be understood to contribute to a wider cultural ecology, one which can be found and emerges from musical engagement as it is is coupled with many other practices, not least linguistic forms of telling about what it is that music does. The narrative about what it is that music does is part, we suggest, of the middle ground of musicking (which includes talk about what it is that music ‘does’, otherwise known as an intervening variable and giving rise to what might otherwise be termed, ‘expectancy effect’ but which in fact is better understood as ‘doing things with music’). So, for example, in relation to mental health and well-being, the BRIGHT study revealed music over time as able to:

• (1) provide a pretext for social relating

• (2) provide opportunities for demonstrating skill

• (3) provide opportunities in which to receive praise

• (4) provide metaphors and subject matter for personal and group-historical narratives (as we saw in the quote above)

• (5) provide means for shifting mood, individually and collectively

• (6) provide opportunities for bodily movement and bodily display, including dance and quasi-dance

• (7) provide opportunities for doing other things (eating and drinking, dressing up, making noise, getting out of the house or ward)

• (8) develop skills that are transferrable to things other than musical activity (such as being able to develop and present a musical/para-musical persona or presentational style)

• (9) provide a means for renegotiating one’s identity and/or role within group culture or organization (e.g., ‘I didn’t know you could sing like that’ or as in taking on a musical persona in variance with the ways one is otherwise known within a group [e.g., ‘I am channelling Elvis’]).

• (10) provide a set of events that can be recalled and thus contribute to a sense of accumulating identity (e.g., ‘You are actually a lot like Elvis’)

• (11) provide opportunities for interaction with others (and thus opportunities to forge relationships)

• (12) provide basic occupation

• (13) provide opportunities for musicianship

• (14) provide opportunities for activities linked to the original musical activities

• (15) provide opportunities for performing/demonstrating success

• (16) provide medium for reframing identities

• (17) provide a means for sharing information that might be harder to share through talk alone

Not one of these items can, in itself, be seen to constitute ‘evidence’ of any form of dramatic (‘statistically significant’) measure of health status change for the entire ‘sample’ in the same time interval, and none would in themselves necessarily lead to any measures of altered and improved health condition. Of course, if we redefine our understanding of ‘evidence’ in ways that focus on singular cases and the development of individual/collective pathways away from illness-situations, then they do indeed ‘evidence’ music’s ability to make a difference. And more delicate forms of empiricism with their focus on flourishing are less costly and easier to implement. They can also be implemented by actual users themselves, i.e., real people with lay forms of expertise (see DeNora [2013a]). Within such a perspective ‘evidence’ is retheorised and in ways that allow researchers, and users, to document moment-by-moment, ‘stiches in time’, in ways that ultimately contribute to fine-grained tapestries depicting change over the medium to long term (months/years). That change is subtle, it is never unidirectional, and can be best understood as a pattern and accumulation of moments of flourishing plus informal learning about how to achieve such moments using music. Each of these moments, each small adjustment, is part, in other words, of a gentle and – in any given moment, perhaps, imperceptible shift, an accumulation of microscopic moments and events that may add up to an enhanced capacity to flourish (over a period of time longer than the duration of an RCT). To pursue the metaphor of ‘under which light’, with ‘what kind of camera shutter speed’, just as with the question of ‘seeing’ a plant grow (the quote from Cleo above about her growth over a year), so too the only way to ‘see’ this change is via time-lapse methodology, in situ and in ‘natural light’. In all cases it is not the music per se that accomplishes this enhancement but rather what is done with, done to, and done alongside musical engagement. It is music plus people plus practices plus other resources that can make a change for the better. In a sense then, music can do nothing and everything. Its potential to promote flourishing, even in extremis, is simply waiting to be tapped.