7Orality and Literacy in Cinema and Television
Abstract: This article first provides an overview about the general linguistic comprehension of orality and literacy. Romance and non-Romance concepts are discussed and put into perspective for research on audiovisual language. These notions are crucial for research on the authenticity of orality in the domain of television and movies. Authentic orality is only present in certain communicative settings, which differ in their ways of address (direct address towards the TV spectators vs. multiple address towards spectators and studio audience). In contrast, all fictional televised or cinematic genres feature fictitious orality, i.e., prescripted dialogue enacted in planned communicative settings. We propose ways to do research on language in audiovisual media, be it authentic or not. Moreover, we present possible fields of analysis together with existing publications, focusing mainly on French television and cinema. The notion of literacy in audiovisual media covers subtitles that have been added subsequently, or that are relevant for the understanding of the narration, as well as written elements playing a role in fictional or non-fictional film.
Keywords: audiovisual media, authenticity, broadcast formats, corpus building, language of immediacy and distance, literacy, orality, semiotic systems
1Introduction
Cinema and television as audiovisual media consist of two semiotic systems: the visual one being represented by the shots, and the aural one by the soundtrack. In the latter, the spoken word plays a highly important role, which will be the main subject of the present contribution. We will touch briefly upon written elements within the visual system.
2Concepts of orality and literacy
In the 1970s, Söll was among the first researchers in the field of Romance linguistics to account for a clear distinction between spoken and written language. In his seminal work, “Gesprochenes und geschriebenes Französisch” [‘Spoken and written French’] (1974), he argues that spoken and written language indeed originate from distinct conceptions in the speaker’s mind. Utterances realized in the code phonique [‘phonic code’] are likely to be conceived in the code parlé [‘spoken code’], whereas utterances realized in the code graphique [‘graphic code’] are likely to be conceived in the code écrit [‘written code’]. To illustrate the differences between the codes, two utterances are arranged in a table:
Table 1: Adapted from Söll (1974, 18)
| parlé | écrit | |
| phonique | sɛpapɔsibl fopaldir | snɛpapɔsibl ilnəfopaldir |
| graphique | c’est pas possible faut pas le dire | ce n’est pas possible il ne faut pas le dire |
The conception in the two codes may differ greatly, especially in the case of French, but to a certain extent also in the other Romance languages. There is a typical affinity between code parlé / code phonique and code écrit / code graphique, since the ephemeral code phonique often calls for spontaneous, unplanned utterances (e. g., informal conversation), whereas the persistent code graphique usually allows for more planning effort and for corrections (e. g., letters). Still, there is the possibility of crosswise correlation: for instance, informal chat conversations are realized in the code graphique but usually conceived in the code parlé; a formal speech is realized in the code phonique but conceived in the code écrit.
In 1985, Koch and Oesterreicher further developed Söll’s theory and presented their model of a continuum of immediacy and distance in language (cf. Fig. 1). The key idea here is that the medial realization of a speech act can only be either phonic or graphic (in the model, below or above the horizontal line), but its conception varies along a continuum between immediacy and distance. These are determined by several factors of the communicative situation, such as the physical and emotional proximity of the persons involved, the communication being private or public, dialogic or monologic, spontaneous or reflected. Any communicative act can thus be situated either on the phonic or on the graphic part of the continuum and according to the aforementioned factors, more towards the side of immediacy or more towards the side of distance.
To situate the language of cinema and television on this continuum, we will have to take into account the authenticity of filmic speech, a complex issue involving spontaneous settings and their respective screenplays. The linguistic aspects of authentic and fictitious speech will be discussed in section 2.
Rather late, in the 1970s, French linguists stopped viewing spoken language as a deficient variety of written language. Apart from theoretical essays – also drawing upon the differences between spoken and written language, on the question whether spoken language is to be considered as an autonomous system, and on the erroneous mingling of diamesic variation and registers – researchers in this field published studies on the specific features of spoken language (cf. the journal Recherches sur le français parlé; Blanche-Benveniste/Jeanjean 1987; Blanche-Benveniste 1997).
The difference between oral and written language production was observed before the early 20th century in the non-Romance context, too. Chafe/Tannen (1987) provide an overview of orality-literacy research. By the beginning of the 1980s, Tannen had concluded from her analyses that culture has an effect on orality and literacy, especially on the structuring of narration and the use of paralinguistic features (Tannen 1982b). To overcome a strict separation of oral and written language, she proposes an orality-literacy continuum similar to Koch and Oesterreicher’s which allows for linguistic features to be present in both modes. As a prominent aspect of spoken language, she names cohesion through paralinguistic and nonverbal features like tone of voice, pitch, speed rate, vocal and facial expression (Tannen 1982a, 41), which is an important point for research on audiovisual media: Analyzing a transcript of the spoken dialogue only can be misleading in that the complexity of semiotics is not respected in its totality.
Chafe (1982) focuses on the process of speaking and writing, an aspect Söll had also drawn upon: Spontaneous speech leaves little time for planning, which makes it prone for mishaps in grammar, coherence and cohesion, less complex thematic and syntactic structuring and less lexical variety, whereas writing completely lacks the usual verbal and nonverbal clues of personal interaction.
The first example of quantitative research on the matter of orality and literacy can be found in the works of Biber (e. g., 1988), who argues that many contradictory findings of prior studies are due to small samples of speech and writing. It is indeed a problem to generalize results from small corpora, yet as long as movies and television broadcasts are not transcribed massively, researchers will have to rely on their own corpus production, and quantitative analyses of audiovisual speech material are restrained.
3Authenticity of orality in cinema and television
Authentic orality means that a speech act is conceived and orally performed by the same person, without being scripted beforehand. This implies that cinema almost totally lacks authentic orality, since most movies are long-term projects with dialogue scripts that actors can only modify to a certain extent. On television, however, there are some non-fictional genres featuring spontaneous spoken language. This concerns mainly programs showing real people’s authentic utterances as statements or comments, or discussions and shows without reworking or cutting afterwards.
Fictitious formats usually do not feature authentic language, but utterances that have been scripted before. Yet the aim of fictitious dialogues is usually not to be perceived as such. A very important aspect of movies is immersion, making the spectators plunge into the setting on screen and giving them the feeling of a real experience. This would be impossible if the language used by the movie characters was implausible, as cinema theorist Mitry already noted in the 1960s:
“Le dialogue de film ne doit pas pour autant être débraillé ou incolore mais spontané. Il doit sortir de la bouche des personnages et non de la mémoire des comédiens. En outre, dans les moments de conversation les plus animés, les individus doivent avoir le temps de penser ce qu’ils disent, tout comme dans la vie réelle. Les mots repris, ânonnés, bafouillés […] ne peuvent qu’ajouter parfois au sentiment de réalité vraie” (Mitry 1963, 326).65
Scriptwriters thus have to produce dialogue with a certain amount of orality features. In terms of Söll, scriptwriters conceive language for the code parlé, write them down in the code graphique, and later actors realize utterances in the code phonique. This number of code transfers makes the genesis of movie dialogue highly complex and is certainly at the origin of an important matter: Any transcription of fiction film language will always differ greatly from a transcription of authentic language. This is, beyond the mentioned code transfers, due to the fact that many orality features are used unconsciously in spontaneous speech, highly dependent on the situative context, which makes them very hard to imitate for fictitious speech: gaps and fillers, grammatical norm deviances, anacolutha, lexical errors and self-correction are for these reasons far less frequent. What can be found, though, are grammatical constructions which are more typical of spontaneous oral speech than of written language, and a lexicon closer to colloquial language or substandard than a written text would feature. In early French cinema, substandard forms in movies were reserved for highly marked communication, e. g., among gangsters, or served to portray the working class – not necessarily in an authentic way (Abecassis 2009, 287). With the arrival of Nouvelle Vague films, the dogma of authenticity changed not only the technical and visual aspects of film production, but also fundamentally the conception of dialogue. It became more and more common to have the characters talk like credible authentic persons (Bedijs 2012, 74ss.).
3.1Audience addressing in audiovisual media
Authentic programs can be classified into two subgroups in correspondence to the addressees: single-audience or multiple-audience communication (Hoffmann 1984). While the first is directed at one specific audience – the television spectators or the film team – the latter is directed at the same time at more than one public. Typical examples of single-audience communication are documentaries in which the speakers talk to other people within the filmed setting, or interviews in which the speaker addresses the interviewer. In contrast, a game show host addresses the candidates, the studio public and the television spectators at once and in different modes. Talk show guests are supposed to address each other and the talk show host like in a natural discussion, while the television spectators are usually not directly addressed and yet the reason for the whole setting.
Even newscasts, a seemingly very rigid genre, can still feature different degrees of immediacy. There is scripted news read out by a talking head who has no role in the previous conception of the text and who displays no personal involvement in the content of the report(s). Other news formats, however, have an anchorman/woman (présentateur/présentatrice de nouvelles) who is usually involved in the scripting and may also bring in their personality when presenting the news. These differences must be taken into account when analyzing the linguistic features of news broadcasts: there will be wide variance according to the degree of immediacy. A recent study on language in French newscast is Garric/Léglise (2007).
Meanwhile, documentaries claim, by definition, to represent unaltered reality. This holds true when the filming takes place covertly, which, in turn, entails ethical matters of privacy (discussed in media studies ever since). Yet hidden-camera programs have constantly been popular, like “La Caméra Invisible” (France, 1960s, succeeded by “La Caméra Cachée”) filming people unbeknownst to them in humorous situations. Without going into detail about the academic discussion on documentary techniques, we may say that the mere presence of a camera alters the setting and thus the linguistic behavior of the involved persons (Chapman 2009, 15; cf. also Labov 1972, 209 on the “Observer’s Paradox”). This is even more true in the case of the recently very popular, so-called reality TV genres like, in France, “Loft Story”, “Star Academy”, “Danse avec les stars”, “Koh-Lanta” and many more (for a detailed discussion of the various documentary and reality genres cf. Hißnauer 2011). A truly hidden camera is the exception; participants are usually aware they are being filmed, and in many genres what is presented as reality has at least partly been scripted beforehand. Thus, when analyzing the language of documentaries and reality genres, we have to distinguish between onscreen dialogue and in many cases an offscreen commentator (added to the soundtrack during postproduction), who reads out a prepared text featuring more or less orality markers in order to sound natural to the spectators.
Interviews involve (at least) one person who asks questions and one or more others who reply. The degree of immediacy varies according to the setting: an interview between a journalist and a politician may be staged formally and may, from a linguistic point of view, be rather distant; an interview between an entertainer and a pop star may be informal and close; emotions in behavior and in language may be involved overtly or covertly. Beyond spontaneous interviews without any preparation, there are those with preconceived questions, sometimes even written down on cards or on the autocue. The interviewer often provides the interview partner with these questions before the shooting to avoid misunderstandings or undesired reactions in a live broadcast situation. Such aspects of preparation situate interviews between conceptional orality and literacy. Hints on how to analyze interview language can be found in Clayman/Heritage (2002); the interview as a mediatic genre is discussed in Andersen (2007).
Most game shows take place in a similar setting: the show studio where the host and the candidates interact with each other and with the studio audience. This setting is being filmed and broadcast to the television spectators, who form another audience. This latter is not secondary in that it is only watching the scene: usually the host addresses the TV audience directly to create a simulation of interactivity. In some shows, the TV audience can actually participate in the event, e. g., by voting for the best candidate. Casting shows like “Popstars”, “Star Academy” and “Loft Story” are a very popular genre featuring audience voting.
Talk shows are a similar case as game shows in that they also feature the host-and-guest setting. Most talk shows have a studio audience (otherwise they must be considered as single-audience communication). Often the situation can be described like an interview that takes place in a studio with an audience that comments on the dialogue with applause or break-ins. This will obviously affect the participants’ linguistic behavior; the host may choose provocative questions to disconcert the guest; the guest may choose to (not) answer questions by using commonplaces and platitudes, technical terminology to create an effect of invulnerability and to avoid objection, etc. Other talk shows are less focused on the structure of question and answer, but rather on conversation between the guests, with the host merely acting as a prompter. The language in talk shows is not only affected by the persons involved, but also by the scope of the format and by the specific topic. In a political talk show with the aim of a heated discussion, the participants will make more use of special language to argue their point, they will make an effort to produce grammatically correct sentences and, to undermine the opponent’s arguments, they might even be offensive.
4Analyzing orality in cinema and television
The most common way to do research on language in audiovisual media is to analyze transcriptions of the speech parts. Linguists interested in working with the collections of public audiovisual media archives should bear in mind that what is stored there is usually the media data itself; transcriptions of the speech have to be prepared in every individual case.
Researchers analyzing language in fictitious films should always be aware of the constructedness of the utterances. Yet what Abecassis states for his analyses of popular language in French 1930s movies holds true for any other era and genre:
“La langue du cinéma est un ensemble où la musique et les images se fondent avec la multiplicité des voix pour participer à l’illusion cinématographique. Malgré leur élément de fiction, les films nous permettent d’observer par le prisme de la caricature les faits langagiers que cinéastes et dialoguistes considèrent caractéristiques de la langue populaire et dont les traces sonores subsistent essentiellement au cinéma” (Abecassis 2008, 14).66
4.1Corpus building
Due to the tension between oral and written code, the shift from orality to literacy entails the well-known problems of conversation analysis: Providing a faithful transcription of the spoken word is nearly impossible, and transcriptions will always be incomplete. Transcribing filmic speech requires much time and effort, all the more if the resulting text is supposed to be very detailed. Thus, a corpus of filmic language will probably comprise only a few movies or some well-chosen excerpts of several films. For qualitative analyses, this is not necessarily a disadvantage, and even small corpora can reflect quantitative linguistic tendencies, but the results should be checked against larger matching corpora of oral speech in order to give an idea of significance.
For these reasons, when elaborating a corpus of transcribed TV or cinema dialogues, the old design and architecture principle “form follows function” is fundamental. This means the researcher should know exactly what she or he wants to use the corpus for, and decide on some methodological aspects according to the research questions (cf., e. g., Korte 42010, 53).
–The decision to include or exclude the pictorial code:
TV and cinema rely on sound and image. Yet including the pictorial code into a linguistic analysis requires not only knowledge about the methods of media studies, but also great effort for elaborating such a semiotically complex corpus. Once the decision is made to include the visual part, the advantage is that the transcribed speech does not have to stand for itself, but can be analyzed in context within the whole complexity of signs. Especially for research questions touching on pragmatics, it can be crucial to consider gestures, postures, facial expressions, patterns of movement, proxemics and all kinds of paraverbal behavior.
–The decision for phonetic or orthographic transcription:
There are several standards for the transcription of conversation for linguistic analysis (cf., as an example, Selting et al. 2009 on the elaboration of GAT 2 conventions; less detailed are the systems of the corpus Valencia Español Coloquial [Grupo Val.Es.Co., n.d.] and of the Corpus de référence du français parlé [Équipe DELIC 2004]). The researcher should choose the suitable set of standards according to the research question. If the work is about phonetic variation (for instance, between characters of different ages, geographic provenance, or social milieus, between movies from different eras, or between real persons on television in different settings), or questions of pragmatics (such as greeting patterns, conversational behavior in talk shows, prosody of political speeches), the transcription requires phonetic details. It can be a very hard task for the researcher to elaborate a phonetically faithful transcription: the sound quality of older records can be poor and she or he must listen again and again to understand correctly, or the dialogue follows the rules of natural conversation involving all kinds of interruptions, overlappings, mumblings etc. that make a recorded colloquial conversation difficult to understand and even more difficult to transcribe.
Nonetheless, the more details a phonetic transcription contains, the more possibilities it offers for further analysis, and the more challenging it becomes for reading. An orthographic transcription makes the above-mentioned analyses impossible, but in contrast can be easily read and processed. For analyses concerning lexis or syntax, this will be the method of choice. In case the elaborated corpus shall be annotated and processed automatically, the researcher must use one single graphic form for all corresponding oral signs – e. g., always transcribe <je suis> even if the phonic code varies between [ʒəsɥi], [ʃsɥi] and [ʃɥi]. There is little room to render phonetic variation, but within certain limits, the researcher can decide to include such information by transcribing variants (for instance, rendering <oui> only when the pronunciation is close to [wi] and <ouais> when it is close to [wɛ]). Such varying transcription should be explained to the reader and used consequently throughout the whole transcribed excerpt.
4.2Fields of research
In the following, we will present some possibilities for research on orality and literacy in television and cinema. Researchers should always bear in mind which data set they use as a corpus. While non-fictional television broadcasts may serve as a faithful resource to analyze linguistic features, fictional formats (be they produced for TV or for cinema) represent authentic language to a lesser extent (cf. above). The analysis of fictitious language can still tell us about the official use of language in different communicative settings and about the general image of certain linguistic features.
4.2.1Phonetics and phonology
Audiovisual media can be used to document diachronic changes in the phonetic system of a language. As an example, the phoneme /ɑ/ had disappeared from Metropolitan French by the beginning of the second half of the 20th century. Since there are only few corpora of spoken French of that era, the analysis of TV recordings could answer the question “When did /ɑ/ disappear and in which communicative contexts did it prove to be most stable?” The corpus should contain similar excerpts from different years; the pieces need not be very long but must contain possible occurrences of /ɑ/ (like [pɑt] as the archaic realization of <pâtes>) or their replacements (in this case, [pat]). Jensen (2004) proposes an analysis of French liaison in different TV genres. Émond/Ménard/Martel (2007) provide an example of the use of speech processing software to investigate prosody in newscasts from Quebec.
4.2.2Lexis
Every corpus of spoken language transcribed following an orthographic standard allows lexical analyses: word frequencies, semantic maps, collocations, contextualization, phraseologisms, use of loanwords and special language, and many more. Such analyses can be conducted for any kind of broadcasting format, provided that the chosen corpus is consistent and can be compared with other data. For instance, a research question could be “Is there a difference between French societal and political talk shows in the use of English loan words?” or “Do politicians use more special language items in written declarations or in speeches?” Several interesting studies on lexis in French politicians’ language use have been conducted in the last years, some of them also including audiovisual media, like Mayaffre (e. g., 2012a; 2012b) in his works on Chirac, Sarkozy and Hollande.
4.2.3Syntax
As for the case of lexis, syntactic analyses require orthographic transcripts. Especially in the field of the conflict between orality and literacy, audiovisual recordings can shed light on constructions typical of oral language. One case to analyze could be the structure of interrogative clauses in French, where we find several possibilities (intonation, inversion, est-ce que, construction with interrogative word) that are distributed very differently according to register and code. A research question could be “Is there a preference for a certain interrogation type in French interviews? Does the pattern vary according to the subject or the involved persons?” Several papers on syntactic issues in French television can be found in Broth et al. (2007) (e. g., Garric/Léglise; Le Bot/Schuwer).
4.2.4Varieties
It is common to have fiction film characters use a more or less colloquial version of standard language with few variational features in order to ensure maximal comprehension for the audience, which is a logical economic aim of cinema producers. Yet there are many cases in which we can witness linguistic variation on different levels (which are not always strictly separated from one another).
According to Koch/Oesterreicher’s model of linguistic variation (cf. 1990, 15), the first variational level is diatopic variation. In French cinema, dialectal features are rather uncommon (cf. Chion 2008). The linguistic centralism of Parisian speech leads to a regular use of this variety as standard for French movies. When dialect is used, it is usually highly marked. An example is the very popular “Bienvenue chez les Ch’tis” (Boon 2008), which features (not very articulated) regional Southern French and (very articulated) Picard (cf. Reutner 2013; Planchenault 2012). Italian cinema makes more use of regional varieties. “Gomorra” (Garrone 2008), for instance, features Napoletan and Casalese dialect in order to make the dialogues appear more authentic – as a consequence, part of the film is subtitled in Standard Italian to be understood by all spectators.
The second variational level involves the social background (diastratic variation). Here, many different aspects come into play: education, friends and family background, gender, age, professional environment and more. All these shape an individual’s linguistic behavior. A society usually knows stereotypes about many linguistic features, e. g., certain phonological traits, grammar mistakes and a heavy use of swear words are connected with a low educational level and a socially disadvantaged environment, whereas distinct pronunciation and the use of complex phrases is connected with higher education and material prosperity. Scriptwriters are aware of such stereotypes and make use of diastratic features to create certain effects, like the substandard speech used in French “banlieue” films since the 1990s (cf., e. g., Fiévet/Podhorná-Polická 2008), adolescent language in French movies across six decades (cf. Bedijs 2012) or français populaire in films on the Parisian working class of the 1930s (cf. Abecassis 2005).
On the third level, situative (diaphasic) variation means that an individual’s linguistic behavior depends on the communicative situation: the persons involved, the subject etc. It is often difficult to consider diaphasic variation in isolation, since the social context (and thus diastratic variation) is always part of the communicative situation. Diaphasic variation can mean the use of formal and/or special language in an office meeting, the use of swear words in a quarrel, the use of a higher voice pitch in a conversation with a child, the use of colloquial language among teenagers, etc. One of the most experimental French movies, “Zazie dans le métro” (Malle 1960), is notable not only for the portrayed Parisian working-class people’s extensive use of non-standard French (mainly français populaire and argot), but also for the little girl Zazie’s use of substandard and swear words, which must have been judged unexpected and inappropriate at least by the contemporary audience.
As to language on television, there is basically no limit to variation as long as the communicative setting allows authentic linguistic behavior. Yet when people are aware of talking in front of a camera, they most probably adapt their language to what they believe is appropriate for this situation – which means that dialect speakers may switch to a less pronounced version of their usual idiolect. If they do not, subtitling is a frequent strategy. Most French TV stations subtitle in Standard Metropolitan French
“[l]orsque les téléspectateurs risquent de ne pas comprendre les propos tenus par une personne qui parle dans un dialecte particulier ou dans une forme de français d’un autre pays, ou tout simplement dans un français non-compréhensible […]” (Mission à la langue française 2014).67
Also, most people will at least try to use less – in their view – stigmatized language like vulgarisms, swear words and grammatical mishaps. In some cases, the editors of the program bleep out such unwanted utterances to avoid complaints.
4.2.5Discourse
An approach to uncover relationships between language use and the state of a society is (critical) discourse analysis. Without going into detail here (for an overview cf. Wodak/Meyer 22009), it basically assumes that our use of language is formed by the conditions we live in, and that our linguistic choices reciprocally shape our social environment. Analyzing language thus allows us to learn about norms and rules of a society, and also about the distribution of power within it. Corpora can tell about this on all linguistic levels – phonological (prosody, use of phonetic features distinct of a certain milieu), lexical (word frequencies, lexicon), syntactic (use of pronouns, structure of impersonal constructions), semantic (use of metaphors and word games, semantic relations, priming), textual (thematic development, overall structure of discourse contributions), and pragmatic (address forms, politeness, face work, communicative behavior like turn taking). A systematic analysis of at least some of these aspects can provide an understanding of underlying social structures. Especially political broadcasts lend themselves to investigation, for instance the TV debates organized between presidential candidates before elections in France: “Which linguistic means do the candidates use in order to come out on top against the adversary? Do male and female politicians use different linguistic strategies to communicate in a debate?” Sullet-Nylander/Roitman (2010) analyze questioning and answering strategies in the TV debate between Royal and Sarkozy, whereas Sandré (2009) draws on turn-taking behavior in the same debate. Some recent studies on the topic of televised political debates use an analysis of several linguistic fields to draw conclusions on the level of discourse (cf. for instance the papers in Burger/Jacquin/Micheli (2011) and Bacot et al. 2010 on various linguistic aspects of political debates in France; Rémi-Giraud 2010 on lexis and semantics; Sandré 2011 on the use of smiles and laughter in political debates; Peñafiel 2011 on the linguistic self-construction of a political leader).
Fictional genres offer perspectives on social conditions as well. As an example, we may use the French genre of banlieue movies since the beginning of the 1990s, which feature many young characters of Arab or African origin. Their language is often designed in a stereotyped way with many features of substandard, argot, phonetic markers of immigrant languages, swear words (cf., e. g., Fiévet/Podhorná-Polická 2008; Bedijs 2012) – all of which are commonly recognized as belonging to a disadvantaged milieu. Comparing these characters’ linguistic behavior to that of others (mostly portrayed as white Français de souche) in the same movies, we can see whether fictitious language use manages to draw a line of power between different social groups, which, in turn, may affect our perspective on the portrayed groups in real life. Another example is Palma-Fahey’s (2012) paper on the construction of social roles and power by the use of vocatives in “Machuca” (Wood 2004) and “Volver” (Almodóvar 2006).
4.2.6Multilingualism
Multicultural societies are nowadays the normal case in all Romance-speaking countries. Many communities of practice use other than the national language. In movies, this has been reflected only in recent years. The already mentioned French banlieue films make use of languages such as Arabic, Romani and others spoken by immigrant minorities. Even though this may be the reality of many French citizens, this linguistic portrait is often used for stereotyped characters from a disadvantaged environment.
The multilingualism of immigrant characters is often presented as code-switching and code-mixing, that is, the alternation of languages in the conversation with complete turns or complete sentences being in one language (code-switching) or the alternation within sentences, inserting phrases or words in another language (code-mixing). To make sure that all spectators can understand the whole dialogue, many movies have subtitles in the multilingual passages. In some cases, though, the director obviously does not deem this necessary. The multilingual dialogue without subtitles creates an effect of foreignness, often connected with the existing stereotypes on the respective community. It is also the reality of many native French not to understand what their immigrant fellow citizens are saying to each other, and this is mirrored in such uncommented multilingual scenes. An example of untranslated code-switching and code-mixing is “L’Esquive” (Kechiche 2004) featuring Arab mixed into French.
Another way of creating foreignness in movies is to make the characters speak the audience’s language, but with a strong accent. Besides the realistic situation of a “foreign” character conversing with a character in the main movie language, the foreign accent often appears when a group of foreign characters has important roles in the movie and subtitling all of their dialogues would distract the spectators from the action. It is also used to create stereotyped effects about speakers of a certain language (cf. Planchenault 2012 on stylization in French movies). In his study on multilingualism in Hollywood movies and the consequent construction of characters, Bleichenbacher (2008b, 18) mentions the use of Mock Spanish – an artificial, ungrammatical variety of English incorporating Spanish accent and loanwords – as a means of “linguistic racism”. Several (US-produced) movies about Mexican immigrants in the United States also create a (questionable) linguistic portrait of the characters having them use a deficient variety of English (“Real Women have Curves”, Cardoso 2002; “Spanglish”, Brooks 2004; “Quinceañera”, Glatzer/Westmoreland 2006).
A different example in French cinema is “L’Auberge Espagnole” (Klapisch 2002). In a community of Erasmus students sharing a flat in the Catalan capital Barcelona, each one speaks their own language – Spanish, French, German, Italian, English and Danish –, they use (broken) English as a lingua franca to communicate with each other, struggle with the national language Spanish and discover that the regional language Catalan is indispensable at university. To deal with the many source languages, the movie is subtitled. One version offers subtitles in the languages heard onscreen without translation, another one offers subtitles translated into the target language, distinguishing the characters by the use of colors (cf. Bartoll 2006).
On television, we mostly deal with multilingualism in newscasts from foreign countries. To ensure the audience’s comprehension, such contributions are usually translated. In contrast to fictional movies, the goal here is not maximal immersion but authenticity, which can be reached by maintaining the original language and placing subtitles at the bottom of the screen. Another translation method typical of TV is voiceover: the spoken translation and the original utterance run simultaneously so that the spectator can hear the original voice. We will come back to subtitling in the section on literacy, and Baños elaborates on the different methods in audiovisual translation ↗20 Audiovisual Translation.
A complex case is the Europe-wide broadcast show “Eurovision Song Contest”. Its hosts address mainly the TV audience (even though there is a considerable audience in front of the stage), usually in English, French and the main language of the host country. Each country’s broadcasting company sends their own commentators who choose to interpret as much as possible, to resume shortly what is being said or to ignore this completely in favor of their very own commentary for the TV audience. At the end of the show, spectators in all countries are invited to vote for their favorite performance. The results of the voting are presented by entertainers or journalists who address primarily the show hosts – often beginning with a ritual formula such as “thank you XY for the great show, you are beautiful, we are having a fantastic evening” –, and then reading out the scores in a ritual way: “Country X – 8 points, Country Y – 10 points, and our 12 points go to … Country Z!” In case this has been said in English, the hosts on stage repeat the number of points in French (“Pays X – huit points”). Yet the traditional bi- or even trilingualism of the whole show has eroded in the last years in favor of English as the main language. In the 2014 show, the French presenter was the only one who did not read out the results in English.
5Literacy in cinema and television
Even though both cinema and television are mainly using aural signs to transport language and visual signs to transport action, they also make use of written language elements. These may serve various aims. We will start describing written language onscreen, incorporated in the filmed setting, and then briefly broach the issue of subtitles, which is discussed in detail in ↗20 Audiovisual Translation.
5.1Written elements in the onscreen setting
Our world is full of written elements: inscriptions and signs on shops, markers on the road, publicity posters, graffiti on walls, etc. We make sense of our surrounding by decoding these written elements. A sign on a shop window or above the door reading “Café” or “Boutique de cadeaux” lets us know what to expect inside. Many road signs include, beyond an iconic picture, written information such as toponyms, speed limits or points of interest. All these written elements – the so-called linguistic landscape (Bleichenbacher 2008a, 189ss.) – appear in the filmed setting, providing an option for the spectator to figure out meaning. They may appear without interventions by the film team, as in news, documentaries and also in movies shot on authentic locations. Movies filmed in studios also need a realistic setting, so there will be a made-up environment including written elements as well. Other common written elements are letters, e-mails, text messages, screenshots, diaries, sticky notes and all kinds of writing that is featured onscreen.
Some of these elements play no special role beyond being part of the illusion of reality. Others are key features to the understanding of a scene. It certainly makes a difference if a character exchanges a few words with a person behind a desk with the sign of a travel agency or of health counseling. Notes on paper may be part of the action of the movie (as the letters that Xavier writes to his girlfriend in “L’Auberge Espagnole”) or include the central message (as in official documents shown in news reports). Even typography and font color are an important hint for the understanding of a scene – imagine a character entering a hotel with a blue or a red light sign (pointing to prostitution): these are also culturally-specific parts of the graphic meaning.
This being said, we can imagine the difficulty of making a movie fully available to a foreign audience. Not only the aural part has to be taken into account; there may be written elements which are fundamental for understanding. The translators have the difficult task to decide how to transfer them from one language to another. In the case of graphics on location, there is seldom a problem. A café can be recognized as such by other elements than the sign; a road sign can often be decoded without translation. Some written elements are translated using a subtitle if the understanding is necessary. Longer written parts are usually read out aloud in the target language (mostly already in the original version) creating an effect of divergence of words and image (cf. Renner 2001), which spectators can usually ignore since their logic expects the text to be read in the target language.
5.2Subtitles and intertitles
Subtitles and intertitles are very prominent written elements in movies and television. Intertitles have been used in movies since the beginning of the cinematic age, when movies were still soundless. Full screens inserted between two movie scenes, they represent a practicable way to add information on the events shown onscreen, details about the actors or retrospection. It is still not uncommon to have intertitles in movies which announce a change of setting (“Paris 1995”, “A year before”, “Six months later in Marseille”…). Such information allows the spectators to immediately locate the following scene and spares the necessity of explicit explanations by the actors. It is also a common means to cope with multilingualism in the narration: An intertitle locating the setting geographically helps the viewer to understand why the characters speak a foreign language in the following sequence, or allows maintenance of the base language of the movie even if the location would logically require the use of a foreign language.
Subtitles have also been part of movies for a very long time. In the early years of cinema, dubbing a movie for a foreign audience required great technical effort. An alternative was to reshoot the whole movie in the target language (the so-called “multiple-language versions”, cf. Crafton 1997, 425), which meant extremely high costs and most often a new cast of actors able to speak the target language. This was only done for the most promising movies. The solution to export movies with a lower budget was to insert subtitles: written text at the bottom of the screen, usually no more than two lines, reproducing the oral dialogue.
There are different types of subtitles (for a detailed typology, cf., e. g., Díaz Cintas/Remael 2014, 14) fulfilling very different tasks:
–Subtitles for the hearing impaired. These subtitles help people with hearing difficulties who master the language of the movie to understand the dialogue.
–Subtitles for a foreign audience. These subtitles help a foreign-language speaking audience to follow the action by reading the dialogue.
Both of the cases mentioned above must cope with the conflict between oral and written language described in 1.1. To be processed rapidly by the reader in spite of the distraction of the moving picture, subtitles must comply with some graphic rules like contrast, font size and font style. Also, the space for subtitles is very limited – in order to not surpass the two bottom lines and to give the reader the chance to read the text in a decent time, the normal case is that the dialogue cannot be rendered completely and must be shortened (cf., e. g., Remael 2010, 15). Orality features such as those mentioned above – anacolutha, fillers, lexical errors – are the first to be eliminated. Next comes the linguistic variation: subtitles tend to be written in standard language, dialects and substandard language is often rendered less faithfully – for several reasons: mostly because they are not easily understood by the whole audience, then because it is sometimes difficult to render oral varieties in written code, and last because the written word is more marked and thus has a higher impact than the spoken one, i.e., a written swear word shocks the audience more than the same word in the oral dialogue (for an analysis of the rendering of emotional language in interlingual subtitles cf. Franzelli 2008).
As opposed to the foreign audience, the hearing-impaired audience does not perceive the background sound of the movie, be it music or sounds made by the characters, cars, planes etc. Subtitles for this audience must, therefore, provide information about these sounds, at least when they are relevant for the understanding of the action. This cuts even more from the available text space (or covers more of the image). To separate dialogue from sounds or unspecific voices, the latter are usually written in square brackets. To make a change of speakers more obvious to the hearing-impaired audience, the text lines use different colors for different characters. A detailed description of these matters can be found in Neves (2005).
–Captions (cf. Ivarsson/Carroll 1998, 97) adding information to the action onscreen. These subtitles function similar to intertitles, but without interruption of the movie. They locate or date the scene, name characters, objects or places when such information is considered important for the spectators to understand the action. Although clues can often be provided through a comment in the dialogue, it is sometimes easier or more natural to insert a text.
–Subtitles translating foreign language elements in the dialogue. This is usually done when written elements are relevant for the action. Many written elements in the linguistic landscape (also called displays, cf. Ivarsson/Carroll 1998, 97; Karamitroglou 2000, 5) stand for themselves or can be understood by a foreign audience without translation, but when their meaning is opaque and still relevant to the understanding of the plot, a textual translation helps the viewer. The same occurs for longer texts such as letters or e-mails, which are either read out aloud by a character or translated in subtitles.
While the first two types of subtitles suppose a shift from the aural to the graphic code, the latter two are situated in different semiotic codes. Translating text that is also shown onscreen is simply an interlingual translation of written language. Adding textual information that is not necessarily redundant to the visual information means creating a new code consisting of scriptural and pictorial elements.
6Perspectives and desiderata
In the previous section, we have given some examples of research questions on audiovisual media and cited only a tiny sample of the published studies on this subject. Yet linguistic research on orality-literacy features in movies and television is not by any means a finished domain. The most crucial problem is probably the lack of corpora that can be analyzed systematically. With the advancing digitalization of audiovisual media, availability of material and tools for research, the task of building and analyzing corpora could soon be easier.
Audiovisual media provide the possibility to analyze orality on every linguistic level – in the case of non-fictional recordings, the material serves as authentic data, whereas fictitious dialogue reflects the common ideas on what language is like at a given time and in a given situation. In this perspective, both fictitious and authentic film speech can provide insights on the state of a language in past times. At the same time, comparing features of spoken language from filmed and non-filmed situations may reveal great differences even if the setting is comparable. Such investigations have not yet been conducted.
Whereas subtitles and additional text is well investigated thanks to researchers in translation studies, onscreen written elements have largely been neglected and would merit further attention; the linguistic features of text elements as well as their relation to the spoken word; their role in the polysemiotic system of audiovisual media as well as the consequences for translation.
A topic that will gain importance in the following years is certainly the role of digital audiovisual media available on the Internet. Broadcasting and movie companies are no longer the only producers of audiovisual content; every individual with a filming device on hand can now create and share videos online via platforms such as YouTube, Vimeo and others. Changing habits in media consumption, the interchangeable role of producer and spectator, the interactivity of online platforms and the influence of other newly-emerging medial genres will modify the medial offerings and their linguistic features. Next-door people talk about their personal hobby into their webcam in order to create video tutorials for interested online users, being part of a worldwide community of nonprofessionals. The questions arising from this evolution are interesting not only for linguists, but also for sociology and media sciences. Interdisciplinary approaches will thus be most promising in this field.
7References
Abecassis, Michaël (2005), The Representation of Parisian Speech in the Cinema of the 1930s, Oxford et al., Lang.
Abecassis, Michaël (2008), Langue et cinéma: aux origines du son, Glottopol 12, 6–16.
Abecassis, Michaël (2009), La représentativité du français parisien dans le cinéma des années 30–40, in: Dorothée Aquino-Weber/Sara Cotelli/Andres Kristol (edd.), Sociolinguistique historique du domaine gallo-roman. Enjeux et méthodologies, Bern et al., Lang, 283–316.
Almodóvar, Pedro (2006), Volver, El Deseo.
Andersen, Hanne Leth (2007), L’interview comme genre médiatique: sous-catégories pragmatiques et leurs traits linguistiques caractéristiques, in: Mathias Broth et al. (edd.), Le français parlé des medias, Stockholm, Acta Universitatis Stockholmiensis, 447–462.
Bacot, Paul, et al. (edd.) (2010), Trente ans d’étude des langages du politique (1980–2010), Mots – Les langages du politique 94, numéro special, Lyon, ENS Éditions.
Bartoll, Eduard (2006), Subtitling multilingual films, in: Mary Carroll/Heidrun Gerzymisch-Arbogast/Sandra Nauert (edd.), Proceedings of the Marie Curie Euroconferences MuTra: Audiovisual Translation Scenarios, <http://euroconferences.info/proceedings/2006_Proceedings/2006_Bartoll_Eduard.pdf> (28.10.2016).
Bedijs, Kristina (2012), Die inszenierte Jugendsprache. Von “Ciao, amigo!” bis “Wesh, tranquille!”: Entwicklungen der französischen Jugendsprache in Spielfilmen (1958–2005), München, Meidenbauer.
Biber, Douglas (1988), Variation Across Speech and Writing, Cambridge, Cambridge University Press.
Blanche-Benveniste, Claire (1997), Approches de la langue parlée en français, Gap/Paris, Ophrys.
Blanche-Benveniste, Claire/Jeanjean, Colette (1987), Le français parlé. Transcription et édition, Paris, Didier Érudition.
Bleichenbacher, Lukas (2008a), Linguistic Replacement in the Movies, Poznań Studies in Contemporary Linguistics 44:2, 179–196.
Bleichenbacher, Lukas (2008b), Multilingualism in the Movies: Hollywood Characters and Their Language Choices, Tübingen, Francke.
Boon, Dany (2008), Bienvenue chez les Ch’tis, Pathé.
Brooks, James L. (2004), Spanglish, Columbia/Gracie Films.
Broth, Mathias, et al. (edd.) (2007), Le français parlé des medias, Stockholm, Acta Universitatis Stockholmiensis.
Burger, Marcel/Jacquin, Jérôme/Micheli, Raphaël (edd.) (2011), La parole politique en confrontation dans les medias, Bruxelles, De Boeck.
Cardoso, Patricia (2002), Real Women have Curves, HBO Films/Newmarket Films/LaVoo Productions.
Chafe, Wallace (1982), Integration and Involvement in Speaking, Writing, and Oral Literature, in: Deborah Tannen (ed.), Spoken and Written Language: Exploring Orality and Literacy, Norwood/New Jersey, Ablex, 35–53.
Chafe, Wallace/Tannen, Deborah (1987), The Relation between Written and Spoken Language, Annual Review of Anthropology 16, 383–407.
Chapman, Jane (2009), Issues in Contemporary Documentary, Cambridge (UK)/Malden (MA), Polity Press.
Chion, Michel (2008), Le complex de Cyrano. La langue parlée dans les films français, Paris, Cahiers du cinéma.
Clayman, Steven/Heritage, John (2002), The News Interview: Journalists and Public Figures on the Air, Cambridge, Cambridge University Press.
Crafton, Donald (1997), The Talkies: American Cinema’s Transition to Sound, 1926–1931, Berkeley/Los Angeles/London, University of California Press.
Díaz Cintas, Jorge/Remael, Aline (2014), Audiovisual Translation: Subtitling, Abingdon/New York, Routledge.
Émond, Caroline/Ménard, Lucie/Martel, Guylaine (2007), Une analyse prosodique des téléjournaux québécois, in: Mathias Broth et al. (edd.), Le français parlé des medias, Stockholm, Acta Universitatis Stockholmiensis, 159–170.
Équipe DELIC (2004), Présentation du Corpus de référence du français parlé, <http://sites.univ-provence.fr/~veronis/pdf/2004-presentation-crfp.pdf> (28.10.2016).
Fiévet, Anne-Caroline/Podhorná-Polická, Alena (2008), Argot commun des jeunes et français contemporain des cités dans le cinéma français depuis 1995: entre pratiques des jeunes et reprises cinématographiques, Glottopol 12, 212–240.
Franzelli, Valeria (2008), Traduire la parole émotionnelle en sous-titrage: colère et identités, Études de linguistique appliquée 150, 221–244.
Garric, Nathalie/Léglise, Isabelle (2007), Aspects syntaxiques et discursifs d’un français parlé des médias: “le discours d’information télévisé”, in: Mathias Broth et al. (edd.), Le français parlé des medias, Stockholm, Acta Universitatis Stockholmiensis, 243–258.
Garrone, Matteo (2008), Gomorra, Fandango.
Glatzer, Richard/Westmoreland, Wash (2006), Quinceañera, Cinetic Media/Kitchen Sink Entertainment.
Grupo Val.Es.Co. (n.d.), Sistema de transcripción, <http://www.uv.es/corpusvalesco/convenciones.html> (28.10.2016).
Hißnauer, Christian (2011), Fernsehdokumentarismus: theoretische Näherungen, pragmatische Abgrenzungen, begriffliche Klärungen, Konstanz, UVK.
Hoffmann, Ludger (1984), Mehrfachadressierung und Verständlichkeit, Zeitschrift für Literaturwissenschaft und Linguistik 55, 71–85.
Ivarsson, Jan/Carroll, Mary (1998), Subtitling, Simrishamn, TransEdit.
Jensen, Ole Kongsdal (2004), Liaison, enchainement et réduction, in: Hanne Leth Andersen/Christa Thomsen (edd.), Sept approches à un corpus. Analyses du français parlé, Bern et al., Lang, 139–163.
Karamitroglou, Fotios (2000), Towards a Methodology for the Investigation of Norms in Audiovisual Translation: The Choice Between Subtitling and Revoicing in Greece, Amsterdam/Atlanta, Rodopi.
Kechiche, Abdellatif (2004), L’Esquive, Lola Films/Noé Productions.
Klapisch, Cédric (2002), L’Auberge espagnole, Bac Films/Ce Qui Me Meut.
Koch, Peter/Oesterreicher, Wulf (1985), Sprache der Nähe – Sprache der Distanz. Mündlichkeit und Schriftlichkeit im Spannungsfeld von Sprachtheorie und Sprachgeschichte, Romanistisches Jahrbuch 36:85, 15–43.
Koch, Peter/Oesterreicher, Wulf (22011 [1990]), Gesprochene Sprache in der Romania: Französisch, Italienisch, Spanisch, Berlin/New York, de Gruyter.
Korte, Helmut (42010), Einführung in die Systematische Filmanalyse, Berlin, Schmidt.
Labov, William (1972), Sociolinguistic patterns, Philadelphia, University of Pennsylvania Press.
Le Bot, Marie-Claude/Schuwer, Martine (2007), Syntaxe du français parlé des médias: le cas de la “relative orpheline”, in: Mathias Broth et al. (edd.), Le français parlé des medias, Stockholm, Acta Universitatis Stockholmiensis, 419–432.
Malle, Louis (1960), Zazie dans le métro, Nouvelles Éditions de Films.
Mayaffre, Damon (2012a), Le discours présidentiel sous la Vème République. Chirac, Mitterrand, Giscard, Pompidou, De Gaulle, Paris, Presses de Sciences Po.
Mayaffre, Damon (2012b), Mesure et démesure du discours. Nicolas Sarkozy (2007–2012), Paris, Presses de Sciences Po.
Mission à la langue française (2014), L’usage du français dans les médias, via private e-mail correspondence, 20.10.2014.
Mitry, Jean (1963), Esthétique et psychologie du cinéma, Paris, Éditions du Cerf.
Neves, Josélia (2005), Audiovisual Translation: Subtitles for the Deaf and Hard-of-Hearing, PhD Thesis, London, Roehampton University, <http://roehampton.openrepository.com/roehampton/bitstream/10142/12580/1/neves%20audiovisual.pdf> (28.10.2016).
Palma-Fahey, María (2012), Exploring the representation of orality: the use of vocatives in two Spanish-speaking films, Machuca and Volver, Sociolinguistic Studies 5:1, 103–126.
Peñafiel, Ricardo (2011), L’image du peuple. Construction de l’ethos plébéien de Hugo Chávez dans l’émission Aló Presidente, Mots – Les langages du politique 96, 29–44.
Planchenault, Gaëlle (2012), Accented French in films: Performing and evaluating in-group stylisations, Multilingua 31:2, 253–275.
Remael, Aline (2010), Audiovisual Translation, in: Yves Gambier/Luc van Doorslaer (edd.), Handbook of Translation Studies 1. Translating and Interpreting, Amsterdam/Philadelphia, Benjamins, 12–17.
Rémi-Giraud, Sylvianne (2010), Sémantique lexicale et langages du politique. Le paradoxe d’un mariage difficile?, Mots – Les langages du politique 94, 165–173.
Renner, Karl Nikolaus (2001), Die Text-Bild-Schere. Zur Explikation eines anscheinend eindeutigen Begriffs, Studies in Communication Sciences 1, 23–44.
Reutner, Ursula (2013), Spécificités culturelles et traduction: l’exemple de “Bienvenidos al Norte”, in: Emili Casanova Herrero/Cesáreo Calvo Rigual (edd.), Actas del XXVI Congreso Internacional de Lingüística y Filología Románicas, València 2010, vol. VIII, Berlin/Boston, de Gruyter, 445–456.
Sandré, Marion (2009), Débat politique télévisé et stratégies discursives: la visée polémique des ratés du système des tours, in: Marcel Burger/Jérôme Jacquin/Raphaël Micheli (edd.), Actes du colloque “Les médias et le politique. Le français parlé dans les médias”, Lausanne, Centre de linguistique et des sciences du langage, <https://www.unil.ch/clsl/files/live/sites/clsl/files/shared/Actes_FPM_2009/SandreFPM2009.pdf> (28.10.2016).
Sandré, Marion (2011), Mimiques et politique. Analyse des rires et sourires dans le débat télévisé, Mots – Les langages du politique 96, 13–28.
Selting, Margret, et al. (2009), Gesprächsanalytisches Transkriptionssystem 2 (GAT 2), Gesprächsforschung 10, 353–402.
Söll, Ludwig (1974), Gesprochenes und geschriebenes Französisch, Berlin, Schmidt.
Sullet-Nylander, Françoise/Roitman, Malin (2010), Voix de campagne présidentielle: quelques observations sur la question et la réfutation dans le débat télévisé Royal–Sarkozy, in: Michaël Abecassis/Gudrun Ledegen (edd.), Les voix des Français à travers l’histoire, l’école et la presse, Oxford et al., Lang, 303–317.
Tannen, Deborah (1982a), The Myth of Orality and Literacy, in: William Frawley (ed.), Linguistics and Literacy, New York, Plenum, 37–50.
Tannen, Deborah (1982b), The Oral/Literate Continuum in Discourse, in: Deborah Tannen (ed.), Spoken and Written Language: Exploring Orality and Literacy, Norwood/New Jersey, Ablex, 1–16.
Wodak, Ruth/Meyer, Michael (edd.) (22009), Methods for Critical Discourse Analysis, London, Sage.
Wood, Andrés (2004), Machuca, Wood Producciones/Tornasol Films.