Volume 11, Number 4, October 2005
Copyright © 2005 Society for Music Theory
The Persona-Environment Relation in Recorded Song
KEYWORDS: Persona, Environment, Subject-Position, Accompaniment, Signification
ABSTRACT: This article proposes a typology of relations between the persona and its environment in recorded popular song, envisaged as the relationship between lyric/melody and accompaniment. It argues that the function of an accompaniment moves from simple pitch/meter orientation, through genre-setting and tone-setting, to support, amplification and explanation of the persona’s situation, and finally to contradiction with it.
 Some years ago, Simon Frith asked the very obvious question, “why do songs have words?”(1)(2) It seems to me that its obverse is equally demanding of attention. Not actually “why do songs have music?,” for that is a little too abstract even for my taste, but the associated “why do songs have accompaniments?” Why, when singing, do people (both listeners and performers) want or need the security to be gained from having that singing accompanied, either by themselves, or by an assortment of others? What does an accompaniment add? The word “accompany” indicates being in the company of something else, with an implied hierarchy of perceptual pertinence, although not necessarily of significance. What accompanies forms a key part of the environment for what is accompanied. Philip Tagg’s semiotic method reads this implied relationship (between what accompanies and what is accompanied) as the embodiment in music of the relationship between an individual and its environment.(3) He observes the normative underpinning by a melody-accompaniment dualism of all popular song, insisting that this textural structure is “what Haydn and AC/DC share in common,”(4) and uses Jan Maróthy to argue that it is as pervasive a musical feature as its visual equivalent, the figure/ground dualism of post-Renaissance European painting. So, if accompaniment equates, at first approximation, to environment, then melody, in Tagg’s understanding, equates to persona. This term I actually derive from the writings of Edward T. Cone, particularly The Composer’s Voice,(5) but I shall depart from him in how I observe the persona to be constituted. Indeed, I depart also from Tagg in that for the analysis of popular song, it is less the melody which is subject to accompaniment, than the result of the activity of singing—Tagg’s work concerns itself only little with lyrics. Eric Clarke’s adoption of a theory of subject position(6) does concern itself with lyrics, and also with accompaniment. He argues that for a song to imply a subject position, for it to encourage a listener to prefer one reading to others, it must be possible to distinguish analytically between what he terms “content” and “technique,” such that the “technique” narrows the possible interpretations a listener may make of the “content.” For Clarke, “content” refers particularly to the realm of the lyrics, while “technique” in principle refers to all other constitutive aspects of a song. Neither Clarke nor Tagg use the term “persona”—Clarke prefers “subject,”(7) while Tagg uses “individual.” Both of these, however, refer to an equivalent identity. I prefer the term “persona” for a number of reasons. Firstly, it reminds us that this identity is fictional (and, frequently, it is passive), a characteristic which is not brought out by the competing terms. Secondly, it is on occasion multiple (particularly when we do not hear only a single voice). Beyond this, it is perhaps just a matter of taste.
 In defining the potential types of relationship obtaining between persona and environment in this article, I adopt a position part-way between those of Tagg and Clarke. Both lyric, its manner of articulation, and its shaping melody can conspire to create the persona, which is inhabited by the individual (or sometimes individuals) who sing to us. The environment within which (or against which) that persona operates is represented by the music which accompanies her/him, and which therefore includes three distinct elements:
Because of its manifestation in sound, we might properly identify this as the personic environment. And, to complicate matters further, because as individuals we are always part of the environment against which we view others, on occasion the melody can, in theory at least, be argued to operate as if it were part of the environment rather than the persona. At root, then, this article proposes a typology for the relationships enacted in song between a persona and its environment, a typology of what an accompaniment may add to a melodic line, which it will illustrate through numerous, varied, examples to demonstrate not only the ubiquity of the relational positions I identify, but also the way that each one is nuanced through its presence in an actual track.(8)
 The impact an environment can have on the persona can range from the
significant to the trivial, as accompanimental textures, harmonies and forms
serve both subtle and blatant functions. Let me dispose of some of the latter
straight away. The need to orient oneself in terms of precise meter and stable
pitch is endemic to anglophone popular musics and, probably, to most European
popular music. I am not convinced that the ability to retain meter and pitch is
normative even amongst highly respected interpreters. Producer Bernard Krause
wrote of singer Patrice Holloway: “So perfect was her pitch [on an unaccompanied
track] that Paul [Beaver] was able to lay a piano track
 A related reason is hardly more interesting. Accompaniments
frequently set the genre of a song, laying out the normative environmental
with all the expectation-related baggage that a theory of genre carries.(10) Thus an adult-oriented rock song, a country song, a punk song, an R’n’B song,
are in large part defined by the instrumentation and sound-sources used to accompany the singer and, in some cases, the style of performance associated
with those instruments individually or collectively. I am highlighting here a
crucial distinction between accompaniments which signify only at the level of
style, and those which signify at the more particularized level of the
individual song. Examples of the former are legion—I shall refer to Patsy
Cline’s “I’m blue again” for its clarity.(11) The song is in 32-bar form, in
 Accompanimental textures can also set the attitudinal tone of a song, laying out a particular manner of approach to which the singer then conforms, as in Bob Dylan’s “Just like a woman.”(14) The accompanying band has a certain rhythmic looseness, most identifiable in very subtle delays of the guitar behind the kit. This comes across most typically as relaxed, but it might be read as hedonistic, as hippy, or even as sloppy, depending on the listener. Whichever, it could be argued to act as a neat metaphor for the falling to pieces of the “little girl” who, with all her insecurities, hides behind the song’s title. If the reliance on generic convention in “I’m blue again” leads any genre-competent listener to expect to hear nothing new in the song, the (relatively) unusual “loose” introduction to “Just like a woman” will probably create in a genre-competent listener an expectation of an idiosyncratic lyric content, and one of a potentially personal nature.
 In other examples of such tone-setting, the accompaniment can function inter-stylistically, to signal that a song will depart somewhat from the style normally associated with a particular band.(15) The prominent Hammond organ at the beginning of Led Zeppelin’s “Thank you,”(16) for example, tells us we are in for an early rock ballad rather than a track in blues-rock, or proto-heavy-metal idiom, a promise redeemed by the earnestness of Robert Plant’s expressions of constancy. In Roy Harper’s “When an old cricketer leaves the crease,”(17) the function of the accompanimental silver band is rather more specific: it encourages a culturally-competent listener to hear the song as nostalgic, in its evocation of a no longer communal, rural English, late summer twilight, echoing the familiar conceit equating the passage of a year to the course of a life.(18)
 So, the function of tone-setting acts to prepare the listener for how to respond to the style, probably from the outset of a song. As far as this function is concerned, the accompaniment will not necessarily intervene thereafter in the expressive meaning of the song. Recall that this is what the majority of accompaniments simply do.(19) And it is here, we may suppose, that a large number of listeners switch off. It seems to me that all will be, on some level, aware of these lower levels of functionality of accompaniments. However, to note the functions which follow, active listening is probably required. Other than this simple distinction, I am not suggesting that the functions I outline are hierarchically related in terms of “importance”—what is important for a listener is for that listener to determine. However, a degree of active listening is required in order to answer the question I set at the opening of this article.(20)
 But as with so many expressive features, we become alert to them when faced
with their subversion (indeed, my comments regarding “Thank you” could be seen
as an unproblematic example of this). The track “‘T smidje,” by Flemish vocal
trio Laïs,(21) begins with
the sort of
 With this function, we begin to tread interesting ground. We begin to
approach that realm where some attention to the detail of the personic
environment is necessary in order to apprehend the expressive richness of the
virtual performances of particular personae. Here we have the received view of
the role of an accompaniment, which is to conform to the ostensible meaning of
the lyrics and, in doing so, to support, or perhaps illustrate, the meaning of
the song. Forms of word-painting are the most obvious devices here, but within
popular music, we must be wary of seeking the sort of notationally-based
word-painting techniques familiar from notated music. In a recorded genre,
word-painting works by reference to the sounds heard, not to their visual
representation. Iggy Pop’s recent song “Whatever”(23) exemplifies this clearly.
The song originates on a lacklustre album, and is ironic in its consistent
recourse to the kind of adolescent disdain indicated by the title (Iggy Pop was
born in 1947). The stylistically conventional overdriven guitar which forms part
of the accompaniment simply drops from the texture on the words “her voice just
fades away.” It is not a true fade, but the parallel is strong enough—by
dropping away, the texture literally illustrates the content of, actually
energizes, the lyric. Indeed, the very obviousness of the device adds to the
rather tongue-in-cheek subject position the song adopts. An analogous, although
not ironic, situation can be found in Joe Cocker’s cover of “With a little help
from my friends.”(24) As in the Beatles’ original, the backing voices change
position during the song from being simply respondents, to actively questioning
the persona’s need for reassurance from others. But, deviating from the
original, in the second bridge (at about 3’46”), the support these “friends”
offer becomes so psychologically strong that Cocker no longer needs to sing the
lyrics his listeners would have known from the Beatles’ version. The “help”
offered by the accompanimental, female, singers here becomes palpable. The key
feature of the environment in which Cocker is singing is formed by these
confidantes. A similar degree of accompanimental support is offered to John
Lennon’s “Imagine”(25) by the production qualities particularly of Lennon’s
accompanying piano. Here, the sonically unfocused quality of the production of
the instrument’s sound supports a similar fuzziness in the singer’s ideology,
which has contributed to the debates about both the degree of realism, and the
self-delusion, which may surround Lennon’s song.(26) This quality is enhanced by
the observation that the melodic line first achieves a measure of
harmonically-supported closure on the word “dreamer.” And we don’t have to wait
for the rise of rock to find such examples. Nat King Cole’s performances of
“When I fall in love”(27) spend most of their course promising how he will
behave when that eventuality comes about. Unexceptionally, the song matches
turns to the minor with recollections of sadness but, at the last, as his
falling in love becomes directed (“when I fall in love
 A rather different example appears in the Yes song “Heart of the sunrise.”(28) According to the lyrics printed in the sleeve, the word “sharp” is clearly marked for attention, being capitalized. Almost without exception, each appearance of the word is immediately followed by a “sharp” sound, i.e. one with a notably crisper attack than its surrounding sounds—on its second appearance (at 4’26”) by a most pointed kick drum, at its third (at 5’43”) by the entire band. Indeed, by enacting the lyric, this example seems closer to what I would describe as “amplifying” the meaning, rather than simply “supporting” it, suggesting that “supporting” and “amplifying” be seen as two stages on a continuum.
 Enactment of a lyric will often take place by means of a track’s form. The
lack of final closure embodied in Janis Joplin’s live performance of “Ball and
chain”(29) serves to prevent the listener from treating the performance as
simply a performance of the song. Here, the accompanying musicians vanish (in
the actual performance, they most likely left the stage) while Joplin delivers
an extempore solo but, in a departure from contemporary practice, they fail to
return to round off the song. This solo serves to enact the lyrics by blurring
the boundary between a (fictionalized) performance and a diatribe on the nature
of contemporary society. In Sonny and Cher’s love song “I got you babe,”(30) the
“us against the world” position which the lyrics promote is maintained musically
by a monotone
 Three further examples will demonstrate the variety of ways simple accompanimental textures are involved in such enactment. Björk’s “Army of me”(33) overturns established gender stereotypes in angrily addressing a weak partner—the absence of a harmonic filler layer within the texture(34) (i.e. the sparsity of the accompanimental strands) removes an element which serves to distance the listener, making palpable the menace expressed. The Beach Boys’ “God only knows”(35) leaves the answer to “what I’d be without you” unanswered in the lyric. However, the song’s conclusion provides an answer, in the three-part quasi-quodlibet, which literally enacts the breaking up that the persona would otherwise endure. In a final example, the persona’s environment is formed by a combination of form and accompaniment, but which seems to go beyond both—this environment for the song “John Gaudie,” by English folk revivalists Whippersnapper,(36) clearly “amplifies” the signification—it makes it larger, sets it on a wider stage. The song concerns our John who breaks out of gaol to go off and write a fiddle tune, which the band then go on to play. We might assume the song to be traditional, and the subsequent tune newly invented. In fact, what happens is a reversal of this—the final tune is traditional, while the song itself is credited to fiddler Chris Leslie. This variety of examples attests to the wide presence of this function.
 So, we move from simply supporting the meaning of the song, to amplifying it, to providing more information than is present in the lyric. There are other ways of amplifying the song’s meaning than this. One concerns lyrics of ambiguous meaning. The issue of ambiguity is problematic because, with many songs, ambiguity is only the result of unfamiliarity with idiom, and is thus dependent on the nature of listeners’ competence. Traffic’s “Hole in my shoe,”(37) though, provides a case which may be more universally valid. It causes trouble to anyone who simply tries to interpret the lyrics at face value—“an elephant’s eye was looking at me from a bubblegum tree” for instance. The accompanimental texture, however, situates the song very clearly within the psychedelic movement, indicating that the lyrics are not to be taken at face value, that the environment is not a realist one. Thus the “100 tin soldiers” do not literally “stand at my shoulder.” Indeed, the prominent psychedelic coding(38) may even imply that the texture, rather than the lyric itself, may be the prime carrier of meaning. This coding is provided primarily by instrumentation (sitar, flute, mellotron) and production techniques (slow chord bending, reverse recording, phasing). Thus we are reassured that the song’s ambiguity is inherent, and is not a feature of our possible lack of competence. In this way its meaning is clarified.
 Meaning is also amplified when an environment provides information more deeply encoded than that of the lyrics, most particularly by its harmonic underpinning, for harmonic setting often implies very particular readings of lyrics. Simple examples are provided by Slade’s “Coz I luv you” and by Jimi Hendrix’ “Hey Joe.”(39) Both songs concern troublesome interpersonal relationships and both are set to open-ended harmonic loops. It is the circular nature of these which illuminates the obsessive nature of both songs, something which is not fully explicit in the lyrics themselves. On “The way that it shows,”(40) Richard Thompson uses harmony to amplify the similar disintegration of a relationship, below the surface, in the obsessive repetition of the sequence Dm–Bm–Gm–E. A sensible sequence in terms of root motion becomes nonsensical (in relation to assumptions about harmonic norms in his repertoire) through its chromaticisms.
 A slightly more complex example is the Beach Boys’ well-known hit
vibrations.”(41) The song opens with a stepwise descent from Em, repeated, before shifting to the
relative major for the chorus. The singer sings of being in receipt of “good
vibrations,” over a pattern which moves upward, from G, to A, to B in readiness
for the second verse. The vector is clear—upward motion, and activity in the
relationship by his partner, towards him. In the centre of the song, the texture
slims remarkably, down to organ and shakers, over which the singer begins to
recognize his culpability—he’s “gotta keep those
 I wrote at the outset that harmonic setting can be vital in its representation of the environment. How, though, to read this? The cornerstone of my understanding equates achievement of I with closure in some sense, and failure to achieve it with lack of closure, modified as necessary by metrical placing. The other key characteristic is the observation that as a mode increases the number of minor intervals above the tonic (as we move from Lydian through to Locrian), the greater the degree of negative emotional quality invested in the environment. This hermeneutic construct is yet to be empirically tested. Deep Purple’s “Child in Time”(42) is one of a body of songs reliant for its effect on the Aeolian cadence. This is the cadential pattern moving VI–VII–i (i.e. F–G–Am in a Aeolian). This cadence(43) in rock is, more often than not, associated with achievement in the face of high odds, what I call its “nonetheless” quality. This is the realm of its signification. The underlying pattern to “Child in Time” is formed of i–VII–i; VI–VII; VII–i. The impending, and ultimately realized, disaster apparent in the lyrics is matched by this constant reiteration, demonstrating an undesired actualization of the inevitable outcome. This sense of the inevitability of an unwanted outcome is crucial to understanding the Coverdale Page track “Take me for a little while.”(44) The verse begins by decorating harmonies of E, moves to a pattern dominated by Am, but then sinks back. A short pre-chorus reiterates the importance of Am, such that the chorus takes up the sequence Am–F–G–G. Failure to resolve to chord I, in this example to resolve to Am at the end of the Am–F–G–G pattern, implies the subversion of an inevitable outcome. Now the subject matter of the song’s lyric concerns the inevitability of failure—at the crucial moment, the singer tells his lover to take him just “for a little while.” Their relationship cannot remain stable, because he’s “growing older.” The harmony, in other words, indicates that “taking him for a little while” is an attempt to subvert the inevitable, i.e. the inevitability of failure, a failure—his demise—which will come about once the final chord I is reached and closure is achieved. For this reason, the track fades without that happening. The Darkness track “Holding my own”(45) uses the same pattern to energize an avoidance of the inevitable. It is less marked: the parallel Am–G–F–G chord sequence accompanies the song’s title, asserting that the persona is “holding his own” in the face of enormous odds; because it is less marked, the sequence functions harmonically as Ionian vi–V–IV–V rather than Aeolian i–VII–VI–VII. The realm of its signification is nonetheless the same.
 We can move further, I’m convinced, from support, through amplification, to explanation: the harmonic underpinning to the Rolling Stones’ “Satisfaction”(46) explains the state Mick Jagger describes. This is a constant state, one of being unable to achieve satisfaction, to achieve closure. Musically, closure is unattainable here because the constant repeated sequence of the open-ended verse is simply a Mixolydian I–IV– the I is so familiar that the closure otherwise attendant on achieving it, is impossible here. The melody, too, is insistent upon its reiterated tonics. No wonder he can’t get no satisfaction—he has nowhere else to go. From a very different genre, Vanessa Carlton’s “Ordinary Day”(47) uses harmony and texture to explain the lyric. The song concerns an encounter with first love, energized from the outset by Carlton’s breathy delivery and an insecure bass which consistently sounds the tonic third rather than the root, providing a picture of (pleasant) insecurity or anticipation. [See Example 3a] After two verses, at 1’46“, the imaginary ”ordinary boy“ is providing security (she’s ”in the palm of your hand“) as the tonic achieves its root [Example 3b]. To mark this coming together, at 2’1”, the two independent parts move in both rhythmic and partial melodic unison [Example 3c]. Subsequently she comes out of what was “just a dream” and we move back to separation and insecurity marked by absence of the harmonic root. However, as she recalls the experience out of the dream world, at 3’20”, she recovers her security. The song thus provides a model of a girl overcoming insecurity through learning to trust a “significant other,” dramatized by the difference between “dream” and “reality” and the transfer of something understood while in the “dream” state into reality. This model cannot be inferred form the lyrics alone, but requires close attention to both the texture and harmonies of her accompaniment.
 In suggesting that an environment may explain the actions of a persona, I return briefly to Clarke’s work on subject position. Translating into the terms I am using in this article, he suggests that certain songs, through the detail of their environments, limit the potential range of listener responses to the issues presented by the persona. Provided we go no further than “limit,” and not as far as “prescribe,” I find this a pretty persuasive argument. He notably focuses on P.J.Harvey and John Parish’s “Taut,”(48) a song which, he argues, “draws the listener in to a close identification with the singer and protagonist, a rather direct confrontation with that potent mixture of infatuation and terrified submission that can turn people into victims.”(49) He demonstrates this most particularly through close attention to the song’s introduction, its combination of particular sounds, their predictability and their connotations, and the way these set up the lyric. In other words, the introduction indicates how we might experience the lyric, once it arrives. However, that said, I would argue that a subject position can become possible even without explicit understanding of the “content,” the lyric. Take the Tri Yann song “Je m’en vas,”(50) “I set off.” The melody to this track is borrowed from the old folk song “The leaving of Liverpool.” In the hands of revivalist groups like the Dubliners or the Spinners, this is a very ordinary song.(51) And Tri Yann’s version starts very much in that vein, as if performed by an unsophisticated musician. We are encouraged to hear it as an inconsequential, anachronistic, amateur performance. But the textural growth which maps the track’s narrative, the staged addition of tight rhythm section, overdriven guitar and prominent pipes, totally changes it for a Celticist audience, at least. This developing texture implies a sense of determination wholly missing from older performances, although both the original song, and Tri Yann’s, are about leaving on a long journey. Thus Tri Yann envelop us in the journey.
 However, it is not only the listener who can be disposed to take up a particular reading of a song. In some instances, the persona represented by the singer can be predisposed to take up a particular reading, with that predisposition being apparent to the listener. In this sort of case, the environment is partially described, as I have suggested already for Joe Cocker’s “With a little help from my friends,” as the singer’s confidante. It effectively provides advice to the persona inhabited by the singer, about how to act in response to the situation which is the subject of the song. A comparison of two tracks will demonstrate how subtle the effect of the accompaniment can be. Indeed, one obvious area of development for the theory I outline here would be a comparison of the range of different types of alternative versions (pastiches, copies, covers, unplugged performances etc.) which exist, observing the changing interplay between persona and environment.
 In her performance of the soul classic
“Stay with me (baby),” Lorraine
Ellison seems almost unconcerned for herself. She is simply pleading for the
continuance of a relationship. She needs to deliver the lyrics with such force,
though, that we might be moved to ask whether such a continuation is in her own
best interests. She points out in the song that she has always been around for “him,” she has proved her devotion, which she now desperately implores him to
reciprocate. The key moment occurs in the chorus, with the way the accompaniment
 What happens here, then, is that the environment provides us with
information which is over and above what we get not only from the lyrics, but
from the singer too. Indeed, it implies different readings of different
performances, confirmation that the signification of a song cannot lie in the
song, but must lie in the way the performance is received. Moreover, it is
information which the singer is probably only party to if she is sufficiently
aware (we can only guess whether the “Lorraine Ellison” persona will actually
heed the accompaniment’s advice). An analogy I have in mind here is that of the
everyday understanding of “body language.” When watching others speak, we can
tell how committed they are to what they are saying, we can tell whether to risk
trusting in them, by how they express themselves, how they accompany their words
with posture and gesture. Simon Frith has argued that even without visual clues,
we can discern whether to trust singers by listening to the non-singing sounds
they make: “Pop songs celebrate not the articulate but the inarticulate, and the
evaluation of pop singers depends not on words but on sounds—on the noises
around the words. In daily life, the most directly intense statements of feeling
involve just such noises: people gasp, moan, laugh, cry
 Philip Tagg, in his lengthy deconstruction of the Abba hit “Fernando,”(56) argues that certain features of that song—melodic intervals, instrumental setting—warn us not to trust the singer’s expression of commitment to the “revolutionary cause,” nostalgic reflection on which is the subject of the song. In such a case, where the persona and the environment are at odds, we are urged to trust the latter, in the same way that we would trust body language over direct speech. This example is by no means alone. Andy Stewart’s self-penned “Donald, where’s your troosers?”(57) is normally taken as a piece of nostalgic whimsy. The song sets up three positions, those of the Scottish yokel (who always wears a kilt), of polite (Edinburgh?) society, and (more hidden) of the (BBC) establishment of which Stewart was himself a part. Polite society is lampooned, and the yokel seems to come off best. However, the expressed distaste of rock’n’roll (the recording of the song itself contains a delightful pastiche of Elvis Presley singing it) is voiced on the part of the establishment. The song is harmonically accompanied by a “double tonic” (i.e. Aeolian i–VII) pattern, which sounds authentically “Scottish.” However, the end of each verse replaces the “expected” VII–i cadence with the V–i of the concert hall—a subtle, but nonetheless forceful, trouncing of the Scottish vernacular with that of high culture. The harmonic environment, then, if we can hear it, tells us not to take the lyrics at face value. In the Boomtown Rats’ “I never loved Eva Braun,”(58) the protagonist is more obviously culpable. He (who can only be identified as Adolf Hitler) brazenly declares that he wasn’t responsible for the results of his actions, but the breakdown and nonchalant whistling (of the sort you might have indulged in as a kid, hands behind your back, head in the air as your misdemeanor is discovered—Who? Me?)(59) which precede the final chorus (at c.3’15”) give the lie to that. And there are other types of example too. The position set up by the Sex Pistols’ “Holidays in the Sun”(60) is also not to be believed. Here is a song which purports to anarchism, a line totally belied by its formulaic rhythmic, harmonic, and articulative approach.(61) A final example is rather bizarre but, I think, particularly informative. I refer to a re-recording by ‘60s balladeers the Bachelors of the Hollies’ 1969 hit “He ain’t heavy.”(62) This is a song which wears it’s heart on its sleeve, inspired by a line from the 1938 film Boys Town, symbolizing comradeship among vagrant boys, and supposedly based on reality. It is a difficult song to bring off, requiring grandeur without pomposity, compassion without sentimentality. This Bachelors’ recent version aims for a big ending, presumably demonstrating the care the singer promises, and his disdain for the cost entailed. In order to bring it off, however, the drummer hopelessly miscalculates his cadential flurry in a rhetorical display empty of content. There may be a risk of overstating the case out of context, so a comparison is in order. In Grandaddy’s “Hewlett’s Daughter,” we can hear the same sort of over-blown, contextually inappropriate drum intervention. However, because this occurs at the end of the bridge between verses, “out of the blue,” rather than at the end of the song, its inappropriateness seems intentional. It seems to be in the service of the anger that is just below the surface of the singer’s declaration that he “should have been your son.”(63) There is no equivalent inappropriateness of position within the form of the song in the Bachelors’ case.
 So, we can declare that the environment sometimes suggests to the listener to occupy a subject position which disbelieves the persona, reminding us that all may not be as it seems. This I relate to one part of a schema put forward recently by Nicholas Cook, who identifies what he calls modes of media pairing(64)—the relationships obtaining between separate media in an artwork. His theoretical model is provided by the Lakoff and Johnson discussion of metaphors and, since metaphor is manifested within a single medium, i.e. within the utterance, it seems reasonable to use his terminology for a single medium (if we insist that persona and environment, notwithstanding the recourse of the former to words, actually constitute a single medium). Cook argues that two concurrent media can inhabit one of three relationships: conformity, complementation, and contest. In the recent examples, above, the views given by the two textural strands (persona/melody and environment/accompaniment) contest each other, and in such cases I think we are always more likely to trust the latter. Perhaps because the accompaniment is wordless and therefore not possibly subject to Frith’s “silver tongue,” the analogy with “body language” seems to me quite fruitful.
 Hitherto, my attention has been focused on the way that the environment may modify our conception of the persona. The environment has always been secondary. What, though, if the two appear more equal? After all, we are all perceived by others as actors within an environment, as well as perceiving ourselves as acting upon it. The second of Cook’s terms describes the situation where the two elements are in an equal, complementary, relationship. The verb to “accompany” implies an hierarchical relationship; it suggests that what is accompanied has a greater role in the identification of what is going on, in the sense of giving identity to it. In complementation, though, this hierarchical relationship comes into question, potentially resulting in an ecologically more sound relationship. The hierarchy may survive, if only because of the distinction between lyric and not-lyric, but it may be somewhat tenuous. In the verse of Jethro Tull’s “Fylingdale flyer,”(65) the rhythmic profile of the accompaniment is unitary, and constantly rubs against that of the parallel multi-tracked voice. The relationship here is complementary, akin to that of a couple ballroom dancing, where one (the melody) leads, but leads only because of the presence of the lyrics [Example 5].
 I come now to two final instances where, again, the simple identification of accompaniment is problematized. In the Jethro Tull example, although there are two identifiable strands, the relationship between them is complementary. Within this broad repertoire of song, can we always posit two strands, can we always distinguish between a persona and its environment (the latter identified as accompanimental)? I think so, but sometimes the voice is far less prominent than we might normally expect. This can happen in terms of production values, as in much of the Rolling Stones’ early material, where what is presented to us is not so much an accompanied lead singer, but almost a communal effort, one facet of which happens to be vocal.(66) Even here, though, I think the voice perceptually dominates simply because of the presence of the verbal strand. A very good test of this position is provided by the very end of King Crimson’s “Larks’ Tongues in Aspic, Part One”(67) in which instrumental interplay covers what appears to be a heated argument. This is in the distance, and the words cannot be made out, and yet by the fact that they are words, they call for our attention. A more common case occurs where a vocal line gets taken over by part of the accompaniment. This is common in examples of “call and response” textures as they develop from African-American genres.(68) It also happens in mainstream rock and pop, as in Wishbone Ash’s “Throw down the sword,”(69) where a bifurcated melody leads into a two-part guitar line. Indeed, perhaps this is potentially what always happens when a solo enters: if we read this solo as a continuation of the vocal melody, at the point of interchange, the expression can be seen as becoming more eloquent, as it moves beyond words. The Carpenters’ “Goodbye to love”(70) exemplifies this more overtly, as the guitar solo actually picks up Karen Carpenter’s phrasing and literally takes her expression up to new heights [See Example 6]. Notwithstanding the fact that Carpenters fans disliked the intrusion of the guitar as stylistically foreign,(71) this reading is certainly available to non-fans. There seems to me a clear difference between this sort of case, where a melody line moves to an instrument for a solo, and the sharing of material between two forces, simply because in this former case, the melody is firmly identified with the persona before the instrument ever gets near it.
 Finally, an obvious question which I have left till last. Is it right to call all the non-melodic parts of a texture “accompaniment”? In other words, is Tagg’s assumption of a melody/accompaniment dualism sufficient? This is actually too large a (new) question for me to more than broach here, but some coverage is necessary. As an assumption, it appears a reasonable one, but there is certainly one significant repertoire where it falls down. This repertoire is formed by instances where, along with a band, the persona of the singer self-accompanies him- or herself on a guitar (self-accompaniment on a keyboard is a different issue, I believe). On Lindisfarne’s “Good to be here,”(72) we simply hear singer Alan Hull, an acoustic guitar (which Hull plays), and an orchestra. The orchestra forms an accompanimental environment which really only supports the persona from the outset, reacting expressively at the “right” points. The guitar is sounded in conjunction with the orchestra, such that its role seems to be accompanimental. However, at the moment at which the singer becomes no longer separated from what he’s describing, the moment when he wakes from what we come to realize is a dream of a ghostly nature, when he comes to describe his wife addressing him in bed, the guitar drops from the texture. It is away only momentarily, but its absence clearly marks that moment of change. If only at this point, the guitar function is clearly separate from that of the remainder of the accompaniment, which does not mark this moment. This usage develops from the singer/songwriter genre, and was taken up by heavy rock in the late 1960s/early 1970s where the guitar, when pitted against the rest of the accompaniment, represents the individual battling against some sort of societal norm, or a body of undifferentiated “others.”(73) As this example demonstrates, in this repertoire, the guitar might more usefully be read as part of the persona.
 So, it seems to me that the function of accompaniments is far from unitary, as their textures, harmonies and forms create environments within which a persona acts. I have isolated seven stages on a continuum, from simple pitch/meter orientation, through genre-setting and tone-setting, to support, amplification and explanation of the persona’s situation, and finally to contradiction, by analogy with body language. Along this continuum, the accompaniment represents an environment which is at first simply inert, in later stages is quiescent, becomes active, and finally oppositional. I have then identified three problematic issues in the simple assumption of a melody/accompaniment dualism: that of complementation; that of the wordless melody; and the non-unitary nature of some accompaniments. And I do not pretend that this list is exhaustive, or even that these are the only points to mark on the continuum. What I hope I have demonstrated, however, is that not only is a reduction of the operative sphere of meaning of a song to that of its lyrics inadequate, but that the same goes for a reduction of such a sphere to that of its lyrics and how they are sung. Even without consideration of production manipulation, concentration on that very rich amalgam which constitutes the track’s personic environment is absolutely crucial if we are to fully experience the expressive richness of popular song.
Allan F. Moore
1. My thanks to respondents at the 7th Finnish Musicology
Congress in Helsinki, 2004, where this was first presented as a keynote paper,
to seminar students at the University of Manchester (February 2005) who helped
me focus its trajectory, and to Anwar Ibrahim for comments made on an early
4. Philip Tagg, Introductory notes to the semiotics of music
(1999), http://www.tagg.org/xpdfs/semiotug.pdf, 38–9, last accessed 19/08/05. Tagg actually writes about
“music” rather than
“song,” but he cites as one identifying characteristic of the “melody” pole of
the dualism that it tends to be “singable.”
5. California University Press, 1982. Cone’s writings on
“persona” are beginning to be taken up, at last, by popular music theorists,
e.g. Matthew Gelbart, “Persona and Voice in the Kinks’ Songs of the Late 1960s,”
Journal of the Royal
Musical Association 128 (2003), 200–41.
7. Eero Tarasti uses both “subject” and “actor” in his semiotic scheme, tracing
the rise of the former in the late eighteenth-century development of periodized
melody in instrumental music. Eero Tarasti, A Theory of Musical Semiotics
(Indiana University Press, 1994), 104.
8. It is of the nature of my discussion that, for many
instances, any sort of representation on the page of the relationship between
the sounds I am discussing is inadequate—there is no substitute for hearing
them. The footnotes therefore contain complete references for my examples; while
some are comparatively superfluous, a sufficient number should be easily
15. Philip Tagg’s semiotic approach identifies as one
feature the genre synecdoche, in which an instrument from a “borrowed” genre
brings with it expectations of the connotations of its parent genre. That may
well be thought to operate here also.
19. Some styles have a greater degree of stereotypicality than others: Motown;
old skool hip-hop; early 1970s Status Quo exemplify this at various levels
(those of label roster; sub-culturally delimited style; idiolect).
24. Joe Cocker, “With a little help from my friends”
(1969), With a little help
from my friends, A&M 1999 (1969) [490 419-2]. I discuss his singing of this
song in greater detail in “The contradictory aesthetics of Woodstock” in Andrew
Bennett (ed.), Remembering Woodstock (Ashgate, 2004), 82. The original
Beatles track from 1967 is found on The Beatles: Sgt. Pepper’s Lonely Hearts
Club Band, EMI 1987 [CDP7 46442-2]. I discuss this further in The Beatles:
Sgt. Pepper’s Lonely Hearts Club Band (Cambridge University Press, 1997), 30–32.
43. See Alf Björnberg, “On Aeolian Harmony in Contemporary Popular Music,”
APSM-Norden, 1989 and Allan F. Moore, “Patterns of harmony,” Popular Music
11/1 (1992), 73–106. Björnberg cites alienation, and fear of it, as a key
associative quality of the Aeolian mode in rock.
52. Lorraine Ellison, “Stay with me baby” (1966),
The best sixties love
album...ever!, Universal n.d. [VTDCD235]. Bette Midler, “Stay with me baby,”
Divine madness, Atlantic 1980 [16022-2].
71. Kevin J. Holm-Hudson, “Your Guitar, It Sounds So Sweet
and Clear: Semiosis in Two Versions of ‘Superstar’,” Music Theory Online
last accessed on 14/1/05.
73. See Rock: the Primary Text, 113. Other examples include the Strawbs,
“New world” (1972), Van der Graaf, “Lemmings” (1971), King Crimson, “Epitaph”
(1969), and the Beatles, “A day in the life” (1967).
My thanks to respondents at the 7th Finnish Musicology Congress in Helsinki, 2004, where this was first presented as a keynote paper, to seminar students at the University of Manchester (February 2005) who helped me focus its trajectory, and to Anwar Ibrahim for comments made on an early version.
Simon Frith, “Why do songs have words?” in Music for Pleasure (Cambridge: Polity, 1988).
As paraphrased in Richard Middleton, Studying Popular Music (Buckingham: Open University Press, 1990), 28.
Philip Tagg, Introductory notes to the semiotics of music (1999), http://www.tagg.org/xpdfs/semiotug.pdf, 38–9, last accessed 19/08/05. Tagg actually writes about “music” rather than “song,” but he cites as one identifying characteristic of the “melody” pole of the dualism that it tends to be “singable.”
California University Press, 1982. Cone’s writings on “persona” are beginning to be taken up, at last, by popular music theorists, e.g. Matthew Gelbart, “Persona and Voice in the Kinks’ Songs of the Late 1960s,” Journal of the Royal Musical Association 128 (2003), 200–41.
Particularly Eric Clarke, “Subject-Position and the Specification of Invariants in music by Frank Zappa and P.J.Harvey,” Music Analysis 18 no. 3 (1999), 347–74.
Eero Tarasti uses both “subject” and “actor” in his semiotic scheme, tracing the rise of the former in the late eighteenth-century development of periodized melody in instrumental music. Eero Tarasti, A Theory of Musical Semiotics (Indiana University Press, 1994), 104.
It is of the nature of my discussion that, for many instances, any sort of representation on the page of the relationship between the sounds I am discussing is inadequate—there is no substitute for hearing them. The footnotes therefore contain complete references for my examples; while some are comparatively superfluous, a sufficient number should be easily traceable to
illuminate my argument. Many are readily available on such web-sites as http://www.allofmp3.com.
Sleevenote to Beaver and Krause, Gandharva, Warner 1971 [9 45663-2].
See Allan F. Moore, “Categorical conventions in music-discourse: style and genre,” Music and Letters 82/3 (2001), 432–42.
Patsy Cline, “I’m blue again” (1959), The one and only Patsy Cline, K-Tel 1994 [ECD3086].
Analogous textural changes are found, for example, in “Today, Tomorrow and Forever” and “Never No More,” to be found on the same album.
Martin Carthy, “The dominion of the sword,” Right of passage, Topic 1988 [TSCD452].
Bob Dylan, “Just like a woman,” Blonde on Blonde, CBS 1972 [CDCBS22130].
Philip Tagg’s semiotic approach identifies as one feature the genre synecdoche, in which an instrument from a “borrowed” genre brings with it expectations of the connotations of its parent genre. That may well be thought to operate here also. http://www.tagg.org/xpdfs/semiotug.pdf.
Led Zeppelin, “Thank you” (1969), Led Zeppelin II, Atlantic 1994 (1969) [7567-82663-2].
Roy Harper, “When an old cricketer leaves the crease” (1975), HQ, Science Friction 1995 [HUCD019].
This might thus be posited as another example of Tagg’s category of genre synecdoche.
Some styles have a greater degree of stereotypicality than others: Motown; old skool hip-hop; early 1970s Status Quo exemplify this at various levels (those of label roster; sub-culturally delimited style; idiolect).
My thanks to Laura Tunbridge for requiring me to clarify this point.
Laïs, “’T smidje,” Laïs, Alea 1998 [WBM21005].
Fairport Convention, “Medley: The lark in the morning,” Liege and lief, Island 1969 [IMCD60].
Iggy Pop, “Whatever,” Skull Ring, Virgin 2003 [B0000D9YE7].
Joe Cocker, “With a little help from my friends” (1969), With a little help from my friends, A&M 1999 (1969) [490 419-2]. I discuss his singing of this song in greater detail in “The contradictory aesthetics of Woodstock” in Andrew Bennett (ed.), Remembering Woodstock (Ashgate, 2004), 82. The original Beatles track from 1967 is found on The Beatles: Sgt. Pepper’s Lonely Hearts Club Band, EMI 1987 [CDP7 46442-2]. I discuss this further in The Beatles: Sgt. Pepper’s Lonely Hearts Club Band (Cambridge University Press, 1997), 30–32.
John Lennon, “Imagine” (1971), Legend, Parlophone 1997 [7243 8 21954 2 9].
This is developed in Keith Negus, Popular Music in Theory (Polity, 1996), 102–6.
For instance, Nat King Cole, “When I fall in love” (1957), Let’s fall in love, EMI 1990 [7243 4 93283 2 7].
Yes, “Heart of the sunrise,” Fragile, Atlantic 1972 [7657-82667-2].
Janis Joplin: Anthology, CBS, 1980 [COL 467405 9]. See the discussion in Allan F. Moore, Rock: the Primary Text (Ashgate, 2001 edition), 96.
Sonny and Cher, “I got you babe” (1965), The best sixties love album...ever!, Universal n.d. [VTDCD235].
Diana Ross, “Ain’t no mountain high enough” (1970), One woman, EMI 1993 [7243 8 27702 2 0].
Cilla Black, “Alfie” (1965), Burt Bacharach and Hal David Songbook, Connoisseur 1988 [VSOP CD 128].
Björk, “Army of me,” Post, One Little Indian, 1995 [TPLP51CD].
See Allan F. Moore, “Analizzare il rock: strumenti e finalità,” Music realtà 62 (July 2000), 95–118.
Beach Boys, “God only knows” (1966), Pet Sounds, Capitol 1990 [CD FA 3298].
Whippersnapper, “John Gaudie,” Promises, WPS 1985 [WPSCD001].
Traffic, “Hole in my shoe” (1968), The collection, Spectrum 2001 [544 558 2].
See Sheila Whiteley, The space between the notes (Routledge, 1992).
Slade, “Coz I luv you” (1971), Greatest Hits, 1999 [537 105-2]. Jimi Hendrix, “Hey Joe” (1966), The best of Jimi Hendrix, MCA 2000 [112 383-2]. See also Rock: the Primary Text, 56.
Richard Thompson, “The way that it shows,” Mirror blue, Capitol 1994 [0777 7 81492 2 4].
Beach Boys, “Good vibrations” (1966), Beach Boys Greatest Hits, EMI 1998 [7243 4 95696 2 1].
Deep Purple, “Child in Time” (1970), Deep Purple in Rock, EMI 1995 [7243 8 34019 2 5].
See Alf Björnberg, “On Aeolian Harmony in Contemporary Popular Music,” APSM-Norden, 1989 and Allan F. Moore, “Patterns of harmony,” Popular Music 11/1 (1992), 73–106. Björnberg cites alienation, and fear of it, as a key associative quality of the Aeolian mode in rock.
Coverdale Page, “Take me for a little while,” Coverdale Page, EMI 1993 [CDEMD1041].
The Darkness, “Holding my own,” Permission to Land, Atlantic 2003 [5050466-7452-2-4].
Rolling Stones, “Satisfaction” (1965), Hot Rocks 1964–1971, Abcko 1986 [844 475-2].
Vanessa Carlton, “Ordinary Day,” Be not nobody, A&M 2002 [493 367-2]. Note that the album contains a cover of “Paint it black.”
Polly Jean Harvey & John Parish, “Taut,” Dance Hall at Louse Point, Island 1996 [CID 8051/524 278-2].
Clarke, “Subject-Position,” 371.
Tri Yann, “Je m’en vas,” Le Pélegrin, Marzelle 2001 [EPC 501632 2].
e.g. Dubliners, “The leaving of Liverpool,” Best of the Dubliners, Kaz 1996 [PDS CD 535].
Lorraine Ellison, “Stay with me baby” (1966), The best sixties love album...ever!, Universal n.d. [VTDCD235]. Bette Midler, “Stay with me baby,” Divine madness, Atlantic 1980 [16022-2].
Simon Frith, Sound effects (London: Constable, 1983), 35.
Beach Boys, “Surfin’ USA” (1961), The Beach Boys Greatest Hits, EMI 1998 [7243 4 95696 2 1].
Chuck Berry, “Sweet little sixteen” (1958), The Best of Chuck Berry, Music Club 1991 [MCCD 019]. Many other examples of the same process exist.
Philip Tagg: Fernando the Flute, IPM, 1991.
Andy Stewart, “Donald, where’s your troosers?” (1960), 20 Scottish Favourites, EMI 2001 [7243 5 32965 2 0].
Boomtown Rats, “I never loved Eva Braun” (1978), Tonic for the troops, Phonogram 1992 [514053-2].
Peter Gabriel’s “Intruder” provides a similar example of this device. Peter Gabriel III, Virgin 1980 [PGCD3].
Sex Pistols, “Holidays in the Sun,” Never mind the bollocks, here’s the Sex Pistols, Virgin 1977 [CDVX2086].
See Rock: the primary text, 130–1.
Bachelors, “He ain’t heavy,” The very best of the Bachelors, Prism 1998 [PLATCD293].
Grandaddy, “Hewlett’s daughter,” The sophtware slump, V2 2000 [WR1012252].
Nicholas Cook, Analysing Musical Multimedia (Oxford: Clarendon, 1998), 99.
Jethro Tull, “Fylingdale Flyer,” A, Chrysalis 1980 [CDP 32 1301 2].
E.g. “Sympathy for the devil,” Beggar’s Banquet, ABKCO 2002 (1968) .
King Crimson, “Larks’ Tongues in Aspic, Part One,” Larks’ Tongues in Aspic, Virgin 1970 [EGCD 7].
The chorus to Chuck Berry’s “Johnny B. Goode” (1958), for example. The Best of Chuck Berry, op.cit.
Wishbone Ash, “Throw down the sword” (1972), Argus, MCA 2002 [088 112 816-2].
Carpenters, “Goodbye to love” (1972), A song for you, A&M 1990 [CARCD4].
Kevin J. Holm-Hudson, “Your Guitar, It Sounds So Sweet and Clear: Semiosis in Two Versions of ‘Superstar’,” Music Theory Online http://www.mtosmt.org/issues/mto.02.8.4/mto.02.8.4.holm-hudson_frames.html, last accessed on 14/1/05.
Lindisfarne, “Good to be here” (1979), The news, Castle 1999 [ESMCD812].
See Rock: the Primary Text, 113. Other examples include the Strawbs, “New world” (1972), Van der Graaf, “Lemmings” (1971), King Crimson, “Epitaph” (1969), and the Beatles, “A day in the life” (1967).
Copyright © 2005 by the Society for Music Theory. All rights reserved.
 Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.
 Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:
 Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.
This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.