Example 1. Didjeridu vamp and variants.
Determinants of Didjeridu Style in Traditional and Popular Yolngu Song
ACCOMPANYING THE DREAMING
Didjeridu From Arnhem Land To Internet - Chapter Four
STEVEN KNOPOFF
Introduction
Alongside impressions of kangaroos,
koalas, boomerangs and a vast out- back, the image and sound of
the didjeridu has fixed itself in the minds of people around the
world as a symbol of Australia and of Aboriginal culture. This
has become all the more so in the past two decades. People who
have never seen or heard a didjeridu in traditional performance
contexts may have encountered it in movie soundtracks or in recordings
by smah but increasing numbers of rock, jazz and new age artists
who have taken up the instrument. As Peter Dunbar-Hall discusses
elsewhere in this anthology, though often regarded as a pan-Aboriginal
instrument, the didjeridu is probably indigenous only to certain
areas of the north coast of the continent (Moyle, 19 8 1). The
didjeridu is called by different names in the various cultures
that use the instrument. The didjeridus played by the Yolngu residing
at Yirrkala, Northeast Arnhem Land, Northern Territory, are called
yidaki.
Made from a variety of different
woods, the didjeridu is simply a hollow tube. The instruments
played at Yirrkata today are generally 1.3 to 1.5 meters long
and made from the narrow trunks of stringy bark eucalyptus (eucalyptus
tetradonta). The instrument maker does not hollow the tube, but
rather seeks out a tree which has already been eaten out by termites.
After the tree is felled and a proper length removed from the
bottom portion of it, the bark and outer layer of wood are peeled
and cut away. Some smoothing out of the inner surface of the blowing
end and optional decorative painting completes the construction
of an instrument.
In traditional performance contexts,
Yolngu do not ascribe tonal pitch value to the sound of the didjeridu
or to the interval forming between the didjeridu and the singer's
voice. Accordingly, the precise tuning of didjeridus is unimportant
in Yolngu culture. I The fact that Yolngu do not ascribe specific
tonal pitch value to the didjeridu does not mean that the instrument's
sound quality is musically unimportant. At each step in the process
of making an instrument, attention is given to providing a sound
that is clear and resonant. This is seen both in the discarding
of tree trunks determined to be too dry to produce a resonant
sound and in the use of a knife to cut away the outer layers of
wood, which renders a lighter, more resonant sound.2, 3
Different sets of playing techniques
are found in cultures who use the didjeridu, but one universal
technique is the production of a constant drone through the use
of circular breathing. While blowing, the player retains small
amounts of air in the cheeks; this air is then used to sustain
the drone when taking quick breaths in through the nose. The basic
blowing technique involves buzzing the lips in a way that is fairly
analogous to Western brass instrument technique. One major difference
is that the embouchure (the position and application of the lips
on the mouthpiece of a wind instrument) of a didjeridu, while
well-controlled, is less taut than the Western brass embouchure.
The resulting greater flapping of the lips helps give the didjeridu
its characteristic rich sound.4 Additional techniques characteristic
of Yolngu didjeridu performance include blowing the first over-
tone; diaphragmatic accents (which tend to raise the pitch of
the fundamental slightly); various types of tonguing; and vocal
sounds, such as singing and guttural shrieks. These techniques
are used in different combinations, simulta- neously affecting
rhythm and timbre.
Musicological writings about traditional use of the didjeridu have focused On the cultures of Arnhemland and Grooyte Eylandt. In this literature (Elkin and Jones 1958; Jones 1967 and 1973; Moyle 1974; Stubington 1978 and 1984) discussions of didjeridu style have generally served two ends. First, basic playing techniques and characteristic rhythmic patterns have been described. Second, these descrip- tions of technique and rhythm have been used to contrast the style of one didjeridu-playing culture with another. For example, the eastern Arnhem Land style, which features rhythmic alternation between the blown fundamental and blown first overtone, is distinguished from the western Arnhem Land style, which uses only the blown fundamental pitch but achieves rhythmic variation through control of vocal resonance.
Discussions along these lines provide
broad cross-cultural comparisons of style, but they do not,explain
stylistic variation within specific cultural areas. The stylistic
use of the didjeridu by Yolngu players is particularly varied.
Jones (1973) recognised this in stating that 'in the northeastern
Arnhem Land area the various forms of blown overtone are added
in, together with all the more western [Arnhem Land] techniques
and the whole range of vocal effects already men- tioned, producing
an amalgam of possibilities that itself divides into many sub-styles'.
(273) The range of techniques and rhythms employed by the Yolngu
is not only varied, but applied in consistently patterned ways.
In order to better understand these patterns of stylistic variation,
a distinction is made between style and determinants of style.
Style refers to the physical aspects of a player's sound and technique,
including the use of particular techniques, rhythmic pat- terns,
timbral quality and overall rhythmic 'feet'. Determinants of style
refers to factors of performance that are external to the actual
production of sound, but which may influence the manner of playing.
Examples include the symbolic meaning of certain sounds or particular
performance protocols.
The first part of this paper is concerned
with the use of the didjeridu in the traditional Yolngu ceremonial
song genre ('clan songs'). Clan song performance comprises a complex
musical system in which didjeridu style,is determined and organised
by many factors. Didjeridu style inceremonial clan song performance
will be discussed in terms of seven determinants: physical deten-ninants,
pe r- formance protocols, ceremonial protocols, symbolic determinants,
aesthetic. de- terrninants, developmental determinants and cross-cultural
determinants. In the second part of the paper, some of the same
determinants will be used to discuss didjeridu performance in
the pop/rock songs of the Yolngu band Yothu Yindi. By contrasting
the use of the didjeridu in these two different performance settings
we may see both some of the changes brought about by the use of
Yolngu didjeridu in a non-traditional context and some of the
aspects of playing that carry over from the traditional to the
commercial performance setting.
Yolngu clan songs
Yolngu clans (b'a'purru) are groups
of patrilineally related kin. Each clan owns an ancestral estate
comprised of land, songs, designs, sacred words, ceremonies and
ceremonial paraphernalia. The ancestral songs performed at public
portions of ceremonies such as funerals and male circumcisions
are referred to (in musicological literature) as 'clan songs'
because each clan is associated with its own song repertoire.
Each clan song repertoire embodies the ancestral history of a
given clan and is marked by distinctive thematic, textual and
musical elements.
During clan song performance, variably
sized groups of men sing and accompany themselves with pairs of
hardwood clapsticks (biLma), while both women and men dance. The
performance also includes the accompaniment of one male didjeriduist
(yidakimi), most often a young adult or teenager.5 Ceremonial
performances are built up from successions of hundreds of short
verses (called yutunggurr, literally 'thigh'). Each verse typically
lasts from fifteen seconds to a little over a minute and there
are short periods of rest or conversation between each verse.
The length of a given performance may vary from two or three hours
to an entire day or evening, depending on the occasion.
Various ancestral spirit-beings associated
with a given clan comprise the explicit subject matter of its
songs. These spirit-beings include flora and fauna, natural phenomena
such as clouds and wind, cultural artifacts (spears, canoe paddles,
speech), great ancestral hunters and ordinary people. The words
of each verse allude to some behaviour or quality of the spirit-being(s)
in a particular ancestral context. The succession of song subjects
over longer stretches of singing alludes to important events that
occurred at particular places in the ancestral past.
Musically, clan song verses are commonly
structured in three sections. In the first, introductory, section,
the lead singer establishes the clapstick pattern, begins singing
lines of text, or communicates (often in singing voice) information
about the song to the other performers. The didjeriduist's entrance
in this section often consists of a brief improvised passage that
begins with one or two short staccato notes (a way of quickly
adjusting the embouchure) and leads to a sustained note of variable
length. While holding this sustained note the did- jeriduist mentally
prepares to begin playing an appropriate rhythmic accompa- niment
to the singer's melodic lines.
The second section comprises the
main body of the verse. In its simplest form, this section involves
a few lines of sung text accompanied by simple, unpattemed clapstick
beating, an appropriate repetitive didjeridu pattern (vamp) and
repeti- tive dance movements that represent some idiomatic behaviour
or quality of the sung spirit-being. Frequently, the verse's main
body is marked by a more complex inner structuring which might
include the alternation between different slapstick and/or didjeridu
patterns, the use of recurring vocal/instrumental re- frains and/or
multi-sectioned choreographic movement. At the end of the main
body of the verse, the clapstick beating, didjeridu playing and
dancing all come to a synchronised halt. In the third, concluding,
section of the verse, singers may continue to sing lines of text,
at this point usually in less metrically-bound phrases.
Clan song verses are constructed
extemporally during performance, drawing from established sets
of key song words, clapstick patterns, didjeridu patterns and
dance movements. Clan song performance involves a significant
degree of improvisation, particularly on the part of the song
leader, who controls the flow from one song subject to the next,
the choice of clapstick pattern and other aspects of performance.
The didjeriduist plays an important accompanying role in the clan
song ensemble. Together with the singer's voice and clapsticks,
the didjeriduist establishes the rhythmic foundation of the verse.
He is also frequently responsible for cueing structural changes
within the verse for the benefit of the other performers. The
song leader, however, is ultimately responsible for the shape
and direction of the performance. The song leader is often a respected
elder or middle-aged clan leader with a large knowledge of song
texts and musical patterns and the ability to organise performances
involving many people and a voice that can hold up during many
hours of daily performance. The younger didjeriduist may be appreciated
and respected for his musical skills, but his role within the
ensemble is clearly that of an accompanists
Within the role of accompanist, the
didjeriduist has a degree of freedom in creating appropriate patterns
for sung verse. In many cases the didjeriduist's improvisation
consists of slight variations upon a brief repeating rhythmic
figure, or vamp. One basic vamp and two syncopative variants are
shown in Example 1. In an appropriate performance context, any
of these rhythms would be interchangeable; the use of one or another
would not effect the perception of stylistic difference. Stylistic
difference - differences in the 'way of playing' - is regarded
here in two senses. In one sense it can involve particular aspects
of rhythm, timbre and technique that are determined by various
ceremonial or performance protocols. In another sense, stylistic
difference may involve more general characteristics in a player's
rhythms, rhythmic feel, or technique.
Physical determinants of style
Both the physics of the instrument
and human physiology are important stylistic determinants. Some
of the physical characteristics of the instrument have been discussed
above. The Yolngu's musical use of the didjeridu is impressively
creative, though by no means exhausts the sounding potential of the instrument. The western Arnhem Lander's rhythmic use of vocal resonance provides one example of didjeridu technique which is not emphasised by Yolngu players from northeast Arnhem Land. It is also possible to play the didjeridu in ways that are not part of any tradition toda . For example, on most didjeridus it is possible, by
tightening the lips, to blow two
or more overtones above the two used by the Yolngu. Yolngu didjeridu
players have developed and maintained a particular range of techniques
that serves their functional and aesthetic musical needs.
Performance protocols as determinants
of style
Some of the standard protocols of
clan song performance directly influence certain aspects of didjeridu
style. One of these involves the use of the blown overtone. In
most clan song verses, the points where clapstick rhythmic changes
take place and the point at which the instrumental parts come
to a stop, are subject to performer discretion. In some circumstances
the song leader or lead dancer provides these cues, but they are
often given through the use of the blown overtone in the didjeriduist's
part. In Transcriptions 1, 3 and 4 the overtone is used to cue
the end of the instrumental and danced portion of the VerSe7 (nb.
all transcriptions are positioned at the end of the chapter).
These types of cadential patterns are so ingrained in clan song
performance that they are often used even where performance cues
are not strictly required; for example, in cases where a set number
of beats determines the end of a verse.
In cases where the didjeriduist provides
performance cues,- the song leader's preferences are still paramount.
One of the didjeriduist'-s tasks as accompanist is to anticipate
the number of vocal phrases the singer wishes to sing in a given
verse. If the singer is unhappy with the point at which the didjeriduist
cues the end of the verse, he may (during the break between verses)
instruct the did- jeriduist to sustain verses for a longer or
shorter period of time.8
Another standard performance protocol
that affects didjeridu playing is that different types of rhythmic
patterns are appropriate to different clapstick patterns. The
song leader changes the clapstick pattern every few verses within
constraints related to the song subject. His choice of clapstick
pattern, in turn, partly determines the range of rhythmic patterns
a didjeriduist may use. Verses using slow, unpattemed stick beating
- which Yolngu refer to as bulnha bilma ('slow clapsticks') -
require a particular type of ametric didieridu style. An instance
of this is in Transcription 1, performed by Dhakaliny Yunupingu
in 1990. One feature of rhythmic style with these slow stick patterns
is that, in the absence of a regular meter, the singer's and didjeriduist's
parts each cue forward motion in the other. Specifically, in the
main body of the verse, the didjeriduist anticipates the onset
of each new vocal phrase with two short notes (indicated by pairs
of sixteenth notes in the transcription). The exact onset of new
vocal phrases may, in turn, be influenced by the sounding of these
two short didjeridu notes. (Note that the beginning of each vocal
phrase in the verse's main body coincides with the didjeridu note
immediately following the pair of sixteenth notes.)
This type of slow, unmetered clapstick rhythm is found in the repertoire of all
Yolngu clans. Analysis of recordings
since 1952 indicate that didjeridu accom- paniments used to accompany
this type of clapstick pattern are a well-conserved element of
performance. Within this context there is room for only minor
variation; for example, the use of one (versus two) brief notes
in anticipation of a new vocal phrase. There are other classes
of named stick patterns that are also accompanied by a restrictive
range of didjeridu styles. These include yindi bilma ('big/important
clapsticks', these are discussed in relation to ceremonial proto-
col, below); and, to a lesser extent, ngarrunga bilma ('walking
clapsticks'), which include a variety of patterned and unpattemed
rhythms in moderate (i.e. 'walking') tempo.
Ceremonial protocols as determinants
of style
During lengthy public ceremonies
there are times when song performance accompanies specific ritual
events and less formal periods when the singing (and dancing)
is itself the main event. Both ceremonial contexts have ramifications
for didjeridu style. At important junctures during ceremonies,
only a narrow range of song subjects, clapstick patterns and didjeridu
styles is appropriate. For example, one periodic ceremonial event,
called gunbur'yun, involves the calling out by one or more mate
performers of sacred names that connect Yolngu to totemic land
sites. The songs that typically culminate in gunbur'yun are called
yindi manikay ('big/important songs'). The clapstick pattern appropriate
to both the act of gunbur'yun and the immediately preceding sung
verses is called yindi bilma ('biglimportant clapsticks') and
consists of rapid, unpattemed beating. The didjeridu pattern accompanying
yindi manikay consists of a simple drone articu- lated by breaths
(often pairs of breaths in quick succession) taken at fairly regular
intervals, but without regard for the timing of the vocal or clapstick
accents.
During less formal portions of ceremonies
there are more opportunities for didjeriduists (as well as singers)
to showcase more contemporary and/or personal styles. This may
be seen especially in portions of informal performance devoted
to 'newsong' verses (yuta manikay). Newsong verses comprise a
unique sub- genre within clan song performance. Whereas all other
clan song verses are believed to have been created by ancestral
beings at the beginning of time and performed in the original
manner to the present day, newsong verses are the acknowledged
creations of Yolngu singers. Newsong verses are inspired by some
contemporary event which, in the singer/composer's mind, is perceived
in relation to an ancestral song subject. For example, on one
occasion several years ago, a family got lost while driving home
through the bush. This inspired the creation of a newsong verse
concerning the ancestral dingo (who perennially wanders 'lost'
through the bush). Innovative aspects of newsong verses may include
invented text alluding to the contemporary event, new clapstick
patterns and new rhythmic counterpoints worked out between singer
and didjeriduist. New song verses are enjoyed by dancers and other
ceremony attendees both because of their frequent rhythmic vitality
and because of their longer duration (generally one minute or
more, compared with fifteen to thirty seconds for most other verses).
Symbolic determinants of style
The sound of the didjeridu is usually
heard as a simple rhythmic drone. At times, however, components
of the sound - blown fundamental, blown overtone, or vocal effects
- may take on specific symbolic meanings. For example, in particular
contexts involving songs about Wuyal, the ancestral honey gatherer,
the fundamental pitch of the didjeridu is understood to represent
the drone of wild honey bees. In certain contexts, a long, blown
overtone is sounded just prior to the start of each sung verse.
Depending on the song subject involved, this sound could mean
different things: for example, the cry of a dolphin or the call-to-performance
of an ancestral didjeridu player associated with the Morning Star.
An example of the latter may be seen in the long blown overtone
at the beginning of the didjeridu part in Transcription 4. Short
vocal shrieks produced simultaneously with the blown fundamental
may be used at any time simply as means of creating rhythmic and
timbral variety. When vocal shrieks are used in the accompaniment
of text about any of a number of ancestral birds, the sound can
be understood as symbolising the call of the given bird. Symbolic
associa- tions such as those described here affect not only listeners'
perceptions, but also the players' selective use of particular
techniques and sounds to accompany certain song subjects.
Aesthetic determinants of style
Some aspects of, the didjeridu's
sound reflect personal or group aesthetics. Personal and group
aesthetics may determine both the general 'groove' or feel of
the basic rhythmic vamp (including the relative placement of overtone
accents on top of or behind the beat) and the relative complexity
of specific rhythmic figures. Some didjeriduists gain reputations
as stylistic innovators and over time may influence others' playing.
In more than one respect, singers' aesthetics also influence didjeridu
style. Song leaders frequently dictate certain didjeridu pat-
terns or ways of playing that they want to hear. At the same time,
a lead singer may be more, or less, tolerant of young, didjeriduists'
stylistic predilections. It is common for a didjeriduist to informally
align himself, for a period of time, with specific singers. In
this context, too, a singer may influence didjeridu style, because
particular styles of playing will be worked out in the course
of frequent improvisational and compositional collaboration. The
degree of aesthetic open- ness or conservatism are traits that
may also apply to entire clans. Some clans' newsong verse performances
exhibit a wide range of clapstick patterns and didjeridu rhythms
while other clans' newsong performances are more limited in their
stylistic range.
Transcriptions 2 and 3 are excerpts
from two performances of a newsong verse concerning an owl-like
bird, tawny frogmouth. The two performances of Tawny Frogmouth,
recorded thirty seven years apart, are by the same singer, Larrtjannga
Ganambarr. This newsong verse was created in the mid-1930s by
the singer's father. The two versions, though different in a number
of respects, are identifiable as the same song by the use of a
recurring three-bar refrain. The refrain is always sung on a single
pitch (the tonic) and always contains the same text (texts are
not indicated in these transcriptions). In both recordings the
didjeriduist plays a different rhythmic pattern during the refrain
than in the rest of the verse.
The earlier Tawny Frogmouth performance (Transcription 2) was recorded at Yirrkala in 1952 by Richard
Waterman. The didjeriduist, Djirrnini
Yunupingu, was a renowned player during the 1950s who featured
prominently in Water- man's recordings. One aspect of Djirmini's
style is the alternate placement of overtone pitches at two eighth-note
and three eighth-note intervals. The result is syncopated with
respect to the regular two eighth-note figures of the singer.
Another frequent, notable feature of Djirrnini's style (though
not in the verse excerpted in Transcription 2) is a relaxed manner
of playing and the use of light, distinctly behind-the-beat, blown
overtones. These aspects of style may be heard in others' playing,
too, from the 1950s through the present. The relaxed, behind-
the-beat rhythmic feet is considered to be an older style of playing.
It is rare in contemporary newsong performance which favours more
precise, on-top-of-the- beat accompaniments.
The more recent Tawny Frogmouth performance
(Transcription 3) was recorded in 1989. The didjeriduist, Gunybi
Ganambarr, plays in a manner that is common in contemporary newsong
accompaniments- the overtone is used for rhythmic syncopation,
but the syncopation occurs within, rather than across, the beat;
the rhythmic feel is not so relaxed as in the earlier recording;
and the overtones are executed squarely in time.
Transcription 4 is of a complete
newsong verse related to the 'Morning Star' song series, recorded
in 1990. The didjeriduist, Dhakaliny Yunupingu, exhibits a style
of playing that could be described as avant-garde. In the context
of Yolngu didjeridu practice, the most striking aspect of Dhakaliny's
style is the use of rapid staccato tonguing on the fundamentals
Additionally, Dhakaliny's style is marked at specific points by
a complex rhythmic counterpoint to the vocal line and by a generally
driving, on-top-of-the-beat rhythmic feel. Dhakaliny's per- sonal
aesthetic is readily discernible in the context of newsotig verses,
but not in verses associated with the stylistically constrained
'slow' clapstick pattern (see Transcription 1).
Development determinants of style
The way that Yolngu learn to play
the didjeridu may help to influence both conservative and innovative
aspects of accompaniment style. On the side of conservatism, Yolngu
are, from childhood, continually exposed to the stylistic models
of their elders. Once a young player begins to perform in ceremony,
he is continually guided by singers and more experienced players
towards the correct (i.e. extant) ways of accompanying various
clapstick patterns and vocal phrasings. On the side of innovation,
the impact of self-training and peer group interaction are both
relevant. Though didjeriduists do not generally perform at ceremonies
until about age fourteen or fifteen, experience with the instrument
begins as child's play much earlier. Boys. (and sometimes girls)
as young as four or five may be seen ambling about blowing on
small-scale instruments. Boys nearing performing age practice
both traditional patterns they have memorised and new patterns
of their own making. They accompany young would-be singers and,
in informal technical duels with other young didjeriduists, they
learn to push their skills and aesthetics to new limits. It may
well be at this developmental stage that generational differences
in style begin to take root.
Cross-cultural determinants of
style
Since the establishment of a non-Aboriginal
mining town (Nhutunbuy) near Yiffkala in the late 1960s, Western
popular music (primarily in the form of commercial audio cassettes)
has captivated Yolngu youth. Yolngu usually expe- rience Western
pop music as passive listeners, though, on occasion, a young didjeriduist
may sit alongside a stereo cassette player improvising rock and
roll rhythms in accompaniment to popular music recordings. 10
We can only speculate about the influence that this type of experience
has upon ceremonial didjeridu style. However, on more than one
occasion, a Yolngu commentator criticised performance involving
contemporary and exuberant vocal and didjeridu rhythms by referring
to it disparagingly as 'rock and roll'. II It is hard to imagine
that there would be no influence. At least to this author's ears,
the rhythmically driving feel of many contemporary newsong verse
accompaniments sound more in line with a pop/rock aesthetic than
earlier, more relaxed styles of playing. Whatever the bases of
the shift in didjeridu style, the change itself is purely at the
aesthetic level. The technology of the instrument has not changed;
neither have the basic playing techniques nor the role of the
instrument within the performance ensem ble.
Yothu Yindi
Yothu Yindi is the most well known
Aboriginal rock band in Australia today. Formed by Yolngu singer/songwriter
Mandawuy Yunupingu in 1986, Yothu Yindi has achieved remarkable
commercial and critical success for their culture- bridging, politically-conscious
songs and performances. The band's members include Yolngu from
Yirrkala and Galiwin'ku (another Yolngu settlement) as well as
non-Aboriginal musicians. Their instrumentation combines standard
pop/rock instruments (electric guitars, keyboards, drum kit) with
Yolngu clap- sticks and didieridu. Yothu Yindi's recorded and
live performances include songs in Yolngu dialects and in English,
with musical influence from three istinct genres: pop/rockl2,
clan songs and a Yolngu recreational song form called dja4pangarri.
13 Many of the band's songs are purely in one of these three styles,
while some combine elements of more than one style.
The effect that performance context
plays upon music style can be seen by considering the stylistic
determinants in Yothu Yindi's didjeridu playing and in their selective
use of the instrument. In the remainder of this section we will
focus specifically on the stylistic use of the didjeridu in Yothu
Yindi's pop/rock songs. These songs comprise the bulk of Yothu
Yindi's recorded repertoires and (compared with the band's recording's
of clan songs and djatangarri) provide the greatest contrast with
didjeridu style in Yolngu ceremonial performance.
In Yothu Yindi's pop/rock material
the use of the didjeridu is made to conform to the performance
protocol and aesthetics of the pop/rock genre in several ways:
by the subdued presence of the fundamental tone in the overall
mix; post-record- ing re-tuning of the didjeridu's pitch; avoidance
of the blown overtone; heavy reliance upon the use of vocal shrieks
for rhythmic fills; and the use of a strictly metronomic rhythmic
feel. One of the protocols of pop/rock musical perform- ance is
the central role that the drums and bass guitar play in establishing
the rhythmic feel of a song. The didjeridu and clapsticks would
assume this role in clan song performance, but in the pop/rock
context - particularly in sections where the bass guitar and lower-pitched
drums are prominent - the didjeridu is relegated to a less central
role within the ensemble.
Unlike the case with clan song verses
where there is internal structural flexibility, all structural
elements of Yothu Yindi's pop/rock songs are worked out prior
to studio recording (perhaps with changes made in post-recording
studio editing). In this context there is no need for the didjeridu
to provide cues for other instruments. This is another reason
for the didjeridu being less functionally important within the
pop/rock performance context. Indeed some of Yothu Yindi's pop/rock
songs do not use the didjeridu at all.
The lack of precise tuning, of the
didjeridu's pitches is a normal'feature of traditional Yolngu
song, but can cause problems in the pop/rock context. For non-Yolngu
listeners, the presence of an arbitrarily pitched didjeridu together
with electric guitars and bass could cause an unwanted perception
of 'out of tunedness'. Because of this, in their pop/rock songs
Yothu Yindi's studio engineers alter the pitch of the didjer-idu's
blown fundamental technologically in order to bring it in line
with the non-Aboriginal instruments (usually to match the tonic
or fifth note of the scale). A perceived need to re-tune the didjeridu
in the pop/rock context is evident in remarks by Mark Moffit,
recording engineer and producer on Yothu Yindi's album Tribal
Voice (I 99 1) 'when it's in tune it really makes a big difference
with a rock track; when it's not it's a terrible sound' (cited
in Neuenfeldt, 1993: 65).
The didjeridu's blown fundamental
is treated in two ways in the pop/rock context. During portions
of songs where the use of bass guitar is prominent the didjeridu
is usually mixed down to the point of near or complete inaudibility.
This allows the sound of the bass guitar line to come through
unimpeded. By contrast, the rhythmic drone of the didjeridu is
quite prominent in certain interludes where the bass guitar drops
out. Thus, just as Yothu Yindi's entire repertoire is divided
into traditional Yolngu and pop/rock songs, one finds that even
within individual pop/rock songs music-stylistic distinctions
can be made between sections that are more or less 'Yolngu-sounding'.
The didjeridu's blown overtone is
scarcely heard in Yothu Yindi's pop/rock songs. As with the selective
use of the blown fundamental, the even scarcer use of the blown
overtone may relate to a desire to avoid conflict with the bass
guitar. The blown overtone on Yolngu didjeridus tends to be an
octave to a tenth above the fundamental (and so usually above
the bass guitar range), but its sound would still conflict with
the bass for two reasons. First, its tone, like that of the bass
guitar, is deep and intense. Second, because the interval between
the didjeridu's fundamental and overtone rarely corresponds to
a Western tempered interval, it would be difficult to bring the
didjeridu's fundamental and overtone pitches simultaneously in
tune with the bass, keyboards and other guitars.
With the avoidance of the blown overtone
and the restrictive use of the blown fundamental in Yothu Yindi's
pop/rock material, the vocal shriek (superimposed on top of a
scarcely audible blown fundamental or remixed together with a
digitally sampled fundamental drone) is a major component in the
band's didjeridu style. The sound of the vocal shriek is timbrally
distinct but does not interfere with the primacy of the bass guitar/bass
drum bottom. Two examples of Yothu Yindi's use of the vocal shriek
may be seen in Transcriptions 5 and 6.15 Transcription 5 is excerpted
from the opening portion of the song Dharpa (Yothu Yindi 1991).
Following a one bar introduction consisting of six vocal shrieks,
the didjeriduist forms a rhythmic line that combines the fundamental
tone with vocal shrieks. At the twelfth bar there is a general
shift in instrumental texture. Here the didjeriduist uses only
the vocal shriek (though a sampled sustained drone is audible
in the background). This eight-bar section ends with a build-up
of rhythmic intensity in the final bar. Transcription 6 is excerpted
from an instrumental interlude in the song Tribal Voice (Yothu
Yindi 1991). Here the didjeriduist uses only the vocal shriek.
As in the previous example, there is a rhythmic build-up in the
final bar of the interlude. One innovation in this section involves
the use of multi-tracked didjeridu: one instrument sounds in the
right channel, the other in the left.
The use of 'behind-the-beat' and
'on-top-of-the-beat' rhythm are both well established in traditional
clan song performance, but only metronomically pre- cise accompaniments
are used in Yothu Yindi's pop/rock songs. In the popular music
realm, metronomic rhythm is not only an aesthetic requirement,
but a measure of a musician's competence. This view may be seen
in Moffit's com- ments on Yothu Yindi's performance skills:
... they are the heavyweight traditional
musicians up there [Arnhem Land]; they're recognised as the leader
and they play all the ceremonies. I've heard didjs before and
it's quite easy for the player to get behind, to drag and this
guy didn't ... (cited in Neuenfeldt, 1993: 66).
Although Yothu Yindi does not perform
at traditional Yolngu ceremonies, some of the, musical decisions
made by the band may be seen as following the same ,ancestral
law' that guides traditional decision making. This includes making
an effort to consult with and gain consensus among all interested
parties before reaching important decisions.16 In the case of
Yothu Yindi this has involved consulting with clan elders regarding
certain elements of their performances; for example, in order
to gain permission to use ceremonial feathered headdresses in
their stage shows. 17 At the same time, in order to 'cross over'
into the commercial music arena, the band has consulted and worked
with people who have experi- ence and vested interests in commercial
record production and distribution. Here, as we have seen, the
perceptions and judgements of others working with the band (such
as studio recording engineers and their perceptions of didjeridu
pitch) have influenced Yothu Yindi's use of the instrument.
Didjeridu style in Yothu Yindi's
pop/rock work is largely disconnected from the wide range of ancestral
spirit-related symbolic associations found in ceremonial performance
contexts (usually the calls of specific birds or animals). But
in this same context there is a new, non-traditional association
connected with the didjeridu's sound: for Yothu Yindi's non-Yolngu
listeners, the didjeridu pro- vides one of the main signifiers
of the band's Yolngu identity.
Yothu Yindi's didjeriduists all learned
to play the instrument in its traditional context long before
Yothu Yindi was formed. These same musicians were also exposed
to Western popular music from an early age. In this sense they,
like all Yolngu, are bi-musical. The possible influence of Western
popular music upon didjeridu playing in ceremonial clan song performance
has already been dis- cussed. In the traditional performance realm,
such influences would be con- strained by the need for the didjeridu's
sound (and style) to support all of the traditional aspects of
the performance genre (the need for structural cueing, the need
to be aesthetically acceptable to elder singers, etc.).
In the popular music context, Yothu
Yindi has been able to use the didjeridu in ways that would not
be possible in traditional performance contexts. As we have seen,
some of the band's contemporary didjeridu stylings involve the
preference for certain traditional techniques (such as vocal shrieks)
over others, the exclu- sive use of on-top-of-the-beat rhythmic
feel and the alteration or extension of the didjeridu's sound
through technological means. But the basic range of traditional
Yolngu playing techniques has so far proven more than adequate
for the purpose of performance with Yothu Yindi.
Conclusions
We have looked at several levels
of clan song performance at which didjeridu style may be determined.
These interrelated determinants are not an exhaustive set, but
are simply intended to provide some filters through which the
complexity of didjeridu style may be viewed. It is clear that
Yolngu didjeridu style does not involve a single manner of playing.
Rather, it is a variable aspect of performance, determined by
aesthetic, functional and symbolic factors. In traditional clan
song performance, didjeriduists (and other performers) enjoy a
degree of creative freedom, yet the genre has, on the whole, proven
to be a robustly conservative cultural practice. By restricting
the contexts in which innovative performance may take place, Yolngu
limit the effect of stylistic variation upon the musical system.
The result of this circumscription is that the greatest degree
of stylistic change affects the aesthetic realm of performance.
By contrast, the functional and symbolic aspects of performance
appear to remain constant.
In a number of ways, Yothu Yindi's
adaptations to the pop/rock performance context can also be viewed
with respect to traditional circumscription of musical style.
In their pop/rock songs the possibility of multitrack recording
and digital sampling and processing have allowed the didjeridia
to be used in some very un-traditional ways. Similarly, aesthetic
problems related to combining the didjeridu with Western pop instruments
have influenced the more prominent use of vocal sounds. Neither
of these factors, however, have altered the basic range of techniques
found in traditional performance.
The circumscription of didjeridu
style in Yothu Yindi's work can also be seen in the fact that
their repertoire includes the use of traditional genres alongside
(and sometimes within) pop/rock songs. In Yothu Yindi's recordings
of clan songs and dja4pangarri the use of the didjeridu is more
or less the same as in traditional contexts. For young Yolngu
listeners, there may be a double-sided message in this, namely,
that pop/rock songs may provide an appropriate avenue for bridging
cultures and experimenting with very new performance ideas; but
traditional songs and performance - including didjeridu accompaniments
- will remain strong and viable, even as Yolngu actively engage
themselves in the creation of commercial music products.
Notes on music transcription
In Transcriptions 1-4, the vocal
parts are given without texts on a six-line staff. The six-note
scale upon which the vocal melodies are constructed does not correspond
to Western scales because Yolngu tuning is both non-tempered and
variable (i.e. the precise tuning of intervals need not be exactly
the same from verse to verse or from performance to performance),
Without concern for exact tuning measurements, the six lines of
the staff represent each of the six discrete scale steps. The
notes represented by the top and bottom lines of the staff are
approximately an octave apart. The uneven spacing of the staff
lines visually reflects the alternately large and small scale
steps that comprise the scale. The sizes of these scale steps
often roughly correspond to 3 and 11/2 Western semi-tones, respectively.
The vocal part in Transcriptions I and 4 use all six notes of
the scale; the vocal parts in Transcriptions 2 and 3 use only
the bottom three notes of the scale. For purposes of reference,
approximate Western tunings of the vocal parts in these particular
recordings are given on a standard five-line staff at the beginning
of Transcriptions 1-4. Upward and downward arrows here indicate
microtonal deviations of less than 1/4 tone from standard Western
pitch.
All didjeridu parts are notated on
two lines. The blown fundamental appears on the bottom line; the
blown overtone on the top line. Circled noteheads indicate vocal
shrieks sounding together with the blown fundamental. Dashed slurs
in the didjeridu parts indicate the use of subtle, untongued articulation.
In these cases, rhythms are created solely by changes in air pressure
and shape of the oral cavity. For purposes of reference, approximate
Western tunings of the didjeridu parts are given on a standard
five-line staff at the beginning of Transcriptions 1-4. No didjeridu
tunings are indicated in Transcriptions 5 and 6. In these excerpts
the pitch of the blown fundamental is inaudible and no blown overtones
were used.
In Transcription I only, vocal phrases
are indicated by notes connected to a single beam. Beams do not
indicate rhythm per se, but rather connect all the notes sung
in one breath. Rhythmic notation of the didjeridu part in Transcription
I is approximate and intended primarily as an indication of accented
note grouping. Visual alignment of noteheads (vertically between
the three parts and horizon- tally within parts) indicates relative
attack points and durations of notes.
In Transcription 1, crescendo/decrescendo
markings in the didjeridu part do not refer to dynamics, but rather
to the relative amount of vocal resonance. The more vocal resonance
that is used, the richer and buzzier is the resulting tone. In
Transcriptions I and 4, double bar lines indicate the dividing
points between introduction and main body and between main body
and vocal coda. The Tawny Frogmouth newsong verse excerpted in
Transcriptions 2 and 3 incorporates a verse/refrain structure.
In Transcriptions 2 and 3, the double bar lines indicate the dividing
points between verses and refrains. In Transcription 2, bracketed
bars in the didjeridu part contain only approximate note values
due to the lack of clarity in the original sound recording.
The transcriptions in this article
were prepared with Coda Music Technology's Finale versions 3.2
and 3.5 for Windows.
Acknowledgement: The fieldwork for this study was generously supported by a Fulbright Award for Australia, research grants from the Australian Institute for Aboriginal and Torres Strait Islander Studies and an Andrew Mellon Predoctoral Fellowship. I would like to thank Gail Rein for commenting on an earlier draft of this paper.
Notes
1. An analogous situation in Western
music may be seen in the use of bongos or conga drums. These instruments
produce quite audibly distinct pitches, yet we have trained ourselves
to ignore the potential tonal relationship between these drums
and other instruments in an ensemble. Instead we hear these pairs
of drums as producing only relative 'high' and 'low' notes.
2. Didjeriduists in the Yirrkala
area sometimes give pet names to instruments which acknow- ledge
the sound quality of the instrument through association with an
ancestral being. Magowan (1994), in discussing the naming of didjeridus
in the Yolngu community at Galiwin'ku, notes that one didjeridu
was named for the Olive Python, Wititj, because 'the twisted trunk
of the tree was said to resemble the snake's body and the. deep
pitch could be likened to the deep and very powerful sound of
the snake heard in the thunder ... [Another didjeridu] was named
after a large swamp bird, the brolga, Gudurrku, as the player
could produce a resonant brolga call from it' (283-284).
3. Tuning measurements of a dozen didjeridus recorded in the Yirykala area in 1989-90 and 1952 show a range of fundamental pitches from 77 Hz (equivalent to a slightly lowered E flat one line below the bass clef) to 99 Hz (a slightly raised G on the bottom line of the bass clef). The intervals formed between fundamentals and first overtones ranged from 1226 cents (a slightly wide octave) to 1686 cents (a slightly flat perfect 12th), with the most common intervals falling in the range between a minor 9th and a neutral 10th.
There is no indication that instrument
makers test for specific musical intervals when looking for and
constructing didjeridus. The approximate range of the fundamental
pitch might be important, however, since very low-pitched instruments
(which tend to be longer and have wider bores) require more air
and are thus more difficult to play and because the overtone on
very high pitched instruments may involve too much air resistance
for the purpose of making graceful transitions between the overtone
and fundamental pitches. There also seems to be some correlation
between the relative height of fundamental pitches and the intervals
between fundamental and overtone pitches. Specifically, the instruments
with lower-pitched funda- mentals tend towards slightly wider
fundamental/overtone intervals while the instruments with lower-pitched
fundamentals tend towards smaller fundamental/overtone intervals.
4. Wiggins (1985) discusses the role
that 'loud, harsh' lip buzzing, vocal resonance and the naturally
convoluted inner surface of the tube play with respect to the
didjeridu's prominent high overtones and inharmonic formants.
5. At certain points during ceremonies,
women perform clan songs as ritualised wailing, called ngdtji
('crying'). Ngdtji is performed by a group of women without clapsticks,
dancing, or didjeridu accompaniment.
6. Some young men, through their
clan lineage and/or song related skills, go on to become song
leaders. For them, a period of time spent accompanying elder singers
serves as an apprentice- ship in singing and song leadership.
7. Unless otherwise noted, all transcriptions
are from recordings made in the Yirrkala area in 1989-90 by the
author.
8. The above discussion has focused
on particular functional uses of the didjeridu's overtone. The
reader should not infer from this that the didjeriduist's use
of the overtone is completely determined by performance protocol,
as this is certainly not the case. Throughout the main body of
a song verse, alternation between the blown fundamental and overtone
may be used to create rhythmic variety and in this sense the way
that a player uses the overtone (as well as vocal shrieks) is
very much a matter of personal style.
9. The assessment of these rapid
staccato figures as strikingly unusual is based both on my own
perceptions of Gakaliny's playing and on comments made to me by
two other Yolngu didjeriduists.
10. Observed during my stay at Yirrkala
in 1989-90.
11. Ibid.
12. The term 'pop/rock' is intended
broadly to include all styles of English language songs - from
country western and folk to heavy metal - that are disseminated
via commercial recordings in Northern Australia.
13. Djatpangarri is a song form that
developed at Yirrkala during the Mission era (late 1930s through
early 1970s). In contrast to the religious, ceremonial nature
of clan songs, djatpan- garri songs are purely recreational in
nature. Originally performed by young men at im- promptu concerts
at Yirrkala's beach carnp, the subject matter of djaipangarri
simultaneously concern a variety of everyday or contemporary phenomena
(birds, the sea tide, cricket players, stage comics) while alluding
to personal relations between individuals through reference to
the names of particular secondary kinship groups (malk) that the
individuals belong to.
14. Sixteen of the twenty nine songs
included on theyothu Yindi albums Tribal Voice and Freedom are
in a pop/rock musical style. The styles used in the remaining
songs include clan song (6), djatpangarri (2), pop/rock and djaipangarri
combined (3) and pop/rock and clan song combined (2).
15. The didjeriduists credited on
the albums Tribal Voice (199 1) and Freedom (1993) are Makuma
Yunupingu, Bunimbirr Marika and Milkayngu Mununggurr.
16. Williams (1985 and 1986) has discussed the importance of consensus-building among interested parties in Yolngu decision-making.
17. Mandawuy Yunupingu, personal correspondence with the author, 1990.
Bibliography
Elkin A. and Jones T. (1958) Arnhem
Land Music Sydney: Oceania Monographs n9.
Jones T. (1967) 'The Didjeridu: Some
Comparisons of Its Typology and Musical Functions with Similar
Instruments Throughout the World', Studies in Music vl.
Jones T. (1973) 'The Yiraki (Didjeridu)
in North-Eastern Arnhem Land: Techniques and Styles' in Berndt
R. and Phillips E. (eds) The Australian Aboriginal Heritage: An
Introduction through the Arts, Sydney: Australian Society for
Education through the Arts in Association with Ure Smith.
Knopoff S. (1992) 'Yula Manikay:
Juxtaposition of Ancestral and Contemporary Elements in the Performance
of Yolngu Clan Songs', Yearbookfor Traditional Music n24.
Magowan F. (1994) 'Melodies of Mourning',
(unpublished D Phil thesis, Oxford University).
Moyle A. (1974) North Australian
Music: A Taxonomic Approach to the Study ofaboriginal Song Performances,
(unpublished) Ph.D. thesis, Melbourne: Monash University.
Moyle A. (I 98 1) 'The Australian
Didjeridu: A Late Musical Intrusion' World Archaeology vl2 n3.
Neuenfeldt K. (1993) 'The Didjeridu
and the Overdub: Technologising and Transposing Aural Images of
Aboriginality' Perfect Beat, vt n2.
Stubington J. (I 978) Yolngu Manikay:
Modern Performances of Australian Aboriginal Clan
Songs Ph.D. thesis, Monash University.
Stubington J. (1984) 'Review of Djambidj: An Aboriginal Song Series From Northern Australia' Australian Aboriginal Studies, v 1.
Waterman R. (1952) Yirrkata Field
Recordings - n3imeo (copies held at the Institute for Aboriginal
and Torres Strait Islander Studies, Canbeff a).
Wiggins G. (1985) The Physics of the Didjeridu (unpublished manuscript, Boston University, College of Liberal Arts).
Williams N. M. (I 985) 'On Aboriginal
Decision-Making' in Barwick D. Beckett J and Reay, M (eds) Metaphors
of Interpretation:Essays in Honour of W. E. H. Stanner, Canberra:
Australian National University.
Williams N. M. (1986) The Yolngu
and Their Land: A System of Land Tenure and the Fight for its
Recognition Canberra: Australian Institute of Aboriginal Studies.
Discography
Yothu Yindi Tribal Voice, Mushroom Records 1991
Yothu Yindi, Freedom, Mushroom Records 1994.