+ All Categories
Home > Documents > feminism

feminism

Date post: 01-Nov-2014
Category:
Upload: denisa-chirita
View: 115 times
Download: 4 times
Share this document with a friend
Description:
paradigms of feminism
Popular Tags:
11
Media texts and gender 67 .~' Ml~dln rrexl6 find (}(JIHIl'li thinkof mllssmedlllcontont nNn unll1ublc cutesory, 'espocially NlnccIts V()hlll1~ IIlId 11I1"l'l1nl vllrlety t:onthlllllllyInol'ou8euncitho boundnl'lcH lII'OIlI1d "II1I1MII1I10tlin" 111'0II\tH,t1~\Hlnuly h/ll'd to tllMocu'n' (McOuntl, 1(JR.,: rlfl) , 1~t.1M"II'ch III IhII' III'GU of l1Iodln output for thllt 111111101' III VIINt, IIIld 11111 11"1C111(10Iv/ll1l"thllt II aUIIIIII'ohl'lllllv8 rlilvl~w 1I1I1II1II1I1n" (h. pOIlNlhl\!ty would dlilll1O1IMtl'llto c~ulIMlllltllll tlIBlhwllv~, Kl.'nclC1I'~d 1'0111111'1111 or IlIadlr, toxts, Alruudy by I\177 II blhllo/,ll"'phy nbunl "tlX Mtc.I'oolYPCII In mlldln output cOlltnlncd mOl'e lhl1l1 1\ thOl\Hllnd c:mll'lQN (COUl'lIlOY nllt! Whlj,plc, 1983), Aside from lIumbers IIndl'Oles, dlfforent modin texlN lIro 111/10 und orten better characterized by their specific narrative and visual conventions which structure the meaning of characters, their mutual relationships and their positions in the 'story'. A range of distinctions such as medium type, genre, formats, target audience etc., differently determine particular representations of gender and make valid generalizations in this area even less feasible. For principal and pragmatic reasons then, this chapter will not offer a general review of the way gender has been encoded in various media texts. 1 Instead, the aim of this chapter is to familiarize the reader with the two basic approaches to the study of media output: content analysis and semiotics. As will become clear later, a semiotic approach fits the theoretical framework of this book best. However, given the frequent occurrence of content analysis of feminist media research and its relevance to specific purposes, its merits and methods will be ~xtensively discussed as well. To facilitate comparison between the two approaches, I shall use a particular genre to illustrate the differences in assumptions, methodology and types of results. In the context of the representation of gender, a focus on advertising seems a particularly appropriate choice. From its rebirth in the late 1960s, the women's movement has singled out advertising as one of society's most disturbing cultural products. Monitoring projects, con'sumer boycotts and other means have been employed to put pressure on advertising agencies to come up with more diverse and less traditional portrayals of women. While the accompanying academic research on advertising waned in the mid-1980s, the subject nowadays seems to have attracted the renewed interest of cultural and feminist critics. As a cultural form, advertising displays a preoccupation with gender that is hardly matched in any other genre. 'In modern advertising, gender is probably the social resource that is used most by advertisers. Thousands of images surround us every day of our lives that address us along gender lines. Advertising seems to be obsessed with gender and sexuality' (Jhally, 1987: 135). This obsession is said to spring from the 'signifying power' of gender, Advertisements and commercials need to convey meaning within limited space and time and will therefore exploit symbols that are relevant and salient to society as a whole. As one of the most deeply felt elements of subjectivity and the social structure, gender provides such symbols most effectively: 'Some- thing that can be conveyed fleetingly in any social situation and yet something that strikes at the most basic characteristics of the individual' III fin QlIl'lIorclulptcl', IIlcdln WtB'CdCl'crlbed us lechnologles of gender, uccomlllodntlng, modifying, reconstructing und producing, disciplining and contradictory renditipns of sexual difference (ef. Chapter 3). One of the sites where we can directly observe how these technologies operate are media 'texts', such as movies, TV programmes, books, newspapers, records etc. When focusing on media output solely, the mai,n task for feminist media research is to unravel both the dominant and alternative meanings of gender encoded in media texts, and their articulation with other discourses such as for instance, ethnicity, class and sexuality. In confronting such issues, feminist media critics have tended to make sweeping statements on media representations of women, femininity and ; /' gender. 'According to feminists, the media-created woman is (1) wife, mother and housekeeper for men, (2) a sex object used to sell products to i.,. men, (3) a person trying to be beautiful for men' (Hole and Levine, 1971: 249). Franzwa discusses television and claims that l televised images of women in large measure are false, portraying them less as they really are, more as some might want them to be.... Television women are predominantly in their twenties. ... portrayed primarily as housewives. ., restricted primarily to stereotyped positions such as nurses and secretaries. ,, portrayed as weak, vulnerable, dependent, submissive and frequently, as sex objects, (1978: 273-4) Davies et al. contend that media and women's magazines in particular purvey images that tell women how to be 'a perfect mother, lover, wife, homemaker, glamorous accessory, secretary - whatever best suits the needs of the system' (1987: 4). While mass media do indeed present us with an incessant stream of gender stereotypes that can evoke curious and , hilariolls us well as indignant or infuriated reactions, general statements HIIQhill' Ihoso IIbove fnll to cupture the specificnature of media texts, The \ Olllhll, ffl/' II1"tllllce, lhut mediu output undcrrepresents women has never , 1:1111,II1U,11,(,1WOI11t\I!'H UCltH'OS "llch liS 1111188 produced romantic fiction, WOIn~II'~ 1I11I"114111~II, HOllpoplll'ns IUld slIch cil1cml\togrnphic expressionsof 1111*10111'1111111 11M Ihe .woUll,ln'M Wlllllplu"' (Bynl'H, 1991.), Moreover, it scems Ulill 1~lIl,Ivl.l(ln I. b"'(~lHlnij IIIIHCI IlIId mol'u .""llllnized' with the growing POllllhll'llv "I .Olll!°1'('11'11"', //1111 IHld 311111(,\ HhowH (l1'iskc,19HH), As fal' as I'll"'" I1IIIIII1"I'H III't' 11II11(!1;I111~d I lit'll, hlt1ll\slvc 8tnlomenls "bout media 1"'l'lI'iI~('IllIi1Llttl\N 01'WOIIJClII11I'C bound t(:1he IrIliell too generul. Just us we clo nol ~peHIi\ot' IhI!!'oolllll1t' or ~ptloch. \)Cloks OJ'oduculion, it is impossible to """"""""""""""""~'~~"~~"'-~"~"'~~~"""""""-""""""""-'.."' "" ," '" ,"'-_., ""~ ~..__..
Transcript
Page 1: feminism

Media texts and gender 67

.~'

Ml~dln rrexl6 find (}(JIHIl'li

thinkof mllssmedlllcontontnNn unll1ublc cutesory, 'espociallyNlnccItsV()hlll1~IIlId 11I1"l'l1nlvllrlety t:onthlllllllyInol'ou8euncitho boundnl'lcHlII'OIlI1d "II1I1MII1I10tlin" 111'0II\tH,t1~\Hlnuly h/ll'd to tllMocu'n' (McOuntl, 1(JR.,:

rlfl) , 1~t.1M"II'ch III IhII' III'GUof l1Iodln output for thllt 111111101'III VIINt, IIIld 11111

11"1C111(10Iv/ll1l"thllt II aUIIIIII'ohl'lllllv8 rlilvl~w 1I1I1II1II1I1n"(h. pOIlNlhl\!ty

would dlilll1O1IMtl'llto c~ulIMlllltllll tlIBlhwllv~, Kl.'nclC1I'~d 1'0111111'1111or IlIadlr,

toxts, Alruudy by I\177 II blhllo/,ll"'phy nbunl "tlX Mtc.I'oolYPCII In mlldlnoutput cOlltnlncd mOl'e lhl1l1 1\ thOl\Hllnd c:mll'lQN (COUl'lIlOY nllt! Whlj,plc,

1983), Aside from lIumbers IIndl'Oles, dlfforent modin texlN lIro 111/10undorten better characterized by their specific narrative and visual conventionswhich structure the meaning of characters, their mutual relationships andtheir positions in the 'story'. A range of distinctions such as medium type,genre, formats, target audience etc., differently determine particularrepresentations of gender and make valid generalizations in this area evenless feasible. For principal and pragmatic reasons then, this chapter will notoffer a general review of the way gender has been encoded in variousmedia texts. 1 Instead, the aim of this chapter is to familiarize the readerwith the two basic approaches to the study of media output: contentanalysis and semiotics. As will become clear later, a semiotic approach fitsthe theoretical framework of this book best. However, given the frequentoccurrence of content analysis of feminist media research and its relevanceto specific purposes, its merits and methods will be ~xtensively discussed aswell. To facilitate comparison between the two approaches, I shall use aparticular genre to illustrate the differences in assumptions, methodologyand types of results.

In the context of the representation of gender, a focus on advertisingseems a particularly appropriate choice. From its rebirth in the late 1960s,the women's movement has singled out advertising as one of society's mostdisturbing cultural products. Monitoring projects, con'sumer boycotts andother means have been employed to put pressure on advertising agenciesto come up with more diverse and less traditional portrayals of women.While the accompanying academic research on advertising waned in themid-1980s, the subject nowadays seems to have attracted the renewedinterest of cultural and feminist critics. As a cultural form, advertisingdisplays a preoccupation with gender that is hardly matched in any othergenre. 'In modern advertising, gender is probably the social resource thatis used most by advertisers. Thousands of images surround us every day ofour lives that address us along gender lines. Advertising seems to beobsessed with gender and sexuality' (Jhally, 1987: 135). This obsession issaid to spring from the 'signifying power' of gender, Advertisements andcommercials need to convey meaning within limited space and time andwill therefore exploit symbols that are relevant and salient to society as awhole. As one of the most deeply felt elements of subjectivity and thesocial structure, gender provides such symbols most effectively: 'Some-thing that can be conveyed fleetingly in any social situation and yetsomething that strikes at the most basic characteristics of the individual'

III fin QlIl'lIorclulptcl', IIlcdln WtB'CdCl'crlbed us lechnologles of gender,uccomlllodntlng, modifying, reconstructing und producing, disciplining andcontradictory renditipns of sexual difference (ef. Chapter 3). One of thesites where we can directly observe how these technologies operate aremedia 'texts', such as movies, TV programmes, books, newspapers,records etc. When focusing on media output solely, the mai,n task forfeminist media research is to unravel both the dominant and alternativemeanings of gender encoded in media texts, and their articulation withother discourses such as for instance, ethnicity, class and sexuality. Inconfronting such issues, feminist media critics have tended to make

sweeping statements on media representations of women, femininity and

;/' gender. 'According to feminists, the media-created woman is (1) wife,

mother and housekeeper for men, (2) a sex object used to sell products toi.,. men, (3) a person trying to be beautiful for men' (Hole and Levine, 1971:

249). Franzwa discusses television and claims that

l

televised images of women in large measure are false, portraying them less asthey really are, more as some might want them to be. . . . Television women arepredominantly in their twenties. . . . portrayed primarily as housewives. . ,restricted primarily to stereotyped positions such as nurses and secretaries. , ,portrayed as weak, vulnerable, dependent, submissive and frequently, as sexobjects, (1978: 273-4)

Davies et al. contend that media and women's magazines in particularpurvey images that tell women how to be 'a perfect mother, lover, wife,homemaker, glamorous accessory, secretary - whatever best suits the

needs of the system' (1987: 4). While mass media do indeed present us withan incessant stream of gender stereotypes that can evoke curious and

, hilariolls us well as indignant or infuriated reactions, general statementsHIIQhill' Ihoso IIbove fnll to cupture the specificnature of media texts, The

\

Olllhll, ffl/' II1"tllllce, lhut mediu output undcrrepresents women has never, 1:1111,II1U,11,(,1 WOI11t\I!'H UCltH'OS "llch liS 1111188produced romantic fiction,

WOIn~II'~ 1I11I"114111~II,HOllpoplll'ns IUld slIch cil1cml\togrnphicexpressionsof1111*10111'111111111MIhe .woUll,ln'M Wlllllplu"' (Bynl'H, 1991.),Moreover, it scemsUlill 1~lIl,Ivl.l(ln I. b"'(~lHlnij IIIIHCI IlIId mol'u .""llllnized' with the growingPOllllhll'llv"I .Olll!°1'('11'11"',//1111IHld 311111(,\HhowH (l1'iskc,19HH),As fal'asI'll"'" I1IIIIII1"I'H III't' 11II11(!1;I111~dIlit'll , hlt1ll\slvc 8tnlomenls "bout media1"'l'lI'iI~('IllIi1Llttl\N 01'WOIIJClII11I'Cbound t(:1he IrIliell too generul.Just us we clonol ~peHIi\ot' IhI!!'oolllll1t' or ~ptloch. \)Cloks OJ'oduculion, it is impossible to

""""""""""""""""~'~~"~~"'-~"~"'~~~"""""""-""""""""-'.."'"" ," '" ,"'-_., ""~ ~..__..

Page 2: feminism

68 Feminist media stu.dies Media texts and gender 69

(Oo/'rhIlUl,1 ItJ7(JI7),~ Thull Ih, 0KtrtlU1ulyOOlldtlllHedform of oommunicll-11011In IIdvClI'lillhlj£IcandNtt.RolfCi'loClptkHWl1ywt;lll to nil I:IKuminllLion of1"111t1l1l11 VUIlII., l",IIt'I'~ I1ndmyth. "01111"010.,110~onclQI',

,III_I n.lhn~ "" 1I(llIlIlt~ III IIII-dill \)\ltpuIIIlClhhul 1*I'OMIUII'1,:11ubout m~eUIIII'~I." 11Hlliniul hml~ lit' work, N~VI:II'th~I(iI.8,I1Qcm'dll1j&LtJMoQu1I1I(1987:IlI:1) III., II\UI.IIIIII III hllw Imlclhi UUtpllt 1'(1ll1tOIlt(} Hoole!y ullderlJus tho1IIIII00IIy III IjllrIIY.I~M01' In\.ldl11 I~X(M,ruiling thQl Issuoof how mass,'(lllIIlIllIlhulihII) pl~rronn. 118' IIIGIIIII\l.II1IJ role In society. The feministItllIlMlnlllMlol111Ii1d",IMof (lOImnnulolitlol1sdiscussed in Chapter 2 assume a1'lItlu,',1III'(IIUhtfCl!w(u'dI'olullon bt:lwt:lon media and society, accusing theII'IIJ8HnlQdln or c(mveying a distorted picture of women's lives andexperloncesund demandinga more realisticreflectioninstead. Massmediaarc thought to produce symbols of reality, expressing in an abbreviatedform the nature of a particular reality (cf. Carey, 1989). However, symbolshave another capacity as well that is often overlooked in transmission

I-'models of communication. They function as symbols for reality, (re )con-I structing reality while simultaneously representing it. The latter view ofL symbolsis central to constructivistviewsof communicationthat perceive

'reality' as the product of the social and sense making activities of humanbeings. Thus, the distinction between transmission and constructivistmodels of communication discussed in Chapter 3 expresses itself also in adifferent prioritizing of the dual capacity of symbols to function simul-taneously as symbols of reality (transmission models) and as symbols forreality (constructi'l'ist models), and in a different conceptualizing of therelation between media output and society. In part, this distinction isreflected in different modes of analysis as well, ranging from contentanalysis' which provides an 'unproblematic' description of media output asconsisting of particular social types and categories, to semiotic analyses ofthe subjective, moral and aesthetic dimensions in media output. Althoughthe interdisciplinary nature of both feminist and cultural studies seems torequire at least some convergence between the two methodologies, it is stillquite unlikely that one will find both content and semiotic analyses used inthe same project, owing to their different disciplinary origins, contentanalysis being a sociological approach and semiotic analyses rooted in thehumanities.

Symbol/! or rQnllty: contant I1l1ulysls

incidence of violence in American television programmes (Gerbner et al.,1980).This list gives only an impression of the kinds of questions addressedby content analysis and is by no means exhaustive. In general its aim is tocompure features of media output with concomitant features in reality.Thus, a typical conclusion of a content analytic study would be that theoccurrence of violence on television grotesquely exaggerates the amountand type of violence one is likely to encounter in real life. In feministresearch, the exemplary conclusion is that media output fails to representthe actual numbers of women in the world (51 per cent) and theircontribution to the labour force.

Formally, content analysis is defined as 'a research technique for theobjective, systematic and quantitative description of the manifest contentof communication' (Berelson, 1952). The respective elements in this bynow classic dictum require some explanation. To begin with, the focus onmanifest media content forms an important limitation of content analysissince the researcher is prevented from reading between the lines of mediaoutput, and is expected not to 'dig' below the manifest level of analysis orto descend to the level of latent meanings and associative conclusions.Only the explicitwords, sentences, texts, images- the signsthat actuallyappear in the media text - are taken into account.To many this mightseemlike an unacceptable limitation since so much of what makes media outputspecial, interesting or pleasurable consists precisely of the latent' andassociative meanings excluded from such an analysis. However, thislimitation can also be considered an asset of content analysis, since itenables a methodology that fulfils traditional scientific requirements for'objectivity' and one that produces results that will be generally accepted as'true' and 'reliable'. A focus on manifest content ensures that differentinvestigators will reach a level of agreement about t~e message under studyand that a repetition or replication of the same project will produceroughly similar results. The inclusion of latent meanings in the analysiswould imply the recognition of individual and cultural-specific interpre-tations of media texts which would lead to disagreements and contradic-tions about the material involved, a truly gruesome prospect for mostcontent analysts.

How is a typical content analysis conducted? For instance, suppose wewere interested in finding out how women and men are portrayed intelevision series of the 1990s. While this might seem a stale question, in thepast decade such questions have hardly been posed in feminist mediaresearch of any significance. One of the reasons for conducting such ananalysis therefore, could lie in the desire to have fresh 'hard data' on theportrayal of women. Secondly, one might be interested in assessingwhether and how transformations in women's position in society arereflected in television series. Another motivation might be to test the moreor less intuitive comments of authors such as Fiske (1988) and Brown

(1990a) that television series are typical women's products. Severaldecisions - whose outcomes depend on the motivation of the research -

1.,,\',\'111/11"/(1""1'(/lid mot!t(~d()ICJIJ~1

CiulIl~nl /\nlll~.I.IN (~rtcmC'1I11,lqyr.c!II) t\88QII.tht:lllJIIl1lros( characteristicsof1111KI-II\nUHHlI~8nf'Itl.~dhl (Hlt~ull, '1:'yplcmll'~MQal'Clh')I'()~11!J111SIn this field orehH 11111/1\\1111\'1/hlll'oII'M. l'IYQlloloUlclIIll1J1CIp!lYMtcul fll.ln(w'eMell'wnl11cn undI\I~JI GI(~t.iU'I'ItIU11111YlIl'h,(y()I' .ilf'lIl'ft.II~GI CCHlIIll'Y music(Saucier,198G),lid V

.

H.

't.

I'hl~.

(.

'(nl1'lI1~'y 1I11d ~h'111pl(\,1V~:J),tolcwlslonnews, commercialsImll ~I!rllli. (~J'IH)Vt8r(H1,11,)80)or Alms (JIIISlc~ll,1987i Rosen, 1973); the

Page 3: feminism

70 Feminist media studies

Vnrlnblll

Cl8r1c1Cilr

IIMill!

()ooupntICIl1

Rosldenco

Marital status

Figure 5.1 Hypothetical coding scheme

need to be made in order to conduct a successful content analysis (Holsti,1969; McQuail, 1987). To begin with one needs to decide from which'universe' or sample of television drama the study will draw: which types ofseries should be included? Is it necessary to include all series or wiJI asample suffice? Consequently, is it necessary to analyse all episodes of theselected series or will a second sample do? Is it to be a study of televisionseries that are broadcast in the 1990s, or will the project be limited to thoseseries that were actually produced in the 1990s? Should foreign series beincluded in the sample or is it a study of domestic television only?Decisions like these have important consequences for the extent to whichthe results of the analysis can be generalized to all series, or domestic seriesonly, police series or soaps only etc.

. The next step in the analysis is to construct a coding scheme and decideon units of analysis. The nature of the coding scheme depends on what weactually want to know about the portrayal of women and men in televisiondrama. Possibly, we might be interested in a number of sociological vari-ables like the presence of women and men, race, age, occupation, residence,marital status. The units of analysis could be the individual characters anda hypothetical coding scheme might look like that in Figure 5.l.

Whill:l this simplified coding scheme uses the most straightforwardC:llttlUOl'iU~ pnAMibla,some of the problems one can run into doing content11I11i1lYNIN uI'CHll.I'IJ!Idyovldent. How, for instance, to classify a character's ageWIlfUlw.~urb d(ojllllngwithIIblot.ll'nphlcaldrlllnfl inwhichthemaincharacter1-fo.llowotll'l'(,m bll.th to burinI?18mll.$not simplifyingrllceby codingit ashll~J\klit whlUI,(jhly'lWIUj! todowith1\doublo occupationof housewifeand1~~,ortM tm' In~tlln~,11SuohI>I'Oblcm~111'0~xl:u~eJ'blltcc1iF one attempts tohn'IHpo'rntill mm'IHCII1II,'oXfol,ll.I.lfC,"f womenline.! I1lliln In TV drama, such"M IlIilh'I~11Y.'Ori) /111(1plyoho!t,g!ejlll IIwlce.up. PhYRic1i1churactcristics likebl'tllllly 1'1 Mtr'~IIQtlh ftol' In.I'IIIIOO, lire subject to differencos in individuall'OI'Oepl:lcHIS,while p.ycholog!clIl flillltllres such as dominance or submis~ive-

Media texts and gender 71

TnbllilS.l Occupation ofmala,\' andfemale.r ill American telavi~'loncommareirll,r -l'oml110 ('" :230)HOU8ewl.fo!illulhorStawnrdo.8ModolCelebrlty/MhIl.tor/dnnccrCook/ml\ld/~ervt\ntSecretory/elerlen]

Millo (iJ..1$~)

1'1,,~bnl1(I/I'nlhGfI'l'tJfll.llbnnlMIlI.toColobrll)'CUIIHtrllcllmlworkerSolOlmUIIBuslnesslntll1

\ PilotCriminalMechanicLawyerInterviewerOther jobs

IIII,~Ii7IiIi65333

27

MIi7533

Other jobs 18

ness are hardly ever unambiguously distributed among television char-acters. Whereas the latter kind of information is often more interesting andrelevant to the interpretation of television characters than more straight-forward sociological characteristics, the reliability of such data is often lowdue to the ambiguity of such features. Thus there is a trade-off in c'ontentanalysis between reliability and meaningfulness.

A following step involves matching the coding scheme with the chosensample of television series. This normally involves a group of coders whohave been instructed by the researcher how to employ the coding schemein such a way that individual variation among coders is eliminated. In fact,tests and figures indicating the reliability of content analyses all stem fromstatistical procedures which measure the degree of consent among coders(d. Holsti, 1969; Krippendorf, 1980).

Advertising

The results of content analysis often take the form of frequency tables likethe one shown in Table 5.1. These data were collected in 1971by Dominickand Rauch (1972).3 From about 1000 commercials shown on New Yorkcity network stations, they selected those in which women appeared forthree seconds or longer, or in which they had one or more lines of speech.A comparative sample of men was taken too. As Table 5.1 shows, womenwere depicted in a smaller variety of occupational roles than men, withmost of them working as a housewife and/or mother. In addition, 75 percent of all commercials in which women appeared promoted productsused in the kitchen or bathroom. In commercials for cosmetics and

personal hygiene women dominated as well, while men on the other hand,held more high status jobs and were seen purchasing high ticket itemssuch as cars. Voice-overs and product representatives were predomi-nantly male.

Vnlull Lobol

1 WomanMar1

1 1laokR Whltb

1 HOU811WlrO2 SlUGoollorII Willi. oollor

Monagement

1 Rural2 Urban

1 Umarried2 Married

Page 4: feminism

72 Feminist media studies

Courtney I1l1dWhipple (J983) report II l1umber of foHow-up studies andl'OI'IIou(l°I\, IIml Q()I1Q/ud"Ihllt t,'cml. III "IV c(')n'morclnl~of tho W70~IInclIQfj()"(iltllm1I1hl\l1"~NIUlllnNH1t1Ylllllh~lu,.hthore880lllUd tobe)n tcmdclI1cyIII pmll'IIY WOIIHmIIIllt!! orhltl wlwldnQ QUIlici,' tho h(1II1Ci,I1IId mcn mOI'(:

ofl~1I III domClHI'It!dClttluQM. 1{tI~(jI1I'Qh IntoOIUIIIj&ca,In Ihtll)OJ'll'Oyulor Ilion1111(1Won'llt!l 111ullvel'tI.\jnlOHl8 In AI11~l'Icl\l1 b\cmcl'tliIl1t~rQ8tmU/jnztl1t.lH1'11(\1111M'I'Iuta, UIWdCII".\'/)/811,1'1,till' Nlfw }'m'!cfJl'unciNew,l'wlu:lc,rouihlye()JI'Qb(~I'III(j.theMe,1'~8nll', J?romthe lutG 1950s'tothe ourly 1980swomen11"(,) II1orlllllsl1181y fIIhownworking outsldo the home, but this number neveroxceeds 25 per cent, Working roles of women have changed a little, withmore women working in professional and middle level business in recentyears. The majority of women are portrayed as non-working, and whereasthis meant portrayal in family roles in the early 1960s, in later years thedecorative function of women has been exploited more often. As in ilie TVcommercials, the goods which women are seen purchasing in advertise-ments relate primarily to domestic duties, beauty and cosmetics. Like TVcommercials, print advertisements show men less often as workers, but thisnumber is still a considerable 50 per cent. Men's work roles in printadvertisements have changed from business to entertainment and sports,and their increasing presence in non-working roles has meant a greateremphasis on their decorative function but not on their role in the family.

The changes in the depiction of women in advertising, especially theirincreased portrayal as professionals working in high status jobs, have beeninterpreted by many researchers as a success for women's liberation and asa measure of social and advertiser acceptance of changing sex roles (cf.Courtney and Whipple, 1983: 19). This rather optimistic conclusion wouldseem to be a direct result of the theoretical assumptions of content analysisand its methodological peculiarities. Advertisements are seen as a reflec-tion of the changing reality of women's social position and of the influenceof the women's movement, presenting us with an image of the 'newwoman':.

She is independent, confident and assertive, finding satisfaction in the world ofwork and recreation, seeking excitement, adventure and fulfilJment. She is a farcry from the consumer. . . finding her satisfaction within a rather smalJ worldand the center of this world is her home. (Cagan, 1978: 8)

There are, however, at least two other stories to be told about the 'new

WOmnJ1', If we do not conceiveof advertising as a reflection of reality, but

fIR 1111"xpression of capitalist consumer culture it is clear that the 'newWOI11 1\11, \s prill1/1rily interesting for her increased purchasing power. HerocclI1()mlo Indopendencesupportsthe capitalist market economy, instead of

1'C'I!othlh1QrJr CYtlnundermining it, as more radical and socialist inclinedr-r.wlnlMl8I11tijhl hnvlS envlRtlAlod.Instead of signifying progress, thel'I'ISIiIIICl6Ior 1.'1'I1Q1Wwoml.ln' in contemporaryadvertisingcan thus beJ)Ol'c:o!ViAdIIHIho QO.optlltloJ1,)f feministideals into acceptablefantasiesofIlIcllvl(hl/'i1rnldclleclussuchlevemontand success.A second problem withthe chcHufulvisl(\Jlsof the 'new woman' is that the image itself is not

'.T

Media texts and gender 73

problemlltized,The methodologyof content analysisonly allowsa recog-nition of thoselelul]'oleswhiehwomenporCoI'm:theevidenceof ndvrartllo,montHportruyhl&tl1<:1'now WOIllIlI1',howe.vcr,royoIIls tlult sheonlYclepflr~s1~H1rglJ1I1\lyf!'Omhet ()ld~J', !nOl'Otfl\tlltleml}1sl~to)'~,

"I'Qk~1for ;Xtll111,ltI,lhl) M111111U1u8",1 cn II\'()ml~t. lho JOllul Ihnnlt.'World""" Stylb' 1'111184'()f u)othllli,ll pl'ocllilmsIIN'th. JOlin! II/H'IIIIIIphl1C1st)phy'thtll. 'A womlU1Hhoululool( fo!'wl.ll'c!to clrCMSII1~for thoorflCCl',This 'philoHOphy' COI1SlJ'UC(Htile! t\fI1co U$ Hl1e)t!U;lfhllPPY ouc/lslon rorwomen to dress up and presenl themselves, Allc.llnd~ed in u photogruphused in one of the Jenni Bames advertisements the woman is portrayedstepping confidently towards the camera in an office environment,observed by a male colleague I'tom behind; but she is not portrayedactually working. To focus on her working role would be to ignorecompletely the dimension of traditional femininity that is exploited in thetext of the advertisement.

Content analysis in general suffers from theoretical and methodologicalproblems like this. It gives precedence to manifest content as the bearer ofme1:lI1ingat the expense of latent content and form, and it assumes thatfrequencies of certain characteristics are valid indicators of meaning.Finally it produces results whose relation to the actual media experience ofproducers and audiences is unclear:

This result [of content analysis]is also based on a form of 'reading' of contentwhichno actual 'reader' would ever, under natural circumstances,undertake. Ina certain sense, the new 'meaning' is neither that of the original sender, or of thetext itself or of the audience, but a fourth construct, whichhas to be interpretedwith care. (McQuail, 1987:184)

Such objections have made content analysis less and less 'fashionable' infeminist media research. Byars (1991) observes for instance that the'images of women' approach to gender in film studies has been rejectedand ridiculed by most feminist film theorists for its lack of theoreticalsophistication. Nevertheless, in certain contexts and for certain purposescontent analysis does yield valuable results. When the research in the areaof gender and media is scarce, as for instance in most developing countries,content analysis is instrumental in providing a general impression of therepresentation of women and men. Also, as a resource for policy andprogramme development - arenas that notoriously prefer 'hard data' -

content analysis is an invaluable means of convincing decision andprogramme makers of the necessity for a diverse portrayal of women andmen. Cast within a solid theoretical framework, content analysis can shedlight on social and cultural matters of representation. The study of culturalindicators is often mentioned as an example, be it not for its relevance tofeminist issues. In this perspective a culture communicates with itself bymeans of its total mass media output with television being the prime massmedium. Through large scale content analyses carried out over severalyears, researchers working in this tradition have tried to show thatAmerican television as a whole carries consistent messages that .devalue

Page 5: feminism

74 Feminist media studies

WO!11qsu, blnQI(~unci the poor by undorrcprescntlng them und by dispropor.tlollilloly IIHIlch1M thom tho vlCtfrl1~ of vloiencil (Oerbner ot nl., 1.980),INtlnlll,r PIII.IClI'119bnv,",bO.CU1eHlnbH8hod by "thaI' l'cscurehl31'Saudfor othorliIalll'~M, II would MalHI bCilllltlc Ih8 p"hH to dony l.hGrOlelVIH1COoC:theseMIHlh1/,111CHI thl!l hll.\8 of' II hlok of Ih(fl()I',Uonl90phll1t1clILIon,Ilnesllilct orthoIII"U\Wt1Itl!J.of! medl" fhrllllllM, IiIl\llur'bHlIne!Incllvlclunl pt'cIg!'ummIil9OJ' thedlln!'t'l"U1l1.Ywith sUl1jGctlv,lI1~dll1dXI~li)rlt;ll1CCl8 or audiences(of,McQunH,191\7),Whllt should be cmphnAh:ocllsthnt tho Ilnnlysis of such brond andwldesf'reud trends In media output does not produce self-evident mean-ings, only data upon which to base a substantial discussion of the question:'So what?' (Cf. Fiske" 1980: 127).

Symbols for reality: semiotics

Semiology or semiotics has become quite popular in feminist mediacriticism because of its ability to unravel structures of meaning beyond themere presence or absence of women in cultural forms. Drawing on variousdisciplines in the humanities, such as philosophy and linguistics, it is not aneasy field to get acquainted with, espeCially since the relevant authorsdiffer in their interpretations and often write in an extremely abstractmanner.4 The purpose of this chapter therefore is not so much to explainthe intricacies of the field as it is to show what semiotics can do in terms ofanalysing meanings'.

The American philosopher Peirce and the Swiss linguist de Saussure areconsidered to have laid the groundwork for semiotics but it was not untilthe 1960sthat the work of Roland Barthes introduced semiologyto wideraudiences applying it to various forms of popular culture, ranging fromtoys, hairstyles and chips to cooking, soap powders and the new Citroencar (Barthes, 1957). As the examples show, almost anything can be con-sidered a sign, but some obvious examples of signs and sign systems are,for instance, words combined in the sign systems of language or imagescombined in the sign systems of art, photography, film, television etc.

,,- Historically, the female body has provided very powerful 'signs'(cf. Warner, 1985). The French nation, for example, is symbolized by'Muriunnc', II virtuous femnle revolutionary. The Statue of Liberty inthe United Stutes. tho first encounter of millions of immigrants withlhlo\AulOl'icUI1 vull1o~ of freodom one! indivIdualism - is a woman. In the

Wo.t:~nlworld ,'u8tloClIHl'oprCIIIIU1l0c]in the imt.lgeof 0.blind'olded womanht11dluUn bl\llmaca,WQII10n Iwve,!llso 8~mbolizec.lless noble ideas, in(~hl'l"'ItUIH~1I'eu IiIMllllle., bc~hlUlu,.with I~ve'smistukoin tustingthe.."piC', In 01111111mcultll tlUlput, rl'.lUJIIICIc11llrllctol'~ Itre ofton constnlcted asIII ~lIfltYI1.Melf "II'tII' 111,,1vletl. .VI"wlilrs f!\l11i1i!lrwith tho codcsof' Anglo-

tf i!I\I"IU.I'fOIHlltIlv!8lulI wUI +know' for In8tl\nc~thul il1 cinema and television a:;~j:)\l\JlihhlC'1mlC!IiJlrl IIrQIsod 111whltG uHUlIlly sigl1U1osinnoeollce und probity,jllllll1l1 tllll'k huh'lid W()ll1cntt.ind to sIgnify clungerand sexuality. In the

11.

!I

Media texts and gender 75

SIgnify by

I!~nntplo

Indox

oonn(lotlon

Prooo..

"make.flrtHyrnptom.cII1I6/ullI

o.nrluuru0\11

Figuro5,2 .TImw(l!\'fUJct.\'ol,t/sm'(lJaryar,ICi82:J$)

..

American soap opora Dynasty, whieh was extremely popular around theworld in the 1980s,this particular set of signswasemployed to setup one ofthe main antagonismsin the series, betweenblonde virtuous Chrystle(played by Linda Evans), married to business tycoon Blake Carrington andusually dressed modestly in pastel colours, and evil Alexis (Joan Collins), jBlake's ex-wife, a brunette often extravagantly dressed in conspicuous icolours like red or black. Media output is ridden with such signs and it is up lto the semiologist to analyse how particular combinations of signs in signjsystems construct particular meanings. In doing so, semiotics appwachesevery sign system as if it were studying the lexicon and grammar oflanguage, focusing on the choice and particular combination of words(signs). All sign systems are treated as 'texts', which explains the commonreference to media output as media texts.

In order to understand how a typical semiotic analysis is conducted, it isnecessary to take a closer look at the key concepts involved. According tode Saussure, who looked at language primarily, a sign consists of twoelements. Its physical appearance in, for instance, a combination of lettersforming the word 'rain' is called the 'signifier' , whereas the concept it refersto - a particular type of weather is called the 'signified'. In the case i:Jfwords the relation between signifier and signified is completely arbitraryand based on convention rather than a self-evident relation.shipbetweenthe four letters r a i n and water falling from the sky. Considering othersign systems, however, the relation between signifier and signified is lessarbitrary. A statue or a picture, for instance, signify by resemblancebetween signifier and signified, whereas red spots on a human body maysignify a given disease. Peirce, therefore, distinguished between threeaspects of signs: iconic, indexical and symbolic (Figure 5.2).

Signs in isolation are hardly ever meaningful, they derive their relevancenot only from the particular articulation of signifier-signified, but also fromtheir relation to other signs. Those relations can be syntagmatic, involvingthe particular combination of signs, or paradigmatic, pertaining to therelation of a sign present within the sign system to its implied absentcounterpart. It is in the latter respect that semiotics can be related to thestructural analysis of folk and fairy tales (Propp, 1923). Structuralismassumes that binary oppositions like female/male, good/evil, black/whiteunderlie sign systems and broader cultural systems like, for instance,kinship relations (Levi-Strauss, 1962). For structuralists any particular sign

Symbol Icon

convontlon rOllombl/Jnoo

wordu 1)11311.11'&statuI

rmJ.llollrn onn 0011

Page 6: feminism

76 Feminist media studies

Slgnltlllr Slgnlflod

oloull up (Inooonly) IntlmMoyrnlldhllll .hot (11100t0'1bodyI por80ntllralatlon.hlpIOl1U@hcn(""!tlng IInd al,ofhat8ro! publlo dlMtmOIllull Bhat(full body or IJlilrOOn! oooial rllintionahipIml down (olllnwrnlookll down) powlilr, Iluthoritypfln up (olHTIOrftlook. UrJ! IIrnlllintlO.,wotlknolll

J~Ii!uro,5.3 CIIIIWQanala.I'and 111()\ICI1Umt.I'(/Jerger, 1982:38)

thus derives its meaning not only from its relation to other signs within thesame system (syntagma) but also from its relation to its absent opposite(paradigma). .

In addition to the particular articulation of signs, 'codes' and 'conven-tions' shared by the members of the culture from which the sign systemoriginates provide clues as to how to understand sign systems. Berger(1982), for instance, gives an example of the way western codes of bodylanguage bestow meaning on camera angles and movements (Figure 5.3).Other codes that usually give meaning to the articulation of signs - as weshall see later in this chapter - are codes of femininity, masculinityandheterosexuality.

Sign, syntagma, paradigma, code and convemion are the key elements inthe different processes of 'signification', the way signs become meaningfulwithin a given culture. Barthes (1957) distinguished between denotationand connotation, or first and second order signification. Denotation is ofparticular relevance to the semiotics of de Saussure which focuses onlanguage. It concerns the direct relation between signifier and signified andits manifest referent in reality and usually involves a description of whatcan be seen or read in a text. The denotative meaning of 'black cat', forinstance, would be an animal with a certain colour. However, the sign'black cat' carries important other connotative meanings as well. In somecountries, like The Netherlands, it is considered an omen of bad luck, butin other countries it is thought of as a signifier of fortune and happiness. Insome feminist circles the black cat is thought of as a signifier of spirituality.Connotation or second order signification thus concerns the latent culturalvaluesarid beliefs expressed by a sign or sign system. This can take on theform of assboiations or of complete narratives, labelled as 'myth' bynllrthes, A r~oosnition andunderstandingof the various codesunderlyingthe purtlou]ur urt!oulllUon Qf ~ignsis necessaryfor secondorder significa-!:Ionto ~oJ'k,Con~~qu(lnny,the !\nnlysisof connotativemeaningrequires aIh(!1'(Hll.lhkntlw!odst'lof tho O~IJtUl'CLhesl,\\nsystemoriginnte~from. Finally,Itl.kla lIud I [IU'(lIry(lP7B) cllst\nQul6hthirdordeu'~Isnlficutionor ideology,omlt\iIIlvlnaClC)unotlltl()\IIInd myth 1'18tht! ItHll1ifest expressions of un11I

.

ICj.I'I~h11Htl'lIGlIIII'.

tr!,.I'l'Inel1,..

11I1 thedomlnonlldcology.TheysummarizeIhocl1tr'I'.nt pl'OC~.MONat NIYI1HlcaClcH1In the Rch~lI1eshown In Fisure 5.4.

WhllGtl1l.1\l1ll1l\Y Hound I'll thoI' ulJstrllcL, semiotic IInulysis can be seen asII rClrIllllllzutl.onof the interl'I'etlltive nctivltles ordinary human beings

Media texts and gender 77

First ordor Secondorder Third ardor

Nature ~ . Culture

Figure 5.4 Processes of signification (O'Sullivan, 1989: 217)

+I

undertake incessantly. Imagine, for instance, a women's group monitoringchildren's books for their portrayal of gender roles. In one book theyencounter the drawing shown in Figure 5.5. The women's group finds thisdrawing sexist and stereotypical for various reasons. They consider dollsand miniature kitchens as toys that are given to girls only, signifying thegirls' futures as mothers and housewives. They argue that the jJicture wouldbe less sexist if a boy was included too and they claim that the drawingwould be emancipatory if there was only a boy playing with the doll and thekitchen. Some women in the group claim that the image of the happyhousewife conveyed by the drawing, does not fully resemble their ownexperience, since some of them do not like kitchen work and others arequite happy when the children (doll) are away playing at the .neighbours.

In such reasoning, all the elements of a semiotic analysis are present. Letus first isolate the signs in the drawing: a little girl, a doll and a miniaturekitchen.s These signs all have iconic elements in the sense that thesignifiers directly resemble the signifieds. The miniature character of thedoll and the kitchen have important indexical elements as well, connectingthe world of toys to the real life existence of mothers and housewives. Thecode of traditional gender roles underlies that connection and is easilyrecognized by members of many cultures and it is precisely what thewomen's group objects to. Moreover, the signs are combined in such a waythat they conveythe imageof a girlplayingcontently, referringto the mythof the 'happy housewife' (second order signification or connotation). ,Another syntagmaticrelation of the signs-the girl fol'instance destroyingthe doll and setting fire to the kitchen - would have radically altered themeaningof the drawing:The paradigmaticrelation of the signs is!

~

7"I

...I~.

+i'\',...

~

.;.

Page 7: feminism

78 Feminist media studies

Figure 5.5

important as well and also recognized in the women's group analysis of thedrawing. Why is a girl playing with a doll, and not a boy? Or why is the girlplaying with a doll and a kitchen and not with a car and a computer? Thesesigns of 'masculinity' are not present in the drawing but they do playasignificant role in the analysis of such a drawing as sexist and stereotypical.Underlying the drawing, one may argue, is the dominant ideology offemininity which structures its overall meaning (third order signification).

Ordinarily, the texts subjected to semiotic analysis are much morecomplex than this drawing and contain elaborate narratives and additional

signs systems such as colours, music and editing codes. In such cases a 'layanalysis' will neither be sufficient nor successful, and semiotics provides ameans for a systematic assessment of the processes of signification in thetext. It is important to realize, that despite the systematic nature of thesemiotic approach, there is no clear methodology of semiotics as there iswith content analysis. One could, however, translate the different elements

of semiotics into systematic 'steps' to carry out the analysis, the first stepbeing the identification of the relevant signs and their dominant aspects(iconic, indexical or symbolic). Then one continues with examining theparadigmatic combination of these signs, for instance by asking what theirabsent opposites are, and how they relate to each other syntagmatically.

~~ "~-......

1

7II~

1.I.

I..I

,.i1!~

,~

Media texts and gender 79

I+

This will also imply an analysis of the codes by which the combination ofsigns is governed. One thus arrives at an understanding of the differentprocesses of signification in the text: denotation, connotation, myth andideology.

These 'results' clarify how single discursive expressions, such as aphotograph or an advertisement, generate meanings, but they provide noinsight into how often that particular process occurs, as would be theconcern of traditional content analysis. Also, one should not expect asemiotic analysis to produce the definitive meaning of a text. Since thecodes that confer meaning on to the syntagmatic and paradigmaticcombination of signs are culturally specific, signification will never becompletely unambiguous or univocal. The multi-accentuality of signs,'consisting in the capacity every sign has to signify more than one meaning,depending on the circumstances of its use' (O'Sullivan et a1., 1989: 144)ensures that the meaning of the text is never nor determined only by thefeatures of the text itself. Even though particular syntagmatic and paradig-matic combinations of signs tend to produce 'closure' of meaning, texts areprincipally 'polysemic' , that is, they contain multiple meanings that need tobe actualized by audiences, involving the culturally competent applicationof codes governing the text. Although the acknowledgement of thepolysemic nature of texts - inspired by post-structuralism (Seiter, 1992) -has recently undermined the semiotic project and directed researchers tothe interpretive activities of actual audiences rather than the processes ofsignification ip the text (Moores, 1990), semiotics remains a powerful toolto understand how sign systems in mass media can evoke emotions,associations, fears, hopes, fantasies and acquiescence.

I

~1

1I,-r

,r

. ,

Advertising

Signification has developed into an art form in advertising, making itprobably the most popular object for semiotic analysis. As a concentratedform of communication, advertising needs to present its message in anextremely short time span, and depends heavily on the successful exploitationof the connotative power of signs. An advertisement cannot afford to becast aside or leafed over; it must stand out in the flow of signs that bombardus daily. For the target group to recognize immediately what is beingexpressed, it is necessary that advertising draws from relatively commoncultural symbols and meanings:

Advertising is itself a 'multiplexing' form that absorbs and fuses a variety ofsymbolic practices and discourses. The substance and images woven intoadvertisingmessagesare appropriated and distilledfrom an unbounded range ofcultural references. Advertising borrows its ideas, its language, and its visualrepresentations from literature and design,from other mediacontent and forms,from history and the future, and from its own experience; then it artfullyrecombines them around the theme of consumption. (Leisset aI., 1986:144-5)

The idiosyncratic nature of semiotic analyses prevents any kind of

'-"-~--~--

Page 8: feminism

80 Feminist media studies

generalization of advertising analysis. In Goffman (1976) and Williamson(1978), for instance, one can find exhaustive analyses of a variety ofadvertisements that belie any attempt at summarizing. Instead, I shallillustrate the semiotic approach by analysing two advertisements featuringwomen from particular ethnic backgrounds. The aim of the analysis is tounravel the ideology structuring the specific articulations of gender andethnicity in these advertisements (third order signification).6

The first advertisement (Figure 5.6), for Japan Air Lines, was publishedin an American news magazine and is aimed at Executive Class businesstravellers, predominantly men. Essentially we see a young Japanesewoman holding an oriental-style teapot in a deferential posture and smilingdirectly at the camera. The advertisement might seem at first to be justanother example of a common representation of women: as the obligingservant attending the needs of (male) passengers. Her plain dress and herfriendly smile, which directly addresses the audience, support such adomestic image. However, other signs in the advertisement do notconvincingly corroborate that connotation: why is she serving tea - and notfor instance coffee or whisky? What is the meaning of the woman we cansee in the background, dressed in a kimono and holding a fan, and why isthere also a fan featnred largely in the lower right corner of theadvertisement, next to the text? These are not mere decorative elementsbut are linked to the central image of the stewardess by the symbol of JAL- a heron - which appears on the fan, the teapot, the stewardess' dress andthe fan of the woman in the background, and by the colour red, whichappears on the fan and also couples the woman in the background to thestewardess. The slogan of Japan Air Lines, printed at the bottom, controlsthe meaning of those signs: 'Attentiveness. With us, it's a tradition.' Andin the text we find another clue: 'attending to the comfort of guests is aheritage refined over a thousand years'. With this emphasis on tradition,

. the teapot, the fan and particularlythe womanin the background begin to. make sense. They are signifiers of tradition, and more specifically of the

tradition of the geishas, the famous Japanese companion ladies whoreceived exquisite education and training to provide intellectual andspiritual pleasure to their male customers. For the western male business

. traveller at whom this advertisement is aimed, the prospect is of a servicewhich to his own culture is both appealing and exotic, especially since theadvertisement also opens up space for fantasies about the kind of servicethe stewardess will offer. The myth of the geisha has always beenambiguous in this respect. While the professional pride of the geishaderives from her intellectual and spiritual capacities, the metier has beensurrounded by suggestions and fantasies of sexual pleasure as well. Theambiguity of the geisha's image is enlarged by the text: 'And most of all,you appreciate it in the way all JAL attendants seem to anticipate yourevery need.'

Through the different orders of signification the advertisement articu-'lates a discourse of gender and ethnicity that constructs Japanese femin-

\

I"IrlI

!:

.II~

I.

i'..Ij"j~,,,'.

Media texts and gender 81~

~

i;j,,

t

1

II~

..

N\0nlivenO5S it begins "'" Inom"nl"Oli are W€loom".-j

Int" 1M "1""'1011$.Jt,L "-,.culiv" CI."" cabin with" 911"'°01

crmrnp"\Jn" \\Ju nolic" it in tM d"lailS: 1)", comlorl.ensulln9deSign Of II.., "",my """I;ng. Tho oleganl pmS'2olalioo 01 YC>tJf

clloic" 01 .Japanes. or Wesler" clItSi"". And m"ol 01 ali.you

appR'f('IHH'1I in 111"wa" nil.tAL "1I."eln,,ls "Hom to anliGipnH1

your.M"Y""",t Because fortho",.I)llen<:hnulo 11"1ClJn~ot'l 01

guesls i5 aherilage ff*ned OW' a It>.ousand ,,""'9. For more,nlorrn"Ii"" C>",JAL. j~"11"" to mce;\IT? OU' Inlmrnaf"'" ,IAL

S"!IVi<;" GUKJI! 01 (<0018:11;.224(1. Ex! 144.'"""'0'"".""'0""'"

A=

..~

fJt

.~

. , ,._,.,.~ , , ~-.,~ , , ,, , , , ~

...

1"II

.I,.

1i

JAPAN AIR L.INES

Attentiveness. With us, it's a tradition.

~.

Figure 5.6

Page 9: feminism

82 Feminist media studies

inity as geared to everything a man could possibly desire. Attentiveness,forbearance and complaisance are suggested characteristics of the sexualityof Japanese women who know from a thousand-year-old tradition how toserve the needs of men. It has been with them so long that it comesnaturally to them, so to speak.

The second advertisement I want to examine (Figure 5.7) was publishedin a Belgian women's magazine and signifying practices similar to those ofthe JAL advertisement 'operate. T~e first order signification or denotationdoes not make much sense. As asserted by the text at the bottom of theadvertisement, Safari is a sweet liqueur with 'the taste of wild nature' andproduced according to an ancient African recipe. There is a bottle and twoglasses of Safari ('African drink') in the centre of the picture. Heterosexua-lity is implied by the two glasses, one a long drink and one on the rocks. AnAfrican woman holding a small tiger in her arms is smiling at us frombehind a blind. In the background we see zebras moving across a plain anda sunset behind distant hills. Black, white and a variety of browns are thecolours that produce coherence in the otherwise dispersed signs of theadvertisement. The brownish tints appear in the liquid, the landscape inthe background, the tiger and the skin of the woman. Nature, drink andwoman are further coupled by the zebra motifs present in the woman'sdress, on the label of the bottle and the zebras themselves. Naturalness andwildness - 'the taste of wild nature' - are thus embodied by the womanholding the tiger, the scenes of nature and the zebras. A final element thatneeds examination is the blind that tries to hide the woman from sight.Why is the woman behind the blind instead of the zebras, for instance?Why is she holding a tiger? What does her smile mean? Is she inviting us oris there danger in her smile as the tiger seems to hint at? The blind preventsus from looking at her directly and symbolically inhibits immediate accessto her. A sense of mystery is created. However, the articulation of genderand ethnicity is significantly different from the JAL advertisement.Whereas they share the exotic quality, African femininity is constructed as .wild and close to nature while Asian femininity is modest and deferential.Both share, however, a reference to tradition. In the Safari advertisementfemininity is linked with an 'ancient African recipe'. The JAL advertise-ment couples femininity with the age-old heritage of the geishas.

The two advertisements have common characteristics: the emphasis ontradition and mystery produces their exotic spirit. Obviously these signi-fiers draw from some common frame of reference that is meaningful in thewhite patriarchal culture they both originate from. This is an issue of thirdorder signification, that is, 'the way that the varied connotations"and mythsfit together to form a coherent pattern or sense of wholeness, that is, theway they "make sense" is evidence of an underlying invisible, organizingprinciple - ideology' (O'Sullivan et al., 1989: 217). The ideology underly-ing the advertisements does not speak for itself, however, and can beapproached from different angles. Judith Williamson (1986), for instance,has interpreted exotic advertisements from a neo-Marxist point of view.

Media texts and gender 83

.'

De smaak van wilde natuur

( Figure 5.7

' '---"""""""~ ""~"~""'" """"""'.."' """""""'"",,".""." <>.'.."""~~,,~~~--'..

Page 10: feminism

84 Feminist media studies

.r

She starts from the assertion that capitalism exploits and controls workersand colonies, suppressing their different interests and value systems. Thefunction of ideology is to contain the resulting antagonism and preventthem undermining the dominant order. This is achieved by creatingdifferences that conceal or transform the true antagonisms: 'The supremetrick of bourgeois ideology is to be able to produce its opposition out of itsown hat' (1986: 100). Williamson argues that 'woman' and 'nationality' arethe main signifiers in this cover-up, suggesting natural and eternaldifferences that cut across class. 'Questions of class power frequently hidebehind the omnipre.sent and indisputable gender difference' (1986: 103).Williamson analyses exotic advertisements to show how particulararticulations of 'woman' and 'nationality' realize the transformation ofstructural antagonisms in mere differences of style and preference. Shecontends that the requirements of the capitalist economy have discardedsuch human values as sensitivity, care, connectedness and respect fornature as useless for public life. However, the common complaints aboutwestern capitalist societies as being cold, unaffectionate and dominated byrationality and considerations of efficiency indicate a dissatisfactionwith the system, and a collective need for the lost values. In commodifiedform these values are still available to us, as the exotic advertisementspoint out:

It is as if western capitalismcan hold up an imageof freedom and fulfil1mentandsay: 'Look our system offers this' . . .Different systems of production which aresuppressed by capitalismare then incorporated into its imagery and ideologicalvalues: as 'otherness', old-fashioned, charming, exotic, natural, primitive,universal. (1986: 112)

The transformation of a collective aversion to the system into consumerneeds and products can be achieved perfectly by the exotic articulation of'woman' and 'nationality' since - accordi~g to Williamson - 'the most likelyOther for a white working-class man, is either a woman. . . or a foreigner,in particular somebody black' (1986: 103).

Williamson's analysis duly points to the overall ideological function ofexoticism, but tends to overlook the specificities in {he phenomenon andthe unmistakably sexual undertones in it. The recurring references to blackfemale sexuality as being temperamental and wild - something to be fearedand desired at the same time - cannot be explained within Williamson'sneo-Marxist framework, but needs rather to be informed by history andpsychoanalysis. In Europe, the fascination with black sexuality has beentraced to the earliest reports about the crusades and has been a continuoustopic in the literature of travellers to Africa and the Orient (Kabbani,1986). The popular and widespread appeal of the image of the 'wild savage'that feeds western discourses of black female sexuality, is said. to haveserved a double purpose for European colonialism. The image of the 'wildsavage' provided a wonderful occasion to fantasize about all that wasforbidden, projecting 'upon colonial people the obscurities of their ownunconscious' (Mannoni, 1950: 19). At the same time, imagining the

I'" ..., ~-t

~

Media texts and gender 85

Semiotics Content analysis

transmissionCommunication model construction

Object latent contentsingle texts

qualitativehumanities

manifest content

large body of texts

quantitativesocial science

Method

Results signification occu rrence

i1-

ij,

!It'I

t1I+

1

t

iIi,'t,

Figure 5.8 Differences between content analysis and semiotics

..

peoples of the colonies as sexually voracious and insatiable legitimizedEurope's colonial enterprise as a project of 'civilization' instead ofeconomic exploitation. Likewise, in the United States the myth ofravenous black sexuality provided an excuse for white slave owners to rapefemale slaves, and made the lynching and castration of black men possibleuntil long into this century (Nederveen Pieterse, 1990).

Although only exemplary, the analysis above of the two exotic advertise-ments does illustrate the advantages and disadvantages of a semioticanalysis. Because of the amount of elucidation and detail entailed, it canonly focus on single texts or a limited sample at the most. While being closeto the 'true' nature of texts and possibly also to actual audience activitiesand interpretations, the volume of 'findings' resulting from semiotics isusually much larger than the text analysed (cf. McQuail, 1987: 189). Mostimportant, the quality of semiotics depends to a large extent on the lresearcher's knowledgeof cultural codes. The exotic advertisements,for I)instance, make much more sense when connected to the economic andhistorical peculiarities of white patriarchal culture and European colonial-

(ism, and will possibly gain from a more detailed psychoanalytic angle aswell (for example, Doane, 1991). A good semiotic analysis thereforequickly develops into a broad cultural critique. ~,

In this chapter, two approaches to the analysis of gender discourse inmedia texts were discussed: content analysis and semiotics. Even whenlimited to the genre of advertising, the overall results of both are hard tosummarize. The usefulness and merits of each approach depend on the

specific questions being asked. The differences between both approacheshave been identified in various ways. Barthes (1957: 120) has claimed that'semiology is a science of forms, since it studies significations apart fromtheir content'. Likewise, Seiter (1992: 31) claims that 'semiotics first askshow meaning is created, rather than what the meaning is'. Looking at thedifferences within feminist media research, it seems that content analysisemphasizes the manifest working and non-working roles women areportrayed in and their visual function as a decorative element while

-

...,

..

.l.

+

,~ ~ ~~..~

Page 11: feminism

86 Feminist media studies

semiotics draws our attention to the power of 'woman' as signifier of .almostanything between virtue and vice, desire and fear. Other differencesdiscussed in this chapter are summarized in Figure 5.8.

(' ' Some researchers have tried to combine elements of content analysis and\ semiotics. Jhally (1987), for instance, tried to quantify the occurrence of

cultural codes like individuality, romance and friendship in advertisementsintended for a female and a male audience. She found that codes of beauty,family and romance occurred more often in advertisements for female

.audiences, while ruggedness and fraternity dominated in advertisementsaimed at men. In general, it seems that in such combined approachescontent analysis is used for inventory purposes while semiotics is employedto give depth to and illustrate the overall quantitative results (cf. Leiss, etal. 1986). Within the theoretical framework underlying this book, it mightbe useful to employ content analysis for inventory purposes. However,conceptualizing gender as a fragmentary discourse that is expressed andreconstructed by technologies of gender such as mass media - a set of

'symbols for reality' - implies a semiotic approach that foregrounds thedifferent signs and contradictory processes of signification in media texts.

Notes

1. The reader with an interest in such review studies is referred to Butler and Paisley, 1980;Ceulemans and Fauconnier, 1979; Gallagher, 1980, 1985. The current impossibility of such areview project is shown by the absence of more recent references in this area.

2. r am indebted to Laura Emmelkamp for these references.3. Table taken from Fiske, 1980: 122.

4. Useful introductions have been produced by Berger (1982), Fiske (1980) and Hartley(1982).

5. The fact that this is a drawing and not a picture can be considered a sign too.6. r collected these and other advertisements by making random clippings from magazines

over a number of years and have not kept track of the precise information concerning theirpublication.

I. rtiI;-

'~

6

Spectatorship and the Gaze

r

!i+LI.iIi...

A core element of western patriarchal culture is the display of woman asspectacle to be looked at, subjected to the gaze of the (male) audience.Pornography is the most obvious genre built on the exhibition of women'sbodies as objects of desire, fantasy and violence, but the 'objectification' ofwomen is not exclusive to pornography. The incorporation of women'sbodies as decorative ingredients in advertisements of drinks, tools and -most notoriously - cars is common practice, while in TV game shows theassistant to the quiz master is the predictably attractive, scantily dressedblonde. Fashion and lingerie photography as published in g'lossy women'smagazines has moved beyond the mere presentation of new styles andlines, and has taken over many of the codes formerly restricted to soft corepornography. Hollywood cinema has a long standing tradition of construct-ing women as a spectacle for voyeuristic pleasure.

This common feature of popular and high culture alike - for in 'art'women's bodies have been exploited in similar ways - suggests that inwestern society to be looked at is the fate of women, while the act oflooking is reserved to men. Even if women do the looking they do not seemto do it through their own eyes. John Berger writes in his classic studyWays of Seeing:

I0I-I

~i~

Men act and women appear. Men look at women. Women watch themselvesbeing looked at. This determines not only most relations between men andwomen but also the relation of women to themselves. The surveyor of woman inherself is male: the surveyed female. Thus she turns herself into an object - andmost particularly an object of vision: a sight. (1972: 47)

i~

...i

+.' Many feminist authors have pointed to the devastating effects that this

'to-be-Iooked-at-ness' might have on ordinary women. Wendy Chapkis(1986), for instance, discusses how cultural messages about beauty andslimness negatively affected her own and other women's sense of well-being in general and of being 'truly feminine' in particular. More recentlyNaomi Wolff (1990) has made similar arguments in her best selling bookThe Beauty Myth. According to Wolff, the cultural industry's prescriptionsfor women to be beautiful and slender - a pleasure to look at - hasproduced a generation of American girls and women who suffer fromeating disorders like bulimia and anorexia nervosa.

Obvious as it may seem that for women there is little good or pleasurablein 'looking' as it is structured in western culture, it is an argument thatdenies the possibilities and historical existence of a positive and -pleasur-

..

'. ~"~~"-"--.F~~~ ~'-'---'~-'~~' ',~ ~.~


Recommended