Abstract
Improvisation is a fundamental aspect of music therapy practice but until recently was much neglected by researchers. This article outlines hitherto unpublished findings from doctorate research completed in 2001, at a time when few investigations were being undertaken. The findings stand the test of time and are detailed in the article as well as updated within the current literature context. The strength of the research is in consideration of underlying temporal, relational aspects of improvisation, which show how improvisations in music and in everyday conversation have both similarities and differences at deeper, structural levels. This research is of interest to current researchers, improvisers and Music Therapists.
The tiniest sound is amplified by intention. Other noises are transformed into counterpoint. The music begins. The musician waits, trying to anticipate and out-think the unthinking but thinkable direction the sounds will take. Construction overtakes the constructionalist, who can only nod approvingly as the piers and girders of musical form slot automatically into place. Here is the invisible handshake, enjoined before a motion was ever formulated.
While all UK-trained Music Therapists improvise, most do not have a discrete theory of improvisation. With renewed, current research interest in improvisation, this article aims to contribute to the literature via findings from PhD research completed in 2001, 1 which apart from one chapter (Sutton, 2007b) largely remained unpublished. In contrast with the situation today, there was little interest in the area; after requests from a number of researchers for further information, it seems timely to present the findings as a separate article, which is one of a series of new publications about improvisation by the author Sutton (in press a, b).
When Prévost wrote of improvisation 2 as ‘the invisible handshake’, he referred to a process of communication rather than of composition. This approach to improvisation is grounded in interactive communication, where the music begins not with the first sound but with the musician’s intention to communicate. This is an approach to definition that lies behind the term ‘free improvisation’, where the musician is ‘free’ to communicate spontaneously, trusting the music will develop and find form. Music Therapists think similarly that every sound (or silence) has communicative potential. The core of Music Therapists’ work is the developing relationship with the client, a fundamentally interactive use of music in improvisation. It was this dimension that was the focus for the research now described, which was studied via musical rather than non-musical phenomena.
Free improvisation surfaced in Europe during the 1960s as a kind of musical-social movement in response to the changing post-war context for music. In free improvisation, there was no pre-planned form or structure, with a creative challenge for the free improvising musician in staying connected musically, as the music unfolded as it was negotiated between those taking part (Bailey, 1993; Munthe, 1995; Prévost, 1995). The end product was a piece of music, usually produced in the presence of an audience, with musicians tending to explore aspects of music related to subtleties of texture and timbre, with a more abstract approach to harmonic elements (rather than rhythmic or metric phenomena frequently seen in free jazz). Free improvisers have often had disagreements about defining what they do, but do agree that the term captures a radical freedom from other forms of improvisation, as outlined by Watson (2013: 112–204).
In asking the question ‘Is musical improvisation like conversation?’ the research examined underlying interactive processes that enable improvisation to take place, in comparison with everyday conversation. This allowed investigation of temporal, interactional features of improvisation that are not only essential to maintaining the ongoing integrity of improvised music and of conversation but also are fundamental to improvisation utilised by Music Therapists. As well as contributing to the literature, these findings stimulate thoughts about improvisation as a method in music therapy. As the research stands the test of time, and because no other researchers have since tackled the topic in this way, older literature references have been included. This enables the reader to follow up original source material 3 to allow an examination of this context. There is also detailed information about the conversational aspect of the research, an area largely unknown to Music Therapists.
As noted, improvisation has now become a focus for contemporary researchers. Some authors have continued to comment on the links between conversation and improvisation, including focusing on free improvisation. Rose (2017) devoted a chapter to the same musicians studied in the author’s research and provided an overview of free improvisers in the United Kingdom as well as exploring the role of improvised music in education and as a way of experiencing oneself. Harris (2013) and Watson (2013) detailed the work of Cornelius Cardew and Derek Bailey, drawing out central themes with interviews from both influential improvisers and their followers. Donnay et al. (2014) noted neural overlaps in terms of processing of language and of music, concluding how rather than specialised areas for each, there are underlying auditory processing and communication systems, and that meaning-making may be situated differently in terms of brain function.
Leading important work being undertaken in Glasgow, MacDonald and Wilson have contributed significantly to the literature, including an emphasis on musicians’ thoughts about and responses to their improvisations, such as making use of focus groups and self-reporting (MacDonald and Wilson, 2005; Wilson and MacDonald, 2012, 2016). Schober and Spiro (2014) noted agreement and disagreement between two improvising jazz musicians about their co-improvised music, wondering if there may be deeper processes behind interaction. Further studies of improvisation have appeared (Caines and Heble, 2015; Lewis and Piekut, 2016), but while sometimes being concerned with interactions, no researchers are currently examining temporal aspects of interaction from both musical and conversational parameters, suggesting the author’s research is still of relevance today.
A thesis by analogy
The communicative potential of music has been compared frequently to the conversational process of talking, with general language parallels to be found in music literature (Aviello, 1994; Clarke, 1989; Hallam et al., 2016; Lerdahl and Jackendorff, 1983; Monson, 1996; Raffman, 2013). In exploring the parallels further, researchers have been challenged to find a method that covers the topic comprehensively and in depth. While the existence of a close relationship between music and language has been recognised, researchers describing these links have done so either in broad terms or with a narrow detailed view (Monson, 1996; Pressing, 1998a, 1998b; Sawyer, 1996, 1999, 2000, 2002). The music-language analogy was widespread both in the literature and in the everyday practice of musicians, but it has produced challenges for researchers, where the prevailing outlook was that perception and experience of music cannot fully be translated into words. Seeger (1977) called this dilemma ‘the linguocentric predicament’, a predicament shared when attempting the translation of human experience into words.
Music is like language because both music and language can communicate some kind of information, passed between people. There are deeper connections between music and language at other levels, as observed by Merker (2006), that ‘music, like language, avails itself of a finite set of musical patterns’ and ‘music and language are intimately related at the deepest level of their generative principles’ (p. 30). Yet music is not like language in the sense that while words and music both convey forms of meaning, the reference and context for meaning is different in each case. While all words do not carry absolute meaning, it is possible to negotiate meaning and to convey concrete information (for instance, requests and answers for the correct time, travel directions, etc.). It is not possible to derive this type of meaning from music, although in other ways music can be as meaningful as language (the transmitting of mood and emotion, for example). If questioned after a piece has finished, most listeners would agree that the experience of receiving the music was meaningful, even if there was disagreement about what exactly the music conveyed. In conversations, words also refer to elements outside the conversational setting itself, in the sense that words can carry meaning that refers to things in the outside world, as well as carrying more than one meaning (Nagy and Scott, 2000).
While conversation research continued to be published widely, this was not the case for musical improvisation at the time of the research, although as noted the situation is now changing. There have continued to be a series of shared problems in undertaking enquiry, beginning with how to precisely determine the area of study. Conversation analysts have not agreed on a singular definition for conversation, and musicians have been unable to resolve the same issue with improvisation (Bailey, 1993; Nettl and Russell, 1998). Disagreement has stemmed from varying theoretical stances, with the method and process of definition influenced by an author’s particular interest. Difficulties in gathering together the various strands of an extensive subject also influenced the ways in which researchers defined the area of study. A broad outlook covered the subject in less depth than a narrower perspective, which in itself was restricted to only an aspect of the whole. The choice was either to concentrate on certain elements or to take in the entire subject from an overall viewpoint. The solution was to consider ways in which conversations and improvisations are managed, how they begin and end and how turns are exchanged.
The research analogy ‘Is music like language?’ was explored by comparing free improvisations 4 with everyday conversation, viewed through the lenses of conversation analysis (CA) and motivic musical analysis (MMA). The research was supervised by a free improvisation and contemporary music specialist and a psycholinguist, both working within the same university. It is relevant to pause here and look at the field of CA, a method of analysis that underpinned the research.
CA
CA is a means of looking at the exchange of information in verbally improvised duets (conversation) in a detailed and subtle way, where conversations are broken down into information-giving units (Nosfinger, 1991). With roots in ethnomethodology, CA observes social interaction in a way that is termed ‘talk in interaction’ (Sidnell and Stivers, 2013). Via this method, interaction becomes the object of study of lived phenomena, particularly but not exclusively in the areas of linguistics, sociology and communication. Conversation analysts search for patterns occurring within an interaction, categorising these as a way of describing different kinds or forms of interaction. CA acknowledges the structure and organisation of speech in interaction (conversation), where single acts (utterances) are broken down into what is termed sequences, enabling sequential analysis of the interaction between speakers. CA aims to stay as close to speaker’s understandings as possible, without interpretation of meaning.
CA particularly focuses on the temporal, sequential management of sounds and silences. In conversation this is words and also sounds made by speakers (such as ‘ah’, ‘mmm’ that have a purpose within talk); the parallel in music is the sounds the players make and the silences framing these sounds. As CA is concerned with the types of exchange that occur, this kind of analysis can incorporate a very small level of detail, which was deemed appropriate also for musical material. Furthermore, CA deals with processes that involve seeking out structuring human exchanges, something that is a good fit for the form-giving exchanges that occur in musical improvisation. Finally, CA is based on analyses of recorded interactions, an approach which was closely aligned with the musical data for the study contained in audio recordings of a variety of improvisations. Video recordings were not used because the amount of additional detail required would have been beyond the remit and resources of the project, but it is something that has potential to be developed in future studies.
Musical analysis and CA
CA deals with small information-giving units. MMA was a good fit for investigating improvisations, making it possible to identify musical units of information as they occur in interaction, a close parallel with what is utilised in CA. MMA is an accepted method of analysing information-giving units in music, as, Dunsby (2002) noted, motif ‘is in fact the shortest complete idea in music’ and ‘has been a crucial element of the whole of modern music-analytical enterprise’ (p. 907). The identification of musical motifs linked to the identification of CA utterances was the main analytic focus, with knowledge that this was not a superficial method obtaining only surface information. Both Shenker (Cooke, 1987: 27–66) and Reti (Cooke, 1987: 67–115) have explored how it was possible to uncover deeper layers of information via MMA, noted in Cooke’s seminal publication, that ‘following Shoenberg’s lead a number of analysts developed quite sophisticated techniques whose purpose was to show that motivic patterns played just as important a role when they were not immediately visible (or audible) on the surface of the music’ (Cooke, 1987: 91).
For the purposes of the research, a motif was defined as a small rhythmic and/or pitch pattern comprising a single statement by one musician that was taken up by others, or by the same musician, such as a group of three notes (e.g. G-G#-A with the rhythm pattern quaver-quaver-semiquaver); this comprised a pattern of sound played by one musician that was then appropriated at other points in the piece, thus echoing the type and form of interactions that conversation analysts observe. These information-giving units were not scanned for meaning but for temporal features of interaction. They were placed in the context of the whole piece structurally and also in terms of perspectives of texture, pace and dynamic. As the CA method is well known in psycholinguistics, it enabled this form of music analysis to accommodate multilayered connections by viewing each decision-making moment in relation to other moments and so on, all within the context of the improvisation. This analysis was context-free (there was no autobiographical information about the musicians) and context-sensitive (the focus was on the sequence of the sounds and silences of the interaction). The method concentrated on the sequential detail and temporal nature of the interaction as the means of revealing more of what musical improvisation is, much as Sacks (1984) summarised that CA was microanalysis that in turn revealed more about the nature of conversation itself. The completed research found the main features of free improvised duets were also the main features of natural conversation, when comparing spoken duets (natural conversation) with musical duets (free musical improvisation).
The analytic method in practice
By using existing findings from CA research, comparisons were made between natural (everyday, spontaneous) conversation and free improvised music. These comparisons focused on the ways in which conversations begin and end, how turn-taking was negotiated and the role and function of silence. Motivic analysis of 10 transcribed free improvised pieces was the basis for a detailed examination of the musicians’ interaction (see Online Appendix 1). Using CA as a focus for this examination, it was shown that free musical improvisation is indeed a form of conversation, with a number of conversation devices present in free improvised music. The analyses also showed free musical improvisation is both similar to and different from conversation. While there were many conversation–improvisation similarities, it was in the differences that the most interesting findings appeared. These differences were where the challenge to theoretical thinking for both language and conversation researchers was found.
It was decided to transcribe improvisations to enable repeated study. Many free improvised musical pieces were too dense and fast-moving for detailed repeated listening or accurate transcription. Therefore, a practical compromise was necessary which resulted in 10 improvisations selected that satisfied the requirement for transcription. All the improvisations were created by experienced and well-known free improvisers, with the main focus on duets. The music was studied through repeated listening to the recordings of the improvisations. Transcriptions, while regarded as ultimately inaccurate versions of all that can happen musically between people, nevertheless provided a score giving the relative placing and order of musical events. Following the CA method, recordings were source material comprised of the improvisations as they unfolded, rather than verbal reports afterwards, which occur at a distance from the process of improvising.
Written records used a stave (to show relative pitch) with temporal aspects indicated by time measured in seconds (instead of metred barlines). Patterns of sound were captured by traditional notation as far as possible, with microtonal or other effects indicated graphically (e.g. for blowing through, or percussive use of, the instrument, creating resonance other than a clear pitch). Motifs were identified by coloured lines and given different letters (A, B, etc.). The analysis was completed by reference to these motifs, following the CA model (e.g. looking for repeated patterns and types of exchange, as will be described in the ‘Findings’ section). The presence of the score provided a constant record of the whole improvisation and enabled the type of detailed and repeated multilayered analysis unavailable if using only audio recordings.
As noted previously, MMA was the basis for investigating the improvisations, following the CA model, focusing on how improvisations began and ended, how turn-taking was managed and how silences were dealt with. The findings are now detailed, with terms explained in footnotes and the following references suggested for further study of the CA method (Atkinson and Heritage, 1984; Gale, 2000; Levinson, 1983; Richards and Seedhouse, 2007; Sacks, 1992; Schegloff, 2006).
Findings
In order to offer the reader potential to delve more deeply into the method used, the following record of findings incorporates working detail of CA. This introduces a new method of analysing improvisation, provides data about more subtle aspects of talk in interaction and enables a view beneath what occurs on the surface, without the added complexity of meaning-making. To enable the text to run freely, footnotes provide a further layer of information with definitions of specific conversational (CA) phenomena. The reader is invited to compare these phenomena with their own experiences of musical improvisation either via the structure of the CA method generally or more specifically through the detailed layers provided by footnotes.
Beginnings and endings
Musicians faced the same problems in beginning and ending the music as speakers starting or ending a conversation. In CA, one important problem was how to agree a conversation has begun, something usually solved by one speaker inviting the other to respond to a cue. Using the ways in which the motivic material was developed between the players as a reference point, it was possible to identify conversation components such as summons-answer, 5 identification 6 and question-answer 7 in the music. Two different approaches emerged. At the beginnings of some improvisations, players exchanged motivic material following the same patterns observed in conversations. In improvisations where the players kept their own motivic material, they continued to maintain the turn-taking aspect of the conversation components. In each case, there was a strong draw towards conversational turn-taking.
Beginnings
One interesting feature emerging from the beginnings of improvisations was that of elongated pairings, 8 where the players extended conversation opening features using musical devices of increasing tension. Where this tension was not immediately resolved (for instance, sustaining dissonance with no resolution to relative consonance), an amount of uncertainty is stimulated in the listener. In music, this is utilised to hold the attention of the audience, yet generally in conversation such uncertainty would be interpreted as a threat to the integrity of the talk.
A parallel to this type of device in conversation is a joke, where tension of uncertainty resolved unexpectedly results in humour, such as the ‘knock-knock’ joke. 9 This particular type of uncertainty works is because it occurs within a predictable format, and the cues given by the first speaker are well known by the second. Uncertainty occurs but the second speaker knows to expect a resolution, which, while expected, will also be unexpected. The unexpected-yet-expected conclusion resolves the tension, perhaps heightened by semantic leaps that have to be made in order to ‘get’ the joke. The informal telling of a joke in this way is also a type of performance within a conversation, with the better the performance by the joke teller, the funnier the joke.
In music, an audience will also expect a degree of uncertainty within performance, that such tension will resolve, although it cannot be predicted how this will happen. With improvisation, the unpredictability increases because like conversation the score is unknown to the performers in advance. In addition, the more predictably a musical uncertainty is resolved, then the less satisfying this will be. Such a resolution could be heard as a cliché.
The analyses revealed clearly defined conversational roles taken by musicians early in the music. In some improvisations, one player was in a listening role, with new musical material being produced primarily by the other player, even taking into account the simultaneous ‘talking’ that went on musically. In other improvisations, both musicians kept independent roles but maintained conversational turn-taking.
Endings
An important finding was that stopping the music was a different process from starting it. Problems encountered when trying to begin a conversation are balanced by difficulties in bringing it to a close, and researchers have identified many ways speakers ensure that there is no more to talk about. Just as a conversation will have a sequence enabling talk to begin, there is also a series of actions that have to take place in order for it to finish. The major difficulty is how to stop the momentum established once the conversation has begun, something also true of the musical improvisations studied.
A significant aspect of the process of ending the music was that of increased exchange of material between the players. Patterns of turn-taking were well established and clearly identifiable. In comparison with beginning the music, there were fewer ways used by musicians to end it. It is perhaps the lack of availability of concrete reference points that influenced this, for example, in conversation it is possible to close by cueing the making of another appointment to meet. There are specific devices such as summary statements that serve to encapsulate what has been talked about, but there are no direct parallels for summary statements in music. However, in some improvisations the musicians naturally returned to earlier motivic material. 10
Other conversation devices were also noted (e.g. continuity statement, 11 topic termination 12 and positive statement 13 ). While exact matches for these devices were not found in the music, the same form of problem solving in order to end the music was seen to be at work. The exchange and use of thematic material was a key device in enabling the music to finish. Material from one player played by the other 14 was a means of musical agreement that no new material would be introduced. With new material, the improvisation could have continued; without any, it became possible to finish the music at this point. In each example, the players decided that the music was at an end. While the reasoning behind these decisions is unknown, the patterns found by conversation analysts provide one feasible explanation.
The conversation concept described by McLaughlin (1984) as openings–closings mirroring 15 provided interesting musical parallels. 16 Comparing the motifs in the third and fourth of the Stockhausen–Parker improvisations revealed an unexpected symmetry that related to slowing the music to a close. An underlying symmetry has also been observed in conversations, helping to balance the processes of beginning and ending the talk. Another symmetry was observed with Dualysis 3, a piece where the musicians’ roles were exchanged. Again there was a balancing of some aspect of the beginning and ending of the music. In the first Stockhausen–Parker piece, there was a symmetry relating to the appearance of the home pitch, suggesting an underlying structural pull towards linking the beginning and end of the music. While it might not be unexpected to see this type of form in pre-composed music, it is surprising that it should exist in the shifting temporal phenomena of free improvisation.
One striking feature was the sudden or unexpected ending. This occurred in three improvisations, 17 which pivoted on the ambivalence and change in function of a silence after the final sound. In the moment after the last sound, an audience might be unsure of what will happen next. Anticipating another sound taking place, the listener has to interpret and re-interpret the silence as time passes. This is also likely to be true for improvisers sensing the potential either for the music to continue or to cease. In not playing or waiting to see what the other musician does, they increase the potency of the silence. At some stage, there is recognition the piece has ended. When eventually broken by audience applause, the silence becomes the final part of the piece, with the ending of the improvisation defined by the action of the audience. This is the point at which the silence becomes the frame for the music (as with the silence at the beginning).
It was hypothesised these types of silence could serve as a contact termination, the point in a conversation at which a final boundary is drawn. In conversation, a sudden ending is rare, an example being as a result of an argument (a telephone slammed down, for instance). A sudden ending in conversation does not occur as frequently as in improvised music. Silences in conversation are considered to be points at which talk could break down, whereas in music there is ambiguity. The silence marking the end of the music is therefore of great significance and is in essence part of the music and musical in itself.
Unlike conversation, musicians also have available a number of musical conventions such as coda or cadence figures to mark the large structural points of which endings are the largest. Company 6 used these in particular. Devices of coda and cadence serve in different ways to slow the momentum of the music. The coda is a means of stretching out the final section of a piece, usually with thematic reference to what has gone before. The pacing and texture of a coda frequently provides contrast, for instance, with less dense and slower moving music that also allows the music to come to a close. The cadence marks the boundary clearly, making a firmly defined end point and with extended cadence common at closing points. Free improvised music is frequently atonal, and while the cadence point is less obvious, nonetheless it can be found in the examples studied. 18 There is no clear parallel between the devices of cadence and coda in conversation, although the use of summary statement fulfils a similar purpose. Like conversation participants, musicians also find ways of drawing together recognition of what has gone before.
Knowledge of or familiarity with existing musical forms has an impact upon musical problem solving during negotiated musical endings. A similar conversational problem-solving history is also present, because everyone has had the early experience of interactive negotiation during their period of language acquisition.
Turn-taking
Close similarities in the process of turn-taking for both conversation and free improvisation were noted. Not only were conversational turn-taking devices found in the improvisations, but the devices could be categorised in detail and identified musically. In particular, devices linked with transition relevant places 19 were found, including changes in intonation, stressing or accenting final notes of a musical statement and increasing the dynamic of the music.
Overlaps 20 were present throughout the examples, and musical equivalents for these were identified. These were used for musical purposes, such as in increasing the tension of the music or frustrating the listener’s expectation of what was to come next. In addition, the devices of repair 21 and recycling 22 were discovered, and it was noted that these were both far more common and more complex than expected. In some ways, it is unsurprising that such parallels can be found, because both improvised music and conversation are ways that people communicate with each other, with the same process of communicating revealed in both talking and improvising. However, the detail and frequency of the parallels is surprising. This provides evidence for a fundamental and complex underlying process of negotiated interaction between people, whether performed in word or music. People work together in interactive settings, finding, checking and agreeing forms of shared meaning in the same ways in both modalities.
Significantly, in spite of the prevalence of simultaneous playing by both musicians, turn-taking mechanisms noted by conversation researchers are identifiable in musical improvisation. When both musicians played at once, it was common for one person to take a more dominant or leading role. This dominance was passed between musicians in a way that utilised conversational turn-taking mechanisms. In the improvisation where there were three musicians, it was still possible to distinguish the dominant and ‘listening/acknowledging’ role being shared in this way. This, too, agrees with the findings of conversation researchers (Heldner and Edlund, 2010).
Silence
At the point of undertaking the study, the topic of silence was less actively researched in the conversation literature, where there was an emphasis on treating silence as a threat to the integrity of the talk. In addition, thinking about silence in music and music therapy was lacking. The research showed how silence in music was regarded quite differently and frequently used as a means of increasing or decreasing tension. While participants in a conversation might find silence uncomfortable (unless they know each other very well), musical silence in free improvisation was heard as part of – and integral to – the whole.
A silence marks the boundary at which a performance begins. Performing musicians place emphasis on the silence into which the first sound is heard and it is unusual to see a piece begin immediately after the musicians have arrived on the concert hall platform. Silence at the beginning of a performance signifies preparation and readiness for the music to come. Performers and audience members alike share this state of expectancy, and concerts begin when this communal state is established. Silence is therefore linked with expectation from the start. It frames performance in this way. Performance in conversation (in the sense that something is going to begin) is usually framed through noting cues such as quality, intensity, tone and intonation of voice, and markers such as the use of ‘well …’. Longer silences also fulfil this function in a similar way to that described as occurring in music. 23 So, while silence in music has an emotional or aesthetic function, it is also framing performance.
In the improvisations studied, parallels were found with the concepts of switching pause, 24 hesitation pause, 25 lapse 26 and pause 27 from CA. The musical analysis produced three new categories of types of silence, namely, micropauses (functional or technical), pause points and fermata. While technical micropauses were due to the physical action of playing the instrument (e.g. breath points for a wind player), it was noted that functional micropauses could be used repeatedly during a passage to slow the overall pace of the music. Pause points were closely related to the structure of the music, with the longer fermata occurring commonly at the ends of pieces. What was most revealing was how silence in music can be integral to its structure.
The closest parallels between silence in conversation and silence in music have related to the process of interaction between participants. The differences outlined above suggested new areas of study for both musicians and conversation researchers, the most promising of which being the potential for adapting the concept of musicians’ discovery of shape, form or order in improvisations to conversations. In music, it is the physical, emotional component of interaction that is to the fore. The analyses undertaken showed how it is possible to study the interaction in detail without making assumptions about the emotional component or its meaning. It is acknowledged, however, that this component is central to what happens when people improvise, and it is also present during conversations. There is potential for researching the emotional component through the use of interviews with participants focused on what they thought was ‘meant’ at different points in a conversation or improvisation. This would be linked to structural points in improvisations or conversations.
It must also be noted that there is no single meaning possible for any one point in a musical improvisation, but a range of possible meanings, which relate to different aspects of the music (e.g. structural, motivic or thematic, textural or dynamic), occurring at different levels. This is true of conversations. There is also the ongoing nature of improvisations to be considered and, as shown in the analyses, musicians utilised both forward and backward references, for instance, restating motivic material from an earlier section of the music at a later stage. Perhaps all these aspects are possible because of the ambiguous nature of musical sounds and silences, an ambiguity that makes possible different kinds of meaning at any one point in the music.
Summary
A major finding of the research was that many of the devices hitherto thought to be the sole preserve of conversation (via CA) are present in improvisations. This is significant because it links temporal, sequential and structural features of conversations and improvisations at subtle, deeper, less overtly visible levels. This provides evidence for factors linking conversation and improvisation in terms of management of interaction when such interaction is novel and spontaneous. It has implications for the study of improvisation generally and the study of improvisation in music therapy particularly. It is a method that is free from interpretation of what meanings may or may not be derived from improvised music. It also offers both researchers and clinicians a means by which they might examine deeper structures within improvisations that relate to the management of interaction across and in time.
Conversation
The identification of many interactional devices present in improvisations offers a means by which improvised duets, trios and larger combinations can be studied. The strength of the methodology was that it recognised the central focus of the music at the same level as it is experienced. It offers a new, more detailed definition of improvisation, that is, interactional and temporal, and within an ongoing, structuring process rather than an end product. Furthermore, the research demonstrated the detail and subtlety of interactional processes within improvisation, something that both supports and grounds the work of improvising Music Therapists. In music therapy terms, it helps clarify the nature of an underlying method, with specific technical aspects alongside theoretical thinking. The methodology used allows research that does not have to encompass meaning because it focuses on sequential and temporal aspects of lived experience. It therefore offers a relevant, useful, practical dimension from which to study improvisation.
Silence
The prevalence of a structural function of silence in the improvisations studied was an unexpected outcome, as was the important role that silence can have during the ending of a piece. We know that silence frames a musical piece, but in examples of an unexpected finish, the silence was interactional, as a means of bridging the gap between performer and listener, ultimately passing the responsibility for the end point to the audience. These differences and similarities merited further detailed consideration (Sutton, 2005, 2007a), also opening up new areas of enquiry for Music Therapists that sit comfortably alongside other approaches, where a focus on temporal, sequential features of interaction can reveal deep structures and fine detail without the need to interpret meaning (although meaning can provide a further dimension).
Interpretation
The task of interpreting what is happening in musical interaction is both part of the therapist’s dilemma and also an impediment to finding any single answer or truth. Much of a therapist’s work is in coming to an understanding of what the music might mean for the client/patient, while knowing that there is no single definitive interpretation, with complex layers of meaning. The dilemma could be addressed by making use of an approach informed by CA, which would be useful in discovering how successfully the musical ‘conversation’ is managed, who dominates, who introduces and/or develops new material, how structural points (including beginnings and endings) are negotiated and so on. The strength of the approach is in the way it does not interpret what the interaction means or might mean, but that it focuses on the local management of the interaction itself. It is not suggested that this method replaces conventional means of assessing interactions in music therapy, but that it be applied as a new perspective.
The invisible handshake
As well as providing a new way of analysing improvisations, the research method has potential to inform the clinical work of Music Therapists. The free improvisation concepts outlined in the ‘invisible handshake’ quotation have encapsulated the overall focus of the thesis:
that there is an intense quality of listening required to perform – and analyse – free musical improvisation;
that form and structure evolve from the interaction between musicians engaged in free musical improvisation;
that the process of interaction is the focus of the free musical improvisation;
that the communicative intent inherent in free musical improvisation is the essential factor of human communication that is seen also during conversation.
In thinking further about this approach to the study of improvisation, it can be noted that the general outlook of musicians who engage in free improvisation is not dissimilar to that of clinical Music Therapists and vice versa. The use of CA methodology highlighted the necessity of focusing on the interactive nature of musical improvisation. In view of this, there is a case for redefining ‘improvisation’ in a way that acknowledges interactional, temporal and sequential processes taking place between people when they improvise together.
For improvisation and music therapy researchers, to omit the interactive focus during a study of improvisation is to omit the central purpose of the music. The research provides a working model offering a means by which this can be achieved while remaining as close as possible to musical phenomena. The major finding was the identification of interactive devices that, while rooted in conversation research, can be applied successfully to musical improvisation. The classification of interactive, temporal musical devices offers improvisation researchers a means by which the examination of improvisation as music-in-interaction is possible. Finally, the research has shown how free musical improvisation is a form of music making that is not only like conversation but comes closest to what conversation is. It is also closer to what occurs in music therapy practice in terms of temporal processes in music therapy improvisation, where invisible handshakes are frequent experiences – experiences which exist incredibly subtly and deeply in the silences before and after improvisers sound each other out.
Supplemental Material
BJM799076_Appendices – Supplemental material for The invisible handshake: A context for improvisation in music therapy
Supplemental material, BJM799076_Appendices for The invisible handshake: A context for improvisation in music therapy by Julie Sutton in British Journal of Music Therapy
Footnotes
Notes
Author biography
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
