The Linguistic Mediation of Film Subtitles

The challenge of film translation took on special complexity with the advent of sound. Major film industries experimented with diverse approaches; initially, dubbing, subtitles and native language translators were tried. In 1929, MGM embarked on an expensive programme to replicate all its feature films in three different linguistic versions and in 1930, Paramount established a studio near Paris to create foreign films in five languages. The British, French, and German industries, meanwhile, followed Hollywood's lead in multiple versions, albeit on a smaller scale, as noted by Douglas Gomery ('Economic Struggle and Hollywood Imperialism; Europe Converts to Sound,' Yale French Studies, Cinema/Sound, no. 60, 1980). In Czechoslovakia, Josef Slechta invented a 'sound camera' which sharply reduced dependence on German and American sound equipment. Eastern European audiences flocked to the movie theatres to hear local stars speak and sing in their native language. In Latin America, similarly, local industries were encouraged by the arrival of sound.

Hoping to break North American domination of their markets, film-makers developed the popular carnival-based chanchada in Brazil and the tango film in Argentina, while Mexico competed with Argentina in supplying Latin America with Spanish-speaking films. Despite earlier European resistance to Hollywood domination, it was only with the coming of sound, ironically, that many Latin American film critics and spectators began to complain about the 'foreignness' of North American films. Silence had the effect of masking the national origins of the films. Thanks to the 'visual esperanto' of silent film, which includes many cross cultural codes, spectators not only read the inter-titles in their own language but also imagined dialogue in their own language. Cinema was retroactively perceived as foreign and colonialist, precisely because other-languaged dialogue destroyed the masking effect of silence. The Brazilian critics' response to sound, for example, are discussed by Ismail Xavier in Setima Arte: Um Culto Moderno (Sao Paulo, Editora Perspectiva, 1978) and Jean-Claude Bernardet and Maria Rita Galvao O National e o Popular tut Cultura Brasileira Sao Paulo, Brasiliense/Embrafilme, 1983, pp. 230-31).

With sound, the transition from an imagined universality into nationality and language difference modified the relationship between spectator and film. Despite the sensation of plenitude engendered by the addition of sound, the change also brought certain psychic losses. With silent cinema, the desiring spectators dreamed, as it were, phantasmatic voices to match the faces of their favourite stars. With sound, spectators were obliged to confront particular voices speaking particular languages not necessarily identifiable with their own. Greta Garbo, it turned out, had an 'attractive' Swedish accent but John Gilbert's voice was 'reedy' and 'unpleasant'. 'Charlot' wasn't French after all--although of course French spectators knew Charlie Chaplin was Anglo-American; we speak here of the psychic regime of 'je sais, mais quand-meme.' The effect of loss was analogous, in some respects, to that experienced by lovers of a novel when dreamed characters are incarnated in a film by specific actors with specific voices and physiognomies. At the same time, silent cinema was retroactively perceived as mute or silent, its 'lack' revealed by the encroaching presence of sound. Silent film inter-titles, as pointed out by Mary Ann Doane in 'The Voice in the Cinema: The Articulation of Body and Space' (Yale French Studies, Cinema/Sound, no. 60, 1980), had the effect of separating an actor's speech from the image of his/her body. The terms proposed to designate the redefined cinematic entity, not surprisingly, celebrate the reuniting of voice and body by emphasising the dialogue track: 'the talkies,' le cinema parlant, cinema falado ('spoken cinema,' in Portuguese). Other designations, less vococentric, were more inclusive in their perception of the role of sound: 'sound cinema,' cinema sonore, and perhaps the most adequate to a medium of images and sounds: the Hebrew kolnoa (sound-voice/movement).

Once it became obvious that the production of multiple foreign-language versions of films was not a viable option, producers, distributors and exhibitors were left with the fundamental choice of either subtitles (dominant in such countries as France and the United States) or post-synchronisation (the standard practice in Italy and Germany). In the case of subtitles, all the processes characteristic of title translation--filtration of meaning through ideological and cultural grids, the mediation of a social superego--operate with equal force. For those familiar with both source and target language, subtitles offer the pretext for a linguistic game of 'spot the error.' It's worth adding that multi-lingual films such as Le Mepris (Contempt, 1963) require some subtitles wherever they are screened. In Contempt, they form an integral part of the signification of a film deeply concerned with diverse 'translations' within the polyglot atmosphere of international co-productions. Italian post-synchronisation eliminated the role of the interpreter, leading Godard to dissociate himself from the Italian version of the film. There would be little point in cataloguing such errors; our intention is only to plot the trajectory of their slippage, the direction of their drift. That the English version of Godard's Masculin, Feminin (1965) translates bruler a napalm ('burning with napalm') as 'burning Nepal' is not, finally, crucial. More significant is the tendency, with French New Wave films, to bowdlerise the French dialogue. (Censorship, for Freud, we are reminded, was a kind of 'translation.') The English subtitles of Breathless, for example, consistently play down the aggressive grossierete of the original. Belmondo's opening 'Je mis con' becomes an inoffensive 'I'm stupid,' and his 'Va te faire foutre!' addressed directly to the camera/audience, is rendered by a desexualised 'Go hang yourself.' The Brazilian subtitles of Woody Allen's Everything You Always Wanted to Know about Sex, in anticipation not only of Brazilian censors but also of an audience not always attuned to the cultural ramifications of Jewishness, elided all that was explicitly sexual or specifically Jewish. The tendency to shy away from sexually connoted words reflects, perhaps, a higher coefficient of puritanism within a society, or at least among its translators. Repressive regimes, meanwhile, have exploited subtitling and dubbing as a mechanism for censorship. To avoid any hint of adultery between Ava Gardner and Clark Gable in Mogambo (1953), the censor-translators of Franco's Spain reportedly transformed the pair into brother and sister, thus arousing audiences with the even spicier theme of incest.

Some films are striking in their omission of subtitles. Film translators tend to be vococentric, concentrating on spoken dialogue while ignoring other linguistic messages such as background conversation, radio announcements and television commercials, not to mention written materials such as posters, marquees, billboards, and newspapers. Thus the spectator unfamiliar with the source language misses certain ironies and nuances, although it should be pointed out that titles can also operate in the opposite direction. For example, certain English subtitles for Godard's Sauve Qui Peut (la Vie) render readable what was barely audible, even for native French speakers, in the original. The non-French speaker, however, misses the play between text and image generated by the written materials pervading Deux ou trois choses queje sais d'elle (Two or Three Things I Know about Her, 1966), a film which might be seen as a gloss on Barthes' dictum that 'we are still, more than ever, a civilization of writing,' as he suggested in 'Rhetoric of the Image' (Image/Music/Text, 1977). The omission of non-vocal linguistic messages can also compromise a film's political tendency, since it is often precisely through such messages that a story is socially or historically contexted. Radio allusions to the war in Algeria in Agnes Varda's Clio de 5 a 7 (Cleo from 5 to 7, 1962), scandalised the partisans of L'Algerie Franchise but went untranslated in the English version. At times, the linguistic strategies of a film compromise its political thrust. In Costa-Gavras' Hanna K (1983), presumably a pro-Palestinian film, Arabic dialogue is left unsubtitled, while English masquerades as Hebrew, thus affirming an Israel-US cultural link while downplaying any specific Jewish-Israeli identity. Subtitles, finally, can inject revolutionary messages into non-revolutionary films. In the late '60s and '70s, French leftists reportedly 'kidnapped' Kung Fu films, giving them revolutionary titles such as La Dialectique Peut-Elle Casser les Briques? ('Can the Dialectic Break Bricks?, 1973) and incendiary subtitles. A sequence of devastating karate blows would be subtitled: 'Down with the bourgeoisie!' thus providing a left political 'anchorage' for what were essentially exploitation films.

The linguistic mediation of subtitles dramatically affects the film experience. For audiences in countries where imported films predominate, subtitles are a normal, taken-for-granted part of the film experience. Literalising the semiotic textual metaphor, spectators actually read films as much as they see and hear them, and the energy devoted to reading subtitles inevitably detracts from close attention to images and sounds. In many Third World countries, in fact, the penetration of subtitled foreign films has indirectly led to the physical neglect of the sound systems in the theatres, exhibitors being guided by the spurious logic that spectators occupied in reading subtitles will not be overly concerned with the quality of sound. In Israel, for example, films spoken in Hebrew have at times been subtitled in Hebrew due to lack of confidence in the sound systems of the movie theatres. In countries such as India and Israel, the spectator is at times confronted with vertical tiers of multi-lingual subtitles. In cases where the films themselves are multi-lingual, subtitles have an effect of homogenisation for the foreign spectator. The exuberant polyglossia of such films as Moshe Mizrahi's Ha-Bayit Berchov Chelouche (The House on Chelouche Street, 1973), which features Hebrew, Arabic, Yiddish, Ladino, Spanish and Russian, or Youssef Chahine's Iskandariyya Leh? (Alexandria Why?, 1979), which deploys the diverse languages spoken in the cosmopolitan Alexandria of the '40s, is 'levelled' by monolingual subtitles when shown abroad.

[The foregoing was extracted and edited from an article by Ella Shochat and Robert Stam entitled 'The Cinema After Babel: Language, Difference, Power,' which was originally published in Screen, Vol. 26, Nos. 3-4, May-August, 1985, pp. 35-59. For another take on film subtitles, see 'Subtitles and Film Marketing for US Audiences,' also on TV Multiversity.]