moive assignment

Tony Yan

Cohen-InterpretingSinginintheRaincopy.pdf

Home >History homework help >moive assignment

Reinventing Film Studies

Edited by Christine Gledhill Professor of Cinemo Studies, Stoffordshire University,

Stoke on Trent, UK

and Linda Williams Director of Film Studies, University of California,

Berkeley, California, USA

~ ARNOLD A member of the Hodder Headline group LONDON Co-published in the United States of America by Oxford University Press Inc., New York

Firstpublished in Great Britain in 2000 by Arnold, a member of the Hodder Headline Group, 338 EustonRood, london NWl 3BH

http://www.arnoldpublishers.com

Co-published in the United Statesof America by Oxford University Presslnc.. 198 Madison Avenue, New York, NYl 0016

Chapter 1© Geoffrey Nowell-Smith Chapler 10 © Christine Geraghty Chapter 18 © Miriam Hansen

Thanks are due to Oxford University Pressand Christopher Williams for permission to reproduce 'Aher the classic, the classical and ideology: the differences of realism' Screen 35(3). 275-93, 1994 ForMi

All rights reserved. No port of this publication may be reproduced or transmitted in any form or by any means, electronically or mechanically, including photocopying, recording or ony informotion storage or retrieval system, without either prior permission in writing from the publisher or a licence permitting restricted copying. In the United Kingdom such licences ore issued by the Copyright licensing Agency: 90 Tottenhom Court Road, london Wl POlP.

The advice and information in this book are believed to be true and accurate at the dote of going to press, but neither the outhors nor the publisher can accept any legol responsibility or liability for any errors or omissions.

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British library

Library of Congress Cataloging-in-Publication Data A cotalog record for this book is available from the library of Congress

ISBN 0 340 67722 B (hbl ISBN 0 340 67723 6 [pb]

1234567B910

Production Editor: Rode Rcdojicic Production Controller: loin McWilliams Cover Design: Terry Griffiths

Typeset in 11 on 13 Minion by Phoenix Photosetting, Chatham, Kent Printed in Great Britain by JW Arrowsmith ltd, Bristol

What do you think about this book? Or any other Arnold title? Pleasesendyour commentsto [email protected]

For Miriam

Contents

List of figures List of contributors

xi Xln

Introdnction I

Part 1: REALLY USEFUL THEORY Editors' introduction 5

1 How films mean, or, from aesthetics to semiotics and half-way back again GeoffreyNowell-Smith 8

2 Why theory? Gill Branston 18

3 Film theory and the revolt against master narratives Bill Nichols 34

4 Case stndy: interpreting Singin' in the Rain Steven Cohan 53

5 Who (and what) is it for? TessaPerkins 76

Part 2: FILM AS MASS CULTURE Editors' introduction 97

6 Dream/Factory Jane M. Gaines 100

7 The publicness of cinema James Donald and Stephanie Hemelryk Donald 114

8 The politics of cultural address in a 'transitional' cinema: a case study of Indian popular cinema Ravi S. Vasudevan 130

viii Contents

9 Reception theory and audience research: the mystery of the

vampire's kiss Henry Jenkins

165

10 Re-examining stardom: questions of texts, bodies and

performance Christine Geraghty

183

Part 3: QUESTIONS OF AESTHETICS Editors' introduction

203

11 After the classic, the classical and ideology: the differences of realism

Christopher Williams 206

12 Rethinking genre Christine Gledhill

221

13 Judging audiences: the case ofthe trial movie Carol f. Clover

244

14 Introducing film evaluation Noel Carroll

265

15 'Style', posture, and idiom: Tarantino's figures of masculinity

Sharon Willis 279

Part 4: THE RETURN TO HISTORY Editors' introduction

297

16 What is film history?, or, the Riddle of the sphinxes ~

Vivian Sobchack 30::;

17 'Animated pictures': tales of cinema's forgotten future, after

100 years of films Tom Gunning I

316

18 The mass production of the senses: classical cinema as vernacular

modernism Miriam Bratu Hansen

332

19 Discipline and fun: Psychoand postmodern cinema Linda Williams

351

Contents ix

Part 5: CINEMA IN THE AGE OF GLOBAL MULTIMEDIA Editors' introduction 379

20 Film theory and spectatorship in the age of the 'posts' Robert Starn and Ella Habiba Shohat ~~

21 Digging an old well: the labor of social fantasy in a contemporary Chinese film Rey Chow 402

22 Facing up to Hollywood AnaM. Lopez 419

23 The end of cinema: multimedia and technological change @Anne Friedberg Index 453

52 Bill Nichols

Rony, Fatimah Tobing, 1996: The third eye: race, cinema and the ethnographic spectacle. Durham, NC: Duke University Press.

Rosenstone, Robert A, 1995: Visions olthe past. London and Cambridge, MA: Harvard University Press. Rotha, Paul and Griffith, Richard, 1950: The film till now and the lilm since then. New York: Funk

and Wagnals. Russo, Vito, 1981: The celluloid closet: homosexuality in the movies. New York and London: Harper

and Row. Said, Edward W., 1978: Orientalism. New York: Pantheon Books. Shah at, Ella and Stam, Robert, 1994: Unthinking Eurocentrism: multiculturalism and the media.

London and New Yark: Routledge. Silverman, Kajo, 1983: The subject of semiotics. New York: Oxford University Press. Silverman, Kaja, 1992: Male subiectivity al the margins. New York: Routledge. Sobchack, Vivian, 1992: The address 01 the eye: a phenomenology ollilm experience. Princeton,

NJ: Princeton University Press. Sobchack, Vivian, 1996: The persistence of history: cinema, television and the modern event.

London and New York: Routledge. Soussloff, Catherine, 1997: The turn to visual culture: on Visual culture and Techniques of the

observer. VisualAnthropology Review 12(11:77-83. Stam, Robert, Burgoyne, Robert and Hitterrnon-lewis. Sandy, [eds}, 1992: New vocabularies in Iilm

semiotics: structuralism, post-structuralism and beyond. London and New York: Routledge. Studlar, Gaylyn, 1988: In the realm 01 pleasure: von Sternberg, Dietrich and the masochistic

aesthetic. Chicago, IL: University of Illinois Press. Taussig, Michael, 1993: Mimesis and alterity: a particular history of the senses. London and New

York: Routledge. Thompson, Kristen, 1981: Eisenstein's Ivon the Terrible: a neoFormalist analysis. Princeton, NJ:

Princeton University Press. Virlllo, Paul, 1988: War and cinema: the logics of perception. New York and London: Verso. White, Hayden, 1987: The content of the form: narrative discourse and historical representation.

Baltimore, MD: The Johns Hopkins University Press. Williams, Linda, 1989: Hard core: power, pleasure and the 'Irenzy 01 the visible'. Berkeley, CA:

University of California Press. Williams, Patrick and Christian, Laura, (eds), 1994: Colonial discourse and post-eolonial theory.

New York: Columbia University Press. Williams, Raymond, 1976: Keywords. London: Collins. Wilson, Rob and Dissanayake, Wimal, [eds]. 1996: Global/locol: cultural production and the

transnational imaginary. Durham, NC: Duke University Press. Wollen, Peter, 1969: Signs and meaning in the cinema. Bloomington, IN: Indiana University Press

[also published 1972]. Wollen, Peter, 1993: Raiding the icebox: reflections on twentiefh-century culture. Bloomington, IN:

Indiana University Press. Wollen, Peter and Cooke, Lynn, [eds], 1995: Visual display: culture beyond appearances. Seottle,

WA: Bay Press. Wood, Robin, 1968: Howard Hawks. London: Seeker and Warburg/New York: Doubleday. Zizek, Slovo], 1992: Looking awry: an introduction to Jacques Lacon through popular culture.

Cambridge, MA and London: MIT Press. Zizek, Slovo], 1994: How did Marx invent the symptom? In Slovo] Zizek [ed.}, Mapping ideology.

London and New York: Verso, 296-331.

4 Case study: interpreti ng Sitiqir.' in the Rain

Steven Cohan

From its original reJease in 1952, when it was overshadowed by the greater success and prestige of An American in Paris released the year before, to its ranking as fourth three decades later in Sight &Sound's listing of the 10best films ever made (Wollen, 1992: 9), to its commodification as the signature film of MGM's Golden Era in a series of anthologies produced by that studio's various corporate successors, Singin' in the Rain has steadily grown in popular and critical appeal. It is still exhibited in an assortment of venues, not only old downtown movie palaces and revival houses, but also museums, home video, cable television, college classrooms, even the Hollywood Bowl in 1990 - all places its creators never anticipated. The film's meanings have undergone considerable revision since it was first made, too. Star and co-director, Gene Kelly, noted in the J 970s, 'A lot of people have called it the first camp picture', and he immediately added his reservation, 'I don't know' (BBC, 1974). This case study on Singin' in the Rain demonstrates how a popular film, which has

endured in value over the years to achieve canonical status as a classic, changes in its meaning, depending on the theoretical methodology applied. Each interpretive framework discussed here brings a different critical strategy to bear upon the film, causing its significance to shift. This is not to suggest that a film can mean just about anything, but that its meanings are determined through interaction with a critical theory. Interpreting therefore amounts to much more than deriving a film's theme; interpretation makes the film an object of inquiry in its own right, a text. Furthermore, as interpreters of a film, we need to recognize that a text has its own specificity but also a porous historical materiality, with regard to its fields of reference when first produced and as it continues to be viewed. A film consequently exists in history as a cultural object as well as a work of art, and this is why Singin' in the Rain can be appreciated today as a classic, as camp, as a commodification of popular entertainment, and as a case study of interpretation.

S4 Steven Cohan

The narrative of Singin' in the Rain Since the most common way audiences make sense of a commercial film is through its narrative, let us begin there. Set in Hollywood during the film industry's transition from silent films to talkies in J 927, Singin' in the Rain opens with the gala premiere of The Royal Rascal.This swashbuckler romance features the two biggest stars working for Monumental Pictures, Don Lockwood (Gene Kelly) and Lina Lamont (Jean Hagen). Before the famous couple can start their next feature, a similar costumer entitled The Duelling Cavalier, the success of The jazz Singer forces the studio to shut down while it converts to sound. Don Lockwood's voice fits comfortably with the new technology, but Lina Lamont's voice - shrill, high, and nasal- does not. The sneak preview of their all-talking The Duelling Cavalier is thus a disaster from start to finish. To save the film, the studio takes advantage of Dori's vaudeville background, converting the historical costume drama into an extended dream sequence set in a backstage musical. To get around the problem of Lina's inadequate voice, a talented contract player, Kathy Selden (Debbie Reynolds), dubs her songs and dialogue. Discovering that someone else's voice is being used in place of hers, Lina instigates an overnight publicity campaign promoting her own singing and dancing talents, foiling the studio's intention of acknowledging Kathy's contribution. Retitled The Dancing Cavalier, the musical version is an unqualified success, and at its premiere, Lina's forged performance is exposed in order to give credit to her voice double, who becomes Dori's new partner) on screen and off. As a means of opening up the narrative of Singin' in the Rain for interpretation, it is

useful to follow the formalist distinction between 'story' and 'plot'. These are not synonymous terms. 'Story' refers to the implied chronological sequencing of events, from which one further infers the logic of cause and effect, temporality, and space governing the 'diegesis' (that is, the spectator's deduction of a film's fictive world inhabited by the characters and appearing fully coherent on its own terms). A story is easily synopsized, reducible to a summary of what happens. 'Plot', on the other hand, refers to the form through which a film narrative actually orders (or even reorders) its story in the process of recounting it. Classical Hollywood cinema of the studio era makes character the prime causal agent unifying a story, and it constructs a linear plot to reiterate this emphasis. A linear plot identifies what amounts to a syntax for the story, organizing it in sequence around four nodal points - 'an undisturbed stage, the disturbance, the struggle, and the elimination of the disturbance' (Bordwell, 1985: 157) - to keep directing attention toward closure as the key to unlocking the story's meaning. In Singin' in the Rain, the showing of a Lockwood-Lamont movie at several points in

the story marks out the plot's linear development. The opening establishes an undisturbed stage with the premiere of The Royal Rascal celebrating the reigning star couple of the silent screen (Figure 4.1). The disastrous preview of The Duelling Cavalier disturbs that stasis by threatening their great popularity, and the dubbing of The Dancing Cavalier crystalizes how that disturbance amounts to a struggle between the two screen stars over their careers. The audience's unqualified enthusiasm for The Dancing Cavalier at its premiere proves that the technology enabling Lina's musical performance can go

Case study: interpreting S;ng;n' ;n the Ra;n SS

undetected, but when the counterfeit is revealed immediately afterward, the plot achieves resolution simply by causing her elimination. The closure restores the undisturbed stage of the opening but with a significant difference, since a new star couple replaces the original one. David Bordwell points out that a classical Hollywood narrative further complicates its

linear development by arranging two plotlines. One revolves around 'heterosexual romance (boy/girl, husband/wife)', while the other one arises from action occurring in the public sphere, be it 'work, war, a mission or quest' (1985: 157). Singin' in the Rain conforms to this convention, too, with a romance plot involving Don and Kathy and a professional plot concerned with his career. The doubling amplifies how Singin' in the Rain represents the industry's conversion to sound technology. The professional plot recounts the historical transformation of the movies from a medium of popular entertainment that divorces image from sound, as epitomized by the disparity between Lina's attractive body and abrasive voice, to one that integrates them, as evident in Don's conversion to musical pictures as well as his change of partners. But the disturbance Lina poses when she will not remain mute, while motivated by the professional plot, obstructs the romance, since she functions as the primary obstacle to Don's relationship with Kathy.

56 Steven Cohan

Dan's first encounter with Kathy encapsulates the disturbance that the double plot contains through Lina's character. Unimpressed when she learns his identity, Kathy mocks Dan's profession. 'The personalities on the screen just don't impress me', she declares. 'I mean, they don't talk, they don't act, they just make a lot of dumb show.... You're nothing but a shadow on film'.' Kathy's description of silent film as 'a lot of dumb show' refers to the pantomime style distinctive of that era's acting; more subtly, it rhymes with Lina's first lines of dialogue just minutes before Don meets his future co- star. Singin' in the Rain pointedly delays letting us hear Lina's voice until after the premiere when, backstage, she complains about not being allowed to say anything to her admiring pubic, adding, 'What do you think I am? Dumb or something?' As used by the double plot, Lina's figure equates one meaning of 'dumb' ('muteness') with another ('stupidity') to underscore the aptness of Kathy's criticism, indicating how and why Lina's bad acting, and not the studio's own inepitude in handling the new technology, obstructs the professional plot. Singin' in the Rain knots its double plot around Lina so that the closure makes it seem as if whatever complications arising from the film industry's tumultuous conversion to talkies have been satisfactorily resolved once Don finds a more authentic partner in Kathy. The final shot shows this couple admiring a billboard advertising their new musical, also entitled Singin' in the Rain. Because the film and its diegesis mesh so perfectly at this end point, any future disturbance seems unthinkable. We can understand why Singin' in the Rain invests so much ameliorating value in the

formation of this new couple by using the insights of structuralism to look beyond the plot's linearity. In a study of the Oedipus myth, anthropologist Claude Levi-Strauss showed how it manages contradictory cultural beliefs about the origins of humankind by building its plot around two incompatible pairs of binary oppositions that appear comparable as antitheses. According to Levi-Strauss, a myth's narrative reconciles cultural contradiction structurally, not logically, because its underlying structure demonstrates that 'contradictory relationships are identical inasmuch as they are both self-contradictory in a similar way' (1963: 216). His analysis has been productively extended to other types of narrative, and two points about the resulting method are worth emphasizing. First, a structuralist analysis takes a lateral perspective upon a plot's linearity in order to examine the binary oppositions arising at every point. Second, in identifying an underlying oppositional structure, this analysis does not dwell on the psychological conflicts between characters but instead seeks to illuminate the conceptual antitheses that the plot enacts. Singin' in the Rain illustrates the usefulness of the structuralist paradigm. Once we

notice how the professional plot pits Don against Lina to initiate the disturbance, certain binarized terms immediately emerge to dominate the entire narrative: body/voice, image/sound, naturalness/engineering, innovation/obsolescence. The film's opposi- tional structure stands out even more clearly if we recall the many musical numbers performed by stars Gene Kelly, Debbie Reynolds, and Donald O'Connor (as Dan's best friend, Cosmo Brown). The numbers supplement but do not advance the plot very much, but they locate the fusion of body and voice, of sound and image, in the characters' joyous act of singing and dancing, and they equate musical performance with

Case study: interpreting Singin' in the Rain S7

innovation and naturalness, juxtaposing these values to the plot's main source of disturbance: obsolescence and engineering. Genre plays a significant role in highlighting this oppositional structure. According to

Rick Altman, who builds upon the tenets of structuralism in his taxonomy of the Hollywood musical, the genre operates from a dual focus that, in concentrating on the couple, obeys a principle of lateral parallelism, not linear sequencing. In a musical, he comments, 'the couple is the plot' (1989: 35), and the genre works in a non-linear fashion to match the male and female leads as a potential couple through paired segments: parallel numbers, scenes, shots, dialogue, settings, and so on. Fundamental to this dual focus is a sexual opposition that associates the male with one cluster of values and the female with another. The couple's eventual union implies reconciliation of various secondary oppositions, intially 'perceived as mutually exclusive', so the couple has the important function of 'reducing an unsatisfactory paradox to a more workable configuration, a concordance of opposites' (1989: 27). Singin' in the Rain structures its narrative according to this generic principle. Dan's

maturity is contrasted with Kathy's youthfulness, his stardom with her fandam, his hammy acting with her criticism, his dancing with her singing, his athletic body with her pure voice. Viewed through this dual focus, Kathy represents the innovation, spontaneity, and artlessness that Don professionally lacks at the start of the film, while the parallelism of their many differences links them as a couple. As a consequence, the opposition structuring the romance plot (masculinity versus femininity) compares with the opposition structuring the professional plot (innovation versus obsolescence), and this dual focus then reconciles the more central opposition motivating the entire narrative: the disjunction of image and sound inherent in the technology of cinema and crucial to the film's historical setting in the industry's conversion to talkies. As Peter Wollen observes, Singin' in the Rain closes its narrative by valuing 'the married couple' and 'the married print' (the industry's term for the joining together of image and soundtrack in the finished product) as metaphors of each other (1992: 55, 57). In their eventual teaming on screen as a couple, Don and Kathy embody the perfect synchrony of body and voice, dancing and singing, which marks the achievement of the talkie, just as the talkie's technological matching of sound and image mirrors the couple's off-screen harmonious blending of masculinity and femininity.

Dubbing, dancing, and deconstruction

The dual focus of Singin' in theRain highlights another opposition underlying the binary pairings of masculine/feminine and innovation/obsolescence as they accrue around the disjunction of sound and image - that of authenticity versus artifice. This latter opposition is crucial to the film's celebration of the musical genre's authenticity as a representational art. However, since a musical's authenticity depends upon effacing the very technological artifice that makes it possible, the binary pairing of authenticity/ artifice also summarizes the textual problematic, inescapable because of the film

S8 StevenCohan

medium, that permeates Singin' in the Rain and cannot be worked out solely at the level of narrative structure. A structure implies stability and containment, but 'the concept of structure', Jacques

Derrida comments, is 'ambiguous' because 'it can simultaneously confirm and shake' (1981: 214). Deconstruction, which has become synonymous with Derrida's name, is a theory of reading a work for its textuality, its destabilizing play of differences, which critiques the logic of structuralism. A deconstructionist analysis traces the textual path through which a binary opposition, in arranging differential terms as a positive and negative relation, inevitably refers to a third term that functions to authenticate the polarity. The third term appears to stand outside the binary pair but is indispensable to its formation, revealing how 'each allegedly "simple term" is marked by the trace of another term' (Derrida, 1981: 221). In Singin' in the Rain dubbing has such an effect of destabilizing the binary logic by which, through the trope of dance, the film hierarchically arranges its oppositions in terms of authenticity or artifice as its primary means of mythifying the musical genre. To be sure, with its Hollywood setting, witty script, and songs recycled from earlier

films, Singin' in theRain might seem too ironically aware of its place in the history of the musical to be doing anything other than demystifying the genre. But while musical entertainment is demystified by the self-reflexive narrative, which becomes increasingly preoccupied with the technology of cinema, it is remystified by the numbers, which efface technology, promoting 'the mode of expression of the musical itself as spontaneous and natural rather than calculated and technological' (Feuer, 1981: 165). Musicals value the spontaneity of singing and dancing over the careful engineering required to perfect a number for filming, giving rise to a number of familiar conventions, well explained by Jane Feuer in her book, The Hollywood musical (1993). These all highlight the effortlessness of performing while underscoring it as performance: the unrehearsed public appearance (as in Kathy's rendition of cSingin' in the Rain' or her closing duet with Don, 'You Are My Lucky Star'); the effacement of choreography to make dancing seem as natural as walking (as in Don's 'Singin' in the Rain') and to disguise the obvious labor involved in perfecting it (as in the eruption of raucous tap-dancing in his and Cosmo's 'Moses Supposes'); the prop dance (as in Cosmo's 'Make 'Em Laugh' or the trio's 'Good Momin"); and a direct address that surpasses diegetic spectatorship (as when the camerawork in 'Make 'Em Laugh' forgets that Cosmo is ostensibly performing for Don's benefit and ceases to follow the latter's spatial perspective). Most of the numbers in Singin' in the Rain are not diegetically motivated as part of a

show but meant to seem spontaneous expressions of joyful emotion and native talent, without overt reference to the technology that exposes the possible fraudulence of movie stardom in the narrative. The numbers establish an implicit comparison with the contrivance of physical action in Don's silent films on the one hand and to the insincerity of his speech when not acting on the other. The one time a number does call explicit attention to the technology of film ~ when Don sings 'You were meant for me' to Kathy on a studio soundstage - it initially highlights cinematic artifice only so that his lyric declaration of love can then disavow it entirely. Don begins the number by using

Case study: interpreting Singin' in the Rain 59

technology (colored lights, a wind machine, and so on) to simulate a suitable romantic atmosphere copied from the movies he stars in with Lina. Once he begins to dance with Kathy, the technology fades from view. Performed on the bare soundstage, their dance confirms their mutual sexual attraction; just as importantly, it supplies the first occasion for making the musical talent each previously displayed apart in a show number seem spontaneous, thereby associating their harmony as a couple with artless musical performance. In their synchronization of voice and body, the numbers offer a view of performing

that, unlike Lina's, succeeds in integrating sound and image, so it follows, as Gerald Mast puts it, that 'the controlling myth of Singin' in the Rain isperformance itself (1987: 265). To support this myth, Singin' in the Rain has to circumvent the film medium's own representation of spontaneity through technology. The production of a musical relied upon the splitting of sound and image while simultaneously engineering the illusion of their unity: the songs and orchestra were prerecorded, often edited from numerous takes, and played back on the set during shooting; dances were usually rehearsed whole but shot in segments, as determined by the various camera set-ups planned for the number's editing; and the sound effects, including the tapping, were postrecorded, sometimes not by the dancers themselves. Thus, while carrying the force of fusing sound and image in a film's exhibition as a completed print, numbers technically depended upon their separation into multiple audio tracks and numerous cinematographic takes during their production. Singin' in the Rain acknowledges the basis of performance in technology with an in

joke that audiences frequently miss. At the point when Kathy dubs Lina's singing and speaking, Debbie Reynolds's voice is itself dubbed by others, Betty Noyes for the song and, even more self-reflectively, Jean Hagen - who plays Lina - for the dialogue. According to co-director Stanley Donen, Reynolds was dubbed in these scenes because of her regional accent, since the voice looped over Lina's 'was supposed to be cultured speech' (Fordin, 1975: 358). If that is the case, why was Reynolds not dubbed as well when Kathy has to sing for Lina 'live'? Presumably because that moment has to appear as the unmasking of Lina's counterfeiting of musical stardom and the simultaneous revelation of Kathy's - and Reynolds's - greater authenticity. The layers of dubbing involved in the manufacturing of Lina's voice for talkies has

fascinated commentators on Singin' in the Rain (Figure 4.2). Wollen points out, 'dubbing, here, represents the cinematic form of writing, through which sound is separated from its origin and becomes a potentially free-floating, and thus radically unreliable, semantic element' (1992: 56). By paradoxically marking the disjunction of sound and image in their technological conjunction, dubbing deconstructs the oppositional logic supporting the myth of performance through the comparable value of the married couple and the married print. Dance, by comparison, supplies Singin' in the Rain with a trope that, by harmonizing voice and body, re-centers the myth. The big production number in Singin' in the Rain, 'The Broadway Ballet', interrupts

the linear trajectory of the plot, stopping it cold for 14 minutes, but the number reinforces the significance of dance as an alternative third term reunifying sound and image. Don explains to studio head R.F. Simpson that there is one number left to shoot

60 Steven Cohon

Figure 4.2 Lina discovers that Kathy isdubbing her voice.

for The Dancing Cavalier. As Don begins to describe it, the film immediately cuts to the number itself. Interestingly, the published screenplay adds: 'This is a big production number shown as it will appear in The Dancing Cavalier' (Camden and Green, 1972:63). However, while set in the 1920s, the number we see takes fuji advantage of 1950s' technology so it in no way duplicates what it would have looked like in 1927. After its conclusion, Simpson remarks, 'I can't quite visualize it. I'll have to see it on film first.' While that line gets the big laugh, Cosmo's retort ismore telling: 'On film, it'll be beller yet.' As the framing dialogue indicates, 'The Broadway Ballet' is another underlined moment of self-reflexivity in Singin' in the Rain, and the gap between what we actually see on the screen and what Don supposedly narrates to Simpson in the diegesis forces us to reconsider what dance means for the film. Because of the way it is set up by Dan's brief introduction, the ballet rhymes with his

voiceover montage about his career at the start of the film, when gossip columnist Dora Bailey asks him to tell his life story to the crowd of fans outside Grauman's Chinese Theater. Summarized by Dan's pretentious motto - 'Dignity, always dignity' - his spoken account represents an elitist class background with his formal training in dance, music, and theater leading to a career in motion pictures playing 'sophisticated' and 'suave' leading men. What we see, though, differs considerably from what he describes.

Case study: interpretingSingin' in the Rain 61

A young Don and his pal Cosmo dance for pennies in a poolroom and sneak into a nickelodeon to see a movie serial; as adults, the two friends entertain in a tavern, do a comic act on amateur night, and perform 'Fit as a Fiddle' in a third-rate vaudeville hall, until they find themselves in Hollywood, where Don accidentally gets a job as a stuntman. The voiceover montage establishes image as truth, demystifying sound as a manufactured representation. However, since Singin' in the Rain recounts the industry's transition to talkies, the film must eventually recuperate sound. The ballet contrasts with the montage because it offers dance as the ultimate wedding of sound and image, enacting what the couple metaphorically represents. This number's own 'star-is-born' narrative parallels the story Don tells at the premiere of The Royal Rascal- he plays a hoofer who rises from burlesque, to vaudeville, to the Ziegfeld Follies - but incorporates sound so that singing and music motivate physical movement in the image as dancing (Altman, 1989: 229), and the dancer himself, combining movement with music and not relying on words (Cuomo, 1996: 43), then embodies the image's unity with sound as grace, coordination, and rhythm. Value is thus restored to sound once it is firmly subordinated to the image through

dance. But dubbing deconstructs the authenticity of dance, too. After all, how can anyone tap-dance in the rain, as Gene Kelly does in this film's famous title number, unless the taps are post-recorded and then added to the image as supplementary sound? That number took a day and a half to shoot but months to find the right sounding taps to use for the dubbing (AMCN, 1994). Far from being a genuine alternative to dubbing, dancing in a musical was produced through the same technology and toward the same end: manufacturing a performance. Consequently, the more Singin' in the Rain self- reflexively highlights the instability that dubbing exemplifies, the more readily it eAlloses the profound rift between sound and image, deconstructing the text's efforts to 'wed' them, technically and metaphorically, in the married print and the married couple. Dance may therefore work as a third term to stabilize Singin' in the Rain's sound/image dichotomy, but it is subject to the text's deconstructive energies too.

Spectatorship and sexual difference

We can understand what else the rhyming of the ballet with the voiceover montage puts at issue by turning to psychoanalytic film theory. When hearing 'The Broadway Ballet' described) Simpson cannot visualize it in his mind) in contrast to the way we as spectators have seen it, and he will not be able to do so until it is filmed. For as Cosmo says, it looks better filmed. In stressing that the ballet is, as far as concerns the diegesis, an imaginary representation right after we have been allowed to see it for ourselves, the exchange between Simpson and Cosmo condenses in a joke what psychoanalytic film theory explains is the source of pleasure in cinematic spectatorship: unconscious identification with the camera through the imagery projected on-screen. According to Christian Metz, in identifying with the act oflooking that characterizes

both spectatorship and camera work, 'the spectator can do no other than identify with the camera, too' (1977: 47). But since the camera is not present during a film's

62 StevenCohan

exhibition, the spectator must accept a substitute or stand-in, in Metz's term, an 'imaginary signifier': the cinematic image. It is 'imaginary' in the sense of being visual and fantasmatic (nothing but 'shadow on film', as Kathy says). Furthermore, the image repeatedly comes between the spectator's look and the camera's, so the viewing situation encourages while continually rending their alignment through the course of a film's unfolding. Mirroring the way that film itself is an edited or 'sutured' series of shots or 'cuts', spectatorship produces a comparably sutured viewer. With each new shot, editing ruptures and re-secures identification with the image; and since how a film addresses a spectator continually interacts with what it shows on-screen, this 'suturing' operation has profound implications. As Kaja Silverman points out, 'the entire system of suture is inconceivable apart from sexual difference' (1983: 221). In explanation, Silverman puts Metz's theorization of spectatorial identification

together with Laura Mulvey's critique of commercial narrative cinema for its perpetuation of an asymmetrical gender dualism. Whereas Hollywood films typically associate masculinity with seeing, in command of an active, voyeuristic look, they just as typically represent femininity as lacking and passive, put on exhibition and looked at. The female figure iseither made an enigma to be investigated, redeemed, or punished, Or she is visually fragmented in a fetishized spectacle - for example, in close-ups that direct attention to glamorized parts of her body. Either representation of femininity makes it 'reassuring rather than dangerous' (Mulvey, 1975: 21), disavowing male lack by encouraging voyeuristic identification with the camera's look at the eroticized female body and by reiterating that identification through the male character's gaze in the diegesis. As amplified by feminism, psychoanalytic film theory reveals additional meanings for

the sound/image dualism of Singin' in the Rain. It explains why the film goes to such lengths to discount, in both its narrative and its numbers, that the male is the initial means of rending sound from image, and why this rupture must then be outwardly attributed to the female. Singin' in the Rain operates upon a viewer by jeopardizing the imaginary unity of sound and image necessary for suture, disrupting the symbolic status of the female as the signifier of male authenticity, but only to restore that function in its closure. Don's voiceover montage detaches his look from the image and dramatizes his masculinity as a state of self-division. After all, in his own biography, he is himself an 'imaginary signifier', a stand-in for a stand-in; he began his career in the movies by replacing a stuntman in silent film. In order to recuperate what it initially reveals is a division in male subjectivity, Singin' in the Rain displaces male lack onto the female so that she gets blamed for the breakdown of sound and image, while he repairs it, restoring the voice to its proper body. As a result of this gendering of the sound/image dualism, Silverman points out about Singin' in the Rain, the male assumes control of the apparatus, defining the female as its object, confining her to representation, and reinvesting a transcendent authority in his voice, aligning it to his look (1988: 47, 57). When Don, Cosmo, and Simpson raise the curtain to let the audience see what it hears- Kathy, the 'real' voice behind Lina's singing performance - hearing becomes dependent upon seeing and sound is married to the image, all controlled by the male look and, once Don orders the audience to stop Kathy from running away, matched to his voice.

Case study: interpreting S;ng;n' ;n the Ra;n 63

Lina's lack of wholeness, because her voice does not match her body, motivates her replacement by Kathy, as we have already learned by analyzing the plot. But now we can draw a further conclusion: this closure takes place only after subjecting the female to a series of substitutions that, far from restoring voice to image, subsumes her disturbing impact. Women are disruptive in Singin' in the Rain because they speak. 'Well, of course we talk', Lina declares after the studio decides to convert The Duelling Cavalier into a talkie. And when they talk, watch out! Kathy's outspoken criticism jeopardizes Don's masculinity by calling attention to its imaginary status as 'nothing but a shadow on film'. Dubbing regulates Kathy's insubordinate voice - diegetically when she ghosts for Lina and extradiegetically in the dubbing of Reynolds - by displacing it from her body. Even more significant, Kathy is replaced as Dan's partner in 'The Broadway Ballet' by a distinctly mute dancer playing the heartless vamp, Cyd Charisse. From the first shot introducing her by way of her long legs, the choreography and camera work fetishize Charisse's body. The female figure is thus punished (Lina) and redeemed (Kathy) in the narrative and then, in the ballet, silenced completely in order to be eroticized as pure spectacle (Charisse) (Figure 4.3). Reversing the lack depicted in the opening voiceover montage, the ballet disavows anxiety about male subjectivity, rendering the female as pure spectacle so that the male appears whole, his body and voice reunified by the repeated song lyric that drives the number, 'Gotta Dance!'.

Figure 4.3 'TheBroadway Ballet'eroticizes the female dancer (Cyd Charisse) as pure spectacle.

64 StevenCohan

Psychoanalytic film theory has been extremely important for film studies in explaining how the cinematic apparatus embeds spectatorial pleasure in the sexual politics of representation. But for all its emphasis on the unconscious mechanism of spectatorial identification, as judith Mayne comments, 'What has been surprisingly absent from much psychoanalytic film theory isan investigation of the waysin which the unconscious refuses the stability of any categorization' (1993: 97-98). Mayne proposes that we instead conceive of 'the cinema as a form of fantasy wherein the boundaries of biological sex or cultural gender, aswell as sexual preference, are not fixed' (1993: 88). Sheplaces emphasis on the psychoanalytic theorization of fantasy not as make-believe, but as a scenario of desire offering variable positions of identification: with the subject of fantasized desire, or with its object, or with its setting in the desire itself, regardless of who is acting on it or receiving it. Fantasy allows for the possibility of shifting between those positions, even when they are gendered and fixed as male or female within the scenario. And, to link Mayne's point back to Mulvey's and Silverman's arguments, if film can be said to draw a viewer into the ideological field of sexual representation as fantasy, then it cannot guarantee for certain the identificatory positions to which viewers will suture. Mayne defines fantasy 'as the staging of desire ... a form of mise-en-scene' (1993: 88),

and this description applies especially well to the musical numbers of Singin' in the Rain - indeed, to its self-reflexivity in general- calling for reassessment of its apparent fixing of male and female identities through the voice/image dualism. For all its disavowal of male lack, the ballet is, ultimately, premised on the hoofer's failure to sustain a heterosexual relation with the vamp; both times they meet in this number, after they dance up a storm, she leaves him for her gangster lover. In fact, only one of the numbers in Singin' in the Rain ('You Were Meant for Me) directly celebrates the heterosexual couple, and even this one begins with Don setting the stage - with his enacting heterosexual desire as a fantasy scenario for a spectator. The other numbers engage even more fluid possibilities of identifying with bodies in

motion and, correspondingly, with dance as a state of mobile desire not tied to the gender and sexual norms legitimated by the active-male/passive-female binary. Don watches Kathy dance with a voyeur's eye in 'All I Do isDream of You' but he takes even more pleasure in dancing by himself in the title number, giving his joy an autoerotic dimension. Similarly, when Cosmo dances to cheer up Don in 'Make 'Em Laugh', he turns himself into a spectacle of energetic, even maniacal, movement that, not coded as feminine, problematizes Dan's male perspective, leading Alex Doty to comment about the difficulty of responding to this number 'as anything but a case of overwrought, displaced gay desire' (1993: 11). The intimation that dance can stage homoerotic desire resurfaces when the two men synchronize their movement as a challenge dance in 'Moses Supposes'. Dance also takes on bisexual implications in 'Good Mornin" when Don and Cosmo perform with Kathy, positioned between them in the dancing every step of the way except in the middle of the number. At that point, each dancer grabs someone else's raincoat, using it as a costume prop for a solo dance that is specifically inflected as a performance of gender: with Kathy's coat, Cosmo dances a Charleston; with Dan's, Kathy dances a hula; with Cosmo's, Don dances a Spanish flamenco. These solos disrupt the presumed correspondence between gender and dancing as enactments

Case study: interpreting S;ng;n' ;n the Ra;n 65

of a fixed sexuality, and they point to 'the interchangeability of Kathy and Cosmo' elsewhere in Singin' in the Rain (Mellencamp, 1991: 6). After this number Cosmo stands in front of Kathy to show Don how her voice can be mimed, and, much later, when her dubbing of Lina is publicly revealed, Cosmo precipitates Kathy's flight from the stage by pushing her aside to sing for Lina himself. The variable positions of sexual identification offered by the numbers extend to the

entire film and may even determine, in subverting, the plot's investment in the romantic couple. Cosmo's ambiguous status as an odd man out whenever in the presence of the two lovers provides an identificatory position that, in triangulating the pairing of Don and Kathy, calls into question the plot's automatic privileging of heterosexual desire. The romantic couple in Singin' in the Rain is not achieved by a process of addition but by one of subtraction, exclusion of the extra male who is interchangeable with the female. The improper coupling of Don and Lina has the same resonance, since she treats Cosmo as if he were her rival, and vice versa. At the start of the film, in short, Don isalready coupled - with Cosmo. Why Cosmo has to be marginalized in the narrative, but not the numbers, does not become clear until we see that he poses an obstacle to the heterosexual masculinity Don represents in his relationship with Kathy, which reiterates the orthodox male/female binary and its erotic premise. Because Singin' in the Rain forms this couple only by replacing a partner of one gender (male) with that of the other (female), Don is not fixed to that position to begin with, and neither is a spectator (Figure 4.4). In drawing this conclusion, I do not mean to say that the film characterizes Don as

homosexual or bisexual, for that is surely not the case. Rather, 1 have wanted to call attention to certain queer tensions regarding gender and sexuality that underlie Singin' in the Rain and help to determine both its representational content and the pleasures spectators may find in the film. These tensions, moreover, are not confined to the film alone but arise from its connections, as a product of the Hollywood studio system, to cultural history. We can begin to account for these tensions by examining Gene Kelly's extratextual importance for the film as both its author and its star.

Gene Kelly's authorship

It may be tempting to get around what the film's self-reflexivity exposes as the inherent instability of cinematic representation, and to evade the claims of psychoanalytic theory regarding its sexual politics and the possibility of fluid positions of identification, by reading Singin' in the Rain solely through the fact of Gene Kelly's authorship. The theory of auteurism interprets a film through its director as 'the expression of his individual personality', demonstrating 'that this personality can be traced in a thematic and/or stylistic consistency over all (or almost all) the director's films' (Caughie 1981: 9). Wollen's insightful monograph on Singin' in the Rain follows this line of thinking to legitimate Kelly, who co-directed the film with Stanley Donen, as its primary auteur. Wollen illustrates his viewpoint with a careful analysis of the title number, interpreting it as 'a summation of Gene Kelly's own work up to that point, echoing and developing numbers he had done in earlier films' (1992: 29).

66 StevenCohan

There is a certain justice in appreciating Singin' in the Rain as a statement of Kelly's convictions about the musical genre and dance. His efforts to make dance cinematic as opposed to photographed theater resulted in an innovative, influential style that reflected his aesthetic views on choreography. Filmed dance, Kelly maintained, has to be conceived of solely in terms of the camera, which sees differently than the human eye watching a live stage show (Hirschhorn, 1974: 98-99). He consequently staged dances so that performers move into or away from the camera, and used editing to enhance the choregraphed movement, all to compensate for 'the two-dimensional limitations imposed by film' (Delameter, 1981: 140). It is therefore consistent with Kelly's ambitions as a director that technical as well as

choreographic innovation receives so much emphasis in Singin' in the Rain, and that the film uses dance to ameliorate its sound/image dichotomy. 'The Broadway Ballet', even

Case study: interpreting S;ng;n' in the Rain 67

more so than the film's other dance numbers, combines tap, jazz, and ballet styles in accord with Kelly's aesthetics and formal training. The fluidity of this choreographic style metaphorically raises 'the larger issue of generic flexibility' (Cuomo, 1996:40), and it further reflects ways in which Kelly's innovations as a director correspond to his character's career within the narrative. Dance, in Cuomo's view, exemplifies how Kelly historically reinvigorated the musical genre, so Singin' in theRain can be read as a 1950s' musical responsive to post-war changes in the industry that were threatening the genre's future and the studios producing them. Dance, the sign of the body's agility at every point in the film, likewise mirrors the generic flexibility of star Don Lockwood as he turns from swashbucklers to musicals. Whereas his athleticism is initially a sign of his inauthenticity, with the change in genres, the same physical movement is set to music and becomes visible proof to the contrary. Singin' in the Rain bears Kelly's authorial signature insofar as it became his signature

film, so much so that on the night he died, for its obituary notice ABC News in the United States reportedly closed its broadcast by playing the title number without any spoken commentary. However, Singin' in the Rain actually makes it difficult to assign authorial intention to a single pair of hands (or feet!). Kelly himself spoke out against auteurism at his 1985API tribute: 'There are no auteurs in musical movies. The name of the game is collaboration.' While true of all studio-produced films, this was especially the case with Singin' in the Rain. It was produced by the Arthur Freed unit at MGM, a team of craftspeople working together under the guidance of Freed and his assistant, Roger Edens. Freed initiated the project because he wanted to produce a film organized around songs he had co-written with Nacio Herb Brown for MGM musicals of the 1920s. The film's script was written by Betty Comden and Adolph Green, Broadway artists under contract to MGM. They wrote the screenplay independently of Kelly - in fact, they recall that it was not originally planned with the star in mind (1972: 4--8), a point that Donen disputes (Harvey, 1973: 5). Most important, because most unusual in Hollywood at that time, Singin' in the Rain was co-directed as well as co-choreographed by Kelly and Donen, a former dancer himself. They often split directing chores when filming Singin' in the Rain, shooting on different soundstages. This is not to suggest that Kelly directed the numbers and Donen the non-musical portions, because their collaboration was interactive at every point. 'Even through rehearsing the numbers, when everything would really get complicated I would be with him,' Donen said of his work with Kelly. 'So it really was a collaboration' (Harvey, 1973: 5). In its diegesis, Singin' in the Rain also views film-making as a collaboration, one

manufacturing an industrial product. The director of the Lockwood-Lamont series is no auteur but a studio employee who keeps referring to R.F. Simpson as 'boss', and who allows Don Lockwood great latitude in writing lines and blocking movement. The director is not present when Simpson listens to a description of 'The Broadway Ballet', so Don seems to be responsible for planning and executing the number. But his best friend Cosmo is also a strong if uncredited influence on The Dancing Cavalier. Cosmo devises the scheme to dub Lina with Kathy's voice, invents the plot for the musicalized The Duelling Cavalier, and comes up with its new title. Indeed, the Don-Cosmo relation bears a striking resemblance to the Kelly-Donen collaboration, not just in

68 StevenCohan

their working relationship but in the auteurist hierarchy that kept the idea man in the shadow of the dancer. Randall Duell, one of the art directors on Singin' in the Rain, describes the partnership in terms that evoke Don and Cosmo's in the film: 'Gene ran the show. Stan had some good ideas and worked with Gene, but he was still the "office boy" to Gene, in a sense, though Gene had great respect for him' (Delameter, 1981: 247). In various television interviews, Kelly spoke of his collaborator as his apprentice,

assistant, or protege. Yet Donen was no 'office boy'. Throughout their collaboration, Kelly staged dance for the camera; he choreographed it in cinematic terms. Donen, far from running the camera in Kelly's place while the latter performed, planned out the dance's relation to the camera; he photographed and edited it in cinematic terms, making full use of his own technical expertise (Charness, 1977: 7, 65). The authorship of Singin'in the Rain is thus a complex matter that perhaps cannot be definitively settled. Kelly's signature nevertheless helps to explain why the film can be taken as his personal statement about cinematic dance, but that is also because he was equally important to the film's production and reception as its star, and his star image adds additional layers of cultural significance to Singin' in the Rain that further contribute to its meaning.

Stardom and culture

'A star image,' Richard Dyer writes, 'consists both of what we normally refer to as his or her "image", made up of screen roles and obviously stage-managed appearances, and also of images of the manufacturing of that "image" and of the real person who is the site or occasion of it' (1986: 7-8). Analyzing a star's image does not examine the 'real person' beneath the persona, in the way that auteurism seeks evidence of an authorial identity as the origin of a film, but reads it as a cultural text in its own right. Using the methodology of cultural studies, star studies concentrate on the discourses surrounding performers and their films, identifying the connections to beliefs and practices that embed cinema in social history. The obituaries on Kelly after his death in 1996 summarize his popular star image as

established in the 1940s and 1950s. They emphasize, to sample just some of the headlines, that he was the 'he-man dancer [who] perfected the filmusical [sic]' (Howard, 1996: 30); that he 'imbued dance with blue collar style' (Wolk, 1996: 2); and that he was 'the athlete who danced our dreams' (Matthews, 1996: 5). Such descriptions illustrate the widespread perception of Kelly's star image as it brought together gender and class to represent male sexuality in a way that, while appearing to give a coherent meaning to his dancing as an expression of virility, produced what Dyer calls a 'clash of codes' (1979: 38). Richard Griffith's comments in a program for the Museum of Modern Art's tribute to

Kelly in the 1960s exemplify, but without acknowledging, the contradictory views of masculinity held together by his star image throughout his career: 'Gene, a superb specimen of manly beauty, doomed, you'd say, to matinee idolatry, has neatly escaped from the trap of dancing and miming in such a way that you would never mistake him

Case study: inierpreting S;ng;n' ;n the Ra;n 69

for anybody but an ordinary joe' (1962: 3). On the one hand, Griffith admits that Kelly's 'dancing and miming' turn him into an erotic spectacle, placing his 'manly beauty' on display to be appreciated as 'a superb specimen'. On the other hand, Griffith insists that Kelly avoids that 'trap' because his films cast him as an 'ordinary joe' to emphasize how the dancer isjust another guy, thereby minimizing his spectacular screen value as a male sex object (Figure 4.5). Like the obituary notices, Griffith's remarks indicate that Kelly was considered to be

the most virile of Hollywood's musical stars, with his manliness characterized through associations with the working class as connoted by macho posturing in place of specific social detail. 'What I wanted to do,' Kelly remembered, 'was dance ... for the common man. The way a truckdriver would dance when he would dance, or a bricklayer, or a clerk, or a postman' (AMeN, 1994). When discussing his screen roles, commentators emphasize the same characteristics: egoism and cockiness, athleticism and physicality, immaturity and ordinariness. Taken together, these attributes formed his screen persona as 'a breezy, brash, unashamedly American urban hero' (Babington and Evans, 1985: 166). But as jerome Delameter also observes, 'The Kelly character is often a city

Figure 4.5 The spectacle of Gene Kelly: matinee idol and 'ordinary Joe' (publicity photograph c. 1950).

70 Steven Cohan

character, though one with a small town soul ... [which reveals] that there is a more personal and friendly nature hiding behind the outward appearance of the city' (J981: 147). The disclosure that a decent, smalltown 'ordinary Joe' lurks inside the pompous, hustling, urban wiseguy occurs in almost every one of Kelly's films, Singin' in the Rain included, and it substitutes an innate moral (and 'American') character for class as the factor delineating his character's social identity. In trying to reconcile what had become incompatible in the post-war national

consciousness, namely, the rural and urban settings of US culture, Kelly's star image further reflects tensions resulting from class difference as it intersected with conflicted beliefs about gender and sexuality. Briefly, the automatic equation of gender (identity) and sexuality (desire) in forming a person's heterosexual or homosexual preference is, as historian George Chauncey documents, a recent invention of US culture, one occurring with the middle class securing its hegemony after World War 2. Before then, the cultural category of 'gender' in working-class ideology did not necessarily correspond to or determine sexual practices, as it had begun to do for the middle class. A 'sissy' was defined by his outward behavior, not by his having male sexual partners, just as a 'manly' man could have sex with men as well as with women without impugning perceptions of his virility (Chauncey, 1994: 13, 100). As Kelly's 'ordinary Joe' persona tries to homogenize working-class identity with

middle-class sexual values, his star image brings into contention what at the time were still distinct class ideologies that differed in their presumptions of the correspondence between gender and sexuality. His gendei ed affiliation with the working class explains why a male sidekick such as Donald O'Connor figures as importantly as a female co-star in the casting of Kelly's films. As well as incorporating the male bonding experiences of World War 2 into his characterizations, the buddy relation provides KelJywith a male partner whose masculinity isplayed for laughs, building up a viewer's identification with the star's dancing as a representation of virility akin to a truckdriver's. According to Clive Hirschhorn, Kelly's biographer:

Instead of alienating his male audiences, which he feared he might do, he made them

identify with him and won them over by the virility of his dancing. There was nothing sissy or effeminate about him, and they relaxed completely in his presence. Gene was

'safe', 'Like a guy in their bowling team - only classier', as Bob Fosse put it. Women found his smile and the cockiness of his personality most attractive and responded to his sex appeal.

(1974:157)

As Hirschhorn's carefully worded appreciation reveals, the erotic dimension of Kelly's dancing is the uneasy meaning that his star image, drawing on cultural assumptions about gender, sexuality, and class that historically converged in the 1940s and 1950s, seeks to manage. His star image projects ambivalence about dance's relation to his screen masculinity, evident in his interviews, which stress that 'dancing is a man's game' (the title of his 1958 TV special), in order to disavow the effeminacy connoted by a male dancer's spectacularity for its corresponding intimations of homoeroticism. "Any man", he maintained, "who looks sissy while dancing is just a lousy dancer"

Case study: interpreting S;ng;n' ;n the Ra;n 71

(Hirschhorn, 1974: 225). The paradoxical basis of Kelly's star image, then, is that while he expresses his virility through a dance, he also worries about losing it in the dance. His typical pairing alongside a more effeminate co-star such as O'Connor gives Kelly's masculinity its social inflection of gender - the working-class man, psychologized by his dependence on post-adolescent male friendships - while allowing for the erotic spectacle of his dancing. However, at the same time that the athleticism of his dancing, like the aggression in his screen persona, tries to compensate for the eroticization of his body, his dance numbers stage a bisexual address to the film audience, because 'the virility of his dancing', as Hirschhorn terms it, is directed at men and women. Kelly's dancing consequently provides his star image with a provocative sexual context that is not as 'safe' as Hirschhorn supposes. Whereas Kelly's MGM films generally promoted his 'reputation as the Douglas

Fairbanks of dance' (Harris, 1996: A-I), when Singin' in the Rain uses actual footage of Kelly's 1948non-musical version of The Three Musketeers (originally made by Fairbanks in 1921) as an inexpensive way to include film supposedly starring Don Lockwood, it does so to distance Kelly's star image from those erotic implications. The inserted scene of Kelly's D'Artagnan now represents Lockwood's hammy acting in The Royal Rascal, the trap of being a 'matinee idol' that, a decade later, Richard Griffith said Kelly could have fallen into if he were not at heart an 'ordinary Joe'. Den's behavior at The Royal Rascal premiere satirizes the character's star image for its pretentious, flamboyant style, and the voiceover montage exposes a working-class origin as the truth concealed by that false persona. AsSingin' in the Rain then proceeds to demystify Don, it remystifies Kelly. In effect, the film incorporates Kelly's own screen past and attributes it to Lockwood in order to revise - through the fictive character's transformation from silent-film swashbuckler to sound-era song-and-dance man, and, correspondingly, from glamour to ordinariness - the star's erotic image. Comparison with the MGM films that, by sexualizing his athletic body through the kind of swashbuckling associated with Lockwood, established his reputation - Du Barry Was a Lady (1943), Anchors Aweigh (1945), and The Pirate (1948) ~ makes Singin' in the Rain's recuperation of Kelly's star image even more pronounced. Singin' in the Rain's revisionist incorporation of Kelly's star image figures into its

mythification of Hollywood; at the same time, the film's continual attention to the production of movies keeps reminding us that the star system cannot be easily separated from its industrial setting. 'It's damn hard work to make a musical', Kelly recalled on television almost two decades after leaving MGM. 'It's hard work, it's as tough as digging a ditch.' Yet moments later he also commented: 'If the audience is aware that you're working hard, then you're not dancing well' (BBC, 1974). While meant to describe the perfectionism a finished number had to display, the comment also applies to the institution of stardom, which disguises the position of performers such as Kelly as industrial employees, commodifying their labor as stardom while effacing the work force that supports it. The title number of Singin' in the Rain, designed to feature Kelly's tap-dancing, relied on numerous artists who remained invisible as laborors: craftspeople and technicians) such as the musical arranger, cameraman, soundrnan, set designer, costumer, not to mention the anonymous crew who lit the set, installed rain pipes

72 Steven Cohan

throughout the street, and even dug ditches, the holes in the ground for puddles that had to match Kelly's steps exactly. The number, furthermore, was shot during the day, with a black tarpaulin covering the backlot set, because the studio wanted to avoid paying the crew double time (Hirschhorn, 1974: 186). Singin' in the Rain shows enough of the invisible work force necessary for star-driven

films such as musicals to suggest that, no less than Kelly's, Don Lockwood's star value stands atop a hierarchy of laborers. Don has a freedom to control his work that Lina, Kathy, Cosmo, and all the artists and crews laboring under him lack, since they are expected to follow directions as obedient personnel. Their labor consequently has a different value for the studio; it ismore easily commodified and effaced as serious work. Thus Kathy easily succeeds Lina as Den's co-star and Cosmo as easily disappears from view when his labor is no longer required. Yet as a star Don is himself a commodity; his value for the studio is as threatened by the failure of The Duelling Cavalier as isLina's. Putting cultural studies in conjunction with Marxism to analyze the labor of stars,

Danae Clark points out that, economically, they 'are actors caught between the forces of production and consumption, between bodily labor and commodified image. A gap exists between an actor's use value (labor) and exchange value (image)' (1995: 21). This gap is essential to the musical genre's myth of performance, in fact, to stardom in general as a cultural phenomenon. Remembering the production date of Singin' in the Rain, we can perhaps see why, historically, the movies' institutionalization of stardom was under particular stress at the time, making that gap between the use value of actors' labor and its exchange value as stardom more evident than usual. By 1952, MGM had replaced studio head Louis B.Mayer with Dore Schary and had begun to drop its big stars as their contracts expired; the House Un-American Activities Committee (HUAC) was investigating the left-wing politics of Hollywood actors such as Gene Kelly (Wollen, 1992: 47-51); the film audience had diminished in size and the studio system was breaking down economically as a result of the Paramount consent decree legally requiring the film companies sell off their theater chains, their main source of profit; and the movies had to compete with a new technology and new type of stardom in television. Using Kelly's star image to cover over the rift between labor and image, which the

female characters expose, Singin' in the Rain plays out these industrial concerns about stardom by projecting them onto fandom. Fandom commodifies stars as images through their consumption, and it is shown to be dangerous in Singin' in the Rain because, in their idolizing of stars, fans collect together as a crowd to form mass culture. Anonymous fans cheer the stars as they enter Grauman's Chinese Theater, but they are safely contained. Later that evening, though, they mob Don and endanger his life. Fans disturb the institution of stardom by forcing recognition of the difference between star labor and its exchange value. One bad movie, and a fan can decide, 'I never want to see that Lockwood and Lamont again', threatening to end the stars' careers and causing the studio to face the prospect of financial ruin. Of course, this account oversimplifies the degree to which the studio system controlled

star labor and profited from actors and fans alike. As Kelly recalled about the contract system, 'we were all indentured servants - you can call us slaves ifyou want - like ball players before free agency. We had seven-year contracts but every six months the studio

Case study: interpreting S;ng;n' in the Ra;n 73

could decide to fire you ifyour picture wasn't ahit' (de Vries, 1994:3). Shifling the source of power from the studio to the fans, Singin' in theRain would nevertheless have usbelieve that the preview audience's mockery of his acting iswhat finally convinces Don to accept Kathy's view that he is nothing but a shadow, just an image: 'Something happened to me tonight ... everything you said about me is true, Kathy. I'm no actor. Just a lot of dumb show. 1know that now.' Enthusiastically taking to the proposal that he rethink his career and make a musical, Don confirms the value of Gene Kelly's star image for Singin' in the Rain. However, even at this crucial moment that meaning does not stay perfectly fixed. At the prospect of Don's returning to his vaudeville roots to sing and dance, Cosmo jests, 'The new Don Lockwood. He yodels. He jumps about to music.' Cosmo's remark implies that the new star persona for Don, like Kelly's, is as much of an erotic spectacle and industrial commodity, and no less of an effacement of the actor's labor, as the old one. And someone else still has to play the music.

Conclusions

Any closure invariably carries with it an expectation of resolution, but I have intentionally resisted synthesizing Singin' in the Rain to offer a totalizing meaning tbat accounts for each possibility of interpretation raised by this case study. We may gather that, thematically speaking, each theoretical perspective explains how Singin' in the Rain is about Hollywood, but that does not go very far in engaging with the specificity of the film, nor can it reconcile what are competing interpretations of the same textual elements, such as the couple, the sound/image dualism, the value of dance, the ballet, the voiceover montage, and so forth. Rather than reaching a definitive statement about this film's meaning, it will be more useful to reach some tentative conclusions about the process of interpretation that this case study has illustrated. Each theoretical perspective yields an analytical methodology, placing Singin' in the

Rain in a different interpretive framework, catching different emphases, and finding different value for the film. We should not rush to conclude, however, that some methods of interpreting are more objective (or 'trustworthy') because they find meanings inherent in the film, while others are more biased (or 'suspect') because they project meanings onto it, since any interpretation has to bring a set of theorized assumptions about the act of interpreting itself to a film in order to make sense of it. Without that informing ground, the film would lack meaning; all we could do is summarize it. Even more importantly, because each theoretical method illuminates the film in

different ways, the one we choose to follow iscrucial not only in terms of explaining how a film means, but in determining what it means to us as critical readers. Interpreting a film delineates what significance can and cannot be attributed to it. So whether we confine our attention solely to textual detail, as in the first half of this case study, or move from there to everwidening cultural and historical contexts, as in the second, the meanings produced have great consequence. There is a politic to any interpretation, however implicit its theoretical reasoning and whatever its analytical method, because how we read film clarifies what we see there and tells us why it matters.

74 Steven Cohan

Endnote

1 All quotations from Singin' in the Rain (1952; MGM; dir. Gene Kelly and Stanley Donen) are my own transcriptions from the MGM-UA videodisc.

References

AFI, 1985: AFllife achievement awards, 1985: Gene Kelly. Worldvision Home Video, 1991. Altman, R. 1989: The American film musical. Bloomington, IN: Indiana University Press. AMeN, 1994: Reflections on the silver screen with professor Richard Brown: Gene Kelly. American

Movie Classics Network, 2 February. Babington, W. and Evans, P.W.1985: 81ue skies and silver linings: aspects of the Hollywood

musical. Manchester: Manchester University Press.

BBC, 1974: An evening with Gene Kelly. Rebroadcast on Turner Classic Movies Network, 23 July 1995.

Bordwell, D. 1985: Narration in the Fiction Film. Madison, WI: University of Wisconsin Press. Caughie, J. [ed.], 1981: Theories of authorship: a reader. London: Routledge, reprinted 1988. Charness, C. 1977: Hollywood cine-dance: a description of the interrelationship of camerawork

and choreography in films by Stanley Donen and Gene Kelly. Dissertation, New York University.

Chauncey, G. 1994: Gay New York:gender, urban culture, and the making of the goy male world 1890-1940. New York: Basic Books.

Clark, D. 1995: Negotiating Hollywood: the cultural politics of actors' labor. Minneapolis, MN: University of Minnesota Press.

Comden, B. and Green, A. 1972: Singin' in the Rain. New York: Viking. Cuomo II, P. 1996: Dance, flexibility, and the renewal of genre in Singin' in the Rain. Cinema

Journal 36(1), 39-54. Delameter, J. 1981: Dance in the Hollywood musical. Ann Arbor, MI: UMI Research Press. Demdc, J. 1981: Semiology and grammatology: interview with Julia Kristeva. In Cobley, P. (ed.),

The communication theory reader. London: Routledge, reprinted 1996, 209-24. de Vries, H. 1994: It's been so entertaining: these days, when there are more MGM stars in heaven

than an earth, there is still Gene Kelly, who is quite happy to debunk some of the myths of the studio's golden age, much as 'That's Entertainment! III' does. Los Angeles Times 1 May, Home Edition: Calendar 3+. Los Angeles Times archives, online 9 August 1996.

Doty, A. 1993: Making things perfectly queer: interpreting mass culture. Minneapolis, MN: University of Minnesota Press.

Dyer, R. 1979: Stars. London: British Film Institute. Dyer, R. 1986: Heavenly bodies. New York: St. Martin's Press. Feuer, J. 1981: The self-reflective musical and the myth of entertainment. In Altman, R. (ed.), Genre:

a reader. Landon: Routledge, 159-74. Feuer, J. 1993: The Hollywood musical, 2nd edn. Bloomington, IN: Indiano University Press. Fordin, H. 1975: The movies' greatest musicals: produced in Hollywood USA by the Freed unit.

New York: Ungar, reprinted 1984. Griffith, R. 1962: The cinema of Gene Kelly. New York: Museum of Modern Art. Harris, S. 1996: Gene Kelly dies: legendary dancer was 83. Los Angeles Times, I May, Home

Edition: A-l . Los Angeles Times archives, online 4 August 1996. Harvey, S. 1973: Stanley Donen. FilmComment 9(4): 4-9. Hirschhorn, C. 1974: Gene Kelly. New York: Sf. Martin's Press, reprinted 1984.

Case study: interpreting S;ng;n' ;n the Ra;n 75

Howard, E. 1996: Gene Kelly: he-men dancer perfected the filmusicol. Memphis BusinessJournal 19 February, 30; Electric library, online 10 August 1996.

levi-Strauss, c. 1963: The structural study of myth. In Structural anthropology, trans. Claire Jacobson and Brooke Grundgest Schoepf. New York: Basic Books, 206-31.

Mast, G. 1987: Can't help sing in'; the American musical on stage and screen. Woostock, NY: Overlook Press.

Matthews, J. 1996: The athlete who danced our dreams. Newsweek 11 February, 5; Electric library, online 10 August 1996.

Mayne, J. 1993: Cinema and spectatorship. london: Routledge. Mellencomp, P. 1991: Spectacle and spectator: looking through the American musical comedy. In

Burnett, R. {ed.]. Explorations in film theory. Bloomington, IN: Indiono University Press, 3-14. Metz, C. 1977: The imaginary signifier: psychoanalysis and the cinema, trans. Celia Britton,

Annwyl Williams, Ben Brewster, and Alfred Guzzetti. Bloomington, IN: Indiana University Press, 19B2.

Mulvey, L. 1975: Visual pleasure and narrative cinema. Visual and other pleasures. Bloomington, IN: Indiana University Press, reprint 1989, 14-26.

Silverman, K. 1983: The subject of semiotics. New York: Oxford University Press. Silverman, K. 1988. The acoustic mirror: fhe female voice in psychoanalysis and cinema.

Bloomington, IN; Indiana University Press. Walk, M. 1996: Gene Kelly dies; imbued donee with blue collar style. Reuters newservice, 2

February. Electric library online, 10 August 1996. Wollen, P. 1992: Singin' in the Rain. london: British Film Institute.

•