Structural Relationships of Music and Images in Music Video

Author(s): Alf Björnberg

Source: Popular Music, Vol. 13, No. 1 (Jan., 1994), pp. 51-74
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/852900 .
Accessed: 30/06/2011 19:29

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=cup. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to Popular
Music.

http://www.jstor.org
PopularMusic (1994) Volume 13/1.Copyright? 1994 Cambridge UniversityPress

Structuralrelationshipsof music
and images in music video1
ALF BJORNBERG

Introduction
In the course of the last decade, the body of writingon music video has grown to
sizeable proportions.The reason forthe presentaddition to this bulk of literature,
in spite of the subject seemingly approaching the state of exhaustion, is that
musical semioticsare stillrarelyapplied to the field. It is a factthatpop and rock
music have always been heavily infused with socially determinedmeaning such
thatan autonomous musical aestheticsappears clearlyinsufficient to explain their
significance; however, to what extent and how this significance linked in with
is
particular musical structures as such is still largely uninvestigated.In my view,
music video may perhaps be less interestingas a phenomenon in itselfthan as
source material for an 'empirical semiotics' of popular music, shedding light on
significationprocesses of a more general applicability.Furthermore,the distinctive
featuresof music video may arguablybe betterexplained on the basis of an under-
standingofthe syntacticalcharacteristics ofpopular music than by prevalenttheor-
ies of postmodernism;the latterappear problematicnot only due to theirspeculat-
ive and unsubstantiatednaturewith regardto media receptionprocesses (cf. Frith
and Horne 1987, p. 11), but theirexplanatoryvalue as regards syntacticfeatures
of music video also seems to be limited(cf. Frith1988, p. 207).
In an earlierwork, I have presented a theoreticaldiscussion of the character-
istics of popular music syntax and their consequences for the analysis of music
video, as well as a general typologyof relationshipsbetween music and visuals
(see Bjornberg1992a). The purpose ofthe presentpaper is to modifyand substanti-
ate in more detail this previous account, by way of an analysis of structuralrela-
tionshipsof music and images in a numberof music videos. This analysis is aimed
at demonstratingboth the range of possible differenttypes of such relationships,
the limitationsimposed upon themby musical syntax,and theirrelative'openness'
to varying kinds of reception and interpretation,depending on the mode of
listening/viewing applied.

Postmodern society, postmodern media, postmodern audiences?

The specificcharacteristicsof (the visual dimension of) music video attractingthe
attentionof writersand scholars may be summarised as the breakdown of linear
of causal logic, and of temporaland spatial coherence. The apparently
narrativity,
widely accepted mainstream line of explanation of these characteristictraitsis
based on the argumentof postmodernism: the 'postmodern condition', pervading
51
52 AlfBjirnberg
all of contemporaryindustrialisedsociety,findsits most adequate representation
in the fragmentary formsof music video, combiningpracticesfromClassical high
art, avant-gardemodernismand popular culture (see, forinstance, Aufderheide
1986; Kaplan 1987; Strom 1989). The evaluation of the consequences of this
situation varies among authors; whereas, for instance, Fiske (1986) regards the
'refusingof sense' as a practicepotentiallyliberatingfrombourgeois hegemony
of meaning, Tetzlaff(1986) is more pessimisticas to the existence of any such
emancipatorypotential. While postmodernismsometimesapproaches the vague,
all-embracingstatus of a Zeitgeist,many scholars relate it, and the characteristics
of music video, to the development of the specificmedia codes of commercial
television (Berland 1986; Jones 1988; Kaplan 1987; Larsen 1987), and especially,
stressingthe advertisingfunctionof music video, to the aesthetics of television
advertising(Allan 1990; Frith1988; Goodwin 1987; Kinder 1984; Laing 1985; Movin
and Oberg 1990; Strom1989 and others). The loss, in thisprocess, of rockmusic's
presupposed 'authenticity'of expression is also commented upon (Grossberg
1988); this is described both in negative terms (Movin and Oberg 1990), and
approvingly,as offeringthe means fora 'celebrationof artifice'(Ihlemann 1992).
The explanations forthe emergence of music video in its specificformalso
include referencesto the significanceof the developmentof technology, both televi-
sion (cable) and video production/post-production technology, as well as that of
musical productionand reproduction(Berland 1986; Goodwin 1987; Ruud 1988).
Ruud (ibid.) also discusses the connectionswith video art, emphasising the aes-
theticqualityof ambiguityinherentin the non-narrativity of music video (see also
Strom 1989).
Several explanationsfornon-narrativity fromthe point of view of the audience
have also been offered.One line of argument apparently lying close at hand
involves the view of the specificitiesof music video as a consequence of a general
tendency,determinedby social and media-technologicaldevelopments, towards
a Benjaminian 'distractedmode of perception',requiringa structureallowing the
user to quickly catch the point of the message (Jones 1988; Larsen 1987; Movin
and Oberg 1990); at times this view is extended into postmodernistspeculation
on the 'schizophrenic subject' (Jensen 1988). Another common explanation
(presupposing the user's devotion of a more continuousattention)emphasises the
functionsof fantasy,escape fromeverydayrealityand imaginaryproblemsolution
provided by the incoherentand fragmentary structuresof music video (Allan 1990;
Aufderheide1986; Morse 1986). The parallelismsbetween these structuresand the
workingsof the primaryprocesses of the psyche have been pointed out (Brown
and Fiske 1987; Ruud 1988), as well as the potentialof fantasyforexploitationfor
commercialpurposes (Kinder 1984). In this contextconnections have also been
made with the need-structuresproduced by the new, 'narcissistic'socialisation
patternssuggested by German socialisationtheory(Forsman 1986; Larsen 1987).

Music: the missing connection

Few of the theoriesrelated above can be totallydismissed as contributionsto the
understandingof the characteristicsof music video. Nevertheless,in spite of the
significantrole of musicin this context(if nothingelse, as the alleged justification
forthe entirephenomenon), most of the authorscitedhave remarkablylittleto say
about it. The opinion has been proposed that,forthe user, the music is somehow
 Music and imagesin musicvideo 53

'dominated' by the visuals (Berland 1986; Kinder 1984; Tetzlaff1986); this, how-
ever, seems to have littlebearing on mattersother than the individual mode of
perceptionand musical preferencesof these scholars. Others point out thathistor-
ically, popular music performancehas always been an audiovisual phenomenon,
combiningmusical sound and visuals (Berland 1986; Goodwin 1987; Laing 1985),
and several authors state thatmusic is primaryin relationto the visuals, although
withoutelaboratingmuch furtheron the natureand consequences of this primacy
(Goodwin 1987; Laing 1985; Strom 1989). Kaplan seems to regard music video
visuals as mainly based on song lyrics(1987, pp. 47f.), restrictingher discussion
of musical characteristicsto littlemore than one half page (1987, pp. 123f.).2
In several instances writerson music video have given indications of the
positionregardingthe conditionsforand/oroperationofmusicalsignification under-
lying their respective treatment of the subject. Ruud (1988) regards signification
in music as mainlydeterminedby social contextand conventionalised.Movin and
Oberg, followingAdorno, characterisemusic as non-referential and experienced
by way of 'emotional empathy' (1990, p. 127), while Brown and Fiske state that
'Rock videos, like rock music, work primarilyon physical sensations and produce
a physicalpleasure thatopposes the common sense of linearnarrativein dominant
ideology' (1987, p. 62). Morse, in her discussion of music video functioningas 'a
formof magical thinking';(1986, p. 24), touches upon aspects of music as per-
formed(the star-singer'creatinga visual world' by virtueof his or her privileged
position in the musical as well as visual mode of communication)but does not
relate this to other musical dimensions. Whereas these various positions indicate
relevantaspects, theyare hardlyexhaustive on the matterof musical signification
in music video, and in most cases seem not to inspire the undertakingof more
detailed analyses.
Allan, in a discussion of correspondences between musical filmand music
video, identifiesthe functionsof music as 'those of grounding,unityand resolu-
tion' (1990, p. 9), while Kinder (1984) states thatthe continuous flow of the music
imposes unity on the video. These are among the rare instances of music video
scholars approaching the significance of structuralaspects of music; another
example is Berland's (1986) somewhat impressionisticaccount of the relationship
between formalsong structureand visuals. Unfortunately,the embryo of struc-
tural analysis inherentin the latterapproach is not followed up, probably due
both to the author's Adornian view of popular music structuresand to the visual
dominationpostulated (cf. above).
On a theoreticallevel, Frith's(1988) discussion of structuralcharacteristicsof
music video amounts to a well-considered counter-argument,based on musical
facts,against postmoderntheoryspeculation. Frithpoints to the general structur-
ing principlesof movementas 'the metaphorforsound' (1988, p. 216) and montage
as representationof rock's musical experientialqualities, and also discusses the
relationshipsbetween musical and visual repetitivity.3 Although the arguments
are coloured by the author's obvious dislike formusic video in its currentlydomin-
ant form,and may be criticisedon some points (cf. Bjornberg 1992a, pp. 382,
386), this articleremains an importantcorrectiveforwriterson the subject. Frith's
argumentshave also been picked up by Nielsen, statingthatmusic video images
are subordinated to the music by way of 'a rhythmicallydetermined pleasure
principle' (1991, p. 299),4and Forsman, who adds to the montage/beathomology
one between sound and an 'associative spatiality'(1991, p. 9).
54 AlfBjdrnberg

However, despite these theoreticalcontributions,detailed analyses of music

videos relatingvisuals to musical structurehave been very sparsely presented.
Ruud (1988) performslengthyanalyses of threevideos; however, only one of these
(Paul Simon's 'Rend and GeorgetteMagrittewith theirdog afterthe war') deals
in greaterdetail with musical (mainly tonal) structure,thus emphasising unique
individualfeaturesratherthan generalstructuralprinciples,and givingan unfortu-
nate bias to the study. Larsen (1987), basing his discussion on Greimas's narrative
theory,analyses the relationshipbetween musical and visual syntaxin Phil Collins'
'Against All Odds'. The author convincinglyrelates visual structureto musical
segmentationand intensity(dynamics and texture)processes. The implied seeds
of a more general application of narrativetheoryto popular music are, however,
not followed up; in a later work (Larsen 1988), the author restrictshis discussion
to Westernart music and its derivativesin Hollywood filmmusic, by implication
dismissing contemporarypopular music as generallynon-narrativeand not sus-
ceptibleto this kind of analysis.5

Popular music formand narrativity

In an attemptto remedysome of the shortcomingsof the literatureon music video
recapitulatedabove, the analyses to be undertakenhere will take as theirtheoret-
ical point of departure the issue of thepotentialforsignification in popularmusic
syntax.That is, I am concerned with 'primary'ratherthan 'secondary' musical
signification(Middleton 1990, pp. 220ff.),and with the significationof syntactical
processes,as opposed to particular,individual musematicmeaning (ibid., p. 235).
I would also like to make a distinctionin favourof the structural ratherthan the
(in a psychoanalytical sense 'more primary', i.e. related to primaryprocesses of
the psyche) 'physical sensation', para-structural significationdimension of music;
although, as Middleton (ibid., p. 219) pointsout, these cannotbe clearlyseparated,
the latteraspect of musical communicationappears as yetto have attractedconsid-
erablymore attentionfromwriterson music video than the former.
Differently phrased, this is a matterof the potentialfornarrativity in music.
In this connection,Middleton suggests a general distinctionbetween threemodes
of constructionof musical syntax:
In contrast to thenarrative
category, whichprivilegesdifference,thereis whatwe can call
an 'epic'mode,wherethefocusis on repetition and variedrepetition;and inbetweencomes
a 'lyrical'
category(marked by symmetrical open/closedand binary (Middleton
structures).
1990,p. 216;myitalics)

The positing of 'narrative'and 'epic' as polar opposites may appear some-

what confusing against the background of the everyday use of these terms in
relationto literatureand film;takingup the geometricalmetaphorsalready hinted
at by Middleton, a betterterminologywould perhaps be using 'linear' instead of
'narrative','circular' instead of 'epic', and maybe 'elliptical' (in the strictlygeo-
metricalsense) for 'lyrical',implyingthe dual aspects of movement and return
inherentin the lattersyntacticalmode. Still,the categorisationappears valid and
potentiallyfruitfulforthe analysis of popular music syntax,and although these
termsin this usage may primarilybe comprehended as ratherabstractstructural
analogies, they nevertheless also arguably indicate experiential correlates to
models of musical syntax,thatis, to musical form,in a general sense. The formal
 Music and imagesin musicvideo 55

structuresof popular music are usually conceived of and described in terms of

standardised, 'neutral' structures,having achieved 'a naturalness' (Berland 1986,
p. 44) to the listener;still,this does not rendertheirintrinsicpotentialforsignifica-
tion non-existent.6
Middleton links the narrativemode of musical syntaxconstructionwith the
absence of repetition,with the extremevariationof the 'infiniteset', in which no
musical elementis repeated, the music continuouslymoving forwardlinearlyand
teleologicallyin a mannerhomologous to the structuresof (verbal/visual)narrative.
In order to investigate furtherthe existence of homologies between music and
narrative,structuralistnarrativetheoryappears to offersome useful concepts.7A
basic distinctionmade withinthisbody oftheoryis thatbetween story,the signified
or contentof narrative,and discourse,the signifieror structureof the narrativetext
(Chatman 1978, p. 19; cf. Barthes 1977, p. 87; Genette 1980, p. 27). While the
semanticprecisionof musical narrativeas to its content,is 'story',is low compared
to verbal or visual narration,musical structuremay be demonstratedto be capable
of exhibitingseveral of the characteristicsconstitutiveof narrativediscourse. In a
somewhat loose but helpfulphrasing,Chatman (1978, p. 25) identifiesthe distinct-
ive featuresof narrativeas 'eventhood, characterhood,and settinghood'. These
concepts appear adequately applicable to musical discourse. A basic featureof
Western music since the Renaissance, and particularlyof popular music, is the
structuraldualism of melody and accompaniment(Mar6thy1974,p. 22; Tagg 1979,
pp. 123f.). As Tagg indicates, this dualism is generallyconceived in terms of a
relationshipbetween figureand background,or between individual/character and
environment/setting. These are frequentlyconsistentlyidentifiablewith separate
musical 'voices'; however, the structuralspecificitiesand polysemic nature of
music also allow forthe same 'voice' to alternatebetween various functions,such
as a guitarfirstplaying a riffas part of the background, then switchingto fillin
the vocal line as a 'secondary character',and subsequently becoming the 'main
character'in a guitarsolo. As forthe qualityof 'eventhood', musical structuremay
appropriatelybe describedas a succession (and/orsimultaneity)of events, ranging
from instantaneous changes to long-termprocesses. From this point of view,
'musical form'is defined by the particularnature of each musical event, by the
temporal density of events and by their distributionthroughoutthe duration of
the piece in question.
For the purpose of investigatingthe degreeofnarrativity of particularmusical
structuresa furtherdistinctionmade by Chatman may be useful. Narrativedis-
course is characterisedas being constructedfromstatements, which may be divided
intostasis('is') statementsand process('does') statements(Chatman 1978,pp. 31f.);
a parallel distinctionis the one made by Greimas between radotageand affabulation
(cf. Larsen 1987, p. 89). Narrativity,in the common-sensemeaning of the term,is
linked to the quantityand densityof process statements:a narrativewhere 'noth-
ing happens' clearlyhas a 'non-narrative'character.Adequate musical analogies
to this categorisationexist both in Ruwet's (1987, p. 16) differentiation between
parametric and non-parametric elements in music, and in Middleton's (1990, p. 215)
distinctionbetween binary/digital and analogueselection.Parametricelements are
characterisedby being constantthroughouta piece of music or switchingbetween
the two poles of an opposition, i.e. being subject to binary alternation,while
non-parametricelements are characterisedby 'a fairlylarge number of internal
distinctionsof the same dimensions', i.e. they tend towards analogue selection.
56 AlfBjirnberg
On the basis ofthe analogy between 'process statement'and 'non-parametricness',
narrativity in music may thus be seen as relatedto the quantitativeand qualitative
significanceof non-parametricorganisation/analogueselection,in agreementwith
the argumentof Middleton related above.
Although,as Ruwet points out, it is impossible to determinea prioriwhether
a particularmusical dimension has a parametricor non-parametricorganisation,
in Western music in general dimensions such as tempo, mode, instrumentation
and timbretend to be parametric,while melodic pitch, surfacerhythmand har-
monic tension are more or less non-parametric.Narrative qualities in Western
music have thus primarilybeen associated with the dimensions of melody and
harmony:melodic processes of apertureand closure combine with harmonicpro-
cesses of tension and release, formingpotentiallylong-ranging,forward-directed
musical structures.Illustrativeexamples of thismay be foundin 'extremelynarrat-
ive' music, such as, forinstance, Wagner's operas; in popular music, however, a
high degree of such narrativityis seldom the case. Generally,popular music is
characterisedby strophicdisposition,i.e. repetitionof a small number of well-
demarcated sections, and symmetrical construction,i.e. larger units being con-
structedby binarycombinationof smallerunits (cf. Bjornberg1992b, p. 4). These
characteristicsimply a predominance of the lyricalmode: the use of symmetrical
structures,whose well-roundednessworks against directionallinearity,is distinct-
ive of this mode, and strophicdisposition involves processes of repeated return
to an experientialfocal point or 'point of departure', processes whose reflective
charactercontrastsboth with the linearityof the narrativemode and with the
short-term repetitivityof the epic category.This lyricalityis not, however, homo-
geneously prevalent across the popular music field; it is modified by different
styles (and individual pieces of music) tending more or less towards linearityor
circularity.It is also importantin an analyticalcontextto take into account the
possibilityforotherdimensions than melody and harmonyto produce narrativity
in popular music, by virtueof being organised non-parametrically.

Analytical concepts and criteria

On the basis of considerations presented in the preceding discussion, in what
follows I will propose a list of analyticaldimensions relevant to the analysis of
popular music formand its narrativecharacteristics.In the subsequent section,
the conceptspresentedwill be applied to the analysis of a numberofmusic videos.

Discursiverepetition/structure
oflyrics/function
This dimension comprises most of the factorscontributingto the constitutionof
musical formin a non-analytical,'intuitive'sense. 'Discursive repetition'is a con-
cept suggested by Middleton, and definedas 'the repetitionof largerunits, at the
level of the phrase, the sentence or even the complete section' (Middleton 1990,
p. 269). Finding the 'main sections' of a piece is largelya question of identifying
the largest sections subject to (discursive) repetition.In principle,this could be
done by using Ruwet's (1987) paradigmaticmethod of analysis; in practice,this
would entail a numberof problemsconcerningthe 'equivalence' of sections (how
much variation is allowed for one section to be considered a 'repetition' of
another?), and an intuitiveclassificationin most cases yields the same results,
 Music and imagesin musicvideo 57

although much faster.Taking the temporalorder of sections and the structureof

lyricsinto account, main and subordinatesectionsmay thenbe classifiedaccording
to theirrespectivefunctions;these functions(cf. BjOrnberg1992b, pp. 4f.) include
verse(V), chorus(C), bridge(B), solo (S),break(Bre), introduction
(In), interlude
(II) and
coda (Co).

Demarcation
A basic, albeit somewhat crude, measure of the distinctnessof demarcationof the
sections of a piece is thenumberofmusicaldimensions changein the trans-
exhibiting
itionfromone sectionto the next:the more dimensions thatare affectedby change
of some kind, the more well-demarcatedthe section will be perceived to be. The
musical dimensions in which changes may occur include lyrics(L), melody(M),
harmony(H), vocal texture(V), accompaniment texture,(A), instrumentation(I),
dynamics(D), tempo(Te) and tonalcentre
(To).

Symmetry
Symmetricalbinaryconstructionbeing the implicitnorm of popular music, asym-
metricalstructureswill be perceived preciselyas deviations froma norm, that is,
occurrences of asymmetryare perceptuallymarked as 'events'. In general, the
experientialeffectof a deviation fromsymmetryis inverselyrelated to its size;
deviations 'below bar-level' not only disturbperiod and phrase structurebut also
regular metre (cf. Bj6rnberg1987, pp. 76f.). The categories of deviation fromthe
symmetricalnorminclude prolongation, (P: addition of metricalunits to a symmet-
rical structure),truncation,
(T: subtractionof metricalunits), and elision,(E: 'over-
lapping' of two symmetricalstructures).

Musematicrepetition
This concept is defined by Middleton (1990, p. 269) simply as 'the repetitionof
musemes'; 'museme' is here to be understood in the sense used by Tagg (1979,
p. 71) as 'the basic unit of musical expression', although, as pointed out by
Middleton (1990, p. 189), 'the nature and size of the museme need to be regarded
flexibly'.In comparisonto discursiverepetition,musematicrepetitionis thus char-
acterisedby the repetitionof smallerunits,and in most cases also a greaternumber
of repetitions;still,the necessaryflexibility
sometimescauses problemsofdelimita-
tion between the two categories,problems which are best discussed in each par-
ticularcontext.Accordingto the dimension affected,musematicrepetitionmay be
classified as melodic(M), harmonic(H), or accompanimental (A); however, in the
interestof clarityof distinctionit appears reasonable to exclude in this analysis
those accompanimentalpartswhich in popular music ordinarilyfeaturemusematic
repetitionto a considerable degree (i.e. drum, bass and chordal accompaniment
parts; cf. 'Motorial flow' below).8

Directionality
This dimensioninvolves the sense offorward-directedness
effectedby the evolving
of melodic and harmonic structuresthroughtime (a more complete designation
58 AlfBjdrnberg
would be pitch-related directionality).The proposal of this concept constitutesan
attemptto summarise the effectsboth of parametricdimensions such as mode
('tonal language') and of non-parametricaspects of tonal organisation.It is thus
related to the amplitude and frequencyof fluctuationsof melodic/harmonic ten-
sion, and partlyanalogous to the of
concept 'prolongation'proposed by Lerdahl &
Jackendoff(1984, pp. 242ff.). While in
directionality principle is a quantifiable
dimension,a detailed quantificationwould seem to entaila high degree of subject-
ive judgement. I will thereforein my analyses only estimatethe degree of direc-
tionalityon a coarse quantitativescale, a high degree being considered as related
both to thetotalnumberofdifferent elements(melodic pitches and chords) used, and
to the degree of adherence to thevoice-leading principles tonality
offunctional (melodic
leading notes, melodic dissonance/consonanceprogressions, 'strong' harmonic
progressions,e.g. involving descending-fifth root progressions, 'regular' use of
chord inversions,etc.). The degree of directionalityis also affectedby deviations
frommetrical symmetry, e.g. prolongationincreasingthe directionality ofa dominant
chord) and by musematic (a
repetition 'strong' progression, when repeated, gradu-
ally loses its sense of forward-directedness).9 In connectionwith the analysis of
directionality,occurrencesof closure(cadential effects),classifiableas melodic(M)
and/orharmonic(H), may also be convenientlyidentified.

Motorialflow
The presence of an accompanimental 'motorial continuum', mainly effectedby
drums,bass and chordalinstruments,approaches the statusofyetanotherpopular
music norm (cf. Brolinson& Larsen 1981, p. 200). Variationsin the motorialflow
are thereforealso perceived as musical 'events'. As is the case with directionality,
the degree of motorialflow is also quantifiablein principlebut open to subjective
judgement; the rough quantitativeestimatein my analyses is based, on the one
hand, on thenumberofcontributing layersofmusicaltexture,and, on the other,on the
temporal and
density regularity of the sound-events (beats) constitutingthe motorial
continuum. It should perhaps be pointed out here that such an analysis presup-
poses a mode of listeninginvolvingthe expectationof an explicit'spelling out' of
motorialflow;such a mode of listeningmay arguablybe assumed to be frequently
applied to contemporarypopular music, in contrastwith, for instance, a jazz-
oriented mode of listeningenabling the experience of intense motorial flow as
implicitin a single melodic line (cf. Durgnat 1971, pp. 36f.).

Dynamics
Changes in the overall dynamiclevel of the music are estimatedon a coarse quant-
itativescale.

Soundprocesses
of theoverallsoundor
This dimension involves changes affectingthecharacteristics
e.g. changes in timbralquality,amount
of thesoundofindividualvoices/instruments,
of reverberationor echo, dynamicbalance between individual voices/instruments,
etc.
 Music and imagesin musicvideo 59

Table1
Narrative/ Lyrical/ Epic/
linear elliptical circular
Discursiverepetition no yes Some
Demarcation variable distinct indistinct
Symmetry low high high
Musematicrepetition no some yes
Directionality high variable low
Motorialflow variable relativelyconstant constant
Dynamics variable relativelyconstant constant
Sound processes short-term long-term long-term
IPF high relativelyhigh low

Individuality factor(IPF)
predominance
This is a rough quantitativemeasure of thesignificance of'individuality'in a musical
piece, based on a reading of the dualism of lead vocal melody and instrumental
accompanimentin termsof the individual/background relationship(cf. above; also
see Bj6rnberg1987, p. 219). It is calculated in three ways: a) as the ratio of the
total duration of vocal phrases to the total duration of sections featuringvocal
parts, i.e. the duration of vocal phrases plus the restsbetween them (IPFvs); b) as
the ratioof the totaldurationof sections featuringvocal parts to the totalduration
of the piece (IPF,,t);c) as the ratioof the totaldurationof vocal phrases to the total
duration of the piece equal to IPFv,. IPF,,).
(IPFvt, between these analyticaldimensions and the three
The general relationships
modes of musical syntacticalconstructiondiscussed above are shown in Table 1.
The table entriesindicate how the analyticaldimensions, each one viewed separ-
ately,are related to the syntacticalmodes; however, due to the multidimensional-
ity of music, actual pieces of music offerseveral possibilities for tendencies in
differentdimensions to work in differentdirections.A thoroughtheoreticalana-
lysis of the overall effectsof such conflictingtendencies being a very complicated
task, in the present contextI will restrictmyselfto a discussion of such conflicts
appearing in the particularexamples to be analysed.
For the sake of clarityof distinction,in what followsI will partiallyeffectthe
change of terminologysuggested above, using the terms 'linear', 'elliptical' and
'circular' when exclusively discussing musical syntax, while retaining the
narrative/lyrical/epic triad when speaking of visuals, combinationsof music and
or
visuals, experientialaspects of musical structures.10 An initialhypothesisunder-
lyingmy analyses is that the visual structures of music video are generallydeter-
mined by, and reflect,the elliptical(i.e. non-linear)natureof popular music syntax,
and the polysemic nature, or 'conditionalreferentiality' (Bjornberg1992a, p. 380),
of musical denotationalsignification,ratherthan a 'postmodern fragmentation'of
visual narrative.However, in light of the range of variation pergeneraet species
characterisingthe syntacticalconstructionof actual popular music, the conclusion
that 'rock music cannot in principlebe illustratedwith logical narrativeprocesses'
(Nielsen 1991, p. 298; my italics) appears too rash; furthermore, visualisation in
music video is of course not necessarilyentirelydeterminedby musical syntax.In
the reciprocal interplaybetween musical and visual signifyingsystems there is
60 AlfBjdrnberg

scope forvarious kinds ofrelationshipsbetween the two (or three,includingverbal

lyrics),and one of the purposes of the analysis is preciselythe investigationof the
practicaland aestheticlimitationson this scope.
The objects chosen for analysis are four recent music videos recorded for
MTV Europe. These have not been selected with an aspiration to any kind of
but with the aim of demonstratinga range of possible
statisticalrepresentativity,
relationships between musical and visual structures.The musical analyses are also
presented graphically(see Appendix, Figures 1-4).

A streetcarnamed Defier
The firstobject selected foranalysis is the video forBruce Springsteen's'Human
Touch'. The song is a 'trad rock' number in medium tempo and 'VCB-form'
(VVCVCBVCC; see Figure 1); however, this is interspersedwith several instru-
mentalsections(In, II and S sections),givingthe song an overalllengthof roughly
6'25". The last C section (preceding the extended instrumentalCo) is prolonged
by way of varied repetitionof the last phrase. Demarcation is generallydistinct,
with changes in vocal texture,instrumentationand accompanimenttexturesup-
portingthe segmentationdefined by lyrics,melody and harmony. Sections are
mostly symmetricallyconstructed,but two-bar prolongationsoccur in the C, B
and S2 sections.Melodic musematicrepetitionis not veryprominent;the V section
melody may be characterisedas using 'varied musematicrepetition',with a small
numberof repetitions.The harmonicstructure,though,utilisesmusematicrepeti-
tion forthe most part of the song: the V and C sections are based on the two-bar
Mixolydianchord sequences I I bVII IV I and I bVII I IVI , respectively,the first
of which is also used in the In, II and S2 sections,while the second appears in the
S1 section. The vocal sections,however, conclude with the sequence I bVII I bVII
IV I I , producinga (weak) harmonicclosure, combined with melodic closure in C
sections. Directionalityis thus fairlylow throughout,except for the B section,
featuringa more active (also Mixolydian) chord sequence; also, there is no final
closure, since the song ends with a fadeout. Motorial flow and dynamics are
highlyvariable,withvariationsreinforcing segmentation.Sound processes mainly
concern the relative dynamic levels of lead vocals and lead guitar. The overall
IPF stated is fairlylow, due to the extended instrumentalsections; however, an
experientiallyrelevantvalue would ratherbe higher,on account of these sections
featuringimprovisatoryvocal elements.
All in all, the song exhibitsstructuralfeaturestypicalof the 'epic' (this time,
in the sense 'heroic narrative')trad rock style characteristicof Springsteen.The
basic conception is circular(the harmonic ostinatos and mainly low-directional
melody), but this is modifiedby its adaptation to an ellipticalframework(clearly
perceptible,albeit weak, harmonicand melodic closures, discursive repetitionof
well-demarcatedsections). The circularcharacter,however, is also manifestin the
relationshipbetween the sections here termed 'verse' and 'chorus': the musical
materialofthe C sectionmainlyamounts to permutedV material(myidentification
of this section as a 'chorus' is based on temporalorder of presentation,closure,
and lyrics,the song's titleappearing at the end of thissection' cf.BjOrnberg1992b,
usually associated with
pp. 2f.). Thus thereis littleof the contrastor directionality
verse-chorusprogressions;the largest contrastin the piece with regard to tonal
processes is produced by the B section.The returns(discursiverepetition)of entire
 Music and imagesin musicvideo 61

sections, being generallytypicalof the ellipticalmode of musical syntaxconstruc-

tion,have the effectof 'grounding'the processual flowby way ofthe 'reconciliation
of difference'functiondiscussed by Middleton (1990, pp. 223f.). Nevertheless,the
song also has a distinctlynarrativequality, effectedby means of a cumulative
succession of multiple circular/elliptical sections modulated by processes of vari-
able motorialflow and dynamics (especially evident in the transitionfrom114and
115,the most climacticpoint in the song), but thisnarrativity is thus of a particular,
static,long-term,slowly evolving nature.1
The visuals of the 'Human Touch' video, as is the case with most music
videos, are to a large extentstructuredby the segmentationof the music; hence
the followingaccount is arranged according to correspondingmusical sections:
In: Nocturnal North American citystreet;approaching streetcar.
VI: Superimpositionof image of Bruce Springsteen(hereafter,Springsteenl) lip-
synchingvocals, shower of sparks fromtrolley-wireon the word 'pretenders';
dissolve to streetcar,with man in window easily identifiableas Springsteen
(Springsteen2).
V2: Intercuttingbetween Springsteenl and passengers of streetcar,including
Springsteen2.
CI: Springsteenl sings against indeterminablegrey-brownishbackground; dis-
solve to streetcarrunning;sparks on the word 'touch'.
Ill: Rapid montage composed of shots of running policemen, guitar-player
(Springsteen3)in tiled pedestrian tunnel or suchlike, Springsteen2in streetcar
and young people in nocturnalcitystreetsettingsilluminatedby flashinglights.
V3: As V2; rapid insertof indistinctclose-up of woman on the word 'tonight'.
C2; Springsteenl singing;lonelywoman in room; Springsteen2in streetcarexchan-
ging smiles with littlegirl.
112:Rapid montage of Springsteen3and street scenes; restless camera; flashing
lights.
B: Montage of Springsteen3lip-synchingand streetscenes; trolley-wiresparks on
the word 'we're' (' . . . all riders on this train').
SI: Montage of Springsteen3 and street scenes, ecclesiastical scenes (funeral,
christening),motherwith baby, littlegirlsdancing and wedding-couple.
113:Montage continued; cut to tilted-camerashot of shadow of streetcarmoving
on housefronts;cut to Springsteen3.
V4, C3, C4, 114:Intercutting between Springsteenllip-synchingand streetcarstop-
ping, Springsteen2gettingoff,crossingstreet,enteringbuilding,climbingstairs,
walking down corridor,knockingon door.
115: Rapid intercutting:Springsteen3 shouts; woman runs towards door, bare-
chested Springsteen3 swings guitar; Springsteen2enters room; Springsteen3;
Springsteen2and woman embrace; Springsteenl, now playing guitar,etc.
116:Springsteenl playing guitar; montage of this, caresses of naked body, and
Springsteen3with hands liftedas if in prayer; cut to trolley-wiresparks with
Springsteen3in background on percussion introto next section.
S2: Montage continued,mostlyfeaturingcaresses and Springsteen3;however, the
video ends with Springsteenl smiling relaxedly and self-confidently into the
camera.
As indicated by this recapitulation,the video is composed of three distinct
visual components: an act of narration,a visual narrative,and a series of discon-
nected scenes in various settings,i.e. in agreement with Kinder's (1984) well-
62 AlfBjdrnberg
known categorisation.Each of these threecomponentsfeaturesits own edition of
Bruce Springsteen: Springsteenl, the balladeer and narrator;Springsteen2,the
actor and protagonistof the narrative;and Springsteen3,the 'guitar hero' from
the streets.The firsttwo components togethermake up the visuals during the In
section,most of the vocal sections and the 114section (which may be regarded as
a short'coda' to the prolonged C4 section), while the thirddominates the instru-
mental sections Ill-3 and S1, as well as the vocal B section. In the montage of the
extended coda section (I15/I16/S2), the threecomponents seem to merge and form
a synthesis. Accordingly,the separate identitiesof the three Springsteens: the
calm, controlledand controllingnarrator,the vulnerable protagonisttravellingto
see his loved one, and the defiantstreetrebel-musician,apparentlymove towards
becoming differentaspects of one and the same identity(the variations of the
clothing of Springsteens 1 and 3 are interestingin this respect), although the
concludingshot of the video indicateswhich of the threeis to be perceived as the
'real' one (cf. Frith1988, p. 216).
In accordance with the general characterof the musical narrativityof the
piece, the visual narrativeis fairlylow on process statements,i.e. not very much
'happens': a man travels by streetcarto meet a woman waiting, they meet and
rejoice. Nevertheless,it formsa coherent'logical narrativeprocess', at the same
time as its openness and ambiguity(who are this man and this woman? where
does he come from?for how long have they been separated? will they have to
part again?) allows fora range of interpretations(cf. Goodwin 1987, pp. 26f.). In
the musical sections high in directionality and/ormotorialflow the visuals evolve
into an 'elliptical' (this time, in the literarysense) narrative,invitingparaphrases
in the directionof 'it's a dangerous world out there, but human touch is still
essential in all stages of life'. Thus a conflictis set up, musically as well as
lyrically/visually, but this conflictis reconciled, visually/narratively as well as
musically,in the coda of the song.12

Blue dolphinbeat
My second example is the video forBryan Adams' 'Thought I'd Died and Gone
to Heaven'. This song is a mainstreamrock ballad in 'VCB-form'(VCVCBVCC;
see Figure 2), where the vocals of the firsteight measures of the V3 section are
replaced by a guitar solo. Demarcation is distinct:the segmentationdefined by
lyrics,melodyand harmonyis reinforcedby changes in vocal texture,instrumenta-
tion and accompanimenttexture.Deviations fromsymmetryare numerous in the
formof 21-barprolongations of V sections (two bars by phrase repetition,one
half 'extra' measure added) and one-bar prolongationsof C sections (11 bar in
C3). Musematic repetitionis sparsely used; the C sections featurethe two-bar,
four-chordvamp II V vi IV I , which may be regarded as a 'border case' of
musematicrepetition(cf. Bjornberg1992b, p. 12), while the VB sections contain a
discreetguitarriff.Experientially,the combinationof In and the firsteightbars of
V1 also produce an effectof harmonicmusematicrepetition,an effectnot present
in the followingV sections. Directionalityis medium-highand fairlyvariable,with
lows in VA sections and a peak in the B section. No unambiguous melodic or
harmonic closure occurs, and the song ends with a fadeout. Motorial flow and
dynamicsare nearlyconstantthroughout,except forbreaks at the ends of sections
and the firstfour bars of the B section. Sound processes include a temporary
echo effecton the firstentryof the lead vocals, and a successive domination of
 Music and imagesin musicvideo 63

accompanimentover lead vocals in the concluding C sections. The overall IPF is

fairlylow, in spite of vocal sections constitutingthe bulk of the piece, on account
of the 'gaps' in the vocal line (IPFvs = .63).
As is the case with most rock ballads, the song is predominantlyelliptically
constructed,with an orderlyrepetitionof clearlydemarcated sections. This, how-
ever, is imprintedwith distinctlylinear traits:the fairlyhigh and variable direc-
tionality,and the frequent deviations from symmetry,13 while motorial flow,
dynamics and sound On
processes support ellipticality. the whole, the processual
characterof the song is thus mainly lyrical,i.e. reflectiveand self-contained,but
with a not insignificantnarrativepotential,although this narrativityis primarily
of a 'local', short-rangenature, and accommodated within the larger elliptical
framework.
The visuals of the video, arranged according to musical segmentation,are
summarised in the following:
In: One continuous shot; camera, directed downwards, moves over grassy
meadow at night but with blue-ish lighting(which pervades the entirevideo);
camera tiltsup to horizontal,dollyingin on Adams, standingin the meadow.
VI: Shot continues; camera stops at m. 14; three dolphins jump out of the grass
and fallback, in slow motion; camera dollies upwards and out; cut (forthe first
time, at m. 19) to medium close-up of Adams; cut to drummer,then Adams,
camera dollying down/in,then out; cut to Adams fromanother angle, camera
moving in; cut to close-up of Adams.
CI: Dissolve to distance shot of entire group, upwards down; camera dollies in
while turningclockwise; dissolve to Adams, camera moving down/out,showing
guitarist;wind moves the previously calm grass; at m. 33 camera stops and
marathon runners (?) startappearing out of the grass at gradually decreasing
distance; the last six or seven of these, who resemble the dolphins by wearing
sweaters pulled up over their heads, disappear at top of picture; camera con-
tinues to move out; cut to drummer,camera moving upwards and out while
turning.
V2: Similarimages of Adams and the othermusicians, camera constantlyin move-
ment; at m. 47 a dolphin jumps up, fallingdown two shots later; last shot in
section featuringbass player, with camera moving in, then around, then
upwards and out.
C2: Similarimages; section ends with fade to black.
B: Fade in to medium close-up of Adams superimposed on meadow with moon
and clouds above; the musicians, superimposed, act swimming- (or stage-?)
divers, fully dressed, at top of picture; fromm. 69 montage as before, with
flyingbirds superimposed fromm. 73.
S: Dissolve to camera dollying in, through shrubbery,to guitaristwith moon
above; camera stops at m. 81, moon turnsinto strobelightand a dolphin jumps
up and somersaults, landing with great splash of water; cut to close-up of
Adams.
V3,: Montage as before;in last shot of section, one dolphin jumps up.
C3: Similarmontage; in second shot of section, three dolphins come down.
C4: Similarmontage.
Co(C5): One continuous shot; dissolve to medium close-up of Adams; camera
moves out, showing the other musicians; at m. 121 five dolphins jump out of
the grass in frontof group, the middle one, correspondingto Adams' position
in the group, jumping later and higherthan the others; at m. 123, the dolphins
64 AlfBjirnberg

having disappeared, the camera sinks, stillhorizontallydirected,down into the

grass.
Obviously, this is not a 'narrative'in any qualified sense. The main visual
components remain the same throughoutthe video: the nocturnalsettingin the
meadow, the blue-ish illumination,and the musicians doing nothing but per-
forming(or, rather,lip- and hand-synchingthe performanceof) the song. The
thingsactually 'happening', i.e. the appearance and movementsof the dolphins,
runners and birds inserted by means of technical special effects,stand out as
isolated occurrencesseeminglyimpossibleto relateto the othervisual components
(or to the song's lyrics)withinany coherentand logical narrativeframework.One
'narrative'elementin the video, however, is constitutedby the almost constantly
moving camera, producing perpetual shiftsin distance and orientation;still,the
'storytold' by these camera movementsremains completelystatic.
Of course, the video is not intended to present any independent narrative,
but to be experienced in connectionwith the music of the song. In spite of my
arrangingthe summaryabove by musical segmentation,the visuals of this video
are not veryclearlystructuredby this segmentation,nor by the motoriallayers of
the music. Visual transitionsare produced by dissolves ratherthan straightcuts
and oftendisplaced in relationto the beat (which is continuouslyspelt out by the
bass, playing straightquavers, and drums). Some transitions(C1-V2, V2-C2; cf.
above) are markedby particularlyenergeticcamera movements,but generallythe
visual processes (besides the camera work, also the slow-motionmovements of
dolphins and runners)emphasise continuityand constancy,in a 'relay' relation-
ship with musical processes (cf. Barthes 1977, pp. 38ff.).In this way, the visuals
underlinethe lyrical,reflectivecharacterof the ellipticalmusical structure,poten-
tiallyopening up vast fieldsof association to the user. On the otherhand, except
forthe work of the 'narrativecamera', the narrativepotentialof the music is not
realised in the video; this,however, is not so much a matterof the images 'domin-
ating' the music as of the visuals focusingattentionon some aspects of the music
while concealing others.

Egyptian divorce
The object formy next analysis is the video forMichael Jackson's'Rememberthe
Time'. The song is a typicalJacksonianhip-hop-influenceddance piece in 'VCB-
form' (VCVCBCCCCC; see Figure 3). Furthermore, there is a soloistic/
instrumentalmiddle segment (the Bre and Il sections; m. 73-104); these are
peculiar to the video version, not being present in the CD album version of the
song. Demarcationis fairlyindistinct,withmostlyonlychanges in instrumentation
(in most cases rather subtle) and vocal texture supporting the segmentation
definedby lyricsand melody. Sections are symmetrically constructedthroughout
the piece, exhibitingno deviations. Harmonic musematicrepetitionis prominent;
except forthe B section and the (implied) staticharmonyof the Bre section, the
entire song is built on a four-barV-i ostinato (with varyingalterationsof the V
chord). Althoughthevocal line shows generalrepetitivetendencies,actual melodic
musematicrepetition(more or less varied) occurs onlyin the Bre,II and concluding
is low, due to the
C sections. In spite of the strongV-i progressions,directionality
large numberof repetitions,and no melodic or harmonicclosure occurs, the song
ending with an echo effecton an altered V chord. Motorial flow and dynamics
exhibit littlevariation, with the exception of the Bre section. Sound processes
 Music and imagesin musicvideo 65

mainly affectpresence and reverberationof the vocals. The IPF is high, due both
to rests between vocal phrases being short and to most sections featuringvocal
melody (Jackson'shabitof singingall vocal partshimselfrendersa clear distinction
between 'lead' and 'background' vocals problematic;the experientialeffectis one
of 'multiplelead vocals').
The song's structuretakes a mainlycircularform:the repeated harmonicriff
and constantmotorialflow, acting against any sense of processual development,
serve to level out contrastsbetween sections; the 'verse'/'chorus'distinctionthere-
foremainlydepends solely on lyricsand, to a limitedextent,melody. This renders
the processual characterof the song epic, i.e. staticand 'mythic'.The Bre and Il
sectionsprovide some potentialforthe kind of long-rangenarrativity discussed in
connectionwith 'Human Touch', but to a considerablylower degree than in that
case.
As in many other Michael Jacksonvideos, the lengthof the 'Rememberthe
Time' video significantlyexceeds that of the song, due to the presence of a
visual/dramaticintroand a (shorter)visual coda; these are denoted 'Vis. In' and
'Vis. Co', respectively,in the followingsummary:
Vis. In: Throne-roomof Pharaoh in ancient Egypt,Pharaoh and Queen seated on
thrones;hooded figureapproaches; Pharaoh asks 'Now, what is it you're going
to do?' (the only instance of spoken dialogue); hooded figuretakes brownish
dust frompouch and spreads it on the floor,where itmoves into circularpattern;
hooded figuresteps onto patternand disappears mysteriouslythroughthe floor,
leaving only his cloak; where he's stood, an amorphous shining figurearises,
eventuallymetamorphosinginto Michael Jackson.The scene is accompanied by
Phrygian'suspense' underscoremusic played by stringsand electricguitar,and
by sound effects.
In: Shots of Jackson, moving rhythmicallyto the beat, alternatingwith coun-
tershotsof Pharaoh and Queen.
Vl: Jacksonlip-synchingvocals; countershotsof Queen; Jacksonkisses her hand;
Pharaoh signals to his guards; Jacksonturnsand runs down the stairsfromthe
thrones,leaving Queen with hand outstretched.
CI: Intercuttingof guards runningor pointing,Pharaoh, Queen swooning; at end
of sectioncut to doorway where Jacksonappears, swirlingon percussionpick-up
to next section.
V2: Intercuttingof guards searching,Jacksonwith passing camel in front,Jackson
in harem, snakes in basket.
C2: Jacksondancing in circleof women; at end of section continuitycut fromveil
of dancing women to drapery.
B: Pan fromdraperyto Queen on bed; zoom out reveals Jacksonat other end of
room; intercuttingof Queen and Jackson.
C3: Queen on balcony, looking out onto pyramids; Jackson approaches; they
embrace and kiss; at end of section cut to column, behind which arm protrudes
making 'Egyptian' gesture.
Bre: Intercuttingof guards and dancers emergingfrombehind columns; intercut-
ting of Jackson and dancers; zoom out fromJackson revealing surrounding
dancers.
I: Dance scene; Jacksonand dancers fromvarious angles.
C4, C5, C6, C7: Dance scene continued; at end of last section cut to close-up of
Jackson.
Co: Jacksonswirls; cut during swirl to different background.
66 AlfBjdrnberg
Vis. Co: (accompanied by percussion underscorein same tempo as song); Jackson
stops and looks around, findinghimselfback in throneroom; Pharaoh in from
left;Jacksonsmiles impudentlyand makes a dash; menacing-lookingguards
approach; Jackson,findinghimselfsurrounded,escapes by swirlingand disap-
pearing mysteriously.
In this case, there is no equivocalityas to the narrativityof the visuals: a
logicallyconsistentnarrative,with a considerable degree of temporaland spatial
coherence, is presented by means of well-establishedfilmiccodes. Nevertheless,
thisnarrativeis clearlydivided into segmentsexhibitingvaryinglevels of 'realism'.
In the first,and 'most realistic'segment,the narrativeevolves in 'real time'; this
segment constitutesthe 'visual intro'. In the second segment, correspondingto
musical sections In-C3, the realismis modifiedby a more ambiguous construction
of temporal continuity.The third segment, startingat the beginning of the Bre
section, dismisses realism completely,suspending the narrativeflow throughout
the remainderof the song; the second shot of the guards in thatsectionconstitutes
the last 'realistic' element. The 'visual coda', by returningto the firstlevel of
realism,resumes and concludes the narrative.
This segmentationofthe visuals is obviouslyrelatedto musical segmentation.
In effect,this video constitutesa condensed version of the classical Hollywood
musical film(as seems to be a favouriteprocedurewithMichael Jackson),underlin-
ing the parallelismsbetween the two formsdiscussed by Allan (1990), and lying
significantly closer to the Hollywood musical than Morse (1986, p. 23) considers
typical of music video. The correlation,characteristicof the musical film,of 'real-
isticnarrative'withabsence ofmusic and of 'fantasy'withitspresence is, however,
modifiedin the video by the establishmentof two levels of realismin the former
mode, one withoutmusic (except forthe underscore) and one with music.14 The
'fantasy'in the video (the dance scene), the transitionsin and out of which are
musically marked by echo effects(in the Bre and Co sections), corresponds in
functionto the musical number of the musical film.However, the connectionof
'level 2' narrativewith circularmusic structuremeans that the latteris not very
clearly realised in the visuals; with a few exceptions, edits and depicted move-
mentsare not stronglyco-ordinatedwith the music. Due to the 'narrativeinterest'
of the images, there is thus, in this particularsegment of this particularvideo,
some reason to speak of 'domination'of the music by the visuals. The only 'narrat-
ive' musical element present here is the high IPF, enabling in a limited way a
musical support forthe evolution of visual narrative.

A space obscurity
My finalanalysis concerns the video forSnap's 'Rhythmis a Dancer'. This song
is a pop-technodance number;'5the overallformalstructuremightbe characterised
as an instance of 'expanded standard form' (cf. Bjornberg1992b, p. 2) with the
disposition VVBV (see Figure 4), although the constructis not supported by the
tonal processes typicalof thatform-type.The S (or B) sectionis composed of eight
bars of instrumentalsolo followedby a sixteen-barrap. Demarcationis of medium
distinctness,the segmentationdefined by lyricsand melody being supported by
changes in instrumentation and accompanimenttexture.All sections are symmet-
ricallyconstructed with no deviations. Musematic repetitionpredominatesin har-
mony and accompaniment structure:the bulk ofthe piece uses the two-barAeolian
chord ostinato i bVI bVII i I, spelt out by a synthesiserriffin quavers. The
 Music and imagesin musicvideo 67

vocal melody, although exhibitingno clear musematicrepetition,works repetitive

tendencies into arch-shaped lines. Directionalityis generallyinsignificant,with a
low in the Phrygian-colouredstatic-harmony B section.The finaltonicminorchord
gives no strong sense of harmonic closure, since it appears in the same metric
position as previouslyin the harmonicriff;the song ends abruptlywith a percus-
sion echo effectdisruptingregularmetreby a shiftto a rhythmof dotted quavers.
Motorial flow is built up in discrete steps up to m. 32, thereafterbeing mostly
constant,with the exception of the Bre sections. Dynamics show littlevariation;
sound processes primarilyconcern the amount and quality of reverberation
applied to vocals and/oroverall sound. The IPF is medium-highas regards both
rests between vocal phrases and relativedurationof vocal sections.
Thus, like the preceding example, this song too exhibitsa circularstructure
based on repetitiveharmony and fairlyconstant motorial flow. One important
difference,though, is that the application of the 'verse/chorusprinciple' is even
less distinctlymanifestin 'Rhythmis a Dancer'. While the vocal melody effectsa
demarcationin the middle of the V section,implyinga differentiation into a 'verse'
(VA) and a 'refrain'(V,), this is not supported by any variation as regards lyrics
between the different VA sections(all lyricsremainidenticalin all threeV sections).
The epic characterof the song is thereforesomewhat more emphasised in compar-
ison with 'Remember the Time'; however, by the alternationof sections and the
variationsof motorialflow, the song stillprovides some (small) potentialforthe
previouslydiscussed long-rangetype of narrativity.
The visuals of the video, arranged as beforeaccordingto musical segmenta-
tion, are summarisedbelow:
In: Establishing shots and show pans of rocket launching site, illuminated by
pale mauve-orangelightingand occasional white flashesand spotlights;rockets
standing in background; smoke; superimposed sped-up moving clouds; female
singer, in tight,shiny,black dress, on slowly-risingconstruction-elevator; sur-
rounding people dressed in similargarmentsof dull metalliclustre.
VI: Similarscenery;threepersons holding huge spheroids over theirheads; mech-
anical body movements; bass player on similar slowly-risingconstruction-
elevator behind and to the leftof the one holding the singer.
V2: Cuts on downbeat of everysecond bar; on m. 33 to computer/paintbox graphics
with symmetricmirroreffect;on m. 35 to differentgraphics featuringmoving
clouds, human silhouette in slow motion and CAD (computer-aided design)
rotatingwheel; on m. 37 to singer;on m. 39 to graphicsresemblingthe firstbut
with two human silhouettesand moving 'atom' symbol;then to intercuttings of
singer,lip-synching and graduallymoving more intensely,and bass player.
Brel: Scenery similarto that of the V1 section.
S: New graphics: superimpositionsof human silhouettesrunningin slow motion,
clouds and nautical charts,changing by quick dissolves in quaver rhythm;on
M. 61 cut to graphics featuringzoom in on human silhouettein slow motion
and CAD rotatingsphere; fromm. 65 montageof these elements,medium-close-
ups of rapper lip-synchingand humans moving mechanically;at end of section
'zoom in cut' on face of rapper on the word 'dancer'.
V3: Scenery similarto that of section V1; several slow pans.
II: Cut to human silhouette/nautical chartsgraphics;on m. 101 cut to bass player.
Bre2:Cuts on downbeat of every second bar to various graphics;on m. 112 cut to
CAD sphere with superimposed clouds.
Co: Sphere recedes into distance, rotating;fade to black.
68 AlfBjdrnberg
As in the case of 'Thought I'd Died and Gone to Heaven', this video is
manifestlynon-narrative.Two main visual components are set up: on the one
hand the launching site scenery,on the otherhand various types of special-effect
graphics, but none of these shows any significanttendency towards narrative
evolution.
Thus, to an even higherdegree than in the BryanAdams example, the visuals
of thisvideo are dependent on the music. The movementsdepicted and the editing
bothworktowardsan emulationofmusicalrhythm,the superimpositionof several
differenttypes of movementreflectingthe latter'smulti-layerednature. Many of
the cuts describedabove as 'on the downbeat' are actuallyplaced fractionally
before
the beat, thus producing an off-beateffectenteringinto a relay relationshipwith,
and reinforcing, the rhythmicflowofthe music. No narrativecomponentsinterfere
with the alignmentof the visuals with the epic characterof the piece. The actual
contentof the visuals potentiallyopens up large fields of association, but these
are not stronglystructuredby the visuals nor by the music of the video.

Towards a typologyof structuralrelationships

On thebasis ofthe precedinganalyses, I would like to propose a tentativetypology
ofstructural betweenmusicand visualsin musicvideo (see Table2).
relationships
The fourtypes indicated in the table are, however, not equally representativeof
contemporary music videos. As Middleton (1990, p. 217) points out, post-
rock'n'rollpopular music has generallytended towards mixturesof the elliptical
and circularmodes of syntaxconstruction.The music/visualsrelationshipsof most
music videos could thereforebe expected to approach one of the two types in the
right-handcolumn of the table. However, this typology,being ratherschematic,
cannot be expected to exhaust the range of possible relationships,and it evidently
needs furthermodificationand differentiation. In this context,it should also be
noted that in relationto the videos analysed here, the concepts 'narrativemusic'
and 'epic music' must be regarded on a relativescale, as each of the foursongs is
based on a clearlydiscernibleellipticalstructure;on the whole, decidedly linear
rarelyappearin popularmusic(cf.above). Thus,it appears
or circularstructures
particularlypertinentforthe furtherelaborationof this typologyto focus on the
ellipticalmode of musical syntaxconstruction,to which no obvious visual counter-
part seems to exist.The musical functionof providing'unity',mentionedby Allan
and Kinder (cf. above) is relevant in this context,but conceived in too general
terms:since music always has a tendencyto effectunityby means of sheer redund-
ancy in comparisonto othercommunicativemodes (cf. Bj6rnberg1992a,pp. 382f.),
the crucial point is ratherhowthis unificationis broughtabout in each particular
case.
It appears, however, to make some sense to draw a parallel between the
axis and Kinder's (1984) aforementioned 'narrative/
linear/elliptical/circular
performance/dreamlike visuals' axis. The alignment of 'linear' with 'narrative',
and of 'circular' with 'dreamlike visuals', seems relevant, as indicated by my
analyses. Furthermore,performancevisuals in music video oftenappear 'neutral'
with relationto the significationof musical syntax,and theymay thus constitute
a 'natural'or privilegedmode of underliningthe reflective,introspective,reconcili-
ation-of-difference characterof the ellipticalmode of musical syntaxconstruction.
These musical characteristics,described by Moore (1993, p. 85) in terms of the
 Music and imagesin musicvideo 69

Table2

MUSIC

Narrative Epic

'Film music' relationship,i.e. Assignmentof 'mood music'

narrativeprocessesin musicand functionto music, i.e. the
Narrative visualsinteracting,
buton terms music providingbackclothto
setby themusicratherthanthe thevisualnarrative
visuals
VISUALS
'Circularisation'
of music, i.e. 'Musicalisation'of visuals,i.e.
Epic focusing of attention on the visuals primarily reinforc-
non-narrativeaspects of the ingmusicalexperience by syn-
music, concealingits narrative aesthetic 'translation' and
potentials emulationof musical experi-
entialqualities

'fictionality'of the popular song as a 'self-containedentity',could thus be viewed

as homologous to musical performancein its functionas an 'irrational'aesthetic
activity,detached fromeverydayreality.
The complexityof the issue should also be acknowledged by takingverbal
lyricsinto account. I have not commentedmuch on the lyricsof the songs analysed
here, since these seem to me to be more importantto an analysis of content
than to one of structure;still,the matterclearlycalls forfurtherattention.In this
connection, it can be noted that music videos with a visual narrativeare often
based on narrativelyrics(or lyricswith some narrative'seed' in them), to a large
extentregardless of the degree of musical narrativity(cf. videos such as Rolling
Stones' 'Under Cover of the Night', 'Small Town Boy' by BronskiBeat, and Tom
Petty's 'Into the Great Wide Open'). Nevertheless, clear correlationsas regards
narrativityin differentsignifyingsystemsappear to exist at least in some musical
genres (notablythe predilectionfor'mythically'conceived narrativelyrics,narrat-
ive videos and 'baroque' musical syntaxin heavy metal music; cf. Walser 1992).

Conclusion
None of the music videos analysed in this article is particularlyexceptional in
respect of aesthetics,neitheras regards music nor visuals; rather,to my view, all
fourexamples are (musicallyas well as visually) well-crafted,and quite effective,
instances of mainstreamformsof expression. The widespread reluctanceto deal
with such 'naturalised' mainstream forms,implicitlydismissing them as unin-
terestingand aestheticallydeficient,seems to be partly rooted in a somewhat
romanticbelief in the subversive power of rock music (see, forinstance, Berland
1986, Tetzlaff1986). However, as Goodwin (1987, p. 31) indicates,the countercul-
tural potential of pop culture is often exaggerated (especially, it seems to me,
by US-American scholars and writers);furthermore, a full understandingof the
subversiveand the exceptionalcannotbe expected to be gained withoutalso exam-
ining the 'ordinary'(cf. Tagg 1982, pp. 63f.).
70 AlfBjirnberg
To be fair,the confusionsand theoreticaldisagreementsdiscussed here con-
cerningthe characteristicsof music video are partiallydependent on the time of
writing,as the dominant formsof music video have changed considerablyover
the years. Furthermore,music video being a veryheterogeneousphenomenon, it
sometimes also appears that 'anything can be proven', depending on how
examples are chosen. This illustratesthe extentto which the fieldof contemporary
popular music is stylistically,socially and semioticallyfragmented;still, despite
postmodernistprophecies of 'the destructionof meaning', users of music video
and otherpopular culturalformscontinueto derive pleasure fromthe production
of (albeittemporaryand volatile) meaning induced by these forms(cf. Straw 1988,
Frith1992).
In spite of the emphasis placed here on textual-structuralanalysis, it would
seem unwise to deny that there are limitationsto the explanatoryvalue of such
an analysis of music video. It appears equally unwise, however, to discard the
possibilities of textual analysis before the nature of these limitationshas been
investigated by way of such an analysis taking all relevant dimensions into
account. Syntacticalcharacteristicsare 'objective possibilities' (Willis 1978, pp.
198ff.),subject to influenceby varyingmodes of perception,contextsof use and
otherfactors;these factors,as well as the ways in which thevisualisationofmusical
structuresis shaped by genre-specificculturalcodes of expression, remain to be
examined in more detail by futureresearch.

Appendix
1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181
Measures
111 I
= 116) lil 1111
(4/4;M 11111l
111II
l ll ll lll 1111l llH Ill 1111
1111
lll
1111
l II
Co

V2 V3 C2 1112 B SI 1113 V4 C3 C4
Discursiverepetition/ nVI 1416 iS2
lyrics/function 11D15
D D
LA LA LAL L LV LA LA H LA LA LV MAH H D
L H M MM M MA MA M A MIM M HI I I
M I H H V H HI H H V H H L I A A AH A A A
DemarcationI I II 1 I I I I II I I

41818
Symmetry 8
10(P212lO(P9818 18 18 110(P)+121418+
10(P4418 14+8I10(P)+8+8(?)
A
M HHH HHH HHH

Musematicrepetition I I I I I I I II I

Directionality/closure
NI H H MH H MH H MH (MH) (H) (H)

Motorialflow I I I II I I I I I I I I I I I I I I

voc.
dom. rev.)
(sh. dom-
gtr
voc.~
Dynamics/ [fade]
soundprocesses = 0.88 = 0.49 = 0.43
IPFv IPFst IPFvt

Human Touch
Figure1. BruceSpringsteen:
 Music and imagesin musicvideo 71
2 2 2 2
1 11 21 4 31 41 51 4 61 71 81 91 4 101 4 111 121
s
I IL I I I I I I I I I
(4/4; = 88) I II II II 1 1
Measuresliii1iiliiii ,i*1i l

Discursive
repetit In V V2 C2 BI s B C3 C4 Co(C5)
lyrics/function C1 i A)lV3B ICo
LV Ic2S LV LI L LI D LI LA LV lc4
LA L
M LI M MA LI M MA LI MA M M M M
I MA H HV MA H HVMA H H H H H
Demarcation HV HV HV

Symmetry8 18+ 10 9 P8+ 110 P19 P4 18 10 PI 9 P 9 10 (?)

M +18
H
Musematicrepetition I I I I I I
I

Directionality/closure

Motorial
flow
I voc:temp.echo;
largerev. accomp.dom.
Dynamics/ I I I fade]
soundprocesses = 0.63 = I0.87 I I = 0.55 I
IPFst
IPFvs IPFvt

Figure2. BryanAdams:Thought
I'd DiedandGonetoHeaven

11 21 31 41 51 61 71 81 91 101 111 121 131

Measures
(4/4;d= 108) II l l II l I I IIIIII I 11111A1
1 II II I Il 11 lli 1 1 1 1 i II I Il
!
Discursiverepetition/In VI CI V2 C2 B C3 Bre 11 C4 C5 I C6 C7
lyrics/function I B 1C3 Bre C4 C C6 C7
L LV LV LV LV LV LI LA LA LV LV
M LV M M LV M MA MA MV M M D M M L L
I M I I M I H H HA I H V I I M M I
Demarcation I I I I I I

8 88+ 8
Symmetry 8 8+ 18 8 18 18 18+ 8 8 8 8 8 8 2
18+
M

Musematicrepetition [.................... .
I I I I I I I . . . .
A-- ,--- J. . . . . ..

Directionality/closure I i
I

Motorialflow I I I I I I I I I I I I I

voc:close voc: veryclose; sh. rev. lead voc

e
Dynamics/ I I I I I 8 I I
= I I fade
soundprocesses = 0.88 IPF,,= 0.88 = 0.77
IPFV,, IPFvt

Figure3. MichaelJackson:
Remember
theTime
72 AlfBjdrnberg

1 11 21 31 41 51 61 71 81 91 101 111
Measures
(4/4;j = 124) 1 1 1 1 l i 11l
l
ll I I 11 I)I
I
ll I II I l II I Il
iJ
l ll
Discursiverepetition/ICn lVI V2
V2 Bre I IS(B)
S(B)
V3
V3
II
Bre2
Bre
Co
I
lyrics/function Bre1 II1
l
LA L L LI LI L L LI
M L M M MA L MV M M MA
M A L A H V L H L A A H
Demarcation I I I (IA)
I

18+ 18
Symmetry 18+ 18 18+ 18 18+ 18+ 18 18+ 18 8 8 2
M
A HHHHHHHHHHHHHHHHHHH ?-4 HHHHHHHHHHHHHHHH
Musematicrepetition I I I I

-
Directionality/closure I
I I I
(H)
Motorialflow
dryer;
voc: more (voc I)
largerev. + echo voc: sh. rev.
resence
Dynamics/ I I I I I I I I I I I
soundprocesses Ifade]
IPF, = 0.79 IPFst= 0.77 IPFv,= 0.61

is a Dancer
Figure4. Snap: Rhythm

Endnotes
1 An earlierversion of this paper has appeared
in the WorkingPaper Series of the Department
ofMusic and Theatre,Universityof Oslo. Parts
of the theoreticaldiscussion are derived from
Bj6rnberg(1992b).
2 For a discussion of the reasons forthe sparse
coverage of the subject of music in the literat-
ure on music video see Bjornberg(1991, pp.
64f.); also cf. Nielsen (1991, p. 297).
3 The relationship between montage and
musical beat is also touched upon by Good-
win, who refersthisto the conditionsof music
video production: 'Directors working under
enormous pressures of time will ( . .. ) often
shoot some material that can be edited at
random to the beat, withoutany great regard
forrealism or narrativeclosure' (1987, p. 26).
4 At the level ofvisual content,Nielsen also con-
nects thiswith Straw's argumentthatthe frag-
mentary eclecticism of music video, rather
than representingempty pastiche, constitutes
withintradi-
ofpopmusichistory
a reconstruction
tional pop song forms: 'The relationship of
song to visuals is ( . .. ) ratherone between
the basic demands of form( . . ) and the het-
erogeneityof codes and visual materialsheld
in play by that form' (1988, p. 258; see also
Berkaak and Ruud 1989).
5 Most of the workscitedhere have been written
by scholars in sociology, filmstudies, literary
studies or mass communicationresearch. Up
to the present, the contributionsof musicolo-
gists to the study of music video have been
few and have also shown a general bias
towards the perspectivesof these otherdiscip-
lines, this situation indicating the current
embryonicstage ofdevelopmentofa structural
semioticsof popular music.
6 Straw's (1988) emphasis on the recoursein the
music video era to traditionalpop structures
(see Note 4) seems to attributea historically
relativelystable and constant, and thus rela-
tively vague and unspecified, nature to this
potential ('the basic demands of form');how-
ever, in view of the significantrange of vari-
ation as regardsformalconstructionin popular
music it appears more relevant to speak of
'forms'ratherthan of one general 'form',and
to investigate in more detail the potential
experiential consequences afforded by this
variation.
7 My earlier discussion of this matter (see
Bjornberg1992a, pp. 380ff.),while remaining,
to my view, theoreticallyvalid, is somewhat
too generalised for the present analytical
purposes.
8 The presence of such parts would seem to
bring popular music in general closer to the
epic category (cf. Table 1 and the analysis
section); it may also be pointed out that the
 Music and imagesin musicvideo 73

regularmetrecharacteristicof most Western big world, littleme'. Existentialfactsdo have

music constitutesa constantelement of muse- a tendency to sound trivial; the interesting
maticrepetition,although on a ratherabstract point, however, is not so much the triviality
level. of the message as the significationpotential
9 The numberof repetitionsis crucial for this (and pleasure) invoked by the specificway in
effect.While a large number of repetitionsof which it is structured.
directionalmusematic structuresreduces dir- 13 Such deviations appear, in the popular music
ectionality,a 'small' number(2-4) oftenserves of the last three decades, to have gradually
to increase it by means of tension/release replaced tonal processes as a means of increas-
effects;cf. Tagg's (1979, pp. 132ff.)discussion ing directionality(in a general sense); whereas
of 'ready, steady, go'-patterns. in a Tin Pan Alley ballad a sectional transition
10 Jones (1988) distinguishesbetween 'mimetic', may be emphasised by a series of II-V progres-
'analog' and 'digital' (non-linear) narrative, sions withchord substitutions,in a rockballad
where television and music video have been one, or a half, extrabar is instead added to a
movingtowardsthe third;however, these con- symmetricalperiod (cf. Bj6rnberg1987, p. 77).
cepts do not appear immediatelytransferable 14 This differentiation of realism levels under-
to discussions of musical syntax(but may per- lines the structural,functionaland experiential
haps be seen as parallel to Middleton's con- differencesbetween traditional film under-
cepts, and mine, as regardsthe ratioof 'narrat- score music and the music of music video;
ive information'to total information). however, a detailed discussion of these issues
11 This may be illustratedby the fact that with cannot be pursued in the present context(on
regard to motorial flow, the 'intro' (i.e. underscore music, see Gorbman 1987).
introductory) section of the song actually 15 The video is based on the '7" edit' version of
extends over the entire first52 measures (cf. the song, with a duration of 3'41"; the CD
Brolinsonand Larsen 1984, p. 345). single also contains a 5'12" version subtitled
12 The essence of the song and the video, like of '12" mix', and a 6'49" 'Purple hazed mix', the
so much of Springsteen'sproduction,appears latter with a considerably differentand less
to amount to what Berland (1986, p. 44) ironic- mainstream-orientedharmonic,rhythmicand
ally describes as 'the oldest tune in the book: formalstructure.

References
Allan, B. 1990. 'Musical cinema, music video, music television',FilmQuarterly,43/3,pp. 2-14
Aufderheide,P. 1986. 'Music videos: the look of the sound', JournalofCommunication, 36/1,pp. 5-78
Barthes,R. 1977. Image- Music - Text(London)
Berkaak, O. A. and Ruud, E. 1989. 'Rockeminner'('Rock memories'), Norveg,32, pp. 205-17
Berland, J. 1986. 'Sound, image, and social space: rock video and media reconstruction',Journalof
Communication Inquiry,10/1,pp. 34-47
Bjmrnberg, A. 1987. En litensdngsomalla andra.Melodifestivalen1959-1983 (A littlesong like all others.
The Melody Festival 1959-1983) (G6teborg)
1991. 'Sign of the times?Om musikvideo och populairmusikenssemiotik'('On music video and the
semioticsof popular music'), Svensktidskrift
fdrmusikforskning, 72, pp. 63-84
1992a. 'Music video and the semioticsof popular music', SecondoConvegnoEuropeodi AnalisiMusicale,
eds. R. Dalmonte and M. Baroni (Trento),pp. 379-88
1992b. PopularMusic Formas NarrativeProcess,Reportseries Ljudbilder,1, Departmentof Musicology,
Universityof G6teborg
Brolinson, P.-E. and Larsen, H. 1981. Rock ... Aspekterpd Industri,Elektronik & Sound, (Rock ...
Aspects of Industry,Electronics& Sound) (Stockholm)
1984. 'FrAnb6rjan. Om "intro" och inledningsfunktioner hos ABBA' ('From the beginning. On
"intro" and introductoryfunctionsin ABBA'), Tvdrspel- trettioen artiklarom musik.Festskrift tillJan
Ling, (G6teborg), pp. 343-51
Brown, M. E. and Fiske, J. 1987. 'Romancing the rock: romance and representationin popular music
videos', OneTwoThreeFour, 5, pp. 61-73
Chatman, S. 1978. Storyand Discourse.NarrativeStructurein Fictionand Film(Ithaca & London)
Durgnat, R. 1971. 'Rock, rhythmand dance', BritishJournalofAesthetics, 11/1,pp. 28-47
Fiske, J. 1986. 'MTV: post-structuralpost-modern',JournalofCommunication Inquiry,10/1,pp. 74-9
74 AlfBjirnberg

Forsman, M. 1986. 'Det eviga nuet' ('The eternalnow'), Filmhdftet, 54, pp. 4-14
1991. 'Bilder av musik - om rockvideonsformsprAk' ('Pictures of music - on the form-languageof
rock video'), Ung Film,9, pp. 4-9
Frith,S. 1988. 'Making sense ofvideo: pop into the nineties',MusicforPleasure(Cambridge), pp. 205-25
1992. 'Det gode, det darligeog det ligegyldige'('The good, the bad and the indifferent'),Mediekultur,
18, pp. 45-56
Frith,S. and Horne, H. 1987. Art intoPop (London & New York)
Genette,G. 1980. NarrativeDiscourse(Ithaca)
Goodwin, A. 1987. 'From anarchyto chromakey:music, video, media', OneTwoThreeFour, 5, pp. 16-32
Gorbman, C. 1987. UnheardMelodies.NarrativeFilmMusic (Bloomington& Indianapolis)
Grossberg,L. 1988. "'You (still)have to fightforyour rightto party":music televisionas billboardsof
post-moderndifference',PopularMusic, 7/3,pp. 315-32
Ihlemann, L. 1992. 'Fra rockkoncerttil musikvideo' ('From rock concertto music video'), Mediekultur,
18, pp. 34-44
Jensen,J.F. 1988 ' ... AND NOW ... THIS:', Kultur,identitet og kommunikation, ed. H.J. Nielsen
(Aalborg), pp. 7-32
Jones, S. 1988. 'Cohesive but not coherent: music videos, narrativeand culture', PopularMusic and
Society,12/4,pp. 15-29
Kaplan, E.A. 1987. RockingAround the Clock. Music Television,Postmodernism and ConsumerCulture
(London & New York)
Kinder, M. 1984. 'Music video and the spectator:television,ideology and dream', FilmQuarterly, 38/1,
pp. 2-15
Laing, D. 1985. 'Music video: industrialproduct,culturalform',Screen,26/2,pp. 78-83
Larsen, P. 1987. 'Bortomberattelsen'('Beyond narrative'),Filmht'ftet, 55-56, pp. 82-96
1988. 'Musik og moderne billedfiktioner'('Music and modern visual fiction'),Kultur& Klasse, 15/4,
pp. 33-53
Lerdahl, F. and Jackendoff, R. 1984. 'An overview of hierarchicalstructurein music', Music Perception,
1/2,pp. 229-52
Mar6thy,J. 1974. Music and theBourgeois,Music and theProletarian (Budapest)
Middleton, R. 1990. StudyingPopularMusic (Milton Keynes & Philadelphia)
Moore, A. 1993. Rock:ThePrimaryText.Developinga MusicologyofRock(Buckingham& Philadelphia)
Morse, M. 1986. 'Post-synchronizing rockmusic and television',Journal ofCommunication Inquiry,10/1,
pp. 15-28
Movin, L. and Oberg, M. 1990. Rockreklamer (Rock commercials)(Viborg)
Nielsen, E. 1991. '"MTV, MTV, MTV get offthe air ... .'", Analyseraf tv og tv-kultur, ed. J.F. Jensen
(Kobenhavn), pp. 281-304
Ruud, E. 1988. Musikkforayet.Om musikkvideo (Music forthe eye. On music video) (Oslo)
Ruwet, N. 1987. 'Methods of analysis in musicology',Music Analysis,6/1-2,pp. 11-36
Straw, W. 1988. 'Music video in its contexts:popular music and post-modernismin the 1980s', Popular
Music, 7/3,pp. 247-66
Strom,G. 1989. Musikkvideo, (Music video) (Oslo)
Tagg, P. 1979. Kojak - 50 Secondsof TelevisionMusic: TowardtheAnalysisof Affectin PopularMusic,
(G6teborg)
1982. 'Analysingpopular music: theory,method and practice',PopularMusic, 2, pp. 37-67
Tetzlaff,D. J. 1986. 'MTV and the politicsof postmodernpop', JournalofCommunication Inquiry,10/1,
pp. 80-91
Walser,R. 1992. 'Eruptions:heavy metal appropriationsof classical virtuosity',PopularMusic, 11/3,pp.
263-308
Willis, P. 1978. Profaneculture(London)

Discography
BryanAdams, 'Thought I'd Died and Gone to Heaven', WakingUp theNeighbours.A & M Records, 397
164-2. 1991
Michael Jackson,'Rememberthe Time', Dangerous.Epic, 465 802-2. 1991
Snap, Rhythm is a Dancer. Logic Records, 665 309. 1992
Bruce Springsteen,'Human Touch', Human Touch.Columbia, 471 423-2. 1992