What is music analysis? Problems and prospects for understanding Aboriginal songs and performance.
Different concerns over music analysis have arisen among historical musicologists specialising in the study of Western art music. Here the issues relate to a rejection of analyses that treat compositions as autonomous works of art, and a desire to deconstruct the ideological underpinnings of classical music. Some music theorists' replies to these criticisms have pointed to diverse and self-reflexive possibilities for music analysis.
Drawing ideas from all of these sources as well as from my own fieldwork and analysis, I conclude that, despite its inherent cultural trappings and other limitations, analysis can and does provide an invaluable means for understanding Aboriginal songs and performance, and that the subjective input of the music analyst--something which is recognised in analyses of Western art music--could be more accepted as a normal part of analysis of Aboriginal (and other non-Western) musics.
Music analysis of various sorts has been integral to European studies of non-Western music since the nascent period of the 1880s, while musical transcriptions of non-Western music found their way both into the notes of voyagers and missionaries and onto the pages of music encyclopaedias well before this time. The single factor that secured an important role for the analysis of music of oral traditions was the advent of recording technology at the end of the nineteenth century. (1) One consequence of recording technology was the reification of musical performance in ways that were previously not possible, and which now rendered the audible output of performance an object of study. (2) Another consequence was the emerging important status of institutions and individuals that collected, owned or analysed recordings.
During the heyday of comparative musicology (through the first part of the twentieth century), analysis comprised something of a sine qua non of musicological studies of non-Western music. The tools of transcription and analysis were commonly regarded as scientific, objective and value-free. However, with the growing body of ethnographic data from many world cultures (which helped draw attention to cross-cultural differences), the growing influence of interpretational methods derived from other disciplines (especially anthropology), and more recently the growing influence of critical theory in musicological circles, ethnomusicologists have come to realise that our analytic tools and methods are neither objective nor value-free, and not necessarily scientific.
Particularly since the 1970s, this shift of attitude has been accompanied by the flowering of many different analytic models and techniques designed to fit better particular music-cultural situations. (3) At the same time, a range of serious ethical questions about the analytic endeavour has been raised, while a significant cross-section of ethnomusicologists has abandoned analysis as a primary means of addressing important issues about music and culture. Some trends in analysis-related work over the past 20 years have included semiological examinations of music and performance, cognitive studies of music and music perception, and computer-aided analyses of pitch and rhythmic aspects of music. James Porter (1995) has noted a recent increase in studies which consider music and performance from both cultural/ethnographic and music analytical perspectives.
This article concerns the role of analysis in studies of Aboriginal music, but some points raised relate more generally to studies of non-Western (and even Western) music. Music analysis ultimately embodies more than just a consideration of sound. Indeed, within studies of traditional Aboriginal songs, consideration of factors such as text, dance, ritual context, and inter- and intragroup politics have come to be regarded as crucial elements for understanding both the internal systematics of song performance and the broader cultural roles of Aboriginal performance practices. Discussion of analysis in this article, however, is limited to its application to musical sound. It is analysis of the aural component of music that has come under particular scrutiny in recent times and one of the main aims of this article is to address some of the major issues of concern. (4)
Australian Aboriginal songs as objects of musical analysis
Very little musicological research on Aboriginal music took place during the first half of the twentieth century. Music analysis from this period was based on small samples of recorded songs and was focused on the characterisation of the musical content of Aboriginal music. A prime example of this may be seen in the work of musicologist E Harold Davies (1927, 1932), who used recorded song items for the purpose of extracting comparative information about musical scales, as well as for descriptions of vocal range and timbre, and melodic contour and ornamentation. Only a few melodic phrases are transcribed in their entirety (without texts), this in order to indicate the scope of melodic variety of Arrernte song (1927:88-90).
The 1950s and 1960s saw some significant work on Aboriginal songs resulting from the sequential efforts of anthropologists and musicologists, in particular Catherine Ellis' work on Desert songs recorded by Strehlow (published in Ellis 1964 and Strehlow 1971), (5) and Trevor Jones' work on Arnhem Land songs recorded by anthropologist AP Elkin (Elkin and Jones 1958). Musicologists Ellis and Alice Moyle each undertook their own extensive recording and analytic projects during this period. Their analytic work provided both the first extensive studies of particular song genres and series, and comparative examinations of different styles within and across broad cultural areas.
Since the late 1970s, music-analytic work on traditional Aboriginal songs has become more consistently specialised, focusing on the conventions of musical style and performance practice associated with specific song genres and/or song series within a single culture (oftentimes, as performed by particular groups or individuals within a given culture). Many studies in recent years have sought to address particular questions regarding performance practice--for example, examinations of fixed versus variable elements of performance, diachronic changes in performance practices, and aspects of rhythmic and melodic alignment, to list just three examples.
Recent analyses of Aboriginal songs have been marked, too, by growing use of both ethnographic and musical data in analyses of songs. This may be seen both in the greater use of ethnographic data by musicologists, and in increasing music-analytical sophistication among anthropologists carrying out research on Aboriginal songs. A growing desire to incorporate more sophisticated analyses of text and dance has also resulted in interdisciplinary collaborations, for example, as between linguists and ethnomusicologists (6) and between ethnomusicologists and choreologists. (7) The past 15 years have also been marked by a growing literature concerning contemporary forms of Aboriginal song, and music analysis has played an important role in some of this literature.
In analytic studies of traditional Aboriginal music to date, the two broad geo-cultural areas that have received the most (though not exclusive) attention are the Central and Western Deserts on the one hand, and Arnhem Land (including Top End areas to the west of Arnhem Land), on the other. Some studies have looked at songs produced in cultures which lie outside of the geographical bounds of the Deserts and Arnhem Land, but which share significant musical characteristics with songs from these areas, for example the Kimberley and Western Gulf of Carpentaria regions. Other analytic studies have considered traditions from cultures which are musically quite distinct from those of the Deserts and Arnhem Land, for example Northern Queensland and Northern New South Wales. (8)
Music analysis as a tool of research: roles and ethics
To begin to understand some of the ways in which Aboriginal music researchers have regarded the use of music analysis, we might look especially to the work of Catherine Ellis, both because she conducted an unusually wide range of analytical projects in different contexts, and because she consistently regarded analysis as a central activity in her work. If Ellis relied upon transcription and analysis, she did not take either their use or usefulness for granted. From the beginning of her work with Aboriginal music, she was aware of the shortcomings of musical transcription, but accepted the compromise between the awareness of detail provided by transcription, and the musician's understanding gained through performance-based learning.
Ellis (1984:149-50) stated that her analyses of Aboriginal songs proceeded, first, on the basis of statements about music by Aboriginal performers, and second, by noting corrections of her own performance errors. Ellis, who came to the study of Aboriginal music from a Western performance background, elsewhere has stressed also the importance of a link between analysis and performer-oriented intelligibility. The performer-driven questions which lay behind her many years of analytic work included the desire to understand how to maintain accurate intonation; how to maintain accurate durational and accentual communication; how to communicate extra-musical and spiritual essences of performance; how to be touched by the spiritual force of performance; and how to experience empowerment through the act of performance (Ellis 1995:203).
There is another sense in which Ellis' work was closely linked to performance: by employing the familiar technique of listening to recordings over and over again as an important step in transcribing and analysing songs. Part of this reiterative process involves singing (i.e. the transcriber's vocal imitation of the recorded music). Accordingly, Ellis' ethnomusicological transcriptions and analyses (like those of countless other researchers) involve an intrinsically performative activity, even in cases where the main interests are not performance-related.
Like the analyses of other researchers of Desert Aboriginal song, Ellis' work often focused on key elements of the performance style, including the juxtaposition of melody and isorhythmic text, and much of this work involved the use of traditional forms of ethnomusicological transcription and analysis. At the same time, Ellis sought to understand certain aspects of performance practice that could not be addressed adequately through traditional means. Examples of these included certain aspects of performers' rhythmic timing (and time consciousness) and the intervallic structures of vocal melodies. In the rhythmic case, she developed various extensions and alternatives to 'traditional' forms of notation--for example, in order to plot and compare pitch change within (and across) rhythmic cells over a series of melodic and textual lines (Ellis 1984), and in order to present statistically detailed information indicating the points at which melodic, rhythmic and textual elements connect and disconnect during performance (Ellis 1992b). (9) In the case of intervallic structures, Ellis relied to varied extents upon machine- and computer-aided measurements of pitch and interval (Ellis 1964, 1965, 1967; Will and Ellis 1994, 1996). (10)
Quite apart from the complex types of analysis just discussed, Ellis also employed analysis for purposes as varied as a pedagogical tool for non-Aboriginal students learning to perform Aboriginal songs, and, together with good recordings in the context of what she referred to as 'recovery fieldwork' (Ellis 1992a), as part of a systematic presentation of lost performance practices to the performers' descendants. But the most prevalent purpose to which Ellis directed her analytical work was to further crosscultural understanding of Aboriginal music. In 1995, she asked:
Does it matter at all what the pitch structure of traditional Central Australian singing is, or how the performers keep accurately in time with each other (simultaneously, as well as with themselves and others over the span of many years of performance of the one song)? I think it does; because in the Western world the value judgment has been made that their music is primitive and its performers are incapable of abstract thought. Nothing could be further from the truth revealed in our research findings on the structural complexities and accuracies of traditional performing. (Ellis 1995:207)
The desire to improve perceptions of Aboriginal culture through descriptions of the complexities of Aboriginal songs and performance is something that Ellis expressed forcefully and often.
Linda Barwick's 1990 article on Central Australian women's performance is of special interest here because it is framed entirely around questions concerning the use of analysis for understanding Aboriginal performance--in particular, 'what can we "know" through musical analysis?' and how can this knowledge be related to performers' knowledge? Here Barwick states that 'analysis is a process of understanding rather than a methodology for producing "truth"'; that one's knowledge or understanding of a music is constantly changing; and that analytical 'results' represent one stage in the continually evolving process of this understanding (1990:60). In her conclusion, she confesses to not being able to completely answer her own questions, but notes that the most striking result of her analytic work was that it led her to perceive the music differently and, more to the point, that it led her towards a greater appreciation for some of the finer aspects of the ways that performers combine text and melody. Paraphrasing Foucault on linguistics, Barwick notes, too, that analysis may be more of a perception than an explanation (1990:76). Her remarks highlight some of the more subjective aspects of analysis, in this case, to point out the constructive role that analysis plays in her coming to grips with an unfamiliar performance tradition.
One aspect of music analysis raised by Barwick--its ability to lead to a different perception of music--is something which Anthony Seeger (1987:102) also regarded as important in his work with songs of the Suya people of Brazil:
Careful music transcription can reveal aspects of the performance that native categories do not highlight. A good musical transcription can raise many questions which may or may not lead to greater understanding of the music, but they are usually worth asking. It is only by the confrontation of our musical parameters ... with the different ones of the Suya that can produce a productive group of questions and a comparative approach to music.
In questioning the 'truth'-producing capacity of analysis, Barwick touches on the issue of subjectivity in analysis. A whole range of questions has been raised about analysis in recent years which, in one way or another, relate to subjective versus objective qualities of analytic practices. These questions are as often as not about ethics as about methodology, and in various contexts have been addressed to scholars working with both non-Western and Western musics.
At this juncture, we might now consider two different debates surrounding the methodology and ethics of music analysis. One of these debates has been carried out by ethnomusicologists concerned with the effects that culturally biased tools of analysis have upon our understanding of non-Western musics. Speaking of the Aboriginal cultural context, and in response to an analysis-based paper given by Allan Marett in 1988 (and subsequently published as Marett 1992), Linda Barwick and Florian Messner pointed out that analysis runs the risk of distorting the performers' experiential world: first, by leaving out many essential performance parameters; second, by transcriptional reduction of a complex simultaneity of interrelationships to a linear sequence; and third, by possible marginalising the performer's experiential world by analytical severing of performance from context (cited in Marett 1991:38). Messner took the extreme view that 'the only valid manifestation of the performance tradition was that of performance itself' (Marett 1991:38). Marett, on the other hand, concluded that in the end it is both true and acceptable that the results of our analysis cannot reflect 'the musical culture itself, but [rather] an image of our own mind ... engaging in an interaction with another culture' (1991:38). In other words, our own personal perceptions and sensibilities inevitably play a key role both in the process of analysis and in the production of analytic results.
In an article published in Ethnomusicology, musicologist Kofi Agawu applied a neo-Schenkerian reductionist analysis to the understanding of Northern Ewe songs. Defending this analytical approach, Agawu claimed not only that analysis can provide worthy knowledge about a musical practice even though Ewe do not conceptualise their songs in terms of concepts employed by Agawu, but also that 'it would be devious to invoke the contentious emic-etic or insider-outsider dichotomy in order to distribute insights that at the end of the day serve "us" not "them"' (1990:229). While not recommending the indiscriminate use of analytic methods employed without reference to indigenous cultural viewpoints, I would agree with Agawu on the point that, if we are truly honest about our endeavours, we must accept that, despite our best wishes otherwise, our work relates firstly to our own agendas, and that recourse to 'emic' statements does not undo (or morally valorise) this basic fact.
The other analysis-related debate involved a group of scholars working in the areas of historical musicology and music theory. (11) In the 1980s, some music scholars influenced by work in areas such as gender studies, critical theory, and cultural studies came to prominence within musicology under the collective banner of 'New Musicology'. Though marked by different interests and theoretic underpinnings, one common aspect of the New Musicologists' approach was a rejection of music analysis which regards compositions as autonomous works of art. New Musicology is interested in re-situating Western art music in its social and cultural nexus. Rather than analyse music in a positivistic and formalistic manner (as is common in 'traditional' music theory and analysis), New Musicologists--while still engaging the tools of analysis to point out certain qualities of a piece of music--might seek to uncover the sexual or other ideological agendas inherent in musical works (as well as in the traditional approaches to music analysis).
Confronted by these new ideas, other music theorists sought to defend the practice of objectively analysing music as music, and in some cases to attack what they regarded as the unfounded philosophical, political and scholarly bases of New Musicology. (12) With the passing of time, much of the earlier bitterness subsided, and New Musicology (now scarcely 'new') has become largely accepted within the mainstream of musicology.
In 1996, at the height of the intra-disciplinary fracas, Music Theory Online published a set of theorists' responses to the challenges put out by New Musicology. Two of these pieces, by theorist Marion Guck and by composer/theorist Joseph Dubiel, served to indicate that some music theorists had already been employing ideas and approaches that were in sympathy with the ideas of some New Musicologists, or that were at least not as strictly formalistic as some of the broader attacks against music theory had implied.
Guck has cultivated a distinctively subjective and interpretive mode of analysis. In her MTO article, she describes in careful detail some of the ways that she relates in a quite physically intimate manner to the opening section of Mozart's A-Major Piano Concerto, K.488. She discusses, for example, the effects that various combinations of notes have simultaneously upon the playing hand and the listening ear; the demands placed upon listening by the 'blockages, delays, displacements, and evasions' (Guck 1996:para. 32, quoting colleague Roland Jordan) in the opening solo bars; and the contrasting feeling of letting go, or of breathing space afforded by the subsequent entrance of the orchestra (para. 32, again quoting Jordan).
In striving for a greater sense of intimacy with the music, Guck (1996:para 34) also stresses the importance of a link between performance and analysis:
I wanted to understand my role as the pianist bringing the sounds into being, the pleasure I take in being completely focused to play each note and figure. Analysis is for me the articulation of a process of growing awareness, increasing closeness, of 'immersion in pleasure'.
Dubiel's MTO article was composed as a rejoinder to criticisms of structural analysis raised by Rose Subotnik (in Subotnik 1988). Dubiel rejects the notion that 'structure' implies something unchangeable, such that 'its internal components and relationships are presumed to have attained something like a status of necessity which disallows alternative versions' (1996:para. 9, quoting Subotnik 1988:101). On the contrary, some music analysis specifically sets out to demonstrate alternative readings of the same music, and, furthermore, Dubiel regards musical structure not so much as some 'quasi-objective property that keeps the work standing but one kind of meaning that we read into it' (1996:para. 18, my emphasis). Dubiel's main point here is that analysis (and structural hearing) ought not to 'draw ... a distinction between sound and meaning' (1996:para. 20) but, rather, should give meaning to sound in a way which includes a consideration of sound itself.
The point I wish to raise with all this is that proponents of both New Musicology and 'New Theory' (for lack of a better term) demonstrate a concern with personal responsibility. Through an insistence on critically examining the ideologies and agendas in our approaches and techniques, New Musicology has provided a means by which we can be aware of and responsible for how we analyse and understand music. Theorists like Guck and Dubiel are also advocating a personal responsibility--not for the purpose of uncovering ideological underpinnings, but for gaining a full awareness of how we hear music and for the means that we employ to describe the experience of our hearing.
Most ethnomusicologists are well aware of the limitations of standard Western notation for the purpose of representing non-Western (or even Western) music in accurate, descriptive transcription. Some of these limitations are often stated at the beginning of a presentation of musical transcription as a sort of technical disclaimer. Sometimes verbal annotations or additions to standard Western notation are used in order to help bridge the gap(s) between what the transcriber regards as the 'actual sound' and what can be understood from the visual representation of the same sound. Such measures might only partly mitigate the perceived shortcomings of the notation; however, a lack of total accuracy in transcription is not problematic as long as the particular features the analyst wishes to consider are rendered accurately. For example, depicting a non-tempered vocal melody in Western chromatic pitches might render all of the pitches 'incorrectly', but could still provide quite accurate information about the contour of the melodic line.
Declared limitations aside, musical transcriptions and related analyses are often at least implicitly presented as providing objective information about a musical performance. Accordingly, even when researchers admit to the shortcomings of their transcriptions, the situation is often problematised as one of accuracy, as opposed to a problem involving the inherently subjective aspects of musical perception, including differences in the perception of individuals operating in different cultural fields.
This is not always the case, however. Richard Moyle, for example, has explicitly intended that his analyses of Kukatja songs from Balgo, Western Australia, should represent how Kukatja themselves conceptualise their own songs and performance. At the same time, he has criticised some of Ellis' analysis for being too complex or abstract, such that it does not relate to the way that performers perceive their own music (1997:185):
Accordingly, I have refrained from detailed formal analysis of the type used by Ellis ... if one pursues analysis to a scale of extreme detail, patterns are evident only on the level of abstraction, it may well be difficult to restore real-world significance to the statistical arrays, and so confirm that these modes of analysis are relevant to a orally transmitted vocal repertoire.
Moyle does not explain why his simpler transcriptions and analyses more closely correspond to the way that Kukatja people conceptualise songs and performance; however, the point is that it is Moyle's intention to make his transcription correspond more closely to the way that Kukatja think about their music, and that, in his judgement in this particular case, that end is achieved through simpler analytic means.
Different sorts of examples may be cited wherein researchers have felt that the relevance of their analytic work was validated in part by Aboriginal performers' explicitly (or implicitly) expressed understanding of various aspects of the researchers' analysis. Marett, for example, confirmed the relevance of a set of analytic diagrams by showing and discussing these with the wangga singer Maralung (Marett 1992:206), and Barwick and Marett (1995:6) have noted that Pitjantjatjara performers' references to Ellis' structural analyses as 'numbers' and 'unsung songs' might indicate a recognition of songs' existence apart from performance--even, perhaps, in analyses.
In my own work with Yolngu song, I have tried to incorporate my understanding of Yolngu musical percepts into a continuous reconsideration of Yolngu song performance, and I try to incorporate any concomitant changes in my own aural perceptions into my transcriptions and analyses. In the end, however, I tend to judge the effectiveness of my analytic work solely on the basis of how well it reflects my own understanding of the music, whether presented in a note-by-note or some synecdochical form. (13) Conceived of and presented in this way, analysis provides readers with essential information about the actual perspective from which various conclusions about a given music are drawn (i.e. the analyst's perspective). (14)
Still considering the notion of taking a personal responsibility for the way that we hear and describe music, I would now like to relate a couple of brief examples of my own analytical experiences working with Yolngu song. The first example concerns the lack of tonal correspondence between either of the didjeridu's two pitches and any of the singers' pitches in Yolngu song performance. Before my first visit to the Yolngu community at Yirrkala (northeastern Arnhem Land), I understood from Jill Stubington's writing (Stubbington 1978) that, despite Richard Waterman's earlier assertions to the contrary (1971:171-2), there was no tonal relationship between the didjeridu and voice. (15) Waterman provided no evidence for his claim, while Stubington based her rejection of Waterman's claim on transcription and analysis of hundreds of recorded song items. If that were not enough, Stubington also pointed to cases where one didjeridu was substituted for another in the middle of a performance and, despite the completely different pair of pitches on the second instrument, the singers' pitches stayed right where they had been before the switch occurred.
When I first arrived at Yirrkala--and despite what I knew from reading Stubington's work--my ears could not help but hone in on the at least apparent tonal relations that often arose between didjeridu and voices. There was a long funeral going on at the time and each night for hours on end I would hear the didjeridu's rhythmically nimble drone sounding a perfect fourth above the tonic of a whole choir of Rirratjingu clan singers. For me this was an aesthetically very pleasing experience. During this time I did witness a couple of cases of didjeridu-switching, which sure enough resulted in some other, arbitrary interval, but I did not care. To my ears, those were aberrations; the perfect fourth was beautiful. At other ceremonies and individual performances I would hear other clearly tonal intervals between didjeridu and singers--not all the time, but often enough, and so I would make a note of them.
Then something else happened: one day--I am not sure exactly when, but perhaps six to eight weeks into my stay at Yirrkala--I realised that for some time already I had stopped hearing tonal relations between the didjeridu and voice. It was not that familiar intervals had ceased to occur, but simply that I had stopped listening to this aspect of the sound. I took this as a sign that I had progressed along some scale of 'hearing-like-Yolngu'.
I raise this example in order to point out, first, that despite my academic understanding of Stubington's reports, I could not honestly report the lack of tonal correspondence until my hearing came around on its own accord. Analysis is very closely linked to the way we hear. Hearing, perhaps more than knowing, has everything to do with analysis. The second point to be made here is that my ears did come around eventually, despite my obsessive interest in perfect fourths, and in this case, simply on the basis of repeated listening; that is, without any supporting statements from Yolngu musicians. Speaking with several performers, I was subsequently able to confirm the 'emic' basis for my own changed perceptions, but the point I wish to draw here is that listening is itself a very powerful tool and, ultimately, I would need to take responsibility for my own analyses based on what I could hear.
The other analytic experience I want to discuss concerns my perception and understanding of Yolngu vocal melodic structure in terms of discrete sets of variably tuned scale steps. Dambu (or an equivalent term liya) translates literally as 'head', but is often translated by Yolngu into English as 'tune'. (16) It refers to the melodic component of sung performance, both in the sense of each extemporised melodic line and in the sense of an underlying skeletal structure that forms the basis for creating lines of sung melody. (17)
In this case, my initial analytical efforts were very much determined by performers' statements concerning melody, in particular the oft-repeated assertion that each clan owns its own 'tune'. The guiding point I drew from this assertion was that, setting aside definitions of what exactly comprised 'dambu', I ought to be trying to hear something distinctly and consistently different about each clan's handling of melody. From performers' statements I also gathered that, whatever dambu was, it could be distinguished from other named aspects of performance, most notably rirrakay (noise, sound, timbre) and bilma (clapstick sound and rhythms). The main things left for possible inclusion in some exclusive notion of dambu were scale and scale tuning, and the contours and rhythms of the brief melodic motifs upon which extemporised melodies are built. At this point, additional performer statements were not forthcoming, so I was on my own.
What I noticed over a period of time was that, at least in the case of clans whose melodies I was at all familiar with, I seemed to be able to distinguish one dambu from another solely on the basis of my awareness of a particular number of scale steps and on reasonably precise or perhaps averaged notions of the interval sizes between scale steps. Whether or not Yolngu perceive dambu in such exclusive terms was one question I could not answer with certainty. In moments when a Yolngu listener's consciousness turns to questions of whose dambu is sounding, it could be that all sorts of cues are relied upon; not only things like rhythms and contours of brief melodic motifs, but even things which could otherwise be defined apart from dambu, such as certain clan-distinct clapstick patterns or, indeed, a clan's idiomatic vocal sound. In normal circumstances, why would a listener exclude useful supporting cues when attempting to identify the ownership of something?
On the other hand, there was a logical efficiency about my honing in on scale and tuning as fundamental elements of dambu, since each clan is usually only associated with one or two such scales (and would refer to separate dambu in the case of two such scales), whereas each dambu can be performed using a relatively large variety of clapstick patterns, melodic rhythms and short-range contours. In any event, the point I wish to make is that in the end I was happy enough to distinguish dambu solely on the basis of what I, as a non-Yolngu listener, could perceive. In a laboratory setting, I have since employed electronic aids in order to determine the average interval size between the scale steps of several dambu that I am familiar with, but I have done this in a way that provides objective measurements of what I subjectively hear, at least as much as objective measurements of what Yolngu do. (18)
In all of this I wish to illustrate how subjective listening experiences can be crucial elements in analytical situations, even, at times, in situations where meaning or understanding is closely linked to performer statements about music.
In a joint article published in Musicology Australia, Ellis and Barwick (1987) point to the uneducated outsider's initial inability to discern 'emically' relevant structural differences from 'etically' discernible (but culturally irrelevant) differences. They state that, by repeated exposure to the musical or linguistic system, the significant boundaries can be better discerned by the analyst. With reference to the interpretive responsibilities of the music analyst, one could add that while the processes described by Ellis and Barwick are essential to our work, one could not expect, in the end, to arrive at the correct understanding of the musical or performative situation but, rather, one of any number of potentially useful explanations.
Statements about music by performers or other cultural insiders are of great importance but cannot in themselves generate the most insightful analyses. Performer statements do not necessarily lend themselves to systematic sets of rules; the determinant 'rules' of performance are always in some state of flux; and it is quite possible to draw different analytic conclusions from the same set of observations, recordings, and performer statements. The situation is in this sense quite comparable to that facing the analysis of Western art music: in both cases, a fuller realisation of the subjective aspect of analysis can avoid misunderstandings about the analytic endeavour.
The taking of personal credit and responsibility for analysis is already well established in some strands of Western music theory. The thought of taking this approach more in studies of Aboriginal music seems somehow awkward or even culturally inappropriate; yet to continue to not do so seems to me to result either in a fundamental dishonesty (pretending that our analyses are more objective or more fully linked to Indigenous ideas than they are), or in a situation where we feel forced to abandon analysis and hence disregard our own hearing of music.
As to the usefulness of analysis, understanding the 'extra-aural' elements of performance--connections to kinship, place, political contestations, and so forth--helps us to put our understanding of music into perspective, and may in many cases be the more important part of our research. In many cases, too, these factors can be addressed without direct reference to the structural or other qualities of musical sound. At the same time, analysis can be laborious to undertake and too often is presented in a manner that is incomprehensible to non-music specialists and in fact unpalatable to almost anyone other than the researcher herself/himself. So why not let analysis fall by the wayside? Could we not, as happens often in the case of popular music studies and even in some studies of other non-Western or Western musics, get straight to the important matters and leave it to the reader to be familiar with the sound of the music under discussion from recordings?
My main response to this, in the context of studies of Aboriginal music by non-Aboriginal scholars, is that analysis, perhaps together with some degree of performance-participation, is the only way that we can directly address how we (as outsiders to the culture) actually hear the music in question. Particularly for those of us who by avocation and profession are musical fetishists, why would we be content to carry out studies of music without critically examining how we hear sound? It is precisely because our musical ears are so culturally biased that we need analysis for a very practical purpose, as a sort of game or exercise to help us hear music in different ways.
Analysis ideally can provide powerful tools for engaging with music and for increasing intercultural awareness and understanding. At the same time it is essential to remember that analysis can only provide part of an understanding of Aboriginal music-making; Aboriginal views on songs and performance must, wherever possible, be taken into account when posing and addressing analytical questions. As non-Aboriginal scholars engaging ourselves with Aboriginal songs, performance and culture through music analysis, (19) we gain the possibility of expanding our understanding not only of Aboriginal music but also of 'music' and 'music analysis' generally. This, too, is an important part of intercultural communication and of the analytic endeavour.
Funding for fieldwork in the Yirrkala area in 1989-90 was generously provided by AIATSIS and by a Fulbright award. I am especially indebted to the many singers and other members of the Yirrkala Yolngu community for the musical and cultural education I have received over a long period of time. I also wish to thank Allan Marett and Peter Toner for their comments and suggestions on this article.
(1.) The possibility (and, ultimately, the inevitability) of repeated listenings of recorded musical performances has changed not only the nature of music analysis, but even the more general ways in which we understand and experience music and performance.
(2.) In the world of Western classical music, the notion of reified musical 'works' as embodied in musical scores is traditionally regarded as distinct from the more fleeting qualities of live concert performance.
(3.) See Herndon (1974) for a good discussion of different approaches to analysis in ethnomusicological studies through the early 1970s.
(4.) Though often considered as something distinct from analysis per se, music transcription is regarded here as a type of analysis, since the notations used to represent musical sound embody important assumptions about the nature of musical sound.
(5.) An earlier music-related publication by Strehlow (1955) included transcriptions of Aranda melodies by Jack Horner of the Elder Conservatorium, University of Adelaide.
(6.) See Clunies Ross and Wild (1982, 1984), Dixon and Koch (1996), Hercus and Koch (1995) and Marett et al (2001). It should also be remembered that many music researchers owe much of their analyses of song texts to the aid of Indigenous linguistic transcribers and translators.
(7.) See Ellis et al (1990) and Marett and Page (1995).
(8.) A sample of published work on traditional music from these areas which contains substantive music-analytic material is given below. Writings that focus on contemporary music are not listed here except where they include analytic comparisons with traditional songs. Central and Western Desert styles (including some genres in the Kimberley and Western Gulf of Carpentaria): Barwick (1989, 1995a), Ellis (1963, 1964, 1965, 1966, 1967, 1968, 1983, 1984, 1985, 1992b), Ellis and Barwick (1987), Ellis et al (1990), Hercus and Koch (1995), Keogh (1989, 1990, 1995), Mackinlay (1998, 2001), McCardell [Prabhu Pritam] (1976, 1980), May and Wild (1967), R Moyle (1979, 1986, 1995, 1997), Tunstill (1987), Wild (1984), Will (1997), and Will and Ellis (1994, 1996); Eastern, Central, and Western Arnhem Land styles (the latter including areas to the west and south of Arnhem Land into the Kimberley): Anderson (1992, 1995), Clunies Ross and Wild (1982, 1984), Corn (2002a, 2002b), Elkin and Jones (1958), Gibson and Dunbar-Hall (2000), Jones (1967, 1973), Knopoff (1997), Magowan (1994), Marett (1991, 1992), Marett and Barwick (1993), Marett et al (2001), Marett and Page (1995), A Moyle (1964, 1974), Stubington (1978, 1982, 1984a), Stubington and Dunbar-Hall (1994), and Toner (2001, 2003); Northern Queensland: Dixon and Koch (1996), Koch (1987a); Northern New South Wales: Gummow (1992, 1995, 2002). This list is not meant fully to represent research on Aboriginal music, analysis-oriented or otherwise. Among the many items not listed here are writings (both by some of the same authors listed above as well as by others) that do not emphasise analytical content, but which are nonetheless informed by music analysis carried out by the respective authors. For a broader picture of the substantial body of research on Aboriginal music from 1975 to 1992, see the annotated bibliographies compiled by Koch (1987b, 1992). For a complete bibliography of Catherine Ellis' and Alice Moyle's prolific outputs, see Barwick (1995b) and Stubington (1984b), respectively.
(9.) The latter article builds on some of the ideas and analyses in a series of previous work by Barwick (1989), Ellis and Barwick (1987), and Tunstill (1987).
(10.) Ellis was not by any means the first to use machine- or computer-aided measurement and graphic notation of musical pitch and interval (the first extensive use of such aids came with Charles Seeger's development of the melograph at UCLA around 1950); nor is she the only musicologist in Australia to have used such aids. What is of particular note with respect to this part of her work was that it led to a series of studies (by Ellis and her colleague Udo Will) which suggested that Desert performers use a linear mode of interval transposition, as opposed to the logarithmic mode generally presumed to be universal for music perception (Ellis 1965; Will and Ellis 1994, 1996; Will 1997). Given the importance that Ellis placed upon the relationship between analysis and performer-oriented intelligibility, it may seem counter-intuitive that she would rely upon purely mechanical means to make sense of musical intervals. Here is where one needs to recall that this work began from a simple desire to analyse precisely melodic intervals in a way that bypasses the shortcomings of Western-biased ears and musical notation. Will (1994, 1996) has vigorously defended the use of objective means of measuring interval and has criticised the biased and imprecise means of traditional musicology.
(11.) In the United States, 'historical musicology' and 'music theory' refer to separate branches of what in Australia is simply referred to as 'musicology'. In Australia, when referring to analysis-intensive work, the term 'music analysis' (vs. 'music theory') is more likely to be used. Most historical musicologists and music theorists are concerned primarily with classical or contemporary Western art music.
(12.) Some of the harshest critics of New Musicology include Pieter C. van den Toorn (1996) and Kofi Agawu (1996).
(13.) Part/whole synecdochical perceptions of music involve grasping 'the whole' or large bits of some music by reference to smaller, more skeletal bits of the actual music or related analytic shorthand. See Judd (1994) for discussion of synecdoche in Western music.
(14.) Like Marett, I have had the pleasure of having certain analytically derived ideas confirmed in conversations with performers. On the few occasions when I have shown examples of my music transcriptions to Yolngu singers, however, the men in question had nothing to say about the shape (or other content) of my drawings. A certain degree of interest was expressed for the simple fact that transcriptions had been produced and it was assumed that representing songs visually on paper might increase both understanding and respect for the songs in the minds of balanda (non-Aboriginal peoples).
(15.) The term 'tonal relationship' is understood here in traditional musicological terms and refers to a correspondence to one or more pitches of a musical scale. The non-tonal correspondence between pitches of the voice and didjeridu provides one of the basic distinctions between eastern and western Arnhem Land performance styles: in western Arnhem Land song genres, the singer's tonic corresponds to the didjeridu's drone pitch.
(16.) The term 'dambu' was spelled as 'dhambu' in earlier published work.
(17.) For discussion of dambu/liya and related understandings of melodic structure in Arnhem Land song, see Anderson (1992), Knopoff (1993, 1998a, 1998b, 1998c), Magowan (1994), and Toner (2001, 2003). Toner (2003) provides especially detailed coverage of both the musical and social implications of liya. This study makes use of digital measurements of melodic pitch and interval, not by assessing the analyst's perception (as in my work with dambu--Note 18) but by extracting objective data directly from recorded material.
(18.) In a laboratory setting, several dozen song items were digitised and saved as sound files. Using a sound editing program, I familiarised myself with these particular items through repeated listening (and repeated viewings, since the dynamic profile of the file is visible while the item is playing). As I listened I created digital loops (varying in duration from 0.10 to 0.50 seconds) at points at which I perceived a relatively steady state pitch corresponding to what I had internalised as a given scale step. Listening to the pitch of a given loop, I matched it by ear, by adjusting the frequency dial on a sweep function generator set to produce a square wave which was played back over speakers. When the pitch of the square wave matched the vocal pitch isolated in the loop, I noted the frequency value (in Hertz) as indicated on a frequency counter linked to the frequency generator. The number of samples taken from each song item varied from ten to fifty. By averaging the frequency values of multiple samples associated with each given scale step (both within and across song items), I derived theoretical averaged tunings for each scale interval and was able to demonstrate consistent differences in the tunings of various clans' interval sets (scales). Yet, ultimately, these intervals reflect the way that I (and not necessarily Yolngu) hear melodic structure.
(19.) The sorts of music research and analysis discussed in this article have (to date) all been undertaken by non-Aboriginal scholars, often but not necessarily in consultation with Aboriginal performers. There are contentious issues surrounding the very fact of non-Aboriginal people generating meaning about Aboriginal music. The problems of who controls and benefits from such meaning cannot easily be addressed within our current political and institutional paradigms (nor within the scope of this article). I would contend, however, that the tools and perspectives of music analysis, even when far removed from Aboriginal conceptualisations, are not in themselves problematic. Reconciling Aboriginal and Western musical perspectives will come with the granting of equal status to multiple voices within the main of our institutions, not through the muting of analytic perspectives at the periphery of academia.
Agawu, K. 1990 Variation Procedures in Ewe Songs, Ethnomusicology 34(2), 221-44.
-- 1996 Analyzing Music under the New Musicological Regime, Music Theory Online 2/4, <http://boethius.music.ucsb.edu/mto/ issues/mto.96.2.4/mto.96.2.4.agawu.html>. [Accessed 3 July 2003.]
Anderson, G.D. 1992 Murlarra: A Clan Song Series from Central
Arnhem Land, PhD thesis, University of Sydney.
-- 1995 Striking a Balance: Limited Variability in Performances of a Clan Song Series from Central Arnhem Land. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 12-25.
Barwick, L.M. 1989 Creative (Ir)regularities: The Intermeshing of Text and Melody in Performance of Central Australian Song, Australian Aboriginal Studies 1989/1, 2-11.
-- 1990 Central Australian Women's Ritual Music: Knowing through Analysis versus Knowing through Performance, Yearbook for Traditional Music 22, 60-79.
-- 1995a Unison and Disagreement in a Mixed Women's and Men's Performance (Ellis Collection, Oodnadatta 1966). In Barwick et al (eds), The Essence of Singing and the Substance of Song, 95-105.
-- 1995b Catherine Ellis: Career History and List of Publications, Papers and Reports. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 209-21.
Barwick, L.M. and A. Marett 1995 Introduction. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 1-10.
Barwick, L., A. Marett and G. Tunstill (eds), The Essence of Singing and the Substance of Song: Recent Responses to the Aboriginal Performing Arts and Other Essays in Honour of Catherine Ellis, Oceania Monograph 46, University of Sydney.
Clunies Ross, M. and S.A. Wild 1982 Djambidj: An Aboriginal Song Series from Northern Australia, companion book to cassette Djambidj: An Aboriginal Song Series from Northern Australia, Australian Institute of Aboriginal Studies, Canberra.
-- 1984 Formal Performance: The Relations of Tune, Text, and Dance in Arnhem Land Clan Songs, Ethnomusicology 28(2), 209-29.
Corn, A.D.S. 2002a Dreamtime Wisdom, Modern-Time Vision: Tradition and Innovation in the Popular Band Movement of Arnhem Land, Australia, PhD thesis, University of Melbourne.
-- 2002b Burr-Gi Wargugu ngu-Ninya Rrawa: Expressions of Ancestry and Country in Songs by the Letterstick Band, Musicology Australia 25, 76-101.
Davies, E.H. 1927 Adelaide University Field Anthropology: Central Australia. No. 4--Aboriginal Songs, Transactions of the Royal Society of South Australia 51, 81-92.
-- 1932 Aboriginal Songs of Central and Southern Australia, Oceania 2(4), 454-67.
Dixon, R.M.W. and G. Koch 1996 Dyirbal Song Poetry: The Oral Literature of an Australian Rainforest People, University of Queensland Press, Brisbane.
Dubiel, J. 1996 On Getting Deconstructed, Music Theory Online 2/2, <http://smt.ucsb.edu/mto/issues/mto.96.2.2/mto.96.2.2.dubiel. html>. [Accessed 3 July 2003.]
Elkin, A.P. and T. Jones 1958 Arnhem Land Music, Oceania Monograph 9, University of Sydney.
Ellis, C.J. 1963 Ornamentation in Australian Vocal Music, Ethnomusicology 7(2), 88-95.
-- 1964 Aboriginal Music Making: A Study of Central Australian Music, Libraries Board of South Australia, Adelaide.
-- 1965 Pre-instrumental Scales, Ethnomusicology 9(2), 126-44.
-- 1966 Aboriginal Songs of South Australia, Miscellanea Musicologica: Adelaide Studies in Music 1, 137-90.
-- 1967 The Pitjantjatjara Kangaroo Song from Karlga, Miscellanea Musicologica: Adelaide Studies in Musicology 2, 171-268.
-- 1968 Rhythmic Analysis of Aboriginal Syllabic Song, Miscellanea Musicologica: Adelaide Studies in Musicology 3, 21-49.
-- 1983 When Is a Song Not a Song? A Study from Northern South Australia, Bikmaus 4(3), 136-44.
-- 1984 Time Consciousness of Aboriginal Performers. In J.C. Kassler and J. Stubington (eds), Problems and Solutions: Occasional Essays in Musicology Presented to Alice M. Moyle, Hale & Iremonger, Sydney, 149-85.
-- 1985 Aboriginal Music: Education for Living, University of Queensland Press, Brisbane.
-- 1992a Living Preservation: Problems of Cultural Exchange with Central Australian Traditional Performers. In A. Moyle (ed.), Music and Dance of Aboriginal Australia and the South Pacific: The Effects of Documentation on the Living Tradition, Oceania Monograph 41, University of Sydney, 155-62.
-- 1992b Connections and Disconnections of Elements of the Rhythmic Hierarchy in an Aranda Song, Musicology Australia 15, 44-6.
-- 1995 Response: Whose Truth? In Barwick et al (eds), The Essence of Singing and the Substance of Song, 201-7.
Ellis, C.J. and L.M. Barwick 1987 Musical Syntax and the Problem of Meaning in a Central Australian Songline, Musicology Australia 10, 41-57.
Ellis, C.J., L.M. Barwick and M. Morais 1990 Overlapping Time Structures in a Central Australian Women's Ceremony. In P. Austin, R.M.W. Dixon, T. Dutton and I.M. White (eds), Language and History: Essays in Honour of Luise Hercus, Pacific Linguistics, Australian National University, Canberra, 101-36.
Gibson, C. and P. Dunbar-Hall 2000 'Nitmiluk': Place and Empowerment in Aboriginal Popular Music, Ethnomusicology 44(1), 39-64.
Guck, M. 1996 Music Loving, or the Relationship with the Piece, Music Theory Online 2/2, <http://smt.ucsb.edu/mto/issues/mto.96.2.2/mto.96.2.2.guck. html>. [Accessed 3 July 2003.]
Gummow, M. 1992 Aboriginal Songs from the Bundjalung and Gidabal Areas of South-Eastern Australia, PhD thesis, University of Sydney.
-- 1995 Songs and Sites/Moving Mountains: A Study of One Song from Northern NSW. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 121-32.
-- 2002 Yawahr: ACorroboree for Everybody, Musicology Australia 25, 48-75.
Hercus, L. and G. Koch 1995 Song Styles from near Poeppel's Corner. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 106-20.
Herndon, M. 1974 Analysis: The Herding of Sacred Cows? Ethnomusicology 18, 219-62.
Jones, T.A. 1967 The Didjeridu: Some Comparisons of Its Typology and Musical Functions with Similar Instruments throughout the World, Studies in Music 1, 23-55.
-- 1973 The Yiraki (Didjeridu) in North-Eastern Arnhem Land: Techniques and Styles. In R.M. Berndt and E.S. Phillips, The Australian Aboriginal Heritage: An Introduction through the Arts, Australian Society for Education through the Arts in association with Ure Smith, Sydney, 269-74.
Judd, R. 1994 Synecdoche in Music, paper presented at American Musicological Society annual meeting, Minneapolis, <http:// www.sas.upenn.edu/~rjudd/syn.html>. [Accessed 3 July 2003.]
Keogh, R. 1989 Nurlu Songs from the West Kimberley: An Introduction, Australian Aboriginal Studies 1989/1, 2-11.
-- 1990 Nurlu Songs of the West Kimberleys, PhD thesis, University of Sydney.
-- 1995 Process Models for the Analysis of Nurlu Songs from the Western Kimberleys. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 39-52.
Knopoff, S. 1993 What's in a Tune? The Relations of Clan Affiliation, Performance Protocol and Musical Style in Yolngu Melodic Construction, paper presented at International Council for Traditional Music, 32nd world conference, Berlin.
-- 1997 Accompanying the Dreaming: Determinants of Didjeridu Style in Traditional and Popular Yolngu Song. In K. Neuenfeldt (ed.), The Didjeridu: From Arnhem Land to Internet, John Libbey and Co./Perfect Beat Publications, Sydney, 39-67.
-- 1998a Yolngu Clan Songs. In A.L. Kaeppler and J.W. Love (eds), The Garland Encyclopedia of World Music, Volume 9: Australia and the Pacific Islands, Garland Publishing, New York, 301-2.
-- 1998b Yolngu Melodic Structures and Notions of the Sacred, Musical Visions '98 conference (co-sponsored by International Association for Study of Popular Music and Arnhem Land Performance), Adelaide.
-- 1998c Scales and Modes as Core Elements of Yolngu Melodic Construction, paper presented at Musicological Society of Australia, 21st national conference, Adelaide.
Koch, G. 1987a Dyirbal Gama Songs of Cape York. In M. Clunies Ross, T. Donaldson and S. Wild (eds), Songs of Aboriginal Australia, Oceania Monograph 32, University of Sydney, 43-62.
-- 1987b ABibliography of Publications on Australian Aboriginal and Torres Strait Islander Music: 1975-1986, Musicology Australia 10, 58-70.
-- 1992 A Bibliography of Publications on Australian Aboriginal and Torres Strait Islander Music: 1986-1992, Musicology Australia 15, 87-95.
Mackinlay, E. 1998 'For Our Mother's Song We Sing': Yanyuwa Women Performers and Composers of a-nguyulnguyul, PhD thesis, University of Adelaide.
-- 2001 'Same but Different': Musical Behaviour, Gender Roles, and the Transference of Power in Yanyuwa Society, Kulele: Occasional Papers on Pacific Music and Dance 3, 65-78.
Magowan, F.C. 1994 Melodies of Mourning: A Study of Form and Meaning in Yolngu Women's Music and Dance in Traditional and Christian Contexts, PhD thesis, Oxford University.
Marett, A. 1991 Wangga Songs of Northwest Australia: Reflections on the Performance of Aboriginal Music at the Symposium of the International Musicological Society '88, Musicology Australia 14, 37-46.
-- 1992 Variability and Stability in Wangga Songs from North-West Australia. In A. Moyle (ed.), Music and Dance of Aboriginal Australia and the South Pacific: The Effects of Documentation on the Living Tradition, Oceania Monograph 41, University of Sydney, 193-221.
Marett, A. and L. Barwick 1993 Bunggridj-bunggridj: Wangga Songs by Alan Maralung, booklet accompanying CD recording by A. Marett and L. Barwick, Smithsonian/Folkways CD SF 40430, International Institute for Traditional Music, Berlin and Washington.
Marett, A., L. Barwick and E. Ford 2001 Rak Badjalarr: Wangga Songs by Bobby Lane, booklet accompanying CD recording by A. Marett and L. Barwick, AIATSIS CD 93-17006-00015C, Aboriginal Studies Press, Canberra.
Marett, A. and J. Page 1995 Interrelationships between Music and Dance in a Wangga from Northwest Australia. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 27-38.
May, E. and S. Wild 1967 Aboriginal Music on the Laverton Reservation, Western Australia, Ethnomusicology 11(2), 207-17.
McCardell, A. [also see Prabhu Pritam] 1976 Rhythm and Melody in Australian Aboriginal Songs of the Western Desert, PhD thesis, University of Western Australia.
Moyle, A.M. 1964 Songs from the Northern Territory, five 12" LP discs and companion booklet, IAS M-001/5, Australian Institute of Aboriginal Studies, Canberra.
-- 1974 North Australian Music: A Taxonomic Approach to the Study of Aboriginal Song Performances, PhD thesis, Monash University.
Moyle, R. 1979 Songs of the Pintupi: Musical Life in a Central Australian Society, Australian Institute of Aboriginal Studies, Canberra.
-- (with the help of Slippery Morton) 1986 Alyawarra Music: Songs and Society in a Central Australian Community, Australian Institute of Aboriginal Studies Press, Canberra.
-- 1995 Singing from the Heart. In Barwick et al (eds), The Essence of Singing and the Substance of Song, 53-8.
-- 1997 Balgo: The Musical Life of a Desert Community, CIRCME, University of Western Australia, Perth.
Porter, J. 1995 New Perspectives in Ethnomusicology: A Critical Survey, Transcultural Music Review 1, <http://www. sibetrans.com/trans//trans1/porter.htm>. [Accessed 3 July 2003.]
Prabhu Pritam [also see A. McCardell] 1980 Aspects of Musical Structure in Australian Aboriginal Songs of the South-West Western Desert, Studies in Music 14, 9-44.
Seeger, A. 1987 Why Suya Sing, Cambridge University Press.
Strehlow, T.G.H. 1955 Australian Aboriginal Songs, Journal of the International Folk Music Council 7, 37-40.
-- 1971 Songs of Central Australia, Angus & Robertson, Sydney. Stubington, J. 1978 Yolngu Manikay: Modern Performances of Australian Aboriginal Clan Songs, PhD thesis, Monash University.
-- 1982 Song Performance and Aboriginal Polity: A North-East Arnhem Land Example, Musicology 7, 84-103.
-- 1984a Review of Djambidj: An Aboriginal Song Series from Northern Australia, Australian Aboriginal Studies 1, 68-76.
-- 1984b Alice M. Moyle: An Australian Voice. In J.C. Kassler and J. Stubington (eds), Problems and Solutions: Occasional Essays in Musicology Presented to Alice M. Moyle, Hale & Iremonger, Sydney, 358-81.
Stubington, J. and P. Dunbar-Hall 1994 Yothu Yindi's 'Treaty': Ganma in Music, Popular Music 13(3), 243-59.
Subotnik, R.R. 1988 Toward a Deconstruction of Structural Listening: A Critique of Schoenberg, Adorno, and Stravinsky. In E. Narmour and R.A. Solie (eds), Explorations in Music, the Arts, and Ideas: Essays in Honor of Leonard B. Meyer, Pendragon Press, Stuyvesant, 87-122.
Toner, P.G. 2001 When the Echoes Are Gone: A Yolngu Musical Anthropology, PhD thesis, Australian National University.
-- 2003 Melody and the Musical Articulation of Identities, paper presented in Australian National University Anthropology Seminar Series, 19 March.
Tunstill, G. 1987 Melody and Rhythmic Structure in Pitjatjatjara Song. In M. Clunies Ross, T. Donaldson and S. Wild (eds), Songs of Aboriginal Australia, Oceania Monograph 32, University of Sydney, 121-41.
van den Toorn, P.C. 1996 Music, Politics, and the Academy, University of California Press, Berkeley.
Waterman, R. 1971 (1955) Music in Australian Culture: Some Sociological and Psychological Implications. In D.P. McAllester (ed.), Readings in Ethnomusicology, Johnson Reprint Corp., New York, 167-74.
Wild, S.A. 1984 Warlbiri Music and Culture: Meaning in a Central Australian Song Series. In J.C. Kassler and J. Stubington (eds), Problems and Solutions: Occasional Essays in Musicology Presented to Alice M. Moyle, Hale & Iremonger, Sydney, 187-203.
Will, U. 1994 Tradition versus Science!? Reflections on Interactions and Responsibilities, World of Music 36(1), 59-65.
-- 1996 Sounds from the Dreamtime and the Pythagorean Paradigm: A Review of L. Barwick, A. Marett and G. Tunstill (eds), The Essence of Singing and the Substance of Song: Recent Responses to the Aboriginal Performing Arts and Other Essays in Honour of Catherine Ellis, World of Music 38(2), 85-9.
-- 1997 Two Types of Octave Relationships in Central Australian Vocal Music?, Musicology Australia 20, 6-14. Will, U. and C. Ellis 1994 Evidence for Linear Transposition in Australian Western Desert Vocal Music, Musicology Australia 17, 2-12.
-- 1996 A Re-analyzed Australian Western Desert Song: Frequency Performance and Interval Structure, Ethnomusicology 40(2), 187-222.
Steven Knopoff is an ethnomusicologist and composer with wide musical interests. His primary research focus over the past 14 years has been Yolngu ceremonial songs and performance, particularly those associated with the community around Yirrkala, NT. Knopoff is a lecturer in music at the University of Adelaide, where he teaches courses in the areas of ethnomusicology, music theory, and popular music and culture. He is also the current president of the Musicological Society of Australia.
Elder School of Music, University of Adelaide. <email@example.com>