CHAPTER 1 INTRODUCTION 1.1. STATEMENT OF THE PROBLEM Since 1994, various pieces of legislation on multilingualism have been introduced that have important consequences for language use and hence interlingual and intralingual transfer generally. Section 6, Act 108 of 1996, of the South African Constitution provides the primary legal and constitutional framework for multilingualism, the use of the official languages, and the promotion and tolerance of South Africa?s linguistic diversity. The Act establishes that all official languages must enjoy ?parity of esteem, be treated equitably and that the status and use of indigenous languages must be enhanced.? The implementation of this language policy based on section 6 of the Constitution (Act No. 108 of 1996) is meant to ?consolidate national unity and democracy in South Africa, provide for the learning of South African languages by all South African citizens in order to promote multilingualism and multiculturalism and to redress the marginalisation of indigenous languages as the dominance of English continues to disadvantage and disempower the majority of the people who are non-English speakers? (Advisory Panel on Language Policy 6 November 2000). In addition, the broadcasting Act No. 4 of 1999, requires that all official South African languages be provided for by the public broadcaster, the South African Broadcasting Corporation. In order to ensure ?parity of esteem and the equitable use of official languages,? the principle of using four categories of languages on a rotational basis was adopted. These language categories are: the Nguni group (Ndebele, Zulu, Xhosa, Swati); the Sotho group (Pedi, Sotho, Tswana); Venda/Tsonga; English/Afrikaans. The languages in the first two groups are grouped together because they are mutually intelligible. However, Venda and Tsonga are independent languages and are not mutually intelligible. English and Afrikaans are paired together because in most cases people who speak English also speak Afrikaans (Advisory Panel on Language 1 Policy 2000) and the latter is closely related to Dutch and Flemish which are Germanic languages like English. This legislation is intended to lead to a transformation of South African television from a vehicle for racist propaganda (Tomaselli and Tomaselli 1989: 107) to a tool for building a multiracial democracy, portraying South Africa's post-apartheid dispensation for a more democratic, inclusive media. My study seeks to find out if the transformation is happening. 1.2. AIM The aim of this research is to carry out a descriptive study to establish the actual practices of subtitling in the South African television broadcasting media, using the soap opera Generations as a case study; how these practices match international, theoretical and methodological practices; and whether they have been affected by changes in legislation calling for the status and use of indigenous languages to be enhanced. Although the research sets out to discuss the actual processes in the subtitling of Generations, it includes an analytical and evaluative component. It examines episodes of Generations for the years, 1999, 2003, 2005 and one from January 2006, looking at the languages spoken in these episodes, the percentage of subtitling in each episode, and the nature of the subtitling in the soap opera. Finally, it assesses what progress has been made towards multilingualism, and subtitling since the two go hand-in-hand because whatever is spoken in the vernacular languages calls for subtitling. . 1.3. RATIONALE Generations is one of the prime-time soap operas shown on SABC 1 every weekday at 20h00. Generations was selected as the focus of this study as it uses several important indigenous languages ? Sotho, Xhosa, Zulu and Tswana and is the only soap opera in which these languages are treated reasonably equitably. It therefore provides us with one 2 of the best examples of multilingualism. In Egoli and 7 de Laan, for example, the principal language used is Afrikaans; in Isidingo and Scandal, it is English; and in Muvhango, it is Venda. Doing this kind of study will contribute to our knowledge about language use and to our understanding of how effectively subtitling is done in South Africa. My reasons for undertaking this study can be summarized as follows: *To my knowledge, no research has been conducted on how these regulations (arising *from legislation) are being carried through. Even the studies mentioned below on this page do not address them. Little or no work has been done on describing or establishing the impact of legislation on subtitling practices or studying actual subtitling practices in South Africa, and evaluating the effectiveness of these practices. Although there has been an increase in subtitling, no proper monitoring has been done on its progression. Hence my interest in it as an area of research. In addition, subtitling is a relatively new discipline in Translation Studies and, therefore, by undertaking this research, I hope to build on previous research and to make a small contribution to developing knowledge in this area in South Africa. The research looks at a number of works related to the current study in order to avoid repeating them. Through such a review of the studies, a gap in literature might be identified. Although there is a growing body of research on subtitling internationally , to my knowledge, little research on subtitling has been done in South Africa. The research that has been conducted so far is work by the following: H.C. Kruger from Potchefstroom University. In her M.A. dissertation entitled The Theory and Practice of Subtitling: The Case of Cyrano de Bergerac, Kruger (2000) examined how the lack of an adequate 3 specific theory for subtitling underlies most problems experienced with subtitling. In her M.A dissertation entitled Translation for the Silver Screen: a Special Case in the Subtitles for the Film ?Le moine et la sorci?r?/Sorceress, Gwendolyn Speeth (1996) from the University of the Witwatersrand examined the differences between the French dialogue and the English subtitles of the film in order to test the hypothesis that subtitling inevitably leads to loss. In his MTech dissertation, entitled The Influence of the Sotho Language and Culture on the Results of the Marketing Questionnaire, N. De Klerk (2002) from Vaal University of Technology examined a marketing questionnaire to determine the need for television programmes with subtitling. The purpose of the study was to determine if language and culture play a role or have an effect on the results of a marketing research questionnaire. S.S. Daswa from UNISA?s Department of African Languages is currently conducting research on The Role of Subtitling in Overcoming Communication Barriers in South Africa. The M.A. dissertation commenced last year (2005). 1.4. CORPUS The study examines subtitling practices on the soap opera Generations and to achieve this objective, the researcher analyses 15 episodes of the programme for the years 1999 (three years after Constitution), 2003 (four years after the promulgation of the Broadcasting Act), 2005 and one from January 2006. 1.5. CONTEXTUALISATION OF THE PROBLEM Broadcasting and Language during the Apartheid Era: In order to clearly understand how broadcasting developed in the country, the following gives the history of public service broadcasting in South Africa. On 29 December 1923, the South African Railways made the first wireless broadcast in Johannesburg. On 1 July 1924 the Associated Scientific and Technical Club of Johannesburg took over these transmissions. On 10 December 1924, the Durban Corporation, and on 15 December, the Cape Peninsula Publicity Association also started broadcasting (Erasmus 2004). At this stage, more English was used in broadcasts than 4 Afrikaans (Hayman and Tomaselli 1989: 29), which probably was the cause of greater state control by the Afrikaner National Party over broadcasting in later years. Generally, the Nationalists regarded Afrikaans as a threatened culture to be kept separate until it had achieved parity with English (Hayman and Tomaselli 1989: 29). Van Zyl and Elion (1989: 193) state that the Afrikaans language had historically been one of the crucial issues in the political development of the Afrikaner nation. Afrikaans developed hand in hand with the National Party, being one of the central factors of identification with the volk. John Reith, the then Director-General of the BBC, came to South Africa in 1934 and recommended a new form of broadcasting and, therefore, the South African Broadcasting Corporation (SABC) was established in 1936 as a Public Service Broadcaster (PSB) through an Act of the South African Parliament (Act 22 of 1936) and was directly controlled by the party in power (Van Zyl and Elion 1989: 196). At first, the SABC only broadcast in the official languages, English and Afrikaans. The first African language transmissions were made in 1940 and broadcast in Zulu, Xhosa and Sotho (Erasmus 2004). Originally modelled on the BBC, the SABC was for years controlled by the Broederbond, a secret Afrikaner society (Oder 1991) created in 1918 to promote Afrikaner interests and to mobilise political support for Afrikaner nationalism. The connection between the SABC and the Broederbond-dominated Afrikaner Nationalists government (Hayman and Tomaselli 1989: 47) became apparent when the latter gained control of broadcasting (Hayman and Tomaselli 1989: 46). Moreover, the SABC Board was dominated by the Broederbond (Hayman and Tomaselli 1989: 55). Drastic changes to the SABC came about when Albert Hertzog became Minister of Posts and Telegraphs in 1958 (Hayman and Tomaselli 1989: 56), the same year that the architect of apartheid, Hendrik Verwoerd, became Prime Minister. Hertzog in turn appointed Piet Meyer to the Chairmanship of the SABC Board. Meyer was also the head of the Broederbond (Hayman and Tomaselli 1989: 59). He had great control over the 5 SABC as Chairman of the Board for 20 years (from 1959 to 1979). The majority of upper management of the SABC were also members of the Broederbond (Tomaselli and Tomaselli 1989: 113). According to Hayman and Tomaselli (1989: 60), there was a very close covert understanding between members of the Broederbond in the SABC and in government. The SABC was used as a tool for the promotion of apartheid and applied policies to help preserve the dominant position of the Afrikaners as rulers, and of whites as the dominant group (Hayman and Tomaselli 1989: 74). Apartheid was based on economic domination based on racial division (Tomaselli and Tomaselli 1989: 97). The apartheid policy, which was implemented by the Nationalist government, stressed not only the separation of whites and blacks, but also ethnic differences within the black community itself, hoping by this means to divide and rule (Tomaselli and Tomaselli 1989: 95). The black audience was segregated by ethnic language (Hayman and Tomaselli 1989: 66) according to homeland areas. Stressing ethnic division and homeland policy, the SABC?s Radio Bantu (formed in 1960) had the task of inducing the majority of black South Africans to accept their ?homeland? status and to view it as independence and development (Tomaselli and Tomaselli 1989: 100).This meant the maintenance of racial discrimination presented as ?ethnic self-determination?. Apartheid, through the application of the homelands policy, attempted to portray the ?ideal? South Africa as a land without blacks (Van Zyl and Elion 1989: 194). In order to have diverse languages represented on the SABC when Meyer was chairman of the Board, the corporation tried to align its philosophy more closely with government policy (Tomaselli and Tomaselli 1989: 89) of creating separate ethnic identities for the several black language groups. Hayman and Tomaselli (1989: 71) remark that separate channels reveal that the intention of the multi-channel policy was not to create a sense of a ?national? culture, but to maintain the divisions of culture, race and class in accordance with the Nationalist 6 government policy, to cement Afrikaner unity, achieve parity with the English, and to ?divide and rule? the black population (Hayman and Tomaselli (989: 49). Television was introduced in South Africa only in the 1970s, mainly due to ideological reasons (Keyan Tomaselli et al. 1989: 154) ? the fear within certain right-wing sectors of the National Party that it would undermine the Afrikaner language and culture. However, the most reluctant of them all was Albert Hertzog. As Minister of Posts and Telegraphs until 1968, he resisted pressure for the introduction of television because he had a deep fear of ?foreign? culture influencing and debasing Afrikaner culture (Hayman and Tomaselli 1989: 57). The country's rulers resisted it for so long because they feared that images of racial mixing and the promotion of the English language would come with it (Oder 1991). Dr Albert Hertzog, a.k.a. ?Dr No.? said that TV was a devilish instrument which could soon replace God in the hearts of the people. British and American programmes were also not advisable, particularly at the time of the Civil Rights movement in the USA. However, his banishment from the National Party (Hayman and Tomaselli 1989: 58) opened the way for the introduction of television. Television only came to South Africa in 1976, many years after it had been installed in other countries (Tomaselli and Tomaselli 1989: 84). Hayman and Tomaselli (1989: 1) remark that broadcasting is one of the main sources of ideology and this is also true of broadcasting in South Africa. The segregation of programmes into separate channels with programmes whose style and content are different is one of the clearest indications of ideological influence in broadcasting. Initially, only one channel, the prime television channel, TV 1, was in operation, divided equally between the two official languages, English and Afrikaans (Tomaselli and Tomaselli 1989: 109). A second channel for black viewers was opened in 1982 (Tomaselli and Tomaselli 1989: 110) and split into two separate channels in 1983. TV2 broadcasting in Zulu and Xhosa and TV3 broadcasting in the Sotho languages started on 1 January 1982. A fourth service, TV4, was started on 30 March 1985 to combat the 7 popularity of BOP TV (Erasmus 2004). Since the audience of TV1 was mostly white and for TV 2/3 black, this inevitably affected the selection of broadcast material for the two channels. While the above literature offers a general description of the evolution of broadcasting services in South Africa, the next section is on the transformation of the SABC. Transformation of the SABC: The linguistic, cultural, ethnic, racial, class and geographical patchwork into which South Africa had been fragmented by apartheid (Teer-Tomaselli and Tomaselli 1994) meant that in 1994, after the elections, an effort was needed to develop a sense of unified identity and nation. The new rulers of South Africa argued that the role of the public broadcaster was to redress the imbalances of the past. SABC television underwent changes that were meant to reflect the new environment. The channels were reconfigured so as to be representative of the changes that had happened. Back then, Ruth Teer-Tomaselli and Keyan G. Tomaselli (1996: 224) wrote that the SABC?s attempts to achieve a balance of programmes for all members of South African society to redress the disparities caused by apartheid years, was a major endeavour. In keeping with its responsibility to address the needs of Euro- and Afro-centric audiences, the SABC should contribute to moulding a national sentiment for a ?collective identity?. Therefore, in 1996, almost two years after the ANC came to power, the SABC reorganized its TV channels, so as to make them more representative of different language groups. This resulted in the downgrading of Afrikaans?s status by reducing its airtime. Therefore, from 1996 onward, the SABC?s channels have been divided as follows: SABC 1, SABC 2 and SABC 3. (The pay channels are SABC Africa and the Africa 2 Africa stations). SABC 1 is a full-spectrum free-to-air channel aimed at younger viewers (SABC Today 2003:7). It broadcasts in English, Zulu, Xhosa, Ndebele and Swati. SABC 2 is also a full- spectrum free-to-air channel, aimed at the whole family. It broadcasts in English, Afrikaans, Sotho, Tswana, Pedi, Tsonga and Venda. SABC 3 is a free-to-air channel that 8 broadcasts in English only. There is another free-to-air station ? namely, e-TV, and there are pay channels in the form of M-Net and DSTV. During the Apartheid era, subtitling on TV was almost non-existent. Many imported programmes were dubbed into Afrikaans, the first being the British series The Sweeney, known in Afrikaans as Blitspatrollie. However, in order to accommodate English speakers, the SABC began to simulcast the original soundtrack of US series such as Miami Vice, which was dubbed into Afrikaans with the original English soundtrack available on Radio 2000 (Erasmus 2004). Although subtitling on TV in South Africa used to be almost non-existent, now many non-English language soap operas have started to display English subtitles as TV stations try to attract viewers from all language groups (http://www.wikipedia.org). Since the 1999 Broadcasting Act was introduced, the use of subtitling has increased in South Africa. On Generations, the percentage has risen from very little subtitling at its inception in 1994, to up to 60 per cent at present according to Matseoane (2006). The figure is still rising and ICASA (Independent Communications Authority of South Africa), which enforces the promotion of multilingualism in the media, requires it to be at 75 per cent by March 2007. With regard to television, the Act also states that an increasing amount of broadcasting air time shall be progressively provided for the African languages and Sign Language/s, up to a point where all official South African languages are accorded an equitable proportion of air time (Advisory Panel on Language Policy 2000). Among other things, the Act establishes that the status and use of indigenous languages must be enhanced (so as to redress the marginalization of indigenous languages) and that the government must take legislative and other measures to regulate and monitor the use of official languages. The regulatory authority sets rules and enforces them, monitors implementation and reviews performance, as a condition for granting broadcasting licences. ICASA has an obligation to monitor and enforce compliance with the constitutional requirement that all 9 official languages be treated equitably, in the public interest. SABC?s licence conditions, which are set by ICASA, regulate the public broadcaster?s programming policy and news editorial policy (Submissions on the SABC?s Application for Licence Amendments to ICASA 2004). ICASA had to enforce these rules strictly because institutions, including the SABC, were not prepared to do so on their own: ?At the outset, we should note with disappointment that the SABC adopts a minimalist approach in its application, attempting to get away with the bare minimum of what is required of them in terms of the law?In fact the SABC has proved consistently that it will take its mandate seriously only when pressure is applied to it, which further makes the case for strong external oversight mechanism coupled with measurable targets set through licensing conditions.?(Submissions on the SABC?s Application for Licence Amendments to ICASA in 2004). ICASA?s contention is that many of the most progressive changes that the SABC has made to its programming regime have been as a result of public or government pressure. ICASA sets conditions ? which the SABC has to meet subject to action being taken against them ? enforces them and monitors their implementation as an essential function for a regulator, to ensure external monitoring and effective mechanisms of redress in the case of targets not being met. Since 1994 expectations have been rising that previously marginalized languages would also benefit from the gains of the transformation agenda (Submissions on the SABC?s Application for Licence Amendments to ICASA on 9 June 2004). ICASA was established by the Independent Communications Authority of South Africa Act 13 of 2000 (ICASA Act), replacing the Independent Broadcasting Authority (IBA). ICASA is the institution contemplated in section 192 of the Constitution of the Republic of South Africa, 1996, which provides that: National legislation must establish an independent authority to regulate broadcasting in the public interest. 10 Accordingly, ICASA has the power, inter alia: to administer the statutory scheme for granting, renewing and amending broadcasting licences, by virtue of section 13(1)(a) of the IBA Act (the Independent Broadcasting Authority Act 153 of 1993). Therefore, ICASA has been given the broad power, when granting a broadcasting licence, to impose any term, condition or obligation appropriate to such licence. According to ICASA: The broadcaster should be prevented from being an exclusive medium for middle and upper classes. It should cater for all classes, and if we may say, be biased positively towards the poorer sections of our society. The public broadcaster should ensure that there are attempts to promote the growth of African languages. The skewed history of the country, which gave rise to African languages being marginalized as mediums of communication needs to be addressed by, among many other players, the public broadcaster. Some time back, Deputy Minister of Education, Mosibudi Mangena (Sunday Times, 22 September 2002) made the following remarks: Many of us have been aware of the marginalization of indigenous languages in the broadcast media. Apart from news and an occasional drama on television, almost all programs are in English. The finding?that 70 per cent of SABC broadcasting is in English is shocking but not entirely unexpected. But the further finding that 78 per cent of those surveyed did not fully understand speeches delivered in English by politicians means that a large percentage of the population is excluded from the democratic process in this country. . The same objection was raised by the Multilingualism Action Group (MAG): A survey by the Pan South African Language Board, PanSALB, found that only 40 per cent of South Africans in general and 22 per cent of African language users are functionally proficient in English (MAG 2002). In addition, MAG (MAG 2002) has also said that it appreciates the sense of deprivation speakers of certain languages feel when they see their languages marginalized and believes that SABC should not have an ?anchor language?.1 1 A term used to refer to a language which serves as an intermediate stage between the source language and the target language when it is for some reason not possible to transfer the source text into the target language (Shuttleworth 1997: 126) 11 According to the group, the domination of English disempowers many people. South Africans need to be exposed more regularly to all the official languages for the sake of nation building and in order to gain greater access to South Africa?s whole cultural heritage. This would be achieved by translating ?unilingual productions? and ?multilingual programmes? into various languages by means of subtitles. An equitable allocation would mean all languages would have an equitable share of limited air time inherent in broadcasting to a multilingual audience. MAG (2002) argues that giving preferential treatment to any one language (as is currently the practice with regard to English) has unacceptable practical and symbolic implications. Subtitling in languages other than English, particularly in the African languages, would make a significant contribution to increased levels of literacy in those languages. Subtitling into languages other than English had the advantage of exposing more Afrikaans and English-speaking South Africans to African languages, and vice- versa. Therefore, subtitling contributes to language learning and thus multilingualism. The regular use of subtitles would help create a ?culture of subtitles, as is common in many other countries? (MAG 2002). The organization is also of the view that in terms of its mandate and the Constitution, and given the commitment to ?equal respect? and ?equitable treatment?, the SABC, in broadcasting to the public, does not require, and in fact should not have, an ?anchor language?. For the purpose of entertainment, information and education, and as far as the SABC?s functions and obligations are concerned, South Africa has 11 ?anchor languages? (MAG 2002). MAG also observed that important programmes such as Special Assignment are only broadcast in English. A clear suggestion was that an attempt should be made to make this available in African languages. As a start, subtitles in African languages should be considered. Moving away from the South African scenario, in other parts of Africa however, the most common pattern with respect to language policy involves the use of one of the country?s 12 mother tongues, a lingua franca and the official language (Mansour 1993: 20). Some of the languages are given the status of national language (Mansour 1993: 73). Many African countries are societies so linguistically complex that multilingual interaction is commonplace (Edwards 1994: 2). Many states in Africa have two official languages ? usually a strong indigenous variety and an important European one for a highly heterogeneous and multilingual population (Edwards 1994: 55). Multilingualism forges unity through policies which suppress linguistic and other differences (Mansour 1993: 100) in order to establish equality between populations of different previous social statuses (Mansour 1993: 109), and there is not much that distinguishes one community from another. After independence, when it came to deciding on an official language, all the Anglophone countries in East Africa elected to keep English as the sole official language (Mansour 1993: 78), while Tanzania promoted Swahili to the status of ?national language? in 1964 and to ?official language?, co-equal with English in 1967 (Mansour 1993: 123). In Africa generally, official languages such as English, French and Swahili are used in administration and (apart from Swahili) internationally. South Africa however is a special case in that, although it has 11 official languages, 10 of them can only be used inside the country and not internationally. Perhaps, it would be more appropriate to call them national languages rather than official languages. Many soap operas in South Africa attempt to include different languages to dramatize a concept of the ?Rainbow Nation?. Dramas are generally subtitled from African languages into English. Some of the soap operas in which there is some subtitling are Egoli, Generations, Isidingo, Muvhango, Scandal, and 7 de Laan. However, as indicated, the present study focuses on Generations because it is the best example of multilingualism in practice since languages such as Zulu, Sotho and Tswana are treated quite equitably. 13 Furthermore, it has a large audience (47.1 per cent of the 18 million people who own television sets in South Africa according to Constance Mokhoantlhe, the Generations Public Relations and Publicity Manager, 12 September 2006). The target market for Generation is LSM 5-8. The LSM 5-8 bracket refers to the middle class. However, our target audience also includes people in the LSM 9-10 bracket. That is, the affluent. Our ideal viewer is a lady who is 25 years old and her whole family ? that includes her parents, grandparents and her siblings or kids. That is because the ideal person on the show is a young woman, Ntombi. Our audience ratings show that South Africa has a total population of 44 million and out of those, only 18 million own television sets. We look at adults over the age of 16 and during the 8.00 O?clock slot, 47.1 per cent of the 18 million viewers are watching the show. While the above literature offers a general background of issues relating to multilingualism and subtitling in South Africa, the next section gives a profile of the soap opera and narrows it to the object of the study, namely, examination of subtitling practices on Generations. This will enable the reader to understand the object of the research thoroughly so that the rationale behind the formulation of the research question can be understood. The background information on Generations also includes comments by reviewers and information about the soap?s key characters. 1.6. BACKGROUND INFORMATION ON GENERATIONS Generations was first broadcast on 3 February 1994 (Anstey Sunday Times 2004), and continued on a weekly basis until 2002 when SABC started airing it from Monday to Friday. It is billed as a soap opera and it airs daily (20h00 on SABC 1). Mfundi Vundla is the creator of Generations and also the executive producer. Generations is one of the most watched soap operas on South African television. Its setting is in the very competitive world of advertising. Since its inception, Generations has been watched by millions of people and the longevity of the show is proof that the viewers cannot get enough of it (Shota 2005). 14 Some of the key characters on Generations are the soap?s veterans Sophie Ndaba (Queen Moroka the receptionist), and Connie Masilo-Ferguson (Karabo Moroka), and reputable actors such as Leleti Khumalo (the lead actor in Yesterday ? nominated for an Oscar in 2005), who plays the character Busi. Protagonists include the characters Karabo and Tau Mogale (Rapulana Seiphemo). There are also the jabbering Ntombi (Sonia Mbele), actor Vusi Kunene (the underworld kingpin Jack Mabaso), Linda Sokhulu (reformed prostitute Cleo), Mike Mvelase, whose character is the strong, silent type Khaphela and the menacing Ngamla (Menzi Ngubane). The characters in Generations reflect the transforming South African society, as the black bourgeoisie is becoming more and more entrenched. It reflects the changing South Africa and the realities of the country today. Generations seeks to portray the emerging rich black people. Generally soap operas in South Africa, including Generations, pride themselves in reflecting this changing society. Vundla is proud of himself for having brought South Africans the first black soap opera in the country. Shota (2005) quotes him as saying: These people we see (in Generations) are the vanguard of the first generation of the African bourgeoisie in the making. It?s a process, it?s happening. Society is transforming and the bourgeoisie is becoming more and more entrenched. They are the torchbearers of a transforming society?and our characters reflect this. Vundla sees himself as a beneficiary of this transforming society, which is depicted by the characters in his soap opera and that, in a way, he is like his characters. He drew his inspiration to create the series from the advertising agency, the Herdbuoys. Generation is a fantasy world of upper middle-class black South Africans in the advertising industry. Mabanga (2002) writes: ?The creation has an obvious element of fantasy and escapism about it. Strangely, this element seems to be trashed for being unreal and misrepresenting the values of the black middle class. But Generations is no more representative than, say, 15 the health education drama represents black communities [sic]. Not all black people live in shacks, are unemployed and have family members afflicted with Aids.? Generations is the kind of soap opera many black people would like to associate themselves with because it depicts blacks in a positive light ? that is rich and successful, and not as the stereotypical poor, suffering human beings. It is a kind of escapism from the harsh violent life of South Africa today. Shota (2005) states that ?when it was launched on 3 February 1994, Generations was a huge success. For the first time, viewers were exposed to black people who lived in mansions, wore Hugo Boss suits, made and broke international deals and, best of all, the women had sexy slinky hair to go with their sexy slinky outfits. Black viewers were hooked.? In the past, the lives of most black people were bleak, with nothing much to aspire to. The airing of the series has given hope to black people ? that they also can aspire to a better life, as middle-class citizens. Anstey (2004) quotes Vundla: . Prior to Generations? first broadcast on 3 February 1994, black people on television were always caught in a quagmire, victims of circumstances beyond their control, depicted only in working-class dramas in which life was just a maze of complications, with no solutions. Watching the latest series of the soap opera, one would conclude that it is the kind of programme that is more likely to appeal to the twenty-somethings because of the age of most of the characters and also because of some of its sub-themes: issues that normal people have to deal with every day ? relationship problems, drug addiction, problems of sharing accommodation with other people, rape, AIDS, alcoholism, scheming, betrayal and so on. Generations steers away from controversial topics such as politics, which soaps such as Isidingo are only too happy to talk about in some of the dialogues. Although there are older actors on the show as well, the majority are in the 20-30 age-group. 16 Subtitled programmes such as Generations are becoming more and more popular thanks also to films such as Yesterday and Tsotsi which have given film subtitling new exposure in South Africa (Kruger 2006). 1.7. THE SOUTH AFRICAN CONTEXT OF GENERATIONS In South Africa, legislation has meant that changes have been made in the area of television in terms of subtitling. That is, through Section 6, Act 108 of 1996, of the South African Constitution and the Broadcasting Act No. 4 of 1999 covered on page 1. However, despite these imperatives, the English language still enjoys ideological dominance in South Africa because in Generations, the translation process entails translating from English into African languages, and then back into English (not into African languages) for subtitling. This current practice, where subtitles are used only to translate from other languages into English is regarded as ?unacceptable? by the Multilingualism Action Group (MAG). (Already covered earlier on pp 11-12). 1.8. POWER RELATIONS IN THE SUBTITLING OF GENERATIONS This dominance of English in the subtitling process of Generations relates directly to issues of power in translation ? dominance of the English language and Anglo-American cultures at the expense of other languages and cultures. This problem is discussed by Niranjana (1992: 48-9) who states that translation studies should consider the question of power imbalance between different languages. Asymmetrical power relationships in a post-colonial context also form the thread of the important collection of essays entitled Post-Colonial Translation: Theory and Practice edited by Susan Bassnett and Harish Trivedi (1999). In their introduction (1999: 13), they see these power relationships being played out in the unequal struggle of various local languages against ?the one master- language of our post-colonial world, English?. The observations of these scholars are relevant to the present study because, despite legislation to redress the problem, inequality between the vernacular languages and English persists. English is valued and the subtitling appears not to be for African audiences. 17 On the subject of ideology and translation, Niranjana (1992: 3) writes that translation reinforces hegemonic versions of the colonised. This leads her to interpret Schleiermacher?s philosophical distinction between domesticating and foreignizing translation in terms of power relations. However, the distinction has been redefined many times by many people, among them Berman (1984), who, writing explicitly about translation and ideology, talks of ethnocentric translation, something also taken up by Venuti (1995). In a way, what is happening on Generations is a form of foreignisation because the colonial language is still dominant. In general, in the soap opera, everything is in English except for the bits that are chosen for translation: a case of codeswitching. Codeswitching is dealt with later in the study. The codeswitching is not really a natural instance of codeswitching. It comes out of the translation process where bits of the text are selected by the translator. They are, in fact, instances of the translation, which is an artificial representation of codeswitching based on the translation process. More significant is the fact that the subtitling is into English instead of into the vernacular languages. English seems to be more important than the vernacular languages and is forced on the majority of the people. The assumption seems to be that everyone understands English and that since South Africa is a multilingual2 country, people who do not understand, for example, Zulu in the soap opera will understand the message through the English subtitles. This would seem to run counter to the understanding of subtitling as a form of language transfer (Luyken et al. 1991: 11) which makes a film or television programme understandable to target audiences who are unfamiliar with the source language in which the original was produced. Furthermore, the subtitling of African languages into English, and not vice versa, foregrounds African languages as the ?foreign other?. 2 The term multilingualism is equally loosely applied to two different types of situations: countries which have more than 2 official languages and countries which have considerable internal linguistic diversity with one super-imposed official language. The latter is the most common pattern in Third World countries, including African countries (Mansour 1993: 2). 18 This foreignisation process also relates to other issues: for example what Nord (1991: 36) says is dominant in a text: what gets translated (what is valued and what is excluded)? However, value has nothing to do with what gets translated in Generations. According to Matseoane (2006), what gets translated is simply what is translatable. And as already discussed, unlike standard translation practice which takes the whole utterance into consideration when translating, in Generations, only a segment of the utterance is translated and subsequently subtitled and even then, the choice of what gets translated is arbitrary. Who does the translation (who controls the production of translation)? Employees of Starke Films do the translation, but ultimately, the authority behind this translation is ICASA because they dictate how much of the episodes should be translated from English into other languages. The producers of Generations have no say in that. The point of all this is to cut down on the ideological dominance of English on the screen, which seems to be a thorn in the side of the government. How is the material translated (what is omitted, added, altered in order to control the message)? In 1969, Hesse-Quack (in Fawcett 1996: 82) saw mass communication as a reflector or controller of society. As the mass media is regarded as the controller of society, offensive statements or topics or words regarded as taboo are usually omitted from programmes, as is the case with Generations. As already stated, the producers of Generations are very careful not to touch on taboo areas. On Generations, the only taboo words found were ?dammit?, ?idiot? and ?bloody?, which were eliminated from the subtitle most of the time. So the way that English dominates on the screen is a form of socio- cultural message and a manifestation of power relations. As can been seen, language and texts represent discourse in its wider sense (Hatim and Mason 1997: 216). In this way, film translation also provides fertile ground for studying the exercise of power in society. This dominance of English in the subtitling process of Generations relates directly to issues of power in translation. It is the use of the English language that sees the dominance of other languages, proving that power lies with the elite. 19 This chapter raised some stimulating discussion on the nature of the subtitling practices in South Africa. The highlight of the discussion has been pointing out how the subtitling practices differ from international ones in that, in South Africa, translation is from English to vernacular and back to English for subtitles. Before concluding the introduction, a reminder to the reader as to why this study was undertaken: A review of studies with the same focus as this study identified a gap in subtitling research as the lack of research in describing or establishing the impact of legislation on subtitling practices or studying actual subtitling practices in South Africa, and evaluating the effectiveness of these practices. The research is significant in that although there has been an increase in subtitling, no proper monitoring has been done on its progression. The study is also noteworthy because it will provide an invaluable contribution to the existing body of knowledge on issues around subtitling. In carrying out this study, the chapters are structured as follows: 1.9. OUTLINE OF CHAPTERS Chapter one was the introduction. Chapter two gives an overview of the literature review on the nature of translation generally and then specifically with reference to South Africa. Because of the nature of this research, which is to examine the actual processes, the approach adopted in this study is a descriptive one. Descriptive translation studies (DTS) is therefore relevant to this study. The chapter begins by discussing descriptive translation studies in general to give a broader understanding of the scientific basis for subtitling. The study then moves to the more specific application of DTS ? that is, Delabastita?s application of DTS to screen translation (1989). According to him, Toury?s tripartite model of ?competence-norms- performance? can be used to analyse screen translation successfully. His inventory of 20 questions are structured after the tripartite model developed by Toury. The analysis of the subtitling will show that it is possible to apply descriptive translation to subtitling. The chapter also touches on the different forms of audiovisual translation; that is, subtitling and dubbing. (The other forms of audiovisual translation are voice-over, narration, audio description, and free commentary). Also covered are translation theory and the inherent constraints of the medium. Finally, the chapter concludes by discussing screen translation with respect to South Africa, as a result of the recent legislation, already covered in this chapter. Chapter Three describes and justifies the qualitative research methodology used to provide answers to the research question. The chapter begins by recapping the research problem being investigated ? namely, that little or no work has been done on describing or establishing the impact of legislation on subtitling practices or studying actual subtitling practices, and evaluating the effectiveness of these practices, in South Africa. Although there has been an increase in subtitling, no proper monitoring has been done on its progression. Hence, the interest as an area of research. Firstly, the chapter justifies the use of qualitative research methods to collect data from the people who had the information and the reasons for the sampling in terms of selecting interviewees. It further describes how data was collected. In this chapter, the study also looks at the languages spoken on Generations and their progression. To do this, manuscripts from 1999, 2003, 2005 and one from 2006 are examined. The second part of the analysis deals with codeswitching and, lastly, with an analysis of the subtitling itself in order to establish the actual practices of subtitling in the South African television broadcasting media, to find out whether these practices match international, theoretical and methodological practices. Chapter Four discusses the findings from the preceding chapters and draws conclusions from them in comparison to the information recorded in the literature review. The conclusions establish whether the actual practices of subtitling in the South African television broadcasting media do match international, theoretical and methodological 21 practices, and whether the percentage of subtitling has increased on the South African television soap opera, Generations, between 1999 and January 2006. Finally, the researcher makes recommendations. With this aim and rationale in mind (covered before the chapter analysis), the following leads us to a literature review which highlights the nature of subtitling. 22 CHAPTER 2 THEORETICAL BACKGROUND TO DTS INTRODUCTION The introduction has provided the background to the study. As a prelude to the description of research methods adopted for this research and to the conclusion, a literature review is done to highlight the major terms of this research ? namely, Descriptive Translation Studies (DTS), Screen Translation and Translation Theory. The review starts by unravelling the concept of DTS. It proceeds to a theoretical consideration of subtitling by different scholars. Finally, the review relates the entire discussion to the South African situation with regard to subtitling practices. 2.1. DESCRIPTIVE TRANSLATION STUDIES: THE DESCRIPTIVE APPROACH The nature of this research is to examine actual subtitling processes (including an analytical and evaluative component at a later stage) and the approach adopted is thereofre a descriptive one. Descriptive translation studies (DTS) is therefore an appropriate framework for the study. DTS attempts to study translation in a scientific way. DTS was responsible for a major shift in Translation Studies because it allowed a broadening of the scope of theoretical studies in the discipline. It marked a shift from earlier prescriptive approaches (Holmes 1988: 94). Theories were mostly concerned with giving guidelines on how people should translate rather than describing the translations themselves. Therefore, DTS does not prescribe but, instead, describes and explains. It is not concerned with how translation should be done but with what is done in translation. Scholars have provided a number of models for researchers to use in analysing translations. This section seeks to provide an understanding of the role of DTS in subtitling. 23 In his Contemporary Translation Theories, Gentzler (1993: 92) describes Holmes?s paper (The Name and Nature of Translation Studies) as ?generally accepted as the founding statement for the field of DTS.? Holmes puts forward an overall framework, describing what translation studies covers. This framework was subsequently refined by the Israeli translation scholar, Gideon Toury, whose translation studies is divided into pure (theoretical and descriptive) and applied (translator training, translation). Descriptive Translation Studies is divided into three branches: these are product-oriented Descriptive Translation Studies, function-oriented Descriptive Translation Studies and process-oriented Descriptive Translation Studies. Product-oriented descriptive translation studies is the branch which deals with the description of existing translations, the description of individual translations and also with the analysis and comparison of different translations (Holmes 1988: 72). Function-oriented descriptive translation studies is the branch which is concerned with the analysis of the function of the target text in the target culture. Process-oriented descriptive translation studies deals with the process that is involved in producing a translation and what happens in the translator?s mind as he translates a source language text into a target language text. While there are a number of theorists on DTS, Gideon Toury has emerged as the most influential and cited by almost all the scholars who venture into the field of translation. He developed Holmes?s approach to descriptive translation further. Toury and Delabastita?a approaches are chosen because of their particular value to this study. In his influential book Descriptive Translation Studies ? And Beyond, Toury (1995: 10) calls for the development of a properly systematic descriptive branch of the discipline in which the findings of individual studies will be inter-subjectively testable and comparable, and the studies themselves replicable (Toury1995: 3). 24 Although Toury focuses initially on the analysis of the translation ?product?, he emphasizes (Toury 1995: 174) that this is simply in order to identify the decision- making ?processes? of the translator. His hypothesis is that the norms that have prevailed in the translation of a particular text can be reconstructed. For Toury, norms mean making generalization regarding the decision-making process of the translator and then ?reconstructing? the norms that have been in operation in the translation and making hypotheses that can be tested by future descriptive studies. These norms are socio-cultural constraints specific to a culture, society and time. They ?determine the (type and extent of) equivalence manifested in actual translations? (Toury 1995: 61). Toury defines norms as: The translation of general values or ideas shared by a community ? as to what is right or wrong, adequate or inadequate ? into performance instructions appropriate for and applicable to particular situations. (Toury 1995: 55) His TT-oriented theoretical framework combines linguistic comparison of ST and TT taking the cultural framework of the TT into consideration. According to Toury, translation research should always begin with the ?observational facts? (1985: 18) and move backwards from there because ?translations are facts of one system only: the target system.? This is the basis of the Descriptive Translation Studies: translation ultimately serves the ?needs of the culture which would eventually host it? (1995: 166). Thus the study of a translation should describe the target text and the differences between it and the source text, in an effort to determine the reason why certain choices were made by the translator. With regard to this study, since translations are ?the facts of only the target system,? the audience, as viewers of film, have to be in a position to understand a television programme or a film. The examination of the ST and the TT reveals the shifts in the relations between the two which have taken place as a result of translation. Toury uses the term ?translation equivalence? (1980: 115 and 1995: 85) differently to its use in prescriptive notions of 25 equivalence. For Toury, it is not a question of whether equivalence exists, but to what degree and what kind of equivalence has been achieved. Therefore, as far as Toury is concerned, equivalence is assumed between a TT and an ST. This is very important because analysis does not then focus prescriptively on whether a given TT is ?equivalent? to the ST. Instead, it focuses on how the equivalence has been realized and is a tool for uncovering ?the underlying concept of translation? [the] derived notions of decision-making and the factors that have constrained it? (1995: 86). Toury (1995: 56-9) sees different kinds of norms operating at different states of the translation process. The basic initial norm refers to a general choice made by the translator. Thus, translators can subject themselves to the norms realized in the ST or to the norms of the target culture or language ? that is, an ST-oriented translation or a TT- oriented translation. Thus, in Generations, the subtitler has had to make a variety of adaptations due to audiovisual constraints and different cultural contexts. The process involved in Generations which involves translation between English and South African indigenous languages, also requires that the translator adapt the language because of differences in syntax. These would be regarded as obligatory shifts (Toury 1995) ? governed by rules of the target system and are of no interest to this study. What is of interest are the optional shifts and these are the shifts which are observed in the analysis of the series. In translation, obligatory and optional shifts are inevitable. Other norms described by Toury, are preliminary norms (1995: 58) and operational norms (1995: 58-9). Preliminary norms consist of translation policy and directness of translation. Translation policy refers to factors determining the selection of texts for translation in a specific language, culture or time. Directness of translation relates to whether translation occurs through an intermediate language, which is therefore very relevant to this study. On Generations, there is indirectness of translation because translation is from English to vernacular languages and then back to English for the subtitles. The vernacular languages, therefore, act as intermediate languages. 26 Preliminary norms are important for this study because, Starke Films, the producers of Generations received a directive from the SABC who commissioned them to make the series. The directive from the SABC is for all the episodes to be multilingual and that the incidence of indigenous languages be progressively increased until it reaches 75 per cent in March 2007, but to reach 65 per cent this year (2006). The SABC had received that directive from the regulator, ICASA. Therefore, with regard to translation policy, the subtitling of Generations should be at 65 per cent this year. However, although the majority of the cast is now black, unlike in the past, a great deal of its dialogue is still carried out in English whilst only bits are code-switched into the vernacular languages. English continues its dominance as even in cases of codeswitching, the OTHER language is always English. There are also whole stretches of text which are exclusively in English. In other soap operas such as Egoli, Muvhango, and 7 de Laan, (with the exception of, for example, Isidingo and Scandal with mostly English dialogues), almost a 100 per cent of the dialogues are in the OTHER languages (whether vernacular or Afrikaans, or a mixture of vernacular and Afrikaans), but with English subtitles. However, in Generations, more English is spoken than any other language, with switches in between to the vernacular languages, which are then subtitled back into English still. So, despite the directive from the ICASA, the level of multilingualism has not yet attained the level envisaged. Proof is in the analysis of episodes from 1999, 2003, 2005 and one from January 2006. The reasons could be partly due to the codeswitching. That is, instead of giving a whole utterance in a vernacular language, for example, the other portion is in English. However, the indigenous languages are slowly closing the gap, as can be seen from the January 2006 example. Recent episodes reveal that subtitling is steadily increasing3 in Generations. What is clear though is that it has not yet reached 65 per cent. However, notwithstanding the steady increase, the major issue still remains and 3 This study is only until January 2006. What has happened after January 2006 can by researched into by another study. 27 that is that subtitling is still all English. Therefore, English is still predominant in the series. In addition, with regard to translation policy, it seems that it is an all English policy because translation is from the vernacular into English and not the other way round. One would therefore beg to ask the question: Whom is it benefiting? People who do not understand English, who are in the majority, can only understand half of what is said. The policy, therefore, benefits only English speakers or those people who understand English. In other multilingual countries such as Belgium, the different speech communities are catered for through the use of bilingual subtitling. In South Africa, however, the language issue is made more complex because the country has 11 official languages and to accommodate all of them is a difficult task. That is probably the reason why a decision was made to subtitle into English. True multilingualism would mean that programmes would be subtitled into Zulu, Sotho, Venda, Xhosa, Afrikaans and all the other official languages with probably each one of them taking turns in respective series ? a very cumbersome proposition. One would therefore ask how far the status quo is advancing the cause of multilingualism. Operational norms consist of matricial norms and textual-linguistic norms. Matricial norms relate to the completeness of the TT. For example, whether there is omission or relocation of passages, textual segmentation, and the addition of passages. Textual- linguistic norms govern the selection of TT linguistic material: lexical items, phrases and stylistic features. These will be dealt with in greater detail in the analysis. 2.2. DELABASTITA?S APPLICATION OF DTS TO SCREEN TRANSLATION We have discussed descriptive translation studies in general and now move to the more specific application of the model to DTS. 28 Those who do refer to translation theory are not all agreed on which is the most suitable to apply to screen translation. However, Delabastita (1989) has applied DTS to screen translation. According to him, Toury?s tripartite model of ?competence-norms-performance? can be used to analyse screen translation successfully. His inventory of questions are structured following the tripartite model developed by Toury. Delabastita states that there are many possible ways of translating a text (theoretical level of competence) (1989: 194). In particular cultural situations, however, one will often observe certain regular patterns of behaviour (empirical level of performance). This allows the scholar to assume an intermediate level of norms which denote particular types of translational bahaviour as more or less desirable (level of required relationships between STs and TTs) (1989: 194). The theoretical level of competence (1989: 194) referred to above relates to the translation potential of films and attempts to establish the various possible ways in which a film can be translated. Delabastita?s model classifies modes of film translation by the operation performed on the verbal and non-verbal signs in both the acoustic and visual channels of transmission. There are five operations traditionally associated with translation. These are (1989: 199): repetitio (repetition ? the film is reproduced unchanged with all its original material features (linguistically, this is a case of ?non- translation?). In Generations, there are a number of instances of borrowed words ? that is, cases of non-translation. For example (p 5) ?n drug addict in my huis nie, geen wonder dit was special nie (p 6), ken a le appointment le client,?go kopana le wena for lunch (p 6),?go re o busy thata (p 24) ; adiectio (addition ? new images, dialogues or sounds have been introduced etc; detractio (reproduction is incomplete ? it implies a reduction or omission); transmutatio (transformation ? the components of the sign are reproduced in a somewhat different order and formation); and substitutio ( substitution ? the sign is replaced with an altogether different sign). The analysis of the study is based on these operations. 29 Delabastita developed this checklist of questions which include some of the following: Which target language has been selected by the translators? What technique is used by the translators in the various types of source text scenes (titles, credits, dialogue, music, verbal text visually presented, etc)? Are there any reductions or any additions? If so, what types of scenes, dialogues have been introduced or deleted? What are the maximum and the average number of characters in the subtitled text? Do syntax and style have a foreign ring? What is the attitude towards loan words and foreign idioms and expressions? How have taboo elements been dealt with? Are the subtitles in one language or more than one? How long are the subtitles, and so forth (Delabastita 1989: 206-110). Certain patterns are likely to manifest themselves if these questions are put not only to the individual film translations but for a whole series of texts. As a result, hypotheses can gradually be formulated concerning the motivations behind the translator?s behaviour. All the questions above are relevant. However, it is not only a question of the motivations behind the translator?s behaviour, but it also has something to do with what is imposed on the translator or on the whole process relating to issues of what is the required percentage of multilingual material. Delabastita?s complete checklist can be found in (1989: 206-210). 2.3. RELEVANCE OF THE MODEL TO THE CASE STUDY As can be seen from the examples from the analysis (below), it is possible to apply descriptive translation to subtitling. By studying whole series of texts, both consistencies and variations of behaviour are discovered, from which norms are deduced ? that is, the systematic occurrence or non- occurrence of specific strategies in translation. Delabastita?s competence-norms-performance inventory of questions has been addressed, although only those questions that were deemed necessary for the study are used. With regard to the competence part of Toury?s model, the different types of screen translation 30 are discussed. Delabastita states that it is best to study whole series of texts (as I have done observing a number of Generations series) on the basis of non-selective corpora in order to determine regularity of behaviour. I am examining the performance aspect of the subtitling that is carried out in Generations. Therefore, I am looking at the operational norms, which involves the comparison of the source text with the translation in terms of Delabastita?s five operations and in terms of the checklist. In looking at the subtitling, first of all, I provide a back translation of the source text and compare the back translation with the target text (the subtitle), which reveals shifts. These shifts are described and possible reasons given ? that is, the norms governing the translation are deduced, with due respect to subtitling constraints and target text conventions and culture. 2.4. SCREEN TRANSLATION Other theoretical discussions that follow highlight the key concepts of screen translation, and translation theory and provide a framework through which a fuller understanding of what is involved in subtitling. Therefore, the theories serve a purpose in this research because they provide a background against which subtitling is understood. Luyken et al. (1991: 180) describe language transfer as ?the process by which a film or television programme is made comprehensible to target audiences who are unfamiliar with the source language in which the original was produced?. Generally, language transfer in films and television programmes is mostly from English into one of the respective vernacular languages (1991: 16). In South African, however, the opposite is the case: language transfer is from the vernacular languages into English. 31 The best-known and most widespread forms of audiovisual translation are subtitling and dubbing. In subtitling, the text is superimposed onto the picture and in dubbing the original voice track of the film or programme is actually replaced by a new one, known as ?revoicing?. Other forms of audiovisual translation are voice-over, narration and free commentary. A dubbed film substitutes the original dialogue with one in another language and offers no immediate basis for comparison between the two, as is the case with subtitling. However, when comparing subtitling to dubbing, Herbst (1995: 257-8) notes that the latter deprives viewers of the opportunity to listen to the foreign language, and this may partly explain why ?the standard of English as a foreign language is so much higher in subtitling countries such as the Netherlands or Scandinavian countries than in Germany,? for instance. 2.5. TRANSLATION THEORY On the question of which language transfer method to choose, audiences are usually in favour of the method they are accustomed to. Generally, older people prefer dubbed programmes (Luyken et al. 1991: 114) because the difficulty in reading subtitles increases with age. People who have a high educational level or a high socio-economic status tend to prefer subtitling. The reasons for this are obvious ? the higher the level of education the better a person?s reading skills and therefore the lower the difficulty in reading subtitles. This makes subtitled programmes easier to understand and to enjoy. The method (of subtitling or dubbing) used will also depend on the genre of television programme. ?Drama or fictional programmes range from Shakespeare to soap operas, from feature films to relayed theatre performances.? A differentiation is made between ?non-contemporary drama,? ?light modern drama,? and ?cultural modern drama.? Dubbing is best in dramas in which the main purpose is to entertain and relax the audience, such as crime and comedy (Luyken et al. 1991: 130). However, subtitling would be more suited to a film, play or series which tries to portray life in a particular country, because the language of that country is an essential part of that cultural experience and it has to be 32 preserved. In short, theorists like Luyken argue that ?light modern drama? should be revoiced while ?cultural modern drama? should be subtitled (Luyken et al. 1991). Therefore, if Luyken?s suggestion is to be taken into consideration, the soap opera Generations would fall into the category of ?cultural modern drama.? The method of Language Transfer is mainly determined by market size and the financial resources available to the broadcasting station concerned (Luyken et al. 1991). The same would apply to South Africa since figures quoted for average costs per hour for subtitling and dubbing in Europe (Luyken et al. 1991: 106) suggest that dubbing is 15 times more expensive than subtitling. Subtitling is becoming a preferred mode of translation not only because subtitles are more economical but also because they are easier to produce. Ultimately, the choice of method seems to be determined largely by audience habits. Viewers in traditionally dubbing countries tend to favour dubbing and those in traditionally subtitling countries find it difficult to enjoy a dubbed film. Luyken et al. (1991: 188) argue for a stronger mix of methods based not on national habits but on programme genre and specific audience profile. Subtitling countries are mainly Greece, Belgium, The Netherlands, Denmark, Sweden, Finland, Portugal and Cyprus. The dubbing countries are France, Germany, Austria, Italy, Spain and Switzerland. Luyken et al. (1991: 32), however, say that Britain and Ireland are neither classical ?subtitling? nor ?dubbing? countries but use these Language Transfer mechanisms as needed and in a mixed manner. Larger linguistic countries prefer dubbing. Subtitling, voice-over and narration are preferred in the smaller ?subtitling countries? to which South Africa belongs. According to Luyken et al. (1991), programmes which should be subtitled are news and current affairs, educational broadcasts, certain drama and life entertainment programmes, music and opera relays, and religious programmes. If the target viewer group for these programmes includes the under 50s, the better educated and more affluent, ?as well as language learners of a foreign language and other intellectual minorities, the hard-of- 33 hearing, and those with an interest in the original language of production, then subtitled versions are particularly likely to be successful among them.? In countries where illiteracy is very high, programmes are revoiced, despite its high cost. Programmes for the very young and the very old, cartoons and puppet shows, science and art programmes, sports and other major public events, variety shows, and drama in which entertainment is the predominant factor are revoiced. If revoiced, drama requires lip- synchronization dubbing while the other cheaper forms of revoicing such as free commentary, voice-over and narration techniques do not (Luyken et al. 1991: 189). Voice-over favours news and children?s programmes. The method of commentary, often used in documentaries, deletes the original off-screen narration and replaces it with narration in the target language. The discussion now shifts to the nature of subtitling, which has received a wide range of theoretical treatments from various theorists. 2.5.1. THE NATURE OF SUBTITLING There is hesitation among translators, and scholars regarding the applicability of the term ?translation? to film (Delabastita 1989: 213). Subtitling has been considered a second- rate, debased form of translation (Hajmohammadi 2004), but is increasingly being recognized as an important area of study. Universities have begun to include it as a subject for study in their curricula, and every year it is the focus of several academic conferences as this relatively new but very rich field of study attracts more and more researchers. Despite this widespread enthusiasm and the potential it represents, it is often assumed in Translation Studies circles that subtitling audiences have a lower cultural standing (Hajmohammadi 2004) than, for example, the readers of literary works in translations, with the result that this highly specialised area of translation has been neglected in the literature. However, Gottlieb (1998b) states that due to the increase in mass communication in more recent times, screen translation has become more important. The interest in screen translation has also been reflected by the fact that some 34 universities (Universit? de Lille, France, the University of Copenhagen, Denmark, and Roehampton University, England) have introduced specialised courses (Gottlieb 1998b).The translators' work is also becoming recognised as their names are mentioned in the films? credits. In 1996, Fawcett (1996: 65-69) noted that very little work had been done in the field of film translation. There might be a number of reasons why film translation has been problematic for translation theory: For example, that subtitling and dubbing are not translating per se and therefore, by implication, cannot be dealt with by translation theory, given the unpredictable nature of the shifts that are likely to arise from the search for synchronization in space and time. These constraints may be one of the reasons that in the transfer of films from one language group to another, the term ?adaptation? is frequently suggested as an alternative to ?translation? (Delabastita 1989: 213). Adaptation means changes are made so that the target text produced is in line with the spirit of the original. A text which is not obviously a translation in the traditional sense is thus created. Adaptation stands on the opposite end of the spectrum to literal translation. Literal or a word-for-word method of translation is rarely applied in translation practice because linguistic differences prevent translators from doing so, especially when the two languages belong to quite different language families. Moreover, adaptation is inevitable in practice if the translation is to maintain the source message's essence, impact, and effect. The emphasis here is on preserving the character and function of the original text, in preference to preserving the form or even the face value semantic meaning, especially where acoustic and/or visual factors have to be taken into account?the main features of this type of adaptation are the use of summarizing techniques, paraphrase and omission (Bastin 1998: 6). Relevance theory (Gutt 1991) as applied to translation studies is clearly relevant to film translation. Fawcett (1996: 79) quotes Gutt?s and Hervey and Higgins?s examples of translations (of a travel brochure and a car information sheet respectively), which make no attempt to reproduce everything in the SL and he notes that ?the demand for 35 translation to convey the same message as the original becomes less likely the more different the context?. If the message of the original is replaced by exactly the same message in the other language, it may not be understood at all. Due to time and space constraints and to different cultural contexts, a variety of adaptations have to be made to the source text in order to avoid incomprehension in the audience. In addition, translation between English and South African languages requires the translator to adapt the language because the languages are so different in structure. For example, Indigenous Language (Tswana): ke lebega jang? Back Translation: I look how? Subtitle: How do I look? ? There is transformation in syntax because English and African languages do not belong to the same language family. Other examples are the following: BABA (Afrikaans p 4): Verbaas my nie (Afrikaans). Back Translation: surprise me not. Subtitle: Doesn?t surprise me. JACK: (p 58): ?Bekungewon? amanga lawo (Zulu). Back Translation: They were not lies those. Subtitle: That wasn?t a lie. There is a transposition of SL elements in the subtitle due to the nature of the target language syntax. In addition, Zulu convention dictates that ?lies? be in the plural, as opposed to English which can be in the singular. There are also lexical differences because English is an ?isolating language? whereas African languages such as Zulu are agglutinating languages (Fromkin and Rodman 1978: 337). That is, in isolating languages, one morpheme equals one word. For example, the morpheme ?child? by itself makes sense. Its Zulu equivalent ntwana does not make sense on its own. An additional morpheme, the prefix um, is required for the word to make sense ? umntwana=child. Besides, verbs in African languages become longer than their English counterparts because of the addition of prefixes and suffixes to the root. 36 Technical and Other Constraints: There are many constraints in the production of subtitles. Only the ones that are important to this study are discussed below. As stated earlier, subtitling differs from literary translation because of time and space constraints, which are brought about by the nature of television and film (Luyken et al. 1991). The constraints include: the screen space available for the subtitle ? subtitles have to fit into an average maximum length of 35 characters (including spaces), the limited time available for and between the subtitles, and the timing of subtitle insertion and removal. No more than two sentences are allowed on the same subtitle. They should occupy one line each, whether they correspond to utterances produced by the same speaker or by different speakers in a dialogue. Apart from the limited available space for a subtitle, the other constraint is the problem posed by having to display the subtitle long enough on the screen for it to be read. Quick changes on the screen lead to insufficient time to read subtitles. A full two-line subtitle must be held on screen for at least 6 seconds to provide ample reading time. It is equally important to keep the same subtitle on the screen for no more than 6 seconds because this would cause automatic re-reading of the subtitle, especially by fast readers. According to research carried out in Belgium on eye movements, if a two-line subtitle remains on the screen for more than six seconds, re-reading of the text will occur (d? Ydewalle et al. 1987, 1989, 1991). The average reading speeds of the viewers is considered to be between 150 and 180 words per minute (Luyken et al. 1991). However, the weakness of these assertions is that it does not take other variables which affect reading speed into consideration. A large number of variables affect reading speeds: level of interest a person has in a subject or topic or their degree of familiarity with a particular programme, the need to process two channels of information simultaneously ? both subtitle and screen image, interest and exposure to a type of discourse (de Linde and Kay 1999). The rate at which the text is presented, the difficulty of the information and whatever action is taking place on the screen can interfere with reading speeds. My observation of subtitles on 37 Generations is that the presentation rate would suit fast readers because the subtitles remain on the screen just long enough to be read by them. Smith (1998: 145) states that combining subtitles allows viewers to make maximum use of the available time, since a two-line subtitle is read faster than two one-liners. Notwithstanding the recommended 6 seconds for a two-line subtitle , this problem of sufficient exposure time for a subtitle is still unresolved (Luyken et al.1991: 186). Different guidelines exist in different countries, such as a minimum of 4 seconds exposure time per subtitle in the United Kingdom, 6 seconds in the Netherlands and up to 8 seconds in some Scandinavian countries, but 51 per cent of the British viewers and 25 per cent of the Dutch viewers older than 50 years of age have problems following them. However, on Generations, there is no official exposure time for subtitles. According to the subtitler, Maggi van Aswegen (2006), the subtitles are flashed on the screen just long enough for them to be read. The problem with the South African case, which makes comparison with other countries difficult, is that on Generations, subtitling is done differently. Because of codeswitching, only a bit of the dialogue is translated, instead of taking the whole utterance into consideration. Criticism generally leveled against subtitling ranges from quick subtitle changes on the screen, not enough time to read the subtitle, to subtitles being distracting, leading to lessened enjoyment. With regard to the cognitive activity required by subtitling viewers, Delabastita (1990: 98) has demonstrated that reading subtitles does not require a conscious cognitive effort on the part of those accustomed to this mode of translation. People who read subtitles do not exhibit the typical eye movement patterns of ordinary reading behaviour. Rather, their eyes tend to make no more than a few quick jumps from one keyword to another. The whole process of subtitle perception tends to be largely automated, so much so that viewers who have no need of subtitles find it hard to avoid reading them. 38 A subtitle should not be retained over a shot change. A subtitle that remains on the screen during a scene or shot change will lead to overlapping (Luyken et al. 1991). Another constraint is that the match between dialogue and picture must be retained when translating. In other words, the appropriate subtitle must appear on the screen at the same time as the appropriate picture. To deal with these problems, translators use specialised software to help synchronise subtitles with both the image and the spoken SL dialogue. They must first determine the precise start of speech. Subtitling software has the possibility to step through any sequence forwards and backwards at any pace. By doing this, the ?in-point?, that is the exact moment of the first sound of the first word can be defined. This is done using time-codes. This allows the subtitler to mark the ?in? and ?out? point of the speech very precisely (SBS 2002). The computer programme facilitates the work by marking the length of the subtitle line. By putting a marker on the line, it indicates the correct length, thus indicating how much the text has to be reduced. Subtitling consists of the transfer of information from one language to another and from the spoken to the written language (Smith 1998). In subtitling, it is inevitable that the subtitler condenses, omits or paraphrases the original dialogue since only in a small minority of cases will it be possible to provide a verbatim translation. This helps explain why when converting the spoken to written word, the volume is typically reduced by one third (Gottlieb1998a: 247). De Linde and Kay (1999) counted that as much as half of the original dialogue is lost due to the spatial constraints. Time and space constraints require that the subtitler make room by deleting something from the original (de Linde and Kay 1999: 154). The target text has to be a mere fraction of the original dialogue as the translator has to abridge the text, sometimes very drastically. Fawcett (1996: 79) says that passages in an SL may be missing from the TL because the space is not available in the target publication, picking out only the minimal information relevant to the situation. What appears in a translated programme is what the translator deems to be relevant. Kovacic (1994) draws on Relevance Theory (Sperber and Wilson 1986) to rationalize the notion of omission and asserts that any deletions can be accounted for by the principle of relevance: 39 When the subtitler is short for space, he/she evaluates the relative relevance of individual segments of a given message. Relying on the viewers? ability to apply adequate cognitive schemata or frames and to draw on either previous information in the story or their general knowledge of the world, the subtitler leaves out the part of the message he/she considers the least relevant for understanding the message in question, for perceiving the atmosphere of a situation or the relationship among the participants involved, and eventually for the general understanding and reception of the story. (Kovacic 1994: 250) In addition, as Ivarsson (1992: 90) points out, fast speakers may say in the space of a few seconds two or three times what can be accommodated in subtitles. Thus, the need to omit some elements of the source text. To achieve text reduction, translators have a range of techniques at their disposal. Gottlieb (1998b) compiled a list of strategies. He distinguishes between different types of text reduction, namely condensation, decimation and deletion. Condensation retains both meaning and most of the stylistic features of the original. In decimation, abridged expressions are used and there is a reduction in content. The cuts in the source language may result in a loss of semantic or stylistic content. The message is conveyed with the help of the other channels (soundtrack or image). Repetitions and filler words can be omitted without loss of information to the audience and are, therefore, examples of deletion. There are many elements of speech which are regarded as superfluous and consequently omittable when converted into written form. The subtitler should, therefore, not attempt to transfer everything, even when this is spatio-temporally feasible. He should attempt to keep a fine balance between retaining a maximum of the original text (essential for the comprehension of the linguistic part of the target film), and allowing ample time for the eye to process the rest of the non-linguistic aural and visual elements. Categories of linguistic items that could be omitted are for example: Padding expressions ? that is, spoken filler words which do not add any new information: for example, ?you know?, ?well?, ?actually? etc; responsive expressions (e.g. ?yes,? ?no,?), ?please?, ?thank 40 you?, ?sorry?. These expressions are recognized and comprehended by the majority of people, and could therefore be omitted from the subtitle (Karamitroglou 1998). Spoken language is generally less formal, highly colloquial and more immediate than its written equivalent. Therefore, grunts, false starts, repetitions colloquialism, non- grammatical utterances, etc are omitted from the subtitle as they are not central to the message of the film (Smith1998: 146). As already mentioned, a film suffers many alterations and cuts and in the case of subtitling, those who follow both SL and the subtitles are aware that they are being offered only a selection of the original information which means that they are not being told everything. Each drama television programme also requires some editing of the original itself, so as to achieve coherence between sound and image (de Linde and Kay 1999: 183). Since film translation is generally sought at speech act level (Gottlieb1998a: 247) and full translation of the spoken discourse in films and television is seldom desirable, intentions and effects are more important than isolated lexical elements. This understanding relates to Nida and Taber?s dynamic equivalence4 which is based upon 'the principle of equivalent effect' (1964:159). In her functional approach, House (1977) argues that ST and TT should match one another in function. Highlighting the functional aspect of translation is important for subtitling because the message to the audience can thus be delivered in the briefest time as possible as the translator would not have to translate everything. Baker (1992) bases her pragmatic equivalence on implicature. Implicature is not about what is explicitly said. The translator needs to work out implied meanings in translation in order to get the ST message across. The role of the translator is to recreate the author's intention in another culture in such a way that enables the TT reader to understand it clearly. Implicature is relevant here because subtitling is polysemiotic. The 4 Dynamic equivalence is defined as a translation principle according to which a translator seeks to translate the meaning of the original in such a way that the TL wording will trigger the same impact on the TC audience as the original wording did upon the ST audience (Nida and Taber 1969/1982: 200). 41 audience can infer a lot from what else is happening on the screen and does not only have to rely on subtitles for meaning. This subject is dealt with below. 2.5.2. FILM AND SEMIOTICS In their contribution, Luyken et al. (1991: 190) state that in the subtitling of a film or of a television programme, the ?message? is expressed by the audiovisual whole, i.e. image, acting, sound and language. That is, film is a polysemiotic medium that transfers meaning through several channels, such as picture, dialogue and music. Thus, the subtitler can rely on the remainder of the soundtrack to carry the full meaning of a film sequence. Since subtitling is polysemiotic (de Linde and Kay 1999), all references to events which are visible to the viewers or to what others have said in the dialogue must, therefore, be eliminated from the subtitles. The integration of the components of subtitled television (image, subtitles and spoken dialogue) combined with viewers? reading capacities determine the characteristics of the medium. That is, there must be some agreement between the subtitles, the spoken source language (SL) dialogue, and the corresponding image5 (Gottlieb1998a). With subtitling, the synchronization of sound and image is made more complex with the addition of a textual component, further illustrating the semiotic interplay in all forms of audio-visual language transfer. Gottlieb (1998a: 245) states that in film and television programmes, the translator has four simultaneous channels to consider: 1. The verbal auditory channel, which includes dialogue and background voices and maybe lyrics. 2. The non-verbal auditory channel, which is made up of natural sound, sound effects, as well as music. 3. The verbal visual channel, comprising the subtitles and any writing within the film, as for example, letters, posters, books, newspapers, graffiti, or advertisements. 5 This is crucial in dubbing where there must be a match between labial consonants. 42 4. The non-verbal visual channel, which includes the composition of the image, camera positions and movement as well as the editing which controls the general flow and mood of the movie. The aim of translation for subtitles is to fulfill their role within this polysemiotic environment. While the audience can enjoy the authenticity of the original dialogue, their ability to take in information is severely tested. In addition to the visual and aural input of the SL version, they have to cope with a sizeable volume of written text, superimposed on the screen (Gottlieb1998a). The experience for the viewers is somewhat disturbed as their eyes are divided between the subtitles at the bottom of the screen and the rest of the image. This constant diversion of focus may result in loss of information which is vital to follow the narrative. Gottlieb (1998b) describes translation for subtitling as a "balancing act" whereby the dialogue is transcribed into lines of text, "conveying a maximum of semantic and stylistic information." The translator not only translates but also resolves which fragments to omit, which of them are irrelevant, and which are vital to the target audience in an attempt to convey 'the core?. To achieve this reduction, translators enter a decision- making process where they determine what has to be translated and what can be left out. This decision is influenced, states Kovacic (1996), by three factors: the type of programme, the target audience and the aesthetic aspect of the language. When translators make decisions about what to translate and what to leave out, Gottlieb (1998a: 247) states that two factors motivate the choice: intersemiotic redundancy and intrasemiotic redundancy. With these terms, Gottlieb notes that there can be redundancy either between channels or within one channel. For example, both visual and auditive channels may convey almost identical information (intersemiotic redundancy) and the spoken word can be regarded as redundant. Alternatively, the same information may be repeated within the sound track. This intrasemiotic redundancy occurs as part of the mode-shift from spoken to written word. Each subtitle has to work both as a unit as well as part of a ?larger polysemiotic whole? (Gottlieb 1998a). 43 2.5.3. FILM AND DIALOGUE Following Peter Newmark?s (1991: 10-13) distinction between semantic and communicative translation, subtitling is a communicative translation in the real sense, and, as such, it is much less concerned with the words of the speaker than with the communicative intention of the speaker. The task of the translator consists of rendering the spoken dialogue into written text. The translator has to produce subtitles which read naturally and are intelligible as a unit. The subtitler should also not depart from the language register of the original (Smith 1998). Subtitles must be broken down into readily digestible chunks (Smith 1998). When a sentence is divided into two lines, it should be according to the natural sentence structure. The audience will clearly find the subtitles more palatable if each of them forms a logical unit in itself. Difficulties in the comprehension of subtitles may be compounded by the fact that subtitles are displayed only for brief moments before they vanish, giving viewers no recourse to previous text while having to process text and image at the same time. Sub- titles have to be read and understood in the few seconds they are visible on the screen. Therefore, subtitles work only when the audience can understand them without difficulty first time round. Subtitlers should have a good idea of their target audience and adapt their language accordingly. But a soap opera is aimed at a much wider audience. It will be inaccessible to many of them unless it is broken down into simply structured subtitles consisting of everyday words (Smith 1998). That is why Fawcett (1996: 79) states that ?Since the stream of speech flows on, the audience cannot be expected to sit and ponder difficult renderings ? otherwise it will lose the subsequent utterances: hence it needs to be able to recover the intended meaning instantly.? Subtitles must not become the prime focus for the viewers. Their function is to aid the audience with the understanding and enjoyment of the film. Good subtitles must remain subordinate to the film and must not interfere too much with the action on the screen. 44 To viewers in subtitling countries, retaining the authenticity of the original production is paramount (Gottlieb 1998b: 310). For these viewers, subtitling is a more authentic mode than dubbing. The audience is not allowed to forget about the foreignness of a translated film and is constantly reminded of its authenticity as it hears the original dialogues throughout the film.The original dialogue allows the target audience to enjoy the voice quality and intonation of the original actors6. In this way, subtitling becomes a form of foreignisation (Szarkowska 2005) where the foreign identity of the source text is highlighted. Foreignisation privileges the source culture and it evokes a sense of ?otherness?, (Szarkowska 2005). Amongst the major methods of translating films, subtitling involves the least interference with the original. Therefore, it is subtitling that contributes to experiencing the flavour of the foreign language and the sense of a different culture more than any other translation mode. This is mainly due to the fact that the original soundtrack and dialogues are not tampered with, unlike the case in dubbing. In subtitling, simpler grammatical and lexical structures are preferred in order to decrease decoding effort. For example, the subtitler should replace passive with active constructions (which also saves space), negative with positive expressions, and should prefer straight-forward questions (Karamitroglou 1998). For example, from Generations: ZANELE (p 20): ? kodwa akulungile ukusola umuntu (Zulu) ? Back Translation: But it?s not right to blame someone? Subtitle: But it?s wrong to blame someone? NGAMLA (p 31): Uyizwephi ke leyo (Zulu)? Back Translation: You heard it where that? Subtitle: Who told you? 6 However, this situation increases the critical power of such people and makes them act as experts in the field of subtitling. Obviously dubbers do not have to confront such a problem, as the audience is not given a chance to compare the original dialogues with their translation. 45 With regard to the translation of cultural terms, the translator on Generations has relied on procedures such as loan-translations (Jakobson 2000: 115), cultural or functional equivalents (for example, the substitution of Mnumzana for Sir), and omissions. Taboos and other 'untranslatables': As swear words are connected to taboo areas, this will differ substantially from culture to culture ? for example, taboo words referring to sex, sex organs, bodily functions, religion, political concerns or offensive statements. This topic is covered by Wehn (2001: 66) who states that scenes containing violence, sex and, in the case of Germany, allusions to National Socialism and the Third Reich are often erased during the dubbing process because they touch on taboo topics. In Generations, a number of taboo words do occur in some of the episodes. It is also usually best to subtitle even the strongest of dialects and sociolect into the standard form of the target language. In his own study, Goris (1993) finds three general processes at play in film translation: linguistic standardisation, cultural naturalisation, and explicitation. Types of Subtitling: Gottlieb (1998a: 247) claims that linguistically, there are two types of subtitling: Intralingual subtitling (in the original language) and interlingual subtitling. Intralingual subtitling includes (a) subtitling of domestic programmes for the deaf and hard of hearing (b) subtitling of foreign-language programmes for language learners. It is vertical, in the sense that it involves taking speech down in writing, changing the mode but not the language. Interlingual subtitling is diagonal in the sense that the subtitler crosses over from speech in one language to writing in another, thus changing mode and language. Subtitles can either be traditional or they can be reduced (containing only key phrases). Traditional subtitles are in full sentences and are synchronized with the original dialogue. This is the most widely used form of subtitling. In reduced subtitling, however, only a summary is given of the dialogue and sometimes even of the action (Luyken et al. 1991). 46 It is usually used for the hard-of-hearing and is a summary consisting of key words or phrases. In bilingual subtitling, the subtitles are given in two different languages. This happens when the target audience comprises people from different language groups. This study examines only traditional subtitles and interlingual subtitling. 2.5.4. FILM AND PUNCTUATION: Paralinguistic signs such as continuity dots, linking dots, dashes and italics are also very important in subtitling. Continuity dots (?) are used to warn the reader to expect more. They are used right after the last character of a subtitle when the subtitled sentence is not finished in one subtitle and has to continue over the consecutive subtitle. ?Linking dots? (...) are used right before the first character of a subtitle when this subtitle carries the follow-up text of the previous uncompleted sentence. Hesitation or insecurity and afterthought can be visualised by inserting an ellipsis (...) ? for example: No? ?But I didn?t lie. Dashes are used in dialogue and are aligned to the left at the bottom of the screen. They are used before the first character of each of the lines of a two-line subtitle to mark a change of speakers in fast speech. Spoken language can express shades of meaning through intonation which can also be done through the use of italics, underlining and exclamation marks. Exclamation marks are used to indicate emphasis or loudness. Italics are usually used to mark foreign words or to emphasise a particular word in the dialogue. Abbreviations help to conserve space on the screen ? for example, through the use of acronyms. Another way to reduce the text volume is to use numbers instead of letters for figures, and apostrophes for abbreviations of auxiliaries such as ?You can?t.? 47 Layout of Television Subtitles: In subtitling countries, including several non-European speech communities (Gottlieb 1998a), subtitles usually consist of one or two lines. As a rule, subtitles are placed at the bottom of the screen and are either centred or left-aligned. If a subtitle only consists of a single line, it should be displayed in the lower line position ? that is, in the second line position of a two-line subtitle (Luyken et al. 1991). Subtitles are usually added to the screen image much later, as a post-production activity. The theories above have assisted in putting screen translation and translation theory into the spotlight and also in understanding how different scholars have treated the latter. This overview is useful when it comes to the analysis in the next chapter. 2.6. SUMMARY The chapter has explored the concept of DTS, discussed screen translation and translation theory and applied it to the current study. It then went on to look at various perspectives on subtitling, ranging from concepts such as?adaptation? (Delabastita 1989: 213), to the problem of sufficient exposure time for a subtitle which is still unresolved (Luyken et al.1991), to the match between dialogue and picture which must be retained when translating. Generally, the literature reviewed above has focused on such themes as norms, competence, performance, translation policy, directness of translation, film and semiotics, the nature of subtitling and so on. The value of this chapter is that it provides an understanding of the research?s major concept under investigation: that is, subtitling. 48 CHAPTER 3 ANALYSIS OF THE SOAP OPERA GENERATIONS INTRODUCTION The two previous chapters provided a background to the study and related research, a discussion of the notion of DTS, and translation theories. This chapter provides a description of the research methods used to collect data from the Generations personnel. The second part of the chapter sets out the methodology used to analyse actual subtitling practices on Generations. It looks at language use in Generations and their increased use over time. Manuscripts from 1999, 2003, 2005 and one from 2006 are examined. The second part of the analysis deals with codeswitching and, lastly, with an analysis of the subtitling itself. The chapter first gives a re-cap of the research problem under investigation. It then describes and justifies the research methodology used to provide answers to the research question. The section covers the choice of research design and methods, sampling methods, and data collection. 3.1. RESEARCH METHODS Research Design: The research is a qualitative study of the soap opera Generations aimed at obtaining and analysing information on the processes and practices of subtitling in respect of the soap opera. Snape and Spencer (2003: 4) state that in qualitative research, the research designs adopt a flexible method of investigation and the researcher conducts inquiry in real-world rather than in experimental manipulated settings. Main data collection methods identified with qualitative research include: observation, in-depth interviews, focus groups, narratives, and analysis of documents and texts. During data collection, researchers use non-leading questioning techniques (Snape and Spencer (2003: 20). Qualitative methods are used to address research questions that require explanation or understanding of social phenomena and their contexts. For example, in this case, processes in the subtitling of Generations. 49 Qualitative research places emphasis and value on the human, interpretative aspects of knowing about the social world and the significance of the investigator?s own interpretations and understanding of the phenomenon being studied (Snape and Spencer 2003: 7) ? for example, whether there has been an increase of subtitling on Generations. Because of its facility to examine subjects in depth, qualitative research provides a unique tool for studying what lies behind, or underpins a decision, attitude, behaviour or other phenomena (Ritchie 2003: 28). . The term ?case study? is strongly associated with qualitative research, although it is used in a variety of ways (Lewis 2003: 51). Features associated with case studies are: the fact that only one case is selected, although it is also accepted that several may be; the fact that the study is detailed and intensive; the fact that the phenomenon is studied in context; and the use of multiple data collection methods (Lewis 2003: 52). Case studies are used where no single perspective can provide a full account or explanation of the research issue and where understanding needs to be holistic, comprehensive and contextualised (Lewis 2003: 52). Population: In this case the people interviewed consisted of the supervisor of translators, the subtitler and the video editor of Starke Films, the producers of Generations. The researcher felt that the population she had targeted was the best to give her the information she was looking for ? the supervisor of translators because he decides what gets translated and what does not; the subtitler has to come up with suitable subtitles, and the video editor because he is in charge of the final product after all the production work has been done. He has to edit the film so that there is coherence between picture, subtitle and dialogue. Therefore, these people are in a position to provide information that would aid the researcher in investigating the processes and practices of subtitling on the soap opera Generations, to find out whether they match international, theoretical and methodological practices. Research Methods: The researcher used in-depth interviews to obtain data from the afore-mentioned people. This was done through the use of an interview guide, which 50 touched on the main topics of the research. According to Ritchie (2003: 27), a major feature of qualitative methods is their facility to describe and display phenomena as experienced by the study population ? in this case, the chief of translators, subtitler and video editor ? answering ?what is?, ?how? and ?why? questions. It therefore offers the opportunity to ?unpack? issues, to see what they are about or what lies inside, and to explore how they are understood by those connected with them. Data collection methods usually involve close contact between the researcher and the research participants, which are interactive and developmental so that emergent issues may be explored (Snape and Spencer (2003: 5). Questions asked were generally related to the major research question that sought to investigate processes and practices in the subtitling of Generations in order to find out whether they match international, theoretical and methodological practices. A tape recorder was used to record the interviews after permission was sought from the respondents. This enabled the researcher to avoid note taking which might have distracted the interview process. The tape recorder also freed the interviewer to be more attentive to the conversation. The research began with a couple of research questions mainly derived from the literature review. Data collection, and, eventually analysis, was guided by those questions. The information presented below is a summary of data collected from the respondents (the supervisor of translators, Mr Matseoane, the subtitler, Maggi van Aswegen and the video editor, Babalo Mpoyiya) after separate interviews with them. Appendix D provides a full transcript of interviews carried with the respondents in January 2006 at the offices of the producers of Generations, the production company Starke Films, at 24 Ditton Road, Auckland Park. 51 3.2. PRODUCTION PROCESSES AND PRACTICES IN THE SUBTITLING OF GENERATIONS The production of Generations starts with the Head Writer. He thinks up the story. He makes the storyline. Then, this storyline is put into what are called treatments ? that is, the basic story plus a little bit of elaboration. The Head Writer just gives the script writers ideas. The treatments are given to the script writers and they write the whole script ? scene for scene. Afterwards, the script goes to the editors. The editors check for flaws, and ensure that the whole story flows. From the editors, it goes to the translators. The translators translate the required sections into the African languages. From there, they do the timing ? to see if there is enough time for each episode because enough time has to be left for advertisements in between the different scenes. And then it goes back to the editors for them to check if the timing is correct. The manuscript is then taken to the SABC studios for the shooting of the episode. Subtitling is a post-production activity. Therefore, the manuscript is only given to the subtitler after production. The subtitler slots in the recaps of the different scenes from the previous episode and does the subtitles. To start with, the whole script is written in English. Some sections of the script are translated into Sotho, Zulu or Tswana and sometimes into Xhosa. There are 3 translators on Generations. One goes to the SABC studios to be present at the production of the programme ? that is, at the filming of the programme. She takes care of the language, ensuring that the actors follow the written script. The presence of the translator in the recording studio is to assist the directors, who are white and are therefore not familiar with the African languages, and to alert them to any errors. The other two translators remain in the office to do the translation. Senior translator Corlett Matseoane chooses the scenes and the dialogues that are to be translated. The choice is not based on promoting any notion such as ?rainbow nation? or the like. The segment chosen for translation is dependent on whether it is viable to translate ? that is, whether it is easy to translate. And of course, it has to involve a scene in which the indigenous languages can be spoken: that is, a scene in which a black person is talking to another black person, or two Afrikaans- speaking people are talking to each other. 52 After the translations, the subtitling process begins. Sometimes, the subtitler simply uses the original English text (what was in the script before translation) for the subtitle. At other times, the subtitler condenses the English original text. At other times, it is clear that the subtitler is condensing the vernacular text, in accordance with subtitling theory practice, to make the subtitles easier for the audience to read. After the subtitler has made the subtitles, the script goes to the video editor, Babalo Mpoyiya. He does the technical aspects of subtitling. He receives two scripts: one from the subtitler, with all the subtitles written on it and the other one from the director of the episode. Each episode gets recorded at the SABC studios and later the tape is sent to the video editor, together with the director?s script because the latter sometimes does three takes per scene and the video editor chooses the best to use on the episode. After Maggi van Aswegen (the subtitler) has made the subtitles, Mpoyiya types them onto a video tape. He digitizes them on a computer from a Beta tape slotted into a Beta Cem SP machine ? that is, he captures the pictures from the tape, into the computer. He then edits the programme by making cuts in order to achieve coherence between picture and dialogue and for rhythm. The cuts mostly include errors made by actors, the count down before ?ACTION?, the boom and the microphone (if seen), camera mistakes and so forth. The cuts are also dependent on the pace of the dialogue. If it involves, for example, two people talking fast in turns as they are walking, then the cuts will be faster than if the dialogue is between two people sitting opposite each other in a restaurant and just talking. After the editing of the whole episode, the subtitles are typed in. He overwrites the subtitle onto the picture on the computer screen. The match between dialogue and picture must be retained. In other words, the appropriate subtitle must appear on the screen at the same time as the appropriate picture. Mpoyiya, therefore, synchronises the subtitles with both the image and the spoken SL dialogue. He first of all determines the precise start of speech by moving the video forwards and backwards. By doing this, the ?in-point?, ? that is the exact moment of the first sound of the first word ? is defined. This is done using time-codes. This allows him to mark the ?in? and ?out? point of the speech very precisely. It facilitates the work by marking the length of the subtitle line. 53 The front title sequence and the end roller are already saved in the template of the computer. After the insertion of the subtitles, the video editor activates the end roller and rolls the names of the people involved in the production of the episode ? that is the names of the actors, the director, translators, head writer, script writers, the producer etc. The names are rolled to signal the end of each episode. The names of the regular actors are already in the file but, the video editor has to type in the names of the supporting cast each time because those come and go and, therefore, never remain the same. The same is true for scriptwriters as they take it in turns to write scripts for respective episodes. The names of directors of episodes also have to be changed each week because there are different directors each week. They take weekly turns to direct the five-day episodes. After the frontal sequence has been put in, the episode is ready to be aired. 3.3. SUMMARY Qualitative research is important because it examines a subject in-depth. It, therefore, offers the opportunity to ?unpack? issues. The research design and methodology described above enabled the researcher to collect data from respondents concerning processes and practice in the subtitling of Generations. In this case, the chief translator, the subtitler and the video editor were interviewed. The information collected has been presented above. 3.4. FREQUENCY OF USE OF VERNACULAR LANGUAGES AND THEIR INCREASE OVER TIME The study now moves on to look at the languages spoken on Generations and their progression. To do this, manuscripts from 1999, 2003, 2005 and one from 2006 are examined. The second part of the analysis deals with codeswitching and, lastly, with an analysis of the subtitling itself. 1999 1999 GENERATIONS 9 EPISODE 1 There are 221 texts (pieces of dialogue) in this episode. Tswana has the highest incidence of codeswitching (16) and has therefore been subtitled more. Examples are on 54 pp 6-13, 32, 41-45. Afrikaans is the next highest with 14 (pp 3-5, 35, 37-40). The only other language spoken in this episode is Zulu, which is spoken three times (pp 15, 16 and 21). There is no subtitling on pp 1-3, 14, 17-20, 22-31, 33-35, 36-38, 41 (out of 45 pages). However, in cases in which a group consists of different races, the language of communication is just English, without any codeswitching, as shown, for example, where Archie (black) is speaking to Sister Angelina (white) pp 25-27, Shaan ( Indian), Kgomotso, Queen (black) and Maxime (white) pp 34-38, Sonny (coloured) and Queen, pp 40-41. Therefore, there is no subtitling in this case. Codeswitching, is where individuals change languages frequently, often within one sentence (Edwards 1994: 2). In this episode, statistics also show that English is dominant: out of a total of 221 pieces of dialogue, only a total of 33 are in the indigenous languages, representing15 per cent of the whole episode. The rest of the dialogue is in English. 1999 GENERATIONS 9 EPISODE 2 There are 229 texts in this episode. Tswana again leads as the most spoken vernacular language, with 30 incidences, and therefore the most subtitled. It is followed by a Zimbabwean language known as Shona (15 incidences), Afrikaans (12) and Zulu (3). The following are examples of the languages spoken in this episode: I am giving some examples at this stage so that the reader has an idea of the languages spoken, but they will be analysed later. The other reason is to show that in the early days of the series, there were even foreign African languages used in Generations, at the expense of local vernacular languages such as Sotho and Xhosa. SIKHALAZO (p 39): Hazvisizvo zvingandi konesa kunakidzwa (Shona)! 55 Subtitle: That?s never stood between me and a good time! (Other examples of Shona are on pp 30, 31, 32, 38, 39). ZINZI (p 23): Oh Archie ? I don?t hate you?The problem is?Nngiyakuthanda (Zulu)? Subtitle: ?I love you? (There is another example on p 23 and another on p 22.) BABA (p 46): So klim in jou bakkie en gaan terug Upington toe (Afrikaans)! Subtitle: So get into your bakkie and go back to Upington! (Other examples of Afrikaans are on pp 33, 35, 36, 46, 47, 48). KARABO (p 4): She decided to take her own life, Mandla! Odirile tse tsothlhe ka boene (Tswana). Subtitle: She did it all by herself! (Other examples are on pp 3, 6, 7-11, 17-18 etc). However, cases in which dialogue is between interlocutors of different races, communication is just in English, without any codeswitching, as shown where Glen (black) is speaking to Sara-lee (white) pp 13-16, Ntsiki (black) is speaking to Shaan (Indian) pp 25-29. Therefore, there is no subtitling in this case. Statistics also show that English is dominant: out of a total of 229 texts, a total of 60 are in the indigenous languages, representing 26 per cent of the whole episode. The rest of the dialogue is in English. 1999 GENERATIONS 9 EPISODE 3 There are 252 texts in this episode. There are 34 instances of codeswitching in Tswana and, therefore, the highest instances of subtitling. Examples are on pp 8, 10, 15-26, 43- 44. Afrikaans is next with 8 (pp 47, 48, 49). Shona and Rastafarian are both spoken two times each (pp 7 and 9; 49 and 51 respectively). There is no subtitling on pp 1-2, 3-6, 11- 14, 27-31, 32-38, 39, 40, 41-42, 46), for the same reasons advanced in the previous episodes. 18 per cent of the episode has been subtitled. 56 1999 GENERATIONS 9, EPISODE 4 This particular episode contains a total of 226 texts. Afrikaans is the most spoken language on this episode (17 texts), some of which are subsequently subtitled into English. Tswana is the next highest with a frequency of 10, followed by Rastafarian (9) and lastly Zulu with 6. Because the vernacular is only a portion of the actual text, only that portion of the text needs to be subtitled. Codeswitching has been noticed especially between people who are close. For example, in this episode, the married couple, Stoffel and Baba usually converse to each other in Afrikaans, as do Queen and Terence, who are also a couple, and Karabo speaks to Mandla in Tswana. Generally, there is quite a lot of codeswitching where black people are talking to each other. Some examples of codeswitching in this episode: STOFFEL (p 21): Shopping. I got you a surprise! Waterblommetjies ? bokse vol. Dis agter op die bakkie (Afrikaans). They were on special. Subtitle: ?boxes of them. They?re on the back of the bakkie. In the above example, only a bit of the text is in Afrikaans, and only this section is subtitled. In the following example, the whole text needs to be subtitled as all of it is spoken in Afrikaans: BABA (p 21): Ag dankie my lam. En ek?s so bly jy?s hier. Nou?s ek sommer meer gerus oor die hof-toe ganery (Afrikaans). Subtitle: Thank you, my dear. I?m glad you?re here. Now I?m more relaxed about going to court. (Other examples of Afrikaans texts are on pp 5, 6, 22, 23, 34, 35, 37). QUEEN: (p 24) Darling! I?ve come to thank you properly. Heavenly! O itsitse jang go re (Tswana) this was my favourite perfume? 57 Subtitle: How did you guess? Statistics however show that English is still dominant: out of a total of 226 texts, 42 are in the indigenous languages, representing 19 per cent of the whole episode. The rest of the dialogue is in English. 1999 GENERATIONS 9 EPISODE 5 In this episode, Afrikaans is again the most codeswitched language, representing 20 incidences, followed by Tswana (15) and Rastafarian (3) from a total of 195 texts. Zulu is conspicuously missing from this episode. In these early years, Zulu appears not to have been ranked highly in the order of things. Tswana appears to have been regarded as more important than all the other African languages because in these earlier series, it is used more often than the rest. Alternatively, the reason could be that the studio did not have suitable people to play Zulu-speaking characters or that perhaps the actors playing the parts were Tswana actors. An example in this episode of Tswana bits which have been subtitled is the following: ARCHIE (p 25): Listen ? I?ve noticed Mandla looking a bit stressed lately. A ke go re o bereka thata, kgotsa go na le sengwe se se mo tshwenyang (Tswana)? Subtitle: Is he working too hard or, is something else bothering him? (Other examples are on pp 9-11, 24-28.) With Afrikaans, in a number of cases, the whole text is in Afrikaans (whereas, with the African languages, it is rare to find the whole texts in the vernacular), for example: STOFFEL (p 5): My magtag! Hy?s die een wat jou in die eerste plek in die hof laat beland het! Hy moet waai. Ek soek nie ?n drug addict in my huis nie (Afrikaans). Subtitle: He?s the one who landed you in court in the first place!...He must go. I don?t want a drug addict in my house. BABA (p 5): Wag nou eers, Stoffie. Ek stem saam hy kan nie bly nie, maar foei tog ? hy voel so goed oor wat hy gister vir my gedoen het (Afrikaans)? 58 Subtitle: Wait first. I know he can?t stay ? but he feels so good about what he did for me yesterday? (Other examples of Afrikaans are on pp 1, 3-5, 6, 16, 30-31, 44, 46). NTSIKI (p 43): Da holy herb gives I and I da munchies, mon. Where be da grub stuff (Rastafarian)? Subtitle: The holy herb makes me hungry. Where?s the food? Here again, in cases where the group consists of different races, the people just communicate in English, without any codeswitching. So there is no subtitling involved in such cases; for example, the conversations between Sarah-Lee and Nstiki (pp 12-13), Glen and Shaan (p 14), Shaan, Ntsiki and Glen (pp 15, 17and19), Baba and Queen (pp 29, 32), Ntsiki and Miles (pp 33-35 etc). Figures however show that English is still dominant: out of 195 texts, 38 are in the indigenous languages, representing 20 per cent of the whole episode. The rest of the dialogues are in English. SUMMARY OF 1999 EPISODES EPISODE 1: Tswana 16 subtitled bits Afrikaans 14 Zulu 3 33 out of 221 texts= 15 per cent EPISODE 2: Tswana 30 Shona 15 Afrikaans 12 Zulu 3 60 out of 229 texts = 26 per cent EPISODE 3: Tswana 34 Afrikaans 8 Shona 2 59 Rastafarian 2 46 out of 252 texts = 18 per cent EPISODE 4: Afrikaans 17 Tswana 10 Rastafarian 9 Zulu 6 42 out of 226 texts = 19 per cent EPISODE 5: Afrikaans 20 Tswana 15 Rastafarian 3 38 out of 195 texts = 20 per cent EPISODE 6: Afrikaans 14 Tswana 2 Shona 2 18 out of 189 texts = 10 per cent From the above statistics, it has been observed that Afrikaans and Tswana traded places as the most codeswitched languages and, therefore, the most subtitled. Surprisingly, Zulu, the language spoken by most black people in South Africa either does not feature at all in most of the episodes as can be seen in episodes 3, 5 and 6, or, if it does feature at all, it is spoken only very few times (episodes 1, 2 and 4). Foreign languages such as Shona and Rastafarian seem to be given even more prominence than Zulu. For example, Shona features 15 times in episode 2 and Zulu is a distant last position with 3. In addition, both Shona and Rastafarian feature in episode 3, but Zulu is absent. There is Rastafarian in episode 5, and Shona in episode 6, but Zulu is absent from both. 60 In a group which consists of different races, the characters in it only speak English. This is probably because people of different races are not expected to speak or understand each others? languages. English is assumed to be understood by everybody. In all the above episodes, English still predominates: in episode one, there are a total of 33 instances of vernacular/Afrikaans (the rest being in English) out of 221 texts. This is on top of the fact that, in these same 33 cases, not even the whole utterance is in the vernacular but is shared with English. The same applies to episode 2 (60 out of 229 texts), episode 3 (46 out of 252 texts), episode 4 (42 out of 226 texts), episode 5 (38 out of 195) and episode 6 (18 out of 189). 2003 However, by 2003, Zulu is more frequent, having displaced Tswana and Afrikaans as the most spoken indigenous language on the soap opera series. An examination of the 2003 Generations 13, Episode 1, reveals the following: of the 250 texts in the episode, 49 are Zulu codeswitches. Even more Zulu-speaking characters have now been added to the series: for example, Angela, Jonathan, Julia and Khaphela, who were not in the 1999 series. Xhosa is second with 12 codeswitches and, therefore, 12 instances of subtitling. Possibly in line with the Constitutional requirement for multilingualism, Xhosa is now included in the series. Tswana and Afrikaans, which hitherto had swopped places for the top slot, (refer to 1999 analysis) are in this episode lying third and last respectively. There are 8 codeswitches in Tswana and, therefore, 8 subtitles. Afrikaans has been relegated to the last position with only one instance of spoken Afrikaans. It is not necessary to analyse the 2003 (and 2005) episodes in the same detail as is done for 1999 because the procedure is all the same and it becomes repetitive. Therefore, what I am presenting here is essentially a summary of the findings. Here is an example of 2003 Generations 13 episode 1. 61 2003 GENERATIONS 13, EPISODE 1 An example of subtitling in this episode is: JONATHAN (p 45): Uma sengiqedile (Zulu) I?m setting up my computer games. Subtitle: As soon as I have finished? ANGELA: Leflat yinhle kunefurniture ephakathi mama (Zulu). Subtitle: Your furniture makes the flat look nice? JULIA: Ngiyabonga (Zulu) my angel. I did the best I could with the cramped space. No subtitle was provided here possibly because the subtitler experts the word ngiyabonga which means ?thank you? to be understood by a majority of the population and according to film theory, words such as ?thanks?, ?no?, ?yes? should not be subtitled as they are usually recognized (Karamitroglou 1998). For the same reasons already mentioned in the 1999 series, mixed-race scenes do not have subtitles for scenes involving, for example, Blue/ Vuyo (p 10), Blue, Vuyo, Tsego (pp 11-12), Shaan/Devan (pp 18-21), Tau/Sonny (pp 18-21), Sonny, Devan, Vuyo, Mandla and Queen (pp 42-44) who are of mixed races. In episode 3, Afrikaans is absent. Again in episode 5, Afrikaans is conspicuously absent, while Zulu dominates the indigenous languages. In episode 6, Tswana has once again eclipsed Zulu. However, there are more codeswitches in Zulu here than in 1999. If anything, it would appear as if a deliberate effort has been made to increase the frequency of Zulu in the series. Xhosa has given way to Afrikaans in this episode. Out of all these languages, only Tswana has been stable because it has been spoken in all the episodes studied, unlike the other languages. However, as can be seen from a summary of the analysis below, the use of English is still predominant. Below is a summary of instances of subtitling in the 2003 Generations 13 episodes: 62 EPISODE 1: Zulu 49 Xhosa 12 Tswana 8 Afrikaans 1 70 out of a total number of 250 texts = 28 per cent EPISODE 2: Zulu 26 Tswana 20 Afrikaans 4 Xhosa 1 51 out of a total of 221 texts = 23 per cent EPISODE 3: Zulu 46 Tswana 13 Xhosa 10 69 subtitles out of a total of 271 texts = 26 per cent EPISODE 5: Zulu 43 Tswana 12 Xhosa 3 58 out of 276 texts = 21 per cent EPISODE 6: Tswana 44 Zulu 38 Afrikaans 7 89 out of 273 texts = 33 per cent 2005 Xhosa (used in the 2003 series) and Afrikaans (used in 1999 and some of the 2003 series), are not used in the first episode examined (2005 Generations 15 episode 76) and have been replaced by Sotho, possibly because of pressure from the regulator, ICASA to 63 ensure that the public broadcaster provides for all the official South African languages in accordance with the 1996 Constitution and the 1999 Broadcasting Act. The following is an example of codeswitching in Sotho, from the episode: THANDISO (p 53): The clients can see different samples pele ba kgetha (Sotho) a badge. Subtitle: ?before choosing ? The frequency of Zulu is relatively high. One of the reasons could be that even more Zulu-speaking characters have been added to the series. These are, Busi, Jack, Cleo, Dudu, Khethiwe, Ntombi and Ngamla, to name but a few. In episode 133, Zulu and Tswana are the only indigenous languages featuring in this episode. With a frequency of 24, Tswana is a distant second to Zulu which is spoken 96 times. In episodes 140 (2005) and 142 (January 2006) Tswana is again a distant second to Zulu. SUMMARY OF THE 2005 EPISODES GENERATIONS 15 EPISODE 76: Zulu 91 Tswana 25 Sotho 8 124 instances of codeswitching in 289 texts = 43 per cent Below are the rest of the statistics: EPISODE 133: Zulu 96 Tswana 24 120 out of 281 texts = 43 per cent EPISODE 140: Zulu 85 Tswana 32 64 Sotho 4 121 out of 281 texts = 43 per cent 2006 EPISODE 142 January 2006: Zulu 114 Tswana 39 153 out of 284 texts = 54 per cent 3.5. SUMMARY In all the episodes studied for 2005, plus one for January 2006, Zulu is by far the most spoken indigenous language, with Tswana a distant second. Afrikaans appears to have been phased out altogether because it is no longer spoken. There are no Afrikaans- speaking characters left on the series either. Languages such as Xhosa and Sotho seem to make an occasional brief appearance only to disappear again. When they do appear, the frequency is low. The enduring indigenous languages seem to be Zulu and Tswana, with the former on top. English continues its dominance as even in cases of codeswitching, the other language is always English. There are also whole stretches of text which are exclusively in English ? for example, in the last episode, conversations between Karabo and Anne (pp 28-32) and, Anne and Jack (pp 16-19). 3.6. ANALYSIS OF CODESWITCHING As can be seen from the dialogues above, codeswitching is part and parcel of conversations on the soap opera Generations. I would now like to give the theoretical basis for this notion and also to provide several examples of the different kinds of switches found in Generations. This codeswitching takes place especially between people who are close, such as couples and between friends. However, generally, there is quite a lot of codeswitching where 65 black people are talking to each other. Although codeswitching on Generations could be a way of accommodating multilingualism, another way of looking at it is that switching is used as a form of identity. By switching from English to the vernacular, a person might be signalling solidarity or co-identity with his interlocutor, while speaking in English only can be interpreted as a desire for distance between the interlocutors (Sebba and Wootton 1998). This then leads us to the theoretical basis for codeswitching which is covered below. It is common to find linguistic alternation occurring within one unit of speech directed to one listener. Speakers may often switch for emphasis because they feel that the mot juste is found more readily in one of their languages than in another, or because of their perceptions of the speech situation, changes in content, the linguistic skills of their interlocutors, degree of intimacy and so on (Edwards 1994: 72). Gumperz (1982: 66) states that in codeswitching: The tendency is for the ethnically specific, minority language to be regarded as the ?we-code? and become associated with in-group and informal activities and for the majority language to serve as the ?they- code? associated with the more formal, stiffer and less personal out-group relations. Many researchers make use of the ?we-code? and the ?they-code? in accounts of codeswitching behaviour (Sebba and Wootton 1998: 262). The concept of ?identity? is often invoked at the same time. In a study conducted on the Jamaican community in London, Sebba and Wootton (1998) found an apparently classic case of the ?we? versus the ?they? ? the ?they-ness? signifying the ?foreignness? of English. In London, London Jamaican is the ?we-code? because it excludes outsiders (particularly white people) (Sebba and Wootton 1998: 264) and its province is the family and peer group, especially during informal conversation. It is used among family and peers in the most intimate discussions and London Jamaican can be used to exclude outsiders but London English cannot. London English is the out group 66 language for the Caribbean community in Britain, and London Jamaican the in-group or ?ethnic? language, (Sebba and Wootton 1998: 275). According to Edwards (1994: 2), there is nearly always shifting among peers but other situations may call for the use of English exclusively. In Generations, when the group involved is of a mixed race, English is generally used. Edward adds that the language switches are non-random and that a switch signifies something. However, as for language changes signifying something, that does not seem to be the case in Generations because switches are dependent on what part of an utterance is easier to translate. Therefore, the switches do not signify anything. Many people who codeswitch are not aware of their behaviour until it is brought to their attention ? and even when it is, they often resist the notion that they really speak that way (Heller 1988: 6). In some communities, codeswitching is not only widespread but also accepted as the normal way of speaking (Heller 1988: 7). Codeswitching does not occur in all multilingual communities and, even in communities where it does exist, not every one codeswitches. Moreover, even among those who do codeswitch, codeswitching does not necessarily occur in all social situations (Heller 1988: 9). Some speakers may operate exclusively within the domain of one language. Codeswitching can be used to appeal to the shared understanding characteristic of co- membership, or to create distance by associating oneself, momentarily, with the out- group (Heller 1988: 83). Codeswitching in the soap opera Generations appears to be arbitrary ? there seems to be no particular pattern to it. The switching is equally from the indigenous languages into English just as it is from English into the indigenous languages. There are also instances where there is no codeswitching at all, where conversations are entirely in one of the indigenous languages, but these are very rare. 67 The switching is mostly intersentential ? switching where the change occurs at a clause or sentence boundary (Edwards 1994: 73) although, there have been many instances of the use of loanwords already referred to earlier on. However, what is beyond doubt is that switching in Generations is mostly between Africans. On codeswitching, Mark Sebba and Tony Wootton (1998) add that ?the social action of shifting identities? in talk is performed linguistically, for example, through choice of lexis, grammar, or by choice of language. (In Generations, codeswitching is mainly through choice of language). Below, the study examines codeswitching. However, not all instances of codeswitching are examined. The ones that are provided here are the ones that the researcher felt had something of interest. From the episodes examined (1999, 2003, 2005, and January 2006), the direction of the switches are as follows: From English into Indigenous Languages: For example in the following: JACKIE (p 41): Speaking of office romance?Ke utlwile ba re (Tswana: I heard) Bali?s hooked up with the Hit Factory?s new receptionist! BUSI (p 56): Stop it, Jack. Please just?stop. Ukhumbul? i-Mauritius, kanye nayoyonke eny? into (Zulu: You remember Mauritius and everything else). SONNY (p 14): I know. But at least they won?t murder any more innocent people. An eye for an eye, and a tooth for a tooth ? dis wat ek nou soek (Afrikaans: That?s what I want now). From the Indigenous Languages into English: DUDU (p 38): Ngabe ngikunikil? i-invitation yami (Zulu: I would?ve given you my invitation). It was a blow out. BABA (p 4): Verbaas my nie (Afrikaans: Doesn?t surprise me). They left in such a hurry she barely had time to pack. 68 QUEEN (p 26): Ke go boleletse Ma (Tswana: I told you Ma) ? no man can resist me. Example of Switches Framed on Both Sides by English LUMKA (p 42): Not quite. Kade ngicabanga kanti futhi (Zulu: I?ve been thinking and) I want to run something by you? KARABO (p 45): She did what she did to punish us ? to punish you ? fela e ne e le choice ya gagwe (Tswana: but it was her own choice)! You are not responsible. SONNY (p 5): And now? Lounging around the house in your nagkabaai (Afrikaans: pyjamas) ? don?t they miss you at that job of yours? Example of Switches Framed on Both Sides by the Vernacular BUSI (p 21): Ngiyamthanda (Zulu: I love him) Karabo. And at the worst moment possible, ngamphendukela (Zulu: I turned on him). KARABO (p 7): Go tshwanetse ga bo go na le explanation ya go re why a sa leka go escapa (Tswana: There must be some explanation why she didn?t try to escape). May be the fire was caused by an electrical fault or something. Gongwe o simolotse a robetse (Tswana: Maybe it started while she was sleeping). In the analysis of Generations, the switches range from emphasis or commenting on information-carrying parts of utterances, to switches to request for information, convey unpleasantness, and so on. The switches are from English to the indigenous languages and from the indigenous languages into English. The content of the utterance is also in English as it is in the indigenous languages. For example: Below are more examples of different kind of switches: Examples of Switches from Indigenous Languages: STOFFEL (p 6): Geen wonder dit was op special nie (Afrikaans: No wonder it was on special). Daylight robbery. I?m taking it back after I?ve told you your fortune, Mister! 69 This is a switch to English, and by ?daylight robbery? the character is making a comment on the content of the utterance. MANDLA: (p 7): Mapodisa a re ga gona evidence ya go re o lekile go escapa (Tswana: The police say there?s no evidence that she even tried to escape). She just willingly let the flames consume her. The switch to English coincides with the unpleasant part of the utterance. ARCHIE (p 24): Yazi inkinga yethu yini (Zulu: You know what our problem is)? We?re too much alike. That?s why it would never work between us. The switch to English is made to comment on a situation of a personal nature. DUDU (p 38): Ngabe ngikunikil? i-invitation yami (Zulu: I would?ve given you my invitation). It was a blow out. The switch to English conveys joy. Example of Switches from English: QUEEN (p 8): Yes. So ke utlwile dilo tse di nasty tse o neng o di bolelela partner ya gago ka nna (Tswana: So I heard the nasty things you told your partner about me). The switch to the indigenous language is to convey displeasure. KARABO (p 20): Why didn?t you talk to me about it? Ke ne ke akanya go re o siame (Tswana: I thought you were okay). The switch to the indigenous language is to reinforce what was said before in English. Examples of Switches from Indigenous languages (Criticism) KARABO (p 53): Ke go bolelletse gore (Tswana: I told you) the gentlemen?s club would fall flat, just like the poetry evening did. You know what your trouble is, Sibusiso? You?re all talk and no action?Go blow your hot air elsewhere. The switch to English is made in order to criticise. LUMKA (p 25): Sibuyela kulokho futhi (Zulu: Are we back on that again)? 70 Can?t we just focus on what we?re doing? Implied criticism is marked by a switch to English. KARABO (p 45): She did what she did to punish us ? to punish you ? fela e ne e le choice ya gagwe (Tswana: but it was her own choice)! You are not responsible. The switch to the vernacular conveys criticism. Examples of Switches in Order to Emphasise KARABO (p 20): No Busi?Thank you. You did a very brave thing. Mme go ne go siame go dira jalo (Tswana: And it was the right thing). The switch to the indigenous language is to emphasise the principal portion of the utterance, which is in English. JACK (p 58): I did love you Busi. In my own way I did. Bekungewon?amanga lawo (Zulu: That wasn?t a lie). The switch to the vernacular is made to emphasize a personal and emotion matter. QUEEN (p 9): E ne e se nna (Tswana: It wasn?t me)! I swear. The switch to English is for emphasis. SONNY (p 14): I know. But at least they won?t murder any more innocent people. An eye for an eye, and a tooth for a tooth ? dis wat ek nou soek (Afrikaans: That?s what I want now). The switch to the indigenous language is made to convey anger at an unpleasant situation. NGAMLA (p 42): Ungaz? ungithenge. Umsebenzi owakho (Zulu: You don?t have to bribe me. The job?s yours). All you have to do is say ?yes.? The switch to English is an addition to the first portion of the utterance. NGAMLA (p 45): Angizange ngibekhon? ekukhuleni kwakho, kodwa manje ngizama ngazo zonk? izindlela. Ngizam? ukukwenz? izinto ngendlel? ehlukile (Zulu: I wasn?t always there for you, but I?m trying to make up for it by 71 doing things differently). Now you throw it back in my face! The switch to English coincides with the unpleasant part of the utterance. BUSI (p 56): Stop it, Jack. Please just?stop. Ukhumbul? i-Mauritius, kanye nayoyonke eny? into (Zulu: You remember Mauritius and everything else). The switch to convey anger is in the indigenous language. SONNY (p 13): And they shot him ? three times?Doesh, doesh, doesh. In cold blood. Hy?s dood, my pal ? nog ?n blerrie crime statistic (Afrikaans: He?s dead ? another bloody crime statistic). The codeswitch into the indigenous language reinforces the chilling statement of the content before it in English. STOFFEL: (p 35): Oor my dooie liggaam gaan my vrou tronk toe (Afrikaans: Over my dead body is my wife going to jail)! I?ll sell the house to pay that fine if I have to. The switch to English is just to reinforce what was said before in Afrikaans. TERENCE (p 26): I ?can?t now. Perfume ya gago?e mpherosa sebete (Tswana: Your perfume is making me feel nauseous). The switch to the unpleasant is in vernacular. BUSI (p 21): Ngiyamthanda (Zulu: I love him) Karabo. And at the worst moment possible, ngamphendukela (Zulu: I turned on him). The switch into English functions as a commentary on the betrayal mentioned in the vernacular. Examples of Switches to Request for information BABA (p 30): Oh my goodness! I?ll never get used to modern technology. Stoffel insists I have it for safety reasons. Waar is die ding tog nou (Afrikaans: Where is that thing now)? The switch to indigenous language is made in search of information. 72 QUEEN (p 10): Wait. O tlile go dirang ka se (Tswana: What are you going to do about this)? The switch into vernacular is to request for information. In the last example below, the request for information is in the vernacular but the switch to English occurs at the point of a desire for confirmation. JACK (p 57): Ubani okutshele lokho? Kungab? uKarabo? Nom? umfowenu? (Zulu: Who put you up to this? Was it Karabo? Your brother?) It has to be because the Busi I know would never? More examples of switches are in appendix A and appendix B. 3.7. ANALYSIS OF ACTUAL SUBTITLING The study now looks at the qualitative and quantitative aspects of the analysis. As already stated earlier in chapter 2, a subtitler should not attempt to transfer everything, even when this is spatio-temporally feasible (Karamitroglou 1998). Which piece of information to omit or include depends on its relative contribution to the comprehension of the message. Categories of linguistic items that could be omitted are as follows ? padding expressions: These expressions are most frequently empty of semantic load and their presence is mostly functional, padding-in speech in order to maintain the desired speech flow. Responsive expressions have been found to be recognised and comprehended by the majority of people (Karamitroglou 1998), when clearly uttered, and could therefore be omitted from the subtitle. However, when they are not clearly uttered or when they are presented in a slang, informal or colloquial version, they are not recognisable or comprehensible and should, therefore, be subtitled. 73 Quantitative and Qualitative Analysis Simpler syntactic structures which are shorter and easier to understand are preferred to complex ones. The translator has also relied on other procedures such as loan-translations (Jakobson 2000: 115) and Delabastita?s (1990: 102) operations: repetitio (repetition), adiectio (addition), detractio (reduction or omission), transmutatio (transformation) and substitutio (substitution). Swear words are connected to taboo areas and are sometimes omitted altogether. Colloquialism, taboo language, non-grammatical utterances, repetition, etc are supposed to be deleted and, therefore, not used in the subtitle (Karamitroglou 1998). SUBTITLING PRACTICES IN INDIVIDUAL EPISODES Examined below are the following episodes: 1999 Generations 9: Episodes 1, 2, 3, 4, 5 and 6; 2003 Generations 13: Episodes 2, 3, 5 and 6; 2005 Generations 15: Episodes 76, 133, 140; and 2006 (January) Generations 15: Episode 142.. The episodes were chosen on the grounds that Starke Films, the producers of Generations only gave me manuscripts for these episodes. More episodes from 1999 are examined than those from 2003 and 2005 because 1999 had more languages and also some interesting examples. In the analysis below, I refer to pages because I am working from the manuscripts. The manuscripts do not deviate from the episodes: the subtitles on the manuscript are the same as the subtitles in the episodes. 1999 GENERATIONS 9 EPISODE 1 BABA (p 4): Verbaas my nie (Afrikaans). Back Translation: surprise me not: Number of characters = 15 (including spaces). Subtitle: Doesn?t surprise me. Number of characters = 19 The syntax elements of the source language, Afrikaans, have been transformed (transmutatio) in the subtitle (target language) in accordance with English syntax. 74 QUEEN (p 11): Boring. O ne a kgona go bua fela ka dikgomo (Tswana). Back Translation: He was only able to talk about cows: Number of characters=35 Subtitle: Could only talk about cattle: Number of characters = 28 ?Could? is a shorter form of ?was only able to? and therefore, the subtitling is in accordance with subtitling theory. Moreover, as can be seen, the number of characters in the subtitle is less than in the back translation. TERENCE (p 13): Ehm? ke emetse call e tswang overseas ? kwa Alaska (Tswana). They are twelve hours behind us? Back Translation: I?m waiting for a call coming from overseas ? from Alaska: characters = 55 Subtitle: I?m expecting an overseas call ? from Alaska: characters = 42 The hesitation in the source text (ehm?) has been omitted (detractio) in the subtitle in accordance with subtitling practices and the phrase ?waiting for? is substituted by a single word ?expecting,? resulting in a shorter subtitle. 2003 GENERATIONS 13 EPISODE 2 KENSANI (p 3): The house is kind of quiet without him. Eseng ka gore o ne a bua thata (Tswana). Except when it came to things he feels passionate about?like art. . Back Translation: Not because he talked too much: characters = 30 Subtitle: Not that he talked much: characters = 23 The subtitler opted for the shorter (detractio) ?that? to the longer ?because.? KHAPHELA (p 13): UMiss Motene wangicel? ukuthi ngingatsheli muntu (Zulu). Least of all you. 75 Back Translation: Miss Motene (asked) begged me not to tell anyone: characters = 40. Subtitle: Miss Motene begged me not to tell anyone: characters = 40 .? The subtitler had a choice between asked/ requested and begged but decided to use the latter possibly because ?begged? adds some urgency or seriousness to the situation as opposed to the more general ?asked? or ?requested.? TAU (p 13): Ke ne ke nagana gore nka go tshepa (Tswana). Back Translation: I thought I could trust you: characters = 27 Subtitle: I thought I could trust you: characters = 27 KHAPHELA: Ungangethemba Mnumzane (Zulu). But I was in a difficult position. Back Translation: You can trust me Sir: characters = 20 Subtitle: You can, Sir?: characters = 12 The subtitler omitted the phrase ?trust me? because it was already referred to by Tau and also by Khaphela in vernacular (see above). So the subtitler can afford to leave it out without any semantic loss. The use of Mnumzana also shows the servant-boss relationship between the two. The subtitler saw it fit to translate it with its cultural equivalent, ?sir,? probably because she believed the majority of people would not understand what Mnumzana means. ZOLEKA (p 28): Angikhumbuli ukuthi kukhona ebesithe sizokwenza Namhlanje (Zulu). Back Translation: I don?t remember that there?s something we said we were going to do today: characters = 73 Subtitle: I didn?t know we had plans: characters = 26 76 The subtitler substituted ?remember? for ?know? and the long noun phrase ?there?s something we said we were going to do? for the brief ?had plans.? The temporal adverb ?today? is omitted possibly because it would be understood from the context itself. ?I didn?t know we had plans? can be understood to mean today. MANDLA (p 28): Ngike ngafona ngifuna ukukutshela ukuthi ngiyeza (Zulu), but you weren?t at Monik?s or at the safe house? Back Translation: I phoned wanting to tell you that I was coming characters= 46 Subtitle: I phoned to say I?d be coming over: characters = 34 ?Wanting to tell you? has been substituted by the shorter ?to say.? The subtitler also abbreviated ?I would? to ?I?d,? but added ?over? perhaps in line with English language convention. MANDLA (p 48): There?s a change of plan. Ngike ngakutshela ukuthi umuhle kanjani namhlanje (Zulu)? Back Translation: Have I told you that you?re very beautiful today? characters = 48 Subtitle: Have I told you how beautiful you look tonight? : characters = 45 The subtitler substituted (substitutio) the SL ?today? for ?tonight? possibly because the encounter took place at night and therefore meaning is to be understood from the context (in this case, the image, since in audio-visual translation, meaning is polysemiotic). Namhlanje ebusuku would be the equivalent of ?tonight?. The English language can afford to specify the time of day without making the syntax longer. ZOLEKA: Yes. Usungitshele kaningi (Zulu). 77 Back Translation: You?ve told me many times: characters = 25 Subtitle: About ten times: characters = 15 The subtitler opted for the shorter syntactic structure by omitting ?you have told me? which had already been referred to. However, she substituted ?many? of the SL for ?ten?. Perhaps, that is to be interpreted to mean ?many? in the target language culture. 2003 GENERATIONS 13 EPISODE 3 ZOLEKA (p 4): Nami ngi yazi (Zulu), but we can?t help ourselves. We can?t stop thinking...Today especially?I?ve been remembering all day. Back Translation: I too know: characters = 10 Subtitle: I know?: characters = 6 It is not clear why the subtitler did not include ?too,? especially that there are not spatial constraints. TAU: (p 14): Ke feditse (Tswana). And no, you?re not disturbing me. You couldn?t if you tried. Back Translation: I?ve finished?: characters = 13 Subtitle: It?s done: characters = 9 The subtitler substituted the longer ?finished? for the shorter lexis ?done.? KARABO ((p 14): You?ve been preoccupied the past couple of days. A ke sengwe se nka go thusang ka sone (Tswana)? Back Translation: Is there anything I can help you with? : characters = 37 Subtitle: Anything I can help with? : characters = 24 The subtitler omitted ?is there? because it is not of any semantic value. The subtitle is easily comprehended even as it is. 78 TAU (p 16): Easier said than done. Go bua nnete, ke nyemile moko (Tswana). Back Translation: To tell the truth?: characters = 17 Subtitle: Frankly, I don?t have the energy: characters = 7 ?To tell the truth? and ?frankly? are examples of padding, whose presence is mostly functional, in order to maintain speech flow. The subtitler has substituted the longer phrase with the shorter lexis ?frankly,? although she very well could have left it out. ZANELE (p 20): ?I feel terrible about that and coming out with this now is only making it worse, kodwa akulungile ukusola umuntu (Zulu) for something that? Back Translation: But it?s not right to blame someone?: characters = 35 Subtitle: ?but it?s wrong to blame someone?: characters = 31 The subtitler substituted a negative expression for a positive one, which is supposed to make a subtitle shorter. VUYO (p 32): Of course I?m going so you?ll have nothing to worry about. Ndizakube ndikhona (Xhosa) to show you all the slick, macho moves. Copy me and you can?t go wrong. Back Translation: I?ll be there: characters = 13 Subtitle: I?ll be right there: characters = 19 The subtitler added ?right?, (adiectio) which is absent from the SL ? perhaps as a way of emphasis. ZANELE (p 41): Ufuna ukuthini (Zulu)? Back Translation: You want to say what? : characters = 20 Subtitle: What are you getting at? : characters = 23 79 The substituted ST ?want to say? is more general than the TT idiom ?getting at? which has negative connotations as well and is more to the point. There has also been a transformation in the syntax of the TT because Zulu and English do not belong to the same language family. ZANELE (p 41): kusho?kush? ukuthi awuzwanga kahle (Zulu). I was just speculating on the cause of the fire. Back Translation: That means?that means you misunderstood characters = 41 Subtitle: You must have misunderstood: characters = 27 Theory says that repetition should be eliminated from the subtitle. The subtitler substitutes it with ?must,? which makes the assertion more certain. 2003 GENERATIONS 13 EPISODE 5 TAU (p 14): More like an expensive risk. Ke ya itse gore o rata go ntsha mo sepitleng se ka gonna partner ya me (Tswana). But it isn?t necessary. Say the word and I?ll find another investor. Back Translation: I know you?re trying to take me out of this difficult situation by becoming my partner: characters = 86 Subtitle: You?re trying to help me by becoming my partner: characters = 47 By omitting ?I know,? possibly because it is just padding, the translator just wants to transmit the core of the message which is just ?You?re trying to help me by becoming my partner.? In addition, the subtitler has substituted the longer phrase ?trying to take me out of this difficult situation? with the shorter ?help.? KARABO (p 15): Not for me thanks. Ke na le tiro e ntsi gompieno (Tswana). Let?s meet at our new bar after work so we can discuss some decorating details and date for the opening. 80 Back Translation: I?ve a lot of work today: characters = 24 Subtitle: I have lots to do: characters = 16 The noun phrase ?of work? has been substituted for a verb ?to do? in the TL. In addition, ?today? is omitted in the subtitle. TAU (p 17): I?ve stalled long enough. Karabo o tshwanela ke go itse gore Tsego o imile ngwanake (Tswana)?I?ll tell her tonight. Back Translation: Karabo is supposed to know that Tsego is pregnant with my child: characters = 63 Subtitle: Karabo deserves to know Tsego?s pregnant: characters = 40 By omitting ?with my child?, perhaps the subtitler just wants to deliver the essence ? that Tsego is pregnant. However, the fact that Tau is the father of the child is important and, perhaps, should have been included in the subtitle. ZOLEKA (p 20): Bengithi ngizokufonela (Zulu). Back Translation: I wanted to phone you: characters = 21 Subtitle: I?ve been meaning to call you: characters = 29 Due to spatial constraints, a subtitler usually opts for a shorter lexis. That is why she opted for ?call? here. In addition, ?I?ve been meaning? in the subtitle gives the impression that it is a thought that the speaker has had in her mind for some time, whereas, reading the SL gives the impression that the thought just occurred once. 2003 GENERATIONS 13 EPISODE 6 TAU (p 15): At last I have you all to myself. Go sengwe se ke batlang go se bua le wena (Tswana). 81 Back Translation: There?s something I need to talk to you about. characters = 45 Subtitle: There?s something we need to discuss: characters = 36 ?I? and ?you? have been substituted for ?we? and the longer ?need to talk about? by the shorter lexis ?discuss.? KHAPHELA ( p 2): ?I didn?t realise Jack had such good taste. TAU (p 2): Ga a na yone (Tswana). He flew one of the country?s top interior designers up from Cape Town. He had everything custom-made. From the counter to the copper? Black Translation: He doesn?t have it (referring to Jack?s good taste mentioned earlier): characters = 18 Subtitle: He doesn?t: characters = 10 Jack?s good taste was mentioned earlier. So even though omitted in the subtitle, it is implicit in ?he doesn?t?. KARABO (p 16): ?There I was going on about how much I trust you and what a good team we make. Vivian couldn?t believe how understanding I am. Ke ne ke sa itse gore o bua ka eng (Tswana). Back Translation: I didn?t know what she was talking about: characters = 40 Subtitle: I had no clue what she was talking about: characters = 40 The shift here is in the substitution of ?didn?t know? for ?had no clue?. TAU (p 16): ke ne ke batla go go bolella (Tswana) dammit. Back Translation: I wanted to tell you dammit: characters = 27 Subtitle: I wanted to tell you?: characters = 20 The swear word ?dammit? touches on a taboo area and is therefore omitted. 82 TAU (p 20): Fela seo ga se reye gore ke ya mo rata (Tswana). I love you dammit. I don?t want anything to do with Tsego Motene. Back Translation: But that doesn?t mean I love her?: characters = 32 Subtitle: That doesn?t mean I have feelings for her: characters = 41 There is a shift here as well because the TT?s ?I have feelings for her? is vague and not as precise as the ?I love her? of the ST. In addition, the subtitle is even longer than the source text. JULIA (p 29): I?m prepared to write articles on any subject you want ? restaurants, fashion shows, noma yini oyishoyo (Zulu). Back Translation: Whatever you say: characters = 16 Subtitle: ?you name it: characters = 11 The subtitle is shorter. JULIA (p 30): Lesilima (Zulu). This idiot! : characters = 10 Subtitle: Idiot: characters = 5 The subtitler made no effort to tone down the insult, which could be regarded as a taboo word, by possibly even finding a similar word in the same semantic field but less offending. ?nor did he omit the word, possibly because it was the only word uttered and the target audience would wonder why no subtitle appeared on the screen when, clearly, something was uttered. ZINZI (p 31): Julia?kwakuhle ukuhlangana nomakhelwane wami lapha (Zulu). Back Translation: How nice it is to meet my neighbour here! : characters = 40 Subtitle: fancy meeting my neighbour here: characters = 31 83 Again the subtitler has opted for the shorter word ?fancy.? SIPHIWE (p 42): I eh?bengiya ekhaya (Zulu)?thought I?d pop in to see if Brad?s in the mood for a quick beer or something. Back Translation: I eh?I was on my way home: characters = 25 Subtitle: I was on my way home: characters = 20 The hesitation has been omitted in the subtitle in accordance with subtitling practices. 2005 GENERATIONS 15 EPISODE 76 KHETIWE: (p 2): Forgive me bhuti? Bengikhathazekile nangawe undiza ngobusuku (Zulu). Back Translation:...brother. I was worried about you flying at night: characters = 47 Subtitle: I?m worried about a night-flight: characters = 32 Bhuti, (meaning brother) which in Zulu culture is used as a form of address to a person you respect, is omitted from the subtitle and the noun phrase ?night-flight? has replaced the ST?s ?you flying at night?. NTOMBI (p 13): Lumka would never purposely hurt you. Ukuthanda kakhulu kabi (Zulu). Back Translation: She loves you a lot: characters = 19 Subtitle: She loves you: characters = 13 Even though there is no danger of the subtitler overshooting the maximum number of characters suggested in subtitling theory, the subtitler chose to omit ?a lot?. She probably thought it was not going to make any difference to the message. CLEO (p 16): Yazi ngilokhu ngicabanga ngoBusi (Zulu)?dead?under all 84 that water? Back Translation: you know, I keep thinking about Busi?: characters = 36 Subtitle: I keep thinking about Busi: characters = 26 The padding ?you know? has been omitted. QUEEN (p 19): This is a celebrity wedding. O tshwanetse go apara sutu (Tswana)? Back Translation: You?re supposed to wear a suit: characters = 30 Subtitle: Wear a suit: characters = 11 The imperative shorter TT ?wear a suit? has replaced the longer source text. NTOMBI (p 20): Akuson iskhathi lesi (Zulu) Queen. Any messages? Back Translation: This is not the time, Queen: characters = 27 Subtitle: Not now?: characters = 7 The shorter temporal adverb ?now? has substituted the longer ?not the time?. Queen has also been omitted. KARABO: (p 27): Hello Tau? Ba mmone. Ba bone Busi?Wa tshela (Tswana). Back Translation: They?ve found, they found Busi?She?s alive characters = 41 Subtitle: They?ve found Busi. Alive: characters = 25 Again repetition is omitted from the subtitle as is the pronoun ?she?. KARABO: (p 28): Ba ne ba itemogela mathata ka di radio tsa bona (Tswana)?They only got into Cape Town harbour early this morning. 85 Back Translation: They had radio problems: characters = 23 Subtitle: Radio problems: characters = 14 ?radio problems? is to be interpreted as ?they had radio problems?. NGAMLA (p 31): Uyizwephi ke leyo (Zulu). Back Translation: You heard it where that? : characters = 23 Subtitle: Who told you? : characters = 12 The subtitle is a straightforward and right to the point question and follows English language syntax, which is different from the vernacular one. LUMKA (p 32): Namhlanje ekuseni?usaphila (Zulu). Back Translation: Today in the morning?she?s still alive: characters = 38 Subtitle: This morning. She?s alive: characters = 25 There is a shift with the omission of ?still? in the subtitle. LUMKA (p 33): Dad I?ve gotta go?She?s back. Ngizokhuluma nawe ( Zulu). later?I will. Back Translation: I?ll talk to you later?: characters = 22 Subtitle: We?ll talk?: characters = 10 Possibly the ST?s ?later? is implicit in the ?will? of the subtitle, since ?will? is indicative of the future. The subtitler also replaced ?I? and ?you? with the shorter ?we?. DUDU (p 35): Ush ukuthi unginika (Zulu) the green light? 86 Back Translation: You mean to tell me you?re giving me?: characters = 36 Subtitle: So I?ve got?: characters = 10 ?You mean to tell me? is padding and can safely be omitted without affecting the meaning of the utterance much. It has been replaced by the shorter ?so? in the subtitle. The function of both is functional, designed to maintain speech flow. JACK (p 38): Lithin igama lale fishing trawler ekutholile konje (Zulu)? Back Translation: What is the name of the fishing trawler that found you? characters = 54 Subtitle: What?s the fishing trawler called? : characters = 33 Again here, the shorter ?called? is preferred to the longer ?what is the name?? BUSI (p 40): Bangithatha bangibeka embhedeni we kaptein njengengane?ngaze ngalala (Zulu)?Someone sat there till I fell asleep. Back Translation: They took me and put me on the bed of the captain: characters = 49 Subtitle: I got the captain?s bunk: characters = 24 The superordinate term ?bed? has been substituted for the more specific ?bunk?. Abbreviation is also achieved by the use of the apostrophe: ?the captain?s bunk? instead of ?the bed of the captain?. NTOMBI (p 42): Ngifusa ukuth?u-Cleo ngab?ulana (Zulu). She makes handling the media look so easy. Back Translation: I?m trying to find out if Cleo could be here: characters = 44 Subtitle: Where?s Cleo: characters = 12 87 A straightforward question has replaced the ST?s round about question. QUEEN (p 42): O ka nne wa tima cell ya gago (Tswana). I can?t exactly unplug the phones. Back Translation: Can you switch off that cell of yours: characters = 37 Subtitle: Switch off your cell: characters = 19 Tswana syntax is different from the English one. NGAMLA (p 50): Bekungaba ysiduphunga kuphela ebesingshiya ama- fingerprints aso (Zulu). Back Translation: It?s only an idiot who can leave his fingerprints: characters= 49 Subtitle: Only an idiot would leave fingerprints: characters = 38 The subtitler has made a decision to include ?idiot? in the subtitle, even though it could be regarded as a taboo word and judged as offending by some people. NGAMLA (p 51): Noma ngab?u-right ? akukho lutho ongalwenza (Zulu). Back Translation: Even if you?re right, there is nothing you can do: characters = 49 Subtitle: even if you?re right, your hands are tied: characters = 41 An idiomatic expression has replaced the source text. LINAH (p 54): Hayi bo! Ke ne ke nahana hore keya ho etsa di badge Fela (Sotho). Back Translation: Goodness gracious! I thought I was going to make badges only: characters = 60 88 Subtitle: I thought I?d only make badges: characters = 30 The sigh expressing surprise is omitted in the subtitle. A sigh can be regarded as empty of semantic value and, therefore, omittable. JANUARY 2006 GENERATIONS 15 EPISODE 142 KHETIWE (p 13): Eyi! Intsha yanamhlanje?izifunela ama-party nje kwaphela (Zulu). Back Translation: Hey, the youth of today ? they want to party only: characters = 47 Subtitle: These young people ? they just want to party: characters = 42 The demonstrative adjective ?these? is an apt substitute for ?of today? since it can also be understood to mean ?of today?. ?Hey? is absent from the subtitle, as it should. JACK: (p 58): I did love you Busi?Bekungewon? amanga lawo (Zulu). Back Translation: They were not lies those: characters = 24 Subtitle: That wasn?t a lie: characters = 17 There is a transposition of SL elements in the subtitle due to the nature of the target language syntax. In addition, Zulu convention dictates that ?lies? be in the plural, as opposed to English which can be in the singular. 2005 GENERATIONS 15 EPISODE 140 QUEEN (p 13): I know it?s not a Gerard Sekoto masterpiece?fela pretenda okare wa e rata (Tswana). Back Translation:?but pretend as if you like it: characters = 29 Subtitle:?but you can at least pretend to like it: characters = 39 89 There is an addition of ?at least? in the subtitle. QUEEN (p 14): Sengwe se se tshwanang le mmereko (Tswana). Back Translation: Something that looks like work: characters = 30 Subtitle: Something that looks a lot like work: characters = 36 The addition of ?a lot? in the subtitle could be for emphasis. KHETHIWE (p 26): The poor girl was hurt but no, um usis Karabo uthi ungay? e London (Zulu), give your daughter space?ubhuti listens. Back Translation: if sis Karabo says ?don?t go to London?? brother listens: characters = 53 Subtitle: ?if Karabo says ?don?t go to London??Tau listens characters = 45 Ubuthi (brother) and usis (sister) are not brother and sister in the literal sense but are a form of address to someone you respect. The subtitler substituted them in preference to actually naming the characters being addressed. KHAPHELA (p 26): Ukhipel? ukudinwa kwakho emqamelelweni futhi (Zulu). Back Translation: You?re taking your frustrations out on a pillow again: characters = 53 Subtitle: You?re venting your frustrations on the cushions again: characters = 54 The SL idiom ?taking out your frustrations? has been replaced by ?venting? which evokes powerful emotions. In addition, the more general term ?pillow? of the SL has been substituted for the more specific ?cushion? which is a kind of pillow. KHETHIWE (p 27): Uthini ngeyethu imibono. Ubhuti wayebambe mina ngenkathi ekhathzeke ngo-Angela, hayi sis Karabo (Zulu). 90 Back Translation: What about our opinion? Bhuti Tau was holding me at the time he was worried about Angela, not Karabo: characters = 100 , Subtitle: What about our opinion. Tau hugged me when he was upset about Angela, not Karabo: characters = 80 Again here, the subtitler chooses the more specific ?hugged? to the general word ?hold? and the stronger and emotional ?upset? as opposed to just plain ?worried?. DUDU (p 29): I?ve been way too morbid since brad left. Ngiding? ukuphuma ngiyo gruva? inkinga ukuthi ? kuphi (Zulu)? Back Translation: I need to get out and go and groove?the problem is where? characters = 56 Subtitle: I need to get out there and groove. Problem is ? where? characters = 52 The subtitler has added ?there? which is not in the source language. NTOMBI (p 37): Ngiyamkhumbula (Zulu), but hey?life goes on. Back Translation: I think about him: characters = 17 Subtitle: I really miss him: characters = 17 ?Really?, which is absent from the SL has probably been added for emphasis and, again, ?miss? is more specific than ?think about?. KARABO: (Tswana p 35): Foo gone wa bua. Back Translation: There, you?re talking: (Can be interpreted as ?you have a point?.): characters = 21 Subtitle: You may have a point: characters = 20 91 The translator has had to make some adaptation because ?There you?re talking? would not be easily understood by the target audience. LUMKA (p 50): Wow?ungithusile (Zulu). Back Translation: Wow?you startled me: characters = 19 Subtitle: you startled me: characters = 15 ?Wow? is omitted from the subtitle as it rightly should. KARABO (p 16): Karabo?intshwarele (Tswana)? Back Translation: Karabo, forgive me: characters = 17 Subtitle: ?I?m sorry: characters = 9 ?Sorry? is shorter than ?forgive.? TAU (p 55): O seke wa tsamaya (Tswana) Karabo. You belong here. Back Translation: Don?t go Karabo?: characters = 15 Subtitle: Don?t go: characters =8 Karabo is omitted from the subtitle. That is the correct thing to do. From the analysis above, what I have picked up is that there are some instances of addition (adiectio), reduction/omission (detractio), transformation (transmutatio) and substitution (substitutio). (Repetition is in the form of English loan words). However, the most frequent strategies employed by the translator in descending order are the following: Detractio (omission = 24 times; reduction = 20 times), substitutio (substitution) = 25, adiectio (addition) = 5, transmutatio (transformation) = 5 92 Inconsistency and Deviation from Theoretical Practice However, the subtitler has shown some inconsistency too. For instance, in the following examples: In Generations, a number of taboo words do occur in some of the episodes. The handling of these in the subtitles has not always been consistent. In the few instances where they are used in the source language, they are usually omitted in the subtitle, although at other times they are included. There is also a tendency to include responses such as ?yes? and ?no? in some subtitles and to omit them in others, and also to include names of people in some subtitles, but to omit them in others and so on. 1999 GENERATIONS 9 EPISODE 1 SONNY (p 5): Net blerrie p?re eet oats en appels (Afrikaans): Number of characters = 35 Subtitle: Only bloody horses eat oats and apples: Number of characters = 38 According to theory, swearing touches on a taboo area and should, therefore, be omitted from a subtitle. Hesse-Quack (1969) saw the mass media as a reflector or controller of society. In the above case, however, the subtitler saw it fit not to omit the swear word ?bloody?, although he should have done so. The same is true of ?dammit? and ?idiot,? which should be omitted from the subtitle as they could be regarded as offending to some people. Although a swear word has been included above, it has been omitted in the following example: TAU (Tswana p 16): ke ne ke batla go go bolella dammit. Subtitle: I wanted to tell you?. MANDLA (p 30): Okay. Truth is ? we have to go. We?re meeting with someone. Lomuntu ucabanga ukwenza idonation (Zulu) to the safe house. Back Translation: This person is thinking of giving a donation?characters = 44 Subtitle: The guy wants to make a donation: characters = 32 93 It is not clear why the subtitler chose the colloquial ?guy?, although subtitling theory states that subtitling should be into standard English and in the register of the SL. 2003 GENERATIONS 13 EPISODE 5 JONATHAN (p 39): Angela, uzwile ukuthi umama utheni (Zulu)? Back Translation: Angela, you heard what mum said: characters = 31 Subtitle: ?you heard what mom said: characters = 23 In this case, she omitted ?Angela?, which is the right thing to do and the more accepted practice than the following example. SIPHIWE (p 48): Ngiyajabula ukukubona nawe (Zulu) Queen. Back Translation: I?m happy to see you too Queen: characters = 30 Subtitle: Nice to see you too, Queen: characters = 26 2003 GENERATIONS 13 EPISODE 6 Other discrepancies observed are in the following where ?yes? is subtitled but ?no? is not in one instance but is in another: TAU (p 50): E?ga ke batle gore o tsamaye (Tswana). Back Translation: Yes. I don?t want you to leave: characters = 25 Subtitle: Yes. I don?t want you to leave: characters = 30 KARABO: (p 51): Nya. Ke lapisitswe ke maaka a gago. Fa o batla go simolla botshelo bo bosha le Tsego, tswella (Tswana). I won?t stand in your way. 94 Back Translation: No, I?m tired of your lies. If you want to start a new life with Tsego, go ahead: characters = 79 Subtitle: No. I?m sick of your lies. If you want to start a new life with Tsego, go ahead: characters = 79 The negative response ?no? has not been omitted from the subtitle. BUSI: (p 59): Cha ? ungahambi (Zulu). Back Translation: No, don?t go: characters = 12 Subtitle: ?don?t go: characters = 8 TAU (p 14): Nya o seke wa tshwenyega (Tswana). With you here I can handle anything. Back Translation: No, don?t worry?: characters = 15 Subtitle: Don?t worry about it: characters = 20 . The negative response ?no? has been omitted from the subtitle., in accordance with subtitling theory. TAU (p 17): E gontse jalo (Tswana). Back Translation: Yes, it?s like that: characters = 19 Subtitle: Yes of course: characters = 13 It is not clear why the subtitler included the ?yes? in the subtitles. What applies to ?no? as in the above two examples, should also apply to ?yes,? if the subtitler is to be consistent. There are no subtitles for the following although, clearly, they needed to be there since not everybody in South Africa understands Afrikaans or Zulu. SONNY (Afrikaans p 7): Hey laaitie, stop complaining. We all have to do our bit. 95 The translator must have assumed that laaitie is common enough a word for it not to be subtitled because the word is an example of SA English. There is no subtitle for it. The following also do not have subtitles: SONNY (Afrikaans p 9): Nou maar give the old ballie a hug. SONNY (Afrikaans: p 46): Bliksem my ou. Now it all makes sense. SIPHIWE: (Zulu p 52): Usho u-Julia Motene? (Back translation: You mean Julia?) JANUARY 2006 GENERATIONS 15 EPISODE 142 As discussed earlier on in the present chapter, scripts are originally written in English and portions translated into indigenous languages and then subsequently subtitled. The comparison of the original dialogue to the subtitle is useful in order to see how the subtitles have deviated from the original English script. In the texts studied, in some cases, the subtitle is exactly the same as the English original. But at other times, the subtitler has had to omit certain elements of the English original in accordance with subtitling practices. In addition, sometimes the translation from the English original to the vernacular is literal. For example: 1. English Original: Thing will get better now. Indigenous Language (Zulu): Izinto sezizobangcono manje (Things will get better now). The only difference is that the lexis is longer because of the affixation of prefixes and affixes to the morphemes. English Original 2. I wonder who voted for Jack. Indigenous Language (Zulu): Kazi ubani ovotele u-jack (I wonder who voted for Jack). Subtitle: I wonder who voted for Jack. 3. English Original: It was Busi. Indigenous Language (Zulu): Bekuwu-Busi. (It was Busi). Subtitle: It was Busi. 96 4. English Original: I?m not talking about you. Indigenous Language (Zulu): Angikhulumi ngawe (I?m not talking about you). Subtitle: I?m not talking about you. 5. English Original: How do I look? Indigenous Language (Tswana): ke lebega jang? I look how? ? There is transformation in syntax because English and African languages do not belong to the same language family. Subtitle: How do I look? 6. English Original: You?ll have to do without me today. Indigenous Language (Zulu): Guzofanele usebenze wedwa namhlanje (You?ll have to work alone today). Subtitle: you?ll have to do without me today (idiom). English Original: Who put you up to this? Indigenous Language (Zulu): Ubani okutshele lokho? (Who told you that?) Subtitle: Who put you up to this? There are many other examples like the ones above where the subtitler just transposes the English original as the subtitle. But where necessary, she does also make changes in accordance with subtitling practices, such as below. II 1. English Original: But what can you do Mnumzana? Indigenous Language (Zulu): Ungenzani Mnumzana (What can you do sir?) Subtitle: What can you do? 2. English Original: I know, but it doesn?t change that I betrayed him. Indigenous Language (Zulu): Ngiyazi, kodwa lokho akushintshi ukuthi ngimdayisile (I know, but that doesn?t change that I sold him ? in other words, betrayed him). 97 Subtitle: I know, but I still betrayed him. 3. English Original: Fine. Are we done here? Indigenous Language (Zulu): Kulungile. Sesiqedile (Alright. Have we finished?) Subtitle: Fine. Are we done? 4. English Original: Now, I?m gonna tell you mine. Indigenous Language (Tswana): Janong ke ya go le bolella ya me (Now I?m going to tell you mine). Subtitle: Now, I?m going to tell you mine? The sociolect ?gonna? is subtitled into standard English in accordance with subtitling theory. 5. English Original: Siphiwe thinks you?re doing this to get me away from him. Indigenous Language (Zulu): Usiphiwe ucabang? ukuthi wenza konke lokhu khona sizohlukana (Siphiwe thinks you?re doing all this so that we separate). Subtitle: Siphiwe thinks you?re doing this to separate us. BORROWED WORDS Still, with regard to translation, there are a lot of borrowed words in all the episodes studied, from 1999 right up to January 2006. Most of the words borrowed from English do have equivalents. Below are some of the examples of borrowed English expressions and words: 1999 GENERATIONS 9 EPISODE 3 Dilo tse di nasty (maswe/ mpe/ bosula), ken e ke sa di mean (ika?l?la)(p 8), Ke go warnile (tlhagisa) go re o lekile go ipolaya (p 17), nkemetse downstairs (kwatlase) (p 23), Maxime a mo designele (direle)?(p 24), le mme wa traitor (mooki, motsietsi) (p 25), dilo tse di pleasant (ntle, monate, molemo)(p 44). 98 1999 GENERATIONS 9 EPISODE 4 Tota ke eng se se go fileng idea (mogopol?/kakany?) eo? (p 25), Re ne re bua ka eng le partner (motlhakanedi,/ molekane) yaka? (p 26), Goreng o tlile early (ka maphakela/ ka go sale gal?/ ka go sale mos?) jaana? (p40). 2003 GENERATIONS 13 EPISODE 2 Go organiser (baakanya, rulaganya) di road shows (pontsho tsa mo tseleng) (p 5). 2005 GENERATIONS 15 EPISODE 142 I-unfair (ngalungile/ ngafanele ) ke leyo Sibusiso (p 3)?kanti uyi- option(ukhetho/lukhetho/ ukuziqomela) ephephile (p 3), uyithatheleni leyonto e- restaurant (indlu yokudlela) (p 6), nna ke ya officeng (ntlo ya tir?/ ofisi) (p 11), I- resolution (umqondo/ isinqumo/ ukuzimisela) yethu (p 41), usho uku-runner (phatha) i-Hit Factory (p 43), kahle kule-position (isikhundla/ umsebenzi) osuyitholile (p 51). In addition, an observation made in the analysis is that the lexis of vernacular languages tends to be longer because words in African languages are agglutinated in that they usually have suffixes and affixes attached to them, unlike English which is an isolating language because a morpheme can stand on its own and make sense. Below are examples of agglutinated indigenous words as opposed to the isolated English ones: Ngizokhuluma nawe= I came to talk to you. -ngikumemezile. Awungiswanga= I called out to you. You didn?t hear me. Ngiyabong? ukuth? ungazisile = Thanks for letting me know. -bengicabang? ukumletha lana e-Jozi = I have been thinking about bringing her here in Jozi. Ngibon?uAnne no Jack Mabaso = I saw Jack and Anne The next chapter draws on major conclusions from the findings in this chapter, and the previous two, and makes recommendations. 99 CHAPTER FOUR 4.1. EVALUATION INTRODUCTION This chapter draws major conclusions from the findings in the previous chapters and makes recommendations. It gives concluding remarks on whether there had been an increase in the subtitling on Generations and on actual subtitling practices on the soap opera. The conclusion first reviews the research objectives and the research question. A short description of the research design and a summary of the major findings of the research follow. The section proceeds to draw attention to how this study closes gaps identified in the literature review, and show the way for a potential area for further research. This research was designed to find out whether there has been an increase in the subtitling of Generations and to examine actual subtitling practices on the soap opera. The research problem was investigated against a background of the impact of pieces of legislation on subtitling practices in South Africa. This research sought to assess what progress has been made towards multilingualism and, therefore, subtitling since the two go hand-in-hand. In addition, the aim was also to establish subtitling practices in South Africa using Generations as a case study and to find out whether these practices match international, theoretical and methodological practices, and to evaluate the effectiveness of these practices. Qualitative research methods were used to obtain information from representatives of Generations. In-depth interviews were carried out with them to investigate subtitling practices on the soap opera. Qualitative methods were further employed in all aspects of the data collection process namely sampling, choice of research methods and data analysis. In addition, scripts were examined to find out whether there had been an increase in subtitling on Generations and also find out actual subtitling practices on the soap opera. 100 4.2. CONCLUSIONS: 4.2.1 CONCLUSION: IMPACT OF LEGISLATION As a public broadcaster, the SABC had to face the challenge of ensuring that it played a role in assisting South Africans to realise one of the positive attributes of the national Constitution, which is the recognition of all African languages as official languages that everyone has the right to use for communication. It has played its part because previously marginalized African languages are now being used on soap operas, including on Generations. Since 1996, almost two years after the ANC came to power, the SABC TV channels are no longer divided on racial lines as in the past, but are now more representative of different language groups. The airwaves are no longer politicized with reference to language, as noted earlier in the introduction. Afrikaans has been downgraded and is now almost on a par with formerly marginalized African languages. On SABC 2, Afrikaans now shares airspace with English, Sotho, Tswana, and languages which were not heard on television during the apartheid era ? Venda and Tsonga. Legislation has had a direct effect on subtitling practices in Generations and, therefore, on South Africa because it has led to multilingualism on the screen and, therefore, to an increase in subtitling, which has been the subject of this study. Findings, however, show that although multilingualism has increased, non- English audiences are still at a disadvantage. The English language still enjoys ideological dominance in South Africa because in Generations, the translation process entails translating from English into African languages, and then back into English (not into African languages) for subtitling. In other countries, as Luyken et al. (1991) have stated, subtitling is from English into vernacular languages, which is what the MAG (2002) has been advocating, as this would have the added advantage of exposing other races to African languages. Therefore, the use of 101 subtitles in African languages could contribute significantly to language learning among South African viewers. As the analysis of different episodes has shown, English is still dominant on the soap opera. According to MAG, this domination of English disempowers many people. MAG (2002) argues that giving preferential treatment to any one language (as is currently the practice with regard to English) has unacceptable practical and symbolic implications since it confers a special status on English. It conveys the notion that English-speaking viewers constitute the main target audience of the SABC. In addition, the exclusive use of English subtitles conveys the mistaken assumption that all or most South Africans understand English. English is only the 5th largest home language in the country (MAG 2002). This amounts to grossly unfair preferential treatment for English, which should be treated like other languages. Moreover, it gives the SABC an elitist image. The writer agrees with the MAG? s (2002) position that an equitable distribution of air time to all the official languages can be achieved if the necessary determination is applied. In this respect, the MAG points out the following: That no language should be singled out for preferential treatment and that the use of subtitles into African languages should be considered. The subtitling practices carried out in Generations and, in South Africa in general, are therefore questionable since subtitling is into English and not into African languages. Because of the wide use of codeswitching in the soap opera, subtitling in Generations is not done in the traditional way of taking the whole utterance into consideration. Therefore, when compared to subtitling practices internationally, the process on Generations is different because it does not occur in the same way as it does in traditional instances of subtitling as only bits of an utterance are subtitled instead of taking the whole utterance into consideration. Since this is a drama that is multilingual, there is a lot of codeswitching in it. However, the 102 instances of codeswitching are not real but are artificially created in order to accommodate the multilingualism. Mansour (1993: 100) contends that multilingualism forges unity through policies which suppress linguistic and other differences in order to establish equality between populations of different previous social statuses (Mansour 1993: 109). South Africa fits in well with these assertions because the advent of multilingualism has helped to suppress linguistic differences. However, it has not done away with them altogether because of the continued dominance of English. To reduce the dominance of English, perhaps South Africa could promote two strong indigenous languages, as Tanzania did with Swahili (Mansour 1993: 123), since most black South Africans are multilingual anyway, and use bilingual subtitling, as in the Netherlands and Israel. Bassnett and Harish Trivedi in Post-Colonial Translation: Theory and Practice (1999: 13) see power relationships being played out in the unequal struggle of various local languages against ?the one master-language of our post-colonial world, English?. The observations of these scholars are relevant to the present study because despite legislation to redress the problem, inequality between the vernacular languages and English persists. English continues to be valued over other languages as is evidenced by the fact that the subtitling is done for English-language speakers but not for African-language speakers. However, the dominance of English was exactly what the Constitution was trying to redress when it passed legislation on the subject in 1996. In a way, what is happening in Generations is a form of foreignisation. The foreign other is what is subtitled and not the colonial language. English seems to be more important than the vernacular languages and is forced on the majority of the people. The assumption further seems to be that everyone understands English and that since South Africa is a multilingual country, people who do not understand, for example, Zulu in the soap opera will understand the message through the English subtitles. 103 Although subtitling on TV in South Africa used to be almost non-existent, today it is that many non-English language soap operas have started to display English subtitles and this has attracted viewers from all language groups to television because of the multilingualism in the programmes, including soap operas. Although many people still prefer dubbing to subtitling, because of the effort involved in reading subtitles, they have come to accept subtitles as a reality on the South African screen because of the government policy of multilingualism. In this way, a culture of subtitles has been established on the SABC. Through the implementation of the language policy, the broadcasting station has attempted to satisfy the diverse needs and tastes of South Africa?s diverse language and cultural communities. Subtitling may be, to a large extent, symbolic but the fact that the use of the vernacular is increasingly used in South African television soapies is in the end a positive. 104 4.2.2 CONCLUSION: MATCHING SUBTITLING ON GENERATIONS WITH INTERNATIONAL PRACTICES Because of the nature of this research, which has been to examine actual processes, the approach adopted in this study has been a descriptive one. Descriptive translation studies (DTS) has, therefore, been relevant to this study. As can be seen from the examples from the analysis, it is possible to apply descriptive translation to subtitling. The analysis has been based on Toury?s model developed by Dirk Delabastita (1989). According to him, Toury?s tripartite model of ?competence- norms-performance? can be used to analyse screen translation successfully. His inventory of questions is structured after the tripartite model. Delabastita?s checklist of questions (1989) falls within the framework of Toury?s Descriptive Translation Studies (DTS). This study has been conducted within this framework. The source text and the translation have been compared in terms of Delabastita?s five operations (adiectio (addition); detractio (omission); transmutatio (transformation), substitutio (substitution) and repetitio (repetition), and in terms of the checklist. Comparing the target text (the subtitle) to the back translation has revealed shifts. The shifts have then been described and possible reasons given. By studying whole series of texts, both consistencies and variations of behaviour have been discovered from which the norms governing the translation have been deduced, with due respect to audiovisual constraints and target text conventions and culture. Therefore, what has been of interest to this study have been the optional shifts, made as the translator saw fit. Translation theory has been applied with respect to shifts in the search for synchronization in space and time. The analysis of Generations has proven that, at times, there are shifts between the subtitle and the original dialogue, and that the translation and the subtitle are not always the same. For example: English Original: Fine. Are we done here? Indigenous Language (Zulu): Kulungile. Sesiqedile (Back Translation: Alright. Have we finished?) Subtitle: Fine. Are we done? English Original: But what can you do Mnumzana? 105 Indigenous Language (Zulu): Ungenzani Mnumzana (Back Translation: What can you do sir?) Subtitle: What can you do? We have also seen from the analysis that translation between English and South African languages requires the translator to adapt the language because the languages are different in structure. There have also been lexical differences because English is an ?isolating language? whereas African languages such as Zulu are agglutinating languages (Fromkin and Rodman 1978: 337). For example (Zulu): ngikumemezile. Awungiswanga= I called out to you. You didn?t hear me. Where South African practices tend to differ from international practices is in the directness of translation. In Generations, translation is done from English to vernacular languages and then back to English for the subtitles. The vernacular languages, therefore, act as intermediate languages. With regard to translation policy, the subtitling of Generations should be at 65 per cent this year (2006), but the analysis clearly shows that that level has not yet been attained (that is in the period covered ? 1999 to January 2006). In addition, in South Africa, it appears that translation policy is an all English policy because of the fact that translation is from the vernacular into English and not the other way round. The question one might ask then is whom the policy is benefiting since people who do not understand English are in the majority and, therefore, can only understand half of what is said? The policy would appear to benefit only English speakers or those people who understand English. As earlier stated in the introduction, a survey by the Pan South African Language Board, PanSALB, found that only 40 per cent of South Africans in general and 22 per cent of African language users are functionally proficient in English (the Multilingualism Action Group (MAG 2002). Although the survey was conducted four years ago, those figures cannot have changed much. With respect to exposure time for a subtitle, we have noted that despite the recommended 6 seconds for a two-line subtitle, the problem of sufficient exposure time is still 106 unresolved (Luyken et al.1991: 186). Different guidelines exist in different countries, such as a minimum of 4 seconds per subtitle in the United Kingdom, 6 seconds in the Netherlands and up to 8 seconds in some Scandinavian countries. In Generations, there is no official exposure time for subtitles. According to the subtitler, Maggi van Aswegen (2006), the subtitles are flashed on the screen just long enough for them to be read. However, the problem with the South African case, which makes comparison with other countries difficult, is that the subtitling (Generations) is done differently. Because of codeswitching, only a bit of the dialogue is translated, instead of taking the whole utterance into consideration. Therefore, on Generations, subtitling does not quite match international methodological practices, nor does it fully do so with respect to theory, because the subtitler has at times been inconsistent. For example, as the analysis has shown, the subtitler has included the swear word ?bloody? in a subtitle but omitted ?dammit? in another; included a ?no? in a subtitle, but excluded it in another, etc. Continuing with subtitling theory, Luyken et al. (1991: 190) in their contribution state that in the subtitling of a film or of a television programme, the ?message? is expressed by the audiovisual whole. That is, film is a polysemiotic medium that transfers meaning through several channels. That is why even in Generations, the subtitler has sometimes also relied on the image to get the message across. Baker?s (1992) input ? implicature ? has also been relevant for this study because the audience has at times been left to infer from what else has been happening on the screen instead of only relying on subtitles for meaning. In conclusion, while the SABC was initially not fully responsive to the projected aims of parity of esteem, the regulations are assisting in the promotion of multilingualism and advancement of subtitling in South Africa. 107 4.2.3. CONCLUSION: ANALYSES In this chapter, the study has also looked at the languages spoken on Generations and their progression. To do this, manuscripts from 1999, 2003, 2005 and one 2006 were examined. The second part of the analysis dealt with codeswitching and, lastly, with an analysis of the subtitling itself. The figures quoted from the examination of scripts ranging from 1999 to January 2006 prove that subtitling has increased. Upon examination of the 1999 series, the range of subtitling for the six episodes studied is from as little as 10 to 26 per cent. 2003 shows an improvement ? that is, a range of 21 to 33 per cent. Looking at the 2005 series, there has been a big jump from the 33 per cent in 2003 to 43 per cent in 2005. The range for the period 2005 to January 2006 is 43 to 54 per cent. These statistics, therefore, prove beyond doubt that there has been an increase in the subtitling of Generations. Therefore, the hypothesis that multilingualism has led to more subtitling has been proven. However, the increase is not up to the level envisaged. So, despite the directive from the ICASA, multilingualism has not yet reached the desired level. In addition, from the episodes studied, the conclusion reached is that the languages on the public broadcaster are not yet equitably provided for although the SABC is trying hard to implement multilingualism. In addition, story-lines in the soap opera are very different to what used to happen in the apartheid era when white characters and story lines dominated television programmes. There has been a shift in recent times with regard to the black/white ratio on Generations. More African characters have been added to the series, thereby attracting a majority black audience to television. On the other hand, white characters have been withdrawn. The only white character remaining is Anne. This is in contrast to the conspicuous absence of blacks as main characters in the stories during apartheid (Van Zyl and Elion 1989: 194) because the soap operas that were broadcast served to reinforce the dominant ideology (1989: 196). Afrikaans is no longer spoken on the series ? unlike in some of the other soap operas in which there are still a number of white actors and some Afrikaans is still sprinkled here 108 and there for good measure: for example the character of George Zamdela in Isidingo speaks Afrikaans once in a while (although he is black) as does, sometimes, actor Clint Blink in Scandal. Therefore, these days, there is a mix of languages and of cultures on the SABC?s soap operas. An increasing amount of broadcasting air time has been provided for the African languages in accordance with the Constitution. In the very latest episodes, the vernacular languages are slowly closing the gap with English. Notwithstanding this, subtitling is still all English. In this regard, MAG (MAG 2002) has argued that the SABC should not have an ?anchor language.? Despite this, English continues to be that ?anchor language? because of its dominance and also due to the fact that everything is subtitled into English. However, multilingualism is not just lip-service in Generations. The SABC is making a serious effort at it. For example, in very recent episodes, there is now more Sotho spoken (northern Sotho) through the characters of Kenneth Mashaba and, the bead worker, Magda. In addition, there is more incidence of southern Sotho through the characters of Mashaba?s children. For Xhosa, the character of Ajax has been added to the series. Although this study only covers the period up to January 2006, it is clear that the required 65 per cent has not yet been reached. In addition, notwithstanding the steady increase, the major issue still remains and that is that subtitling is still all English. The section concludes by asking whether it is realistic to expect the SABC to subtitle a soap opera into the different indigenous languages, considering that South Africa has 11 official languages. Masenyama (2005) carried out interviews at the SABC on the role of language in the creation of a national identity. There were contrasting views on the subject as some respondents, such as the commissioning editor of SABC 3 (Tim in Masenyama 2005)7, believe that local languages may increase identity awareness, as all groups of people feel 7 The name given in Masenyama?s M.A. dissertation without further details. 109 included in the broader South African framework and feel that they are respected as equal citizens of the country. Other commissioning editors believed that the use of various languages by the SABC is a hindrance to the creation of a unified South African national identity. One commissioner, Mary, argued that because of its diversity, only a single language, like English, can unite the country. Teer-Tomaselli (2001:118) justifies the dominance of English on SABC programmes by arguing that English is the core language understood as a second language of choice by most South Africans. However, given South Africa?s past history, perhaps it is understandable that some people feel strongly about the issue of promoting indigenous languages. CODESWITCHING: We have seen that codeswitching is part and parcel of conversations on the soap opera Generations. Although codeswitching on Generations could be a way of accommodating multilingualism, another way of looking at it is that switching is used as a form of identity. Sebba and Wootton (1998) have suggested that by switching from English to the vernacular, a person might be signalling solidarity or co- identity with his interlocutor. In addition, according to Edwards (1994: 2), there is nearly always shifting among peers but other situations may call for the use of English exclusively. He adds that language changes made are non-random and that a switch signifies something. However, this does not appear to be the case in Generations because switches are done randomly. Senior translator Corlett Matseoane chooses the scenes and the dialogues that are to be translated. The segment identified for translation is randomly selected and is dependent on whether it is easy to translate. Of course, it has to involve a scene in which the indigenous languages are spoken: that is, a scene in which a black person is talking to another black person, or two Afrikaans-speaking people are talking to each other. Codeswitching in the soap opera Generations appears to be arbitrary ? there seem to be no particular pattern to it. The switching is equally from the indigenous languages into 110 English just as it is from English into the indigenous languages. There are also instances where there is no codeswitching at all, where conversations are entirely in one of the indigenous languages, but these are very rare. The other observation made is that in the days when Afrikaans was spoken on the series, there was not much codeswitching as the entire text would be in Afrikaans, especially where it involved the main Afrikaans-speaking characters in those days, the couple Stoffel and Baba. However, there was and, there still is codeswitching where it involves African languages in that part of the text is in vernacular which has to be translated into a subtitle, while the rest of the text is in English. In instances where there is a white person ? whether Afrikaans speaking or not ? speaking to a black person, the dialogue is conducted in English. Generally, there is codeswitching when a black person is speaking to another black person. This codeswitching, which leads to the subtitling, is not always of people conversing in the same language, possibly as a way of promoting familiarity with each others? cultures in accordance with the provisions of the Constitution. In the 1999 series, one does not see much of that, but in the later series, especially the 2005 and 2006 ones, Zulu-speaking characters converse freely with Tswana-speaking ones. The best examples are the servant-boss pairs: Khaphela/Tau, KhetiweTau; the couple Tau/Karabo; friends Busi/Karabo and so on. However, English continues its dominance as even in cases of codeswitching, the other language is always English. And as earlier stated, there are also whole stretches of text which are exclusively in English. In other soap operas such as Egoli, Muvhango, and 7 de Laan, (with the exception of, for example Isidingo and Scandal with mostly English dialogues), the dialogues are in the other languages (whether vernacular or Afrikaans, or a mixture of vernacular and Afrikaans), but with English subtitles. However, with Generations, more English is spoken than any other language, with switches in between to the vernacular languages, which are then subtitled back into English still. 111 What is beyond doubt, though, is that subtitling has increased as a result of more codeswitching. Subtitling Practices: With regard to the analysis of scripts to examine actual subtitling practices, the researcher found that space constraints do not seem to be much of a concern for the subtitler because she does not have to subtitle the whole dialogue as is normal subtitling practice, as. In addition, the subtitling is not very difficult probably because the bits that are chosen for translation by the translator (subsequently to be subtitled) are the parts that are not too challenging to translate. In the texts studied, in many cases, the subtitle is exactly the same as the English original. But at times, the subtitler has had to omit certain elements of the English original in accordance with subtitling practices.That is, the English original text or the vernacular languages themselves are condensed into English subtitles, to make them easier for the audience to read. In addition, sometimes the translation from the English original to the vernacular is literal. As can be seen from the analysis, what is translated and subtitled are not token words but real sensible texts and serious dialogues. However, the subtitling in Generations has at times been inconsistent. As discussed before, as a general rule, there seem to be parts where the subtitles appear to follow international theoretical practices: for example, the examples given in the analysis (pp 74-92). Then, there are instances where the subtitling has deviated from international practice by, for example, including swear words, including instances of inconsistency on the part of the subtitler (for example, pp 92-96). Having looked at the subtitling practices, this is what I have found: generally, in terms of subtitling, it is done well, but because of the different kind of subtitling in that not the entire programme is subtitled, there is more latitude to subtitle in a particular way. In addition, what we have here is translation being used to increase multilingualism in an artificial way. 112 There is still a dominance of English: the script is written in English and English is selected for translation (subtitle). In conclusion, in terms of answering the major research question posed earlier, the findings indicate that there is a strong relationship between multilingualism and the progression of subtitling in South Africa and also that subtitling practices on Generations differ from traditional subtitling practices. The findings indicate that the SABC has shown its willingness to make the station a vehicle for the promotion of indigenous languages. But English still enjoys ideological domination. Closing Gaps: A gap in subtitling in South Africa was identified and established as the lack of research on establishing the impact of legislation on subtitling practices or studying actual subtitling practices, and evaluating the effectiveness of these practices. Although there has been an increase in subtitling, no proper monitoring had been done on its progression. To my knowledge, no research had been conducted on how these regulations are being carried through, inclusive of the works mentioned above. A review of studies with the same focus as this study carried out by H.C. Kruger, Gwendolyn Speeth, N. de Klerk and S.S. Daswa, identified a gap in subtitling research. None of their studies have researched on the impact of legislation on subtitling practices in South Africa and on actual subtitling practices on a soap opera in South Africa. Therefore, there was need to carry out a research on these issues. This research has added a piece to the puzzle by researching the progression of subtitling on South African television through the examination of scripts ranging from 1999 to January 2006, and examined subtitling practices through interviews with representatives of Generations. Though this research is limited to the study of one soap opera, it adds another significant dimension to research in subtitling. Therefore, it has gone a long way in closing the gap identified earlier on in the literature review. Concluding Remarks: Even though the study adds a new dimension to research, it is also limited. Therefore, in order to complement the current research, there is a need to 113 research on the feasibility of bilingual subtitling on SABC using two strong indigenous languages as suggested earlier on. This will go a long way to diminishing the dominance of English on South African television. 4.4. RECOMMENDATIONS There should be more subtitling into African languages on a rotation basis. There is Muvhango, in which the predominant language is Venda (with English subtitles) and, 7 de Laan, where Afrikaans is mostly spoken (with English subtitles). Why not make Generations into something where Tswana or Zulu speakers can have access to this programme if they do not understand English? Consideration should be given to subtitling all of the English in the programme into the vernacular languages. The SABC should produce a programme that is really multilingual in the true sense of the word. Having a programme where just certain portions are translated into the vernacular and subtitled back into English is not a truly multilingual exercise. True multilingualism means that all languages should be treated equally and South Africa is still a long way from achieving that goal. END 114 APPENDIX A MID- TURN SWITCHES By mid-turn switch is meant that the switch is framed on both sides by either a vernacular language or by English. For example in the first two examples, English is framed on both sides by a vernacular language. However, in the third example, the vernacular is framed on both sides by English. GENERATIONS 15 EPISODE 142 JANUARY 2006 BUSI (p 21): Ngiyamthanda (Zulu: I love him) Karabo. And at the worst moment possible, ngamphendukela (I turned on him). SIPHIWE (p 24): Yini (Zulu: What?). You obviously have an opinion about this. Asiwuzwe-ke (Let?s hear it). JACKIE (p 41): Speaking of office romance?Ke utlwile ba re (Tswana: I heard) Bali?s hooked up with the Hit Factory?s new receptionist! LUMKA (p 42): Not quite. Kade ngicabanga kanti futhi (Zulu: I?ve been thinking and) I want to run something by you? TAU (p 50): I organised champagne, but no glasses. Fela seo se ka se re thibele (Tswana: That shouldn?t stop us) from celebrating properly. BUSI (p 58): Someone saw you with her, Jack. La ehotel. Nina nobabili kudala lento niyihlanganisa, ningenz?isilima-nizam? ukuthath? inkampani yami. (Zulu: Here, at the hotel. You two planned this. You fooled me, tried to take away my company). I couldn?t let it happen again. I couldn?t let Anne win again. 2005 GENERATIONS 15 EPISODE 76 THANDISO (p 53): The clients can see different samples pele ba kgetha (Sotho: before choosing) a badge. 2003 GENERATIONS 13 EPISODE 5 KARABO (p 15): Not for me thanks. Ke na le tiro e ntsi gompieno (Tswana: I have lots to do today). Let?s meet at our new bar after work so we can discuss some decorating details and date for the opening. 115 2003 GENERATIONS 13 EPISODE 3 VUYO (p 32): Of course I?m going so you?ll have nothing to worry about. Ndizakube ndikhona (Xhosa: I?ll be right there) to show you all the slick, macho moves. Copy me and you can?t go wrong. 2003 GENERATIONS 13 EPISODE 2 KENSANI (p 3): The house is kind of quiet without him. Eseng ka gore o ne a bua thata (Tswana: Not that he talked much). Except when it came to things he feels passionate about?like art. 1999 GENERATIONS 9 EPISODE 1 KARABO (p 7): Go tshwanetse ga bo go na le explanation ya go re why a sa leka go escapa (Tswana: There must be some explanation why she didn?t try to escape). May be the fire was caused by an electrical fault or something. Gongwe o simolotse a robetse (Maybe it started while she was sleeping). MANDLA (p 7): If you want to ease your conscience by assuming she was asleep, go ahead. But we both know it was suicide. O ipolaile (Tswana: She killed herself) ? and I could have stopped her. TERENCE (p 13): Nka se kgone tonight. Ehm? ke emetse call e tswang overseas ? kwa Alaska (Tswana: I can?t tonight?I?m expecting an overseas call ? from Alaska). They?re twelve hours behind us, so ga ba kwa tirong gone jaanong (so they?re not at work right now). KARABO (p 45): She did what she did to punish us ? to punish you ? fela e ne le choice ya gagwe (Tswana: but it was her own choice)! You are not responsible. 1999 GENERATIONS EPISODE 5 QUEEN (p 11): What? Must be a slow news-day if they print a photo of someone in such a ridiculous outfit! Ke ile go bolaya that Maxime (Tswana: I?m going to kill that Maxime)! He put her in this to get back at me for stealing you from him! SONNY (p 46): Baba, sorry to say, maar ek vrek van die honger (Afrikaans: but I?m dying of hunger). And I have to get back to the bar. 1999 GENERATIONS EPISODE 4 SONNY (p 5): And now? Lounging around the house in your nagkabaai (Afrikaans: pyjamas) ? don?t they miss you at that job of yours? 116 APPENDIX B FINAL-TURN SWICHES By final-turn switch is meant that the switch is at the last part of an utterance. GENERATIONS EPISODE 142 (January 2006) KARABO (p 20): No Busi?Thank you. You did a very brave thing. Mme go ne go siame go dira jalo (Tswana: And it was the right thing). LUMKA) p 24): Never mind. Umqond?omuhle lowo (Zulu: It?s a good idea). LUMKA (p 25): Sibuyela kulokho futhi (Zulu: Are we back on that again)? Can?t we just focus on what we?re doing? JACKIE (p 38): A o gopotse (Tswana: Do you miss) Bradley that much? DUDU (p 38): Ngabe ngikunikil? i-invitation yami (Zulu: I would?ve given you my invitation). It was a blow out. NGAMLA (p 42): Ungaz? ungithenge. Umsebenzi owakho (Zulu: You don?t have to bribe me. The job?s yours). All you have to do is say ?yes.? NGAMLA (p 44): Cha. Uzosebenz? e-Hit Factory (Zulu: You take over The Hit Factory) full time or, there?s not deal. NGAMLA (p 45): Angizange ngibekhon? ekukhuleni kwakho, kodwa manje ngizama ngazo zonk? izindlela. Ngizam? ukukwenz? izinto ngendlel? ehlukile (Zulu: I wasn?t always there for you, but I?m trying to make up for it by doing things differently). Now you throw it back in my face! NGAMLA (p 45): Sekwanele (Zulu: Enough), Lumka. I told you where I stand. The choice is yours. SIPHIWE (p 48): Cha Lumka, ufun? ukuthi uphatheke kabi yikhon? ungeke umbuze lutho (Zulu: he wants you to feel bad so you won?t question him). He?s so damn clever. Laying a guilt trip on you! KARABO (p 53): Ke go bolelletse gore (Tswana: I told you) the gentlemen?s club would fall flat, just like the poetry evening did. You know what your trouble is, Sibusiso? You?re all talk and no action?Go blow your hot air elsewhere. BUSI (p 56): Stop it, Jack. Please just?stop. Ukhumbul? i-Mauritius, kanye nayoyonke eny? into (Zulu: You remember Mauritius and everything else). 117 JACK (p 57): Ubani okutshele lokho? Kungab? uKarabo? Nom? umfowenu? (Zulu: Who put you up to this? Was it Karabo? Your brother?) It has to be because the Busi I know would never? JACK (p 58): I did love you Busi. In my own way I did. Bekungewon?amanga lawo (Zulu: That wasn?t a lie). 2005 GENERATIONS 15 EPISODE 76 NTOMBI (p 13): Lumka would never purposely hurt you. Ukuthanda kakhulu kabi (Zulu: She loves you). CLEO (p 16): Yazi ngilokhu ngicabanga ngoBusi (Zulu: I keep thinking about Busi) ?dead?under all that water? 2003 GENERATIONS 13 EPISODE 2 KHAPHELA (p 13): UMiss Motene wangicel? ukuthi ngingatsheli muntu (Zulu: Miss Motene begged me not to tell anyone). Least of all you. MANDLA (p 48): There?s a change of plan. Ngike ngakutshela ukuthi umuhle kanjani namhlanje (Zulu: Have I told you how beautiful you look tonight)? 1999 GENERATIONS 9 EPISODE 1 BABA (p 4): Verbaas my nie (Afrikaans: Doesn?t surprise me). They left in such a hurry she barely had time to pack. MANDLA (p 6): Ke nna mong wa ntlo (Tswana: I?m the owner of the house). Or what?s left of it ? which isn?t much. MANDLA: (p 7): Mapodisa a re ga gona evidence ya go re o lekile go escapa (Tswana: The police say there?s no evidence that she even tried to escape). She just willingly let the flames consume her. MANDLA (p 42): In the fireproof safe ? where she knew it wouldn?t be damaged. Ke tshaba le go e bula (Tswana: I?m almost scared to open it). KARABO (p 44): Sophie o ne a lwala, baby (Tswana: Sophie was sick). She let one unfortunate incident rule her entire existence. 118 1999 GENERATIONS EPISODE 6 SONNY (p 13): And they shot him ? three times?Doesh, doesh, doesh. In cold blood. Hy?s dood, my pal ? nog ?n blerrie crime statistic (Afrikaans: He?s dead ? another bloody crime statistic). SONNY (p 14): I know. But at least they won?t murder any more innocent people. An eye for an eye, and a tooth for a tooth ? dis wat ek nou soek (Afrikaans:That?s what I want now). 1999 GENERATIONS 9 EPISODE 5 STOFFEL (p 5): Verdomp! (Afrikaans: Damn). Those waterblommetjies are from our Y2K stockpile! STOFFEL (p 6): Geen wonder dit was op special nie (Afrikaans: No wonder it was on special). Daylight robbery. I?m taking it back after I?ve told you your fortune, Mister! ARCHIE (p 25): Listen ? I?ve noticed Mandla looking a bit stressed lately. A ke go re o bereka thata, kgotsa go na le sengwe se se mo tshwenyang (Tswana: Is he working too hard, or is something else bothering him)? KARABO (p 27): Ke go boleletse (Tswana: I told you) ? you?re only drawn to Zinzi?s wildness, her impulsiveness, her sense of fun? BABA (p 30): Oh my goodness! I?ll never get used to modern technology. Stoffel insists I have it for safety reasons. Waar is die ding tog nou? (Afrikaans: Where is that thing now?) 1999 GENERATIONS EPISODE 4 TERENCE (p 26): Okay ? so tell me. Re ne re bua ka eng le partner yaka (Tswana: What did my partner and I talk about)? TERENCE (p 26): I ?can?t now. Perfume ya gago?e mpherosa sebete (Tswana: Your perfume is making me feel nauseous). STOFFEL (p 37): Don?t tell me that half-gebakte (half-baked) Rasta?Ek dog jy?t ges? hy mag nie sy voete daar sit nie (Afrikaans: I thought you said he wasn?t allowed to set foot in there)! 1999 GENERATIONS EPISODE 3 QUEEN (p 8): Yes. So ke utlwile dilo tse di nasty tse o neng o di bolelela partner ya gago ka nna (Tswana: So I heard the nasty things you told your partner about me). 119 KARABO (p 15): Oh o setse o tsogile (Tswana: You?re up already). I thought I?d spoil you with breakfast in?What are you doing? MANDLA (p 17): How could you judge? O ne o sa mo itse (Tswana: You hardly knew her). KARABO (p 20): Why didn?t you talk to me about it? Ke ne ke akanya go re o siame (Tswana: I thought you were okay). KARABO (p 20): O mmone kae (Tswana: Where did you see him)? You didn?t go to Glen?s, did you? MATLAKALA: (p 22): Oh. Well ? hide me, quickly! Ba ntshetse morago (Tswana: They?re after me)! MATLAKALA (p 24): Of course not! Ga o nke o seba (Tswana: You never gossip). QUEEN (p 26): Ke go boleletse Ma (Tswana: I told you Ma) ? no man can resist me. 1999 GENERATIONS EPISODE 2 KARABO (p 3): She decided to take her own life, Mandla! O dirile tse tsotlhe ka boene (Tswana: She did it all by herself). QUEEN (p 9): E ne e se nna (Tswana: It wasn?t me)! I swear. QUEEN (p 10): Wait. O tlile go dirang ka se (Tswana: What are you going to do about this)? ARCHIE (p 24): Yazi inkinga yethu yini (Zulu: You know what our problem is)? We?re too much alike. That?s why it would never work between us. KGOMOTSO (p 35): Oor my dooie liggaam gaan my vrou tronk toe (Afrikaans: Over my dead body is my wife going to jail)! I?ll sell the house to pay that fine if I have to. 120 APPENDIX C REVIEWS OF GENERATIONS Editor Babalwa Shota Watches As New ?Bourgeoisie? Wash Dirty Linen On Screen WEEKEND POST 15 OCTOBER 2005 The longevity of the show ? and the fact that it is South Africa?s most watched daily drama ? are proof that millions of South Africans cannot get enough of the scheming, the backstabbing, the make-ups and break-ups, the business deals, and the tragedies. Vundla is known for not mincing words. He is bold and not afraid to pat himself on the back for bringing South Africans the first black soap opera in the country. In a news report, the executive producer and creative force was quoted as saying: ?These people we see (in Generations) are the vanguard of the first generation of the African bourgeoisie in the making. It?s a process, it?s happening. Society is transforming and the bourgeoisie is becoming more and more entrenched. They are the torchbearers of a transforming society, and our characters reflect this. When it launched on February 3, 1994, Generations was a huge success. For the first time viewers were exposed to black people who lived in mansions, wore Hugo Boss suits, made and broke international deals and, best of all, the women had sexy slinky hair to go with their sexy slinky outfits. Black viewers were hooked. SUNDAY TIMES 28 March 2004. A Mirror for the Rainbow Nation By Gillian Anstey A Decade of Democracy South African soap operas have not only become a growing industry; they pride themselves on reflecting our society. Mfundi Vundla, creator of SABC1's Generations is bolder. "I came along and I showed Africans who are in complete control of their lives... I am probably a beneficiary of this transforming society. In a way, I am like my characters." 121 Prior to Generations ' first broadcast on February 3 1994, black people on television were always caught in a quagmire, victims of circumstances beyond their control, depicted only in working-class dramas in which, Vundla says, "life was just a maze of complications, with no solutions.? Now it's a different picture. Generations characters Karabo and Tau are more than simply the South African versions of the characters in the US soap The Bold and the Beautiful. [The series] mirrors South Africa's evolving democracy. We deal with the realities of South Africa. [It] reflects the issues of the country. Soapies claim to be sticking to their brief of reflecting the changing South Africa, yet each is different ? which perhaps is an acknowledgment of the many faces of the country. Today, South African soap opera is an industry, a source of immense entertainment for millions and a breeding ground for academics to delve into for its depiction of our society. The Stuff Soaps are Made of (MAIL AND GUARDIAN 31 May 2002) A strangely evolved curate?s egg that has spewed both trashy froth and classy entertainment, writes Thebe Mabanga Generations set out to create a fantasy world of upper black middle-class South Africa in the advertising industry. The setting in that industry came about when its creator, Mfundi Vundla, looked to his brother Peter Vundla?s business, the advertising agency Herdbuoys. The creation has an obvious element of fantasy and escapism about it. Strangely, this element seems to be trashed for being unreal and misrepresenting the values of the black middle class. But Generations is no more representative than, say, the health education drama Soul City represents black communities. Not all black people live in shacks, are unemployed and have family members afflicted with AIDS. The truth is that both shows, and television in general, work because they depict some kind of extreme. Soul City seeks to depict a low-income end of black society. Generations seeks to portray the emerging rich and black. 122 AM Archive - Monday, 14 August 2000. AM Archives Africa correspondent Sally Sara writes: In South Africa, a new generation of television soap operas is breaking down the racial barriers. The new series are a far cry from the dramas of the apartheid era when white characters and story lines dominated the small screen. This is the new sound of South African soap opera. Now, more than ever before, story lines featuring black actors are appearing on the small screen. The mainly white soap operas of the old South Africa left little room for black actors. But now the opportunities are growing. The increasing numbers of black actors are attracting strong audiences. It's becoming a popular mix of voices and cultures. ***** Since its inception, Generations has aired on SABC 1. Generations is watched by millions of viewers not only in South Africa but in neighbouring countries as well. It remains one of South Africa's most watched soap operas. Generations is set in the hectic cut-throat world of advertising. www.joburg.org.za/2003/apr/apr30_soap ***** We are seeing more and more subtitles being used on television in South Africa as TV stations try to attract viewers from all language groups. In addition, films like ?Yesterday? and ?Tsotsi? have given film subtitling new exposure in South Africa (Seminar on localization to celebrate SATI?s 50th anniversary in support of the Year of African Languages) www.translators.org.za/pretoria.pdf - ***** Americanised drama that has become the staple of most S. African soaps like "Egoli", "Generations" and the worst of them all, "Backstage. These shows aim to scandalise every development in the plot and fall into typical and conventional soap opera narrative modes. IMDb (1998) www.imdb.com/title/tt0373543/usercomments 123 APPENDIX D GENERATIONS INTERVIEWS CORLETT MATSEOANE (TRANSLATOR, SUPERVISOR OF TRANSLATORS, LANGUAGE ADVISOR, ARTISTIC ADVISOR). He has been a translator for about 16 years. Q: TO WHAT EXTENT IS MULTILINGUALISM IMPLEMENTED? A: To a great extent because the SABC demands that we should put a lot of vernac in our programmes. Q: WHICH LANGUAGES ARE SPOKEN IN GENERATIONS? A: Tswana, South Sotho, North Sotho and Zulu. Q: WHY THESE PARTICULAR LANGUAGES? A: Well, it all depends on the characters that we get. If a character is fluent in South Sotho, then we have to give him Sesotho. If he speaks Tswana, it will be Tswana. If he speaks Zulu, it will be Zulu. Q: UNDER WHAT CIRCUMSTANCES ARE THESE LANGUAGES INTRODUCED? A: As I said, it depends on the characters. Q: WHICH LANGUAGES ARE CHOSEN FOR SUBTITLING? A: Ya, they are all chosen for subtitling, irrespective of whether it is Sotho, Zulu, Tswana ? they are all chosen. In fact, everything that is in vernacular should be subtitled. Q: TO WHAT EXTENT DOES SUBTITLING OCCUR? WHAT percentage OF AN EPISODE IS SUBTITLED? Episode is Subtitled? 124 A: At the moment (January 2006), SABC demands that we should be at something like 65per cent for all the languages. Now, I would say that we are at 60 per cent. We are at 60 and at the end of March next year, we should be at 75. Q: IS THE SUBTITLING UNIFORM OR VARIES FOR EACH EPISODE? A: It is the same ? it?s uniform ? 60 per cent. Well, it might differ because in some instances, you might find that there is a lot of English. So, obviously the vernac will not be as much as in the other episode where there is a lot of vernac. So, it all depends, but mostly, it should be at 60 to75per cent. Q: WHY IS IT THAT SOMETIMES YOU USE A LOT OF VERNACULAR AND SOMETIMES A LOT OF ENGLISH? A: Because it depends on the scenes in an episode. If perhaps there are white characters in the episode, then definitely, there will be a lot of English in that episode. If there is less English in some episodes, there will be more vernac and less English. So, the percentage of subtitling will definitely be high in episodes that have less English . Q: YOU DO NOT SEEM TO SUBTITLE ALL DIALOGUE. IT SEEMS AS IF PARTIAL SUBTITLING OCCURS, WITH SOME BITS SUBTITLED AND OTHERS NOT. WHO MAKES THE DECISION AS TO WHAT GETS SUBTITLED? AND WHY DO YOU CHOOSE THOSE PARTICULAR BITS? A: I chose the scenes and the dialogue that is to be translated and that has to be subtitled. And that depends still, depends on the characters. If a scene involves a white and a black, the scene will be entirely in English because a white person is not expected to understand a black language. So, we won?t subtitle the scene because it will be using English. But if a scene involves a black and a black, then of course that scene should be subtitled. Q: ON AVERAGE, HOW MANY SUBTITLES ARE THERE PER EPISODE? A: As I have mentioned, plus or minus 60 per cent. It is quite difficult to say we have got so much per episode. But in terms of percentage, I would say about 60 per cent. Q: O.K, let me put it this way. Say, for example we have got two characters who speak vernac. Irrespective of whether it is a South Sotho speaking to a North Sotho, or a North Sotho speaking to a Zulu, we are entitled, and forced in fact, to 125 translate almost the whole scene in vernac. So, the whole scene will be translated in vernac where it is possible. And it means, the number of subtitles will be a lot in that scene because we are supposed to translate the whole scene in vernac, but now, to count how many subtitles would be in that scene is quite difficult. Q: NOW, WHEN YOU SUBTITLE, HOW LONG DO THE SUBTITLES REMAIN ON THE SCREEN? I AM ASKING THIS BECAUSE OF THE READING SPEEDS OF PEOPLE. IS THE TIME ENOUGH FOR THEM TO BE ABLE TO READ BEFORE THE SUBTITLE DISAPPEARS FROM THE SCREEN? A: Yes. We do take that into consideration. Hence you find that we translate the intended meaning. We don?t translate what is exactly on the script. We translate the meaning. So the meaning shortens what is on the script. So, in other words, the subtitle will be shorter than what the characters are saying?the time it remains on screen is enough to be read. Q: ARE THOSE WHO DO THE TRANSLATION TRAINED TRANSLATORS OR ARE THEY JUST PEOPLE WHO HAPPEN TO KNOW THE RESPECTIVE LANGUAGES? A: Well, I was trained by the SABC personnel some time back ? in 1984, to be a translator, to be a language advisor, to be artistic advisor. So I am specializing in that. So I am a trained translator. Q: AND THE OTHERS? A: the others are also trained. Q: HOW MANY TRANSLATORS WORK ON GENERATIONS? A: We are 3. One is in the studio, taking care of the language, seeing to it that the actors are saying the correct thing that is written on the script and to assist the directors, who are white. So they do not understand black languages. So what happens is that this one in the studio is the one that takes care of the language to see to it that they [actors] speak the correct thing that is written on the script and to tell the directors whether it is right or wrong. And two of us remain in the office doing the translation. 126 Q: ARE YOU FULLTIME EMPLOYEES? A: Yes we are all fulltime employees. Q: AND IS THE SUBTITLING DONE HERE? A: Yes, the subtitling is done here at Ditton Road, 24 [Auckland Park]. Q: IS THE SUBTITLING DONE BY THE TRANSLATORS WHO WORK FOR SABC OR BY AN OUTSIDE COMPANAY? A: No, not by an outside company. It is people that work for GENERATIONS. They are employed fulltime by GENERATIONS. Q: WOULD YOU SAY THAT SUBTITLING HAS INCREASED ON GENERATIONS, LET?S SAY IN THE LAST FIVE YEARS? A: Yes, it has increased because of the mandate that the SABC got from ICASA that all programmes are supposed to broadcast in all the languages. Since the inception of GENERATIONS some time back, it was mainly in English and little bits of vernac. Now, since the mandate was adopted by ICASA and the government, we are forced now to increase the number of black languages in our drama. Q: SO THAT IS WHY THE NEED TO SUBTITLE? A: That?s right ? to subtitle. Yes. Q: I WOULD LIKE TO STUDY EPISODES FROM THE BEGINNING OF THE SERIES UP TO NOW. A: Unfortunately, we don?t have episodes that we started with sometime back. We have only mid and recent episodes. The next interviewee is the video editor. He does the technical aspect of the subtitling. Q: WHAT IS YOUR NAME AND WHAT ARE YOU HERE? A: My name is Babalo Mpoyiya. Here, I work as the video editor for GENERATIONS. That means I put the video together, for the viewers to see at home. Specifically about subtitling, each episode gets recorded at the SABC and they send the tape to me and I log, together with the director?s script because the director sometimes does three takes per scene and then prefers one or two that he would want me to print into my system, that he wants to use on the episode. So what I do, I log everything into my system. And then after I have typed these into my system, I digitize them. That means, I take the tape and stick it inside this big machine. This is a Beta 127 Cem SP machine. This is a Beta tape?it is recorded not according to scenes. Scene one might be the last to be recorded, or might be in the middle. It is not necessarily, they don?t necessarily record chronologically. So after I digitize, that is recording, that is some kind of capturing from the tape, into the computer. I move my mouse like this, in order to see it. After digitizing, I start editing. These are the edits. Q: WHAT DO YOU EDIT OUT? A: Mostly it?s when the actor has done a mistake or, the count down, 5, 4,3,2..ACTION and then I cut out that bit. And then I start from the actors when they start talking or when they start doing the action. I cut out. Sometimes, there is a camera mistake?the camera, the boom, the mic [if they can be seen]. Those are the things I cut out. And then I make the cuts a bit smooth and then look at the rhythm of the episode. If may be the episode is too slow, my cuts will be a bit slow, depending on the dialogue of each scene. If the dialogue is of two people sitting opposite each other, in a restaurant, just talking, it is slower than the scene where people are walking up and down, talking. That one is a bit pacy. Q: SINCE THEY DO NOT COME IN CHRONOLOGICAL ORDER HOW DO YOU KNOW WHICH TAKE IS FIRST, SECOND AND SO ON? A: I read the director?s script and then the director will give me the cuts?to cut from shot one to shot two. The director indicates ? this is shot two. Short two ends here and then short three, the director will indicate. Then after editing the whole episode, I put in, I type the subtitles. Q: HOW DO YOU DO THAT? A: Subtitles, it?s just simple. I have got some sort of a template of the subtitles. This is for the subtitles. What I do is I overwrite this subtitle over, for example, this girl and then I type it in. Q: WHO GIVES YOU THE SUBTITLES OR DO YOU MAKE THEM YOURSELF? A: From the script department. They send me these subtitled scripts, which is different from the director?s script. Then I set it in. Q: SO YOU ALREADY HAVE THE SOFTWARE FOR SUBTITLING? A: No, we create the subtitles. Like this template we create here and then I save it in 128 some sort of a file? I have got a lot ? the end roller, everything. The front title sequence is in the same. We create it ourselves and then we save it in the computer. After the subtitles, I put in the end roller and roll the names. Then you know that when the names are rolling up, it is the end of the episode. I put that in and then also there, I have to type in the names of the supporting cast. Q: DON?T YOU ALREADY HAVE THE NAMES OF THE ACTORS? A: No, I get a list from the SABC because I don?t know who has been cast to act on that episode that day, because the supporting cast is per episode. It is unlike people that you see like the KHAPHELAs, TAUs who are [regulars]. So they give me this list. So I just type them in and then the directors, because directors change. One director may be directs five episodes per week or six and then next week it?s another director. The following week it?s another director. And then the writers, there is one writer per episode. I type in?[the scriptwriter?s name and] then put in the frontal sequence. Basically, editing is what you see at home. I do what you see at home. All that you see at home has been put together here ? putting the episode together in order for you to see a good quality [film]. MAGGI VAN ASWEGEN - SUBTITLER I do the subtitles. I do recaps of the different scenes ? the stuff that you see before the show begins. And also the stuff that you see in the magazines ? that, this is going to happen in the next scene or in the next episode. I put it on the website every week. I also do the website questions. Q: HOW DO YOU SUBTITLE ? IS IT FROM THE TRANSLATED SCRIPT OR FROM THE ORIGINAL? A: What happens is, the whole script is written in English and then what happens is that we have people that translate here. So, they translate either into Sotho, Zulu or Tswana - sometimes Xhosa. Then, what we do, after they have done the translation, then we take the English text that we have got, then, we just try and shorten it so that there is not too much to read on the screen. So we take the whole sentence that was written and then we try and condense it to make it shorter so that it?s not too much to read on the screen. That is my job. Q: SO, HOW LONG DOES THE SUBTITLE REMAIN ON THE SCREEN? A: Oh, seconds ? few seconds. Basically, it?s just as long as that person speaks that the subtitle is on. Obviously, they have to move on as the dialogue moves on, because you don?t want to see a subtitle that the one man is speaking and then the next person is talking. So 129 it is only while that person is talking that it is flashed on the screen and then it goes on to the next person?s dialogue. Q: WHAT IS THE RELATIONSHIP BETWEEN GENERATIONS AND THE SABC? Q: We are just a production company that works for SABC. SABC has a stake in the company itself. We just produce Generations for SABC. We are just a separate production company. We are not part of SABC. Q: WHAT IS THE WHOLE PROCESS THAT IS INVOLVED IN SUBTITLING ? A: There is a huge process. First, we start off with the head writer. He thinks up the story. He makes the story line. Then, this storyline is put into treatments ? it?s the basic story plus a little bit of elaboration. They just give the writers an idea. Then, it is given to the script writers and they obviously write the whole script ? scene for scene. Then, it goes to the editors. The editors check that it is correct, that the whole story is flowing etc. From the editors, it goes to the translators. Then the translators do their bit- it is translated into the African languages. From there, they do the timing ? to see if we have enough time for each episode on the screen because you have to leave enough time for advertisements in between the different scenes. And then it comes back to the editors and then they check the timing to see that it is okay. From there, it comes to me (subtitler) for the recaps from the previous episode and then also to do the subtitles. But then, the subtitles can come in right at the end?the subtitles come in when it is finished being filmed. Once it is filmed then the subtitles come on. That is basically the last thing that happens before it is aired. Q: YOU SAY YOU ARE THE ONE WHO SHORTENS THE SENTENCES. HOW DO YOU DO THAT? A: They give you a long sentence. I?m trying to think of a simple example. For example, ?Mary had to go to the shop yesterday?. That is the sentence that you get the writers have written. Now, they?ll put in the Zulu at the top, whatever the sentence might be for the Zulu, and then what we have to do to put it on the screen we try and look what words we don?t?[need]. So what we can say is: ?Mary went to the shop?. The fact that she went yesterday is not that important and the fact that she has to go is not that important. Mary went to the shop ? end of story. So it?s a much shorter sentence. It is much easier for people to read on the screen. If you put a long sentence like ?Mary had to go to the shop yesterday?, it takes a long time to read and then by that time, the next person is already talking. So you have to keep it very short so that it?s time for the next person?s dialogue to come on the screen. Q: SO YOU YOURSELF HAVE GOT NOTHING TO DO WITH MAKING UP A SUBTITLE? A: No, you don?t make a subtitle. A subtitle in our business is basically just condensing the dialogue that is already in the script. So, you just make it shorter ? you don?t have to think up? You condense the script, because obviously it?s written in English and our subtitles are 130 in English. So we can just condense it. May be if the whole script had been written in Zulu, then you would need somebody to translate it and then think up the script or condense it from Zulu. But now, it is already written in English. So all we do we just condense and make it shorter and easy to read. Q: HAS THE SUBTITLING OF GENERATIONS INCREASED IN THE LAST FIVE YEARS? A: They have tried to increase the use of African languages in the production and I think it used to be 45per cent African languages. They?ve now gone up to 65per cent. So definitely it has increased. So obviously if the program is in an African language and you have people that do not understand you need to have the subtitles. So it has definitely increased. That was a requirement from the SABC, that the African languages are used more. Q: HAS THIS GOT TO DO WITH THE REQUIREMENT OF THE CONSTITUTION? A: It might be because of the constitution ? I don?t know. It might be also that they are just trying to accommodate more the rural Africans that don?t understand English. That they are actually trying to draw them into the focus group as well?make them part of the target? because you have a lot of people that don?t understand English, especially if you go out, way out, there is no concept of English and so then the African languages are more functional because they can understand. Q: WHEN TRANSLATING FROM ENGLISH INTO FOR EXAMPLE ZULU, IS THE WHOLE ENGLISH PART TRANSLATED INTO THE VERNACULAR? A: If this is viable to translate ? If it carries within an African language. If it doesn?t, then they will just leave it in English. If it carries well in an African language ? If there is a nice translation for it, then they?ll use it. Otherwise, they just leave it in English. 131 REFERENCES Anstey, Gillian. (2004). A Mirror for the Rainbow Nation. In Sunday Times 28 March 2004. Baker, Mona. (1992). In Other Words: a Coursebook on Translation. London: Routledge. Basnett, S. and Trivedi, Harish. (1999). Post Colonial Translation: Theory and Practice. London: Routledge. Bastin, George. (1998). In: Mona Baker (ed.) Routledge Encyclopaedia of Translation Studies. London and New York: Routledge. Berman, A. (1984). L?preuve de l?etranger: culture et traduction dans l?Allemagne romantique. Paris: Gallimard. Daswa, S.S. (2005). The Role of Subtitling in Overcoming Communication Barriers in South Africa. Submitted in partial fulfilment of the requirements for the degree of Master of Arts. De Klerk, N. (2002). The Influence of the Sotho Language and Culture on the Results of the Marketing Questionnaire. Submitted in partial fulfilment of the requirements for the degree of Master of Technology. Delabastita, D. (1989 in Babel 35 4).Translation and Mass Communication: Film and TV Translation as Evidence of Cultural Dynamics pp 192-215. Delabastita, D. (1990). Translation and the Mass Media. In S. Bassnett and A. Lefev?re (ed.) Translation, History and Culture. London and New York: Pinter Publishers. De Linde, Z. and Kay, N. (1999). The Semiotics of Subtitling. Manchester: St. Jerome Publishing. Doke, C.M. et al (2005). English/Zulu and Zulu/English Dictionary. Johannesburg: Witwatersrand University Press. d?Ydewalle, G?ry et al (1987). Reading a Message When the Same Message Is Available Auditorily in Another Language: The Case of Subtitling. In Eye Movements: From Physiology to Cognition. Amsterdam: Regan & L?vy-Schoen pp 313-321. d?Ydewalle, G?ry et al (1991).Watching Subtitled Television: Automatic Reading Behaviour. In Communication Research 18:5 October 1991 pp 650-666. Edwards, John. (1994). Multilingualism. London and New York: Routledge. Erasmus, Frans. (2004). The History of South African Broadcasting. 132 Retrieved November 2006 from http:// www.pumamouse.com/SABbrbHistory Fawcett, P. (1996). Translating Films. In G. Harris (ed.) On Translating French Literature and Film. Amsterdam: Editions Radopi B.V. Fromkin, V. and Rodman, R. (1978). An Introduction to Language. 2nd ed. New York: Holt, Rinehart and Winston. Gentzler, E. (1993). Contemporary Translation Theories. London and New York: Routledge. Goris, Olivier. (1993). The Question of French Dubbing: Towards a Frame for Systematic Investigation. In Target 5:2 pp 169-90. Gottlieb, H. (1998a). Subtitling. In M. Baker (ed.) Routledge Encyclopaedia of Translation Studies. London and New York: Routledge. Gottlieb, H. (1998b). Subtitling ? A New University Discipline. In Dollerup, C. and Loddegaard, A. (ed.) Teaching Translation and Interpreting. Amsterdam: John Benjamins Publishing Company. Gumperz, John. (1982). Discourse Strategies. London and New York: Cambridge University Press. Gutt, E. (1991). Translation and Relevance: Cognition and Context. Oxford. Blackwell. Manchester: St. Jerome. Hajmohammadi, Ali. (2004).The Viewer as the Focus of Subtitling: Towards a Viewer Oriented Approach.Translation Journal, Vol.8, No.4. October 2004. Retrieved 15 June 2006 from http:// accurapid.com/journal/30 subtitling.htm Hatim, Basil and Mason, Ian. (1997).The Translator as Communicator. London: Routledge. Hayman, Graham and Tomaselli, Ruth. (1989). Ideology and Technology in the Growth of South African Broadcasting, 1924-1971. In Ruth Tomaselli, Keyan Tomaselli and Johan Muller (ed.) Currents of Power: State Broadcasting in South Africa. Cape Town: Anthropos pp 23-75. Hayman, Graham and Tomaselli, Ruth. (1989). Broadcasting Technology as an Ideological Terrain: Some Concepts, Assumptions and Problems. In Ruth Tomaselli, Keyan Tomaselli and Johan Muller (ed.) Currents of Power: State Broadcasting in South Africa. Cape Town Anthropos pp 1-19. Heller, Monica. (1988). Codeswitching. Berlin, New York and Amsterdam: Mouton de Gruyter. 133 Herbst, Thomas. (1994). Linguistische Aspekte der Synchronisation von Fernsehserien. T?bingen: Niemeyer. Hervey, S. and Higgins, I. (1992). Thinking Translation: a Course in Translation Method: English to French. London and New York: Routledge. Hesse-Quack, O. Schwachen zu beheben. In Fawcett, P. (1996). Translating Films. In G. Harris (ed.) On Translating French Literature and Film. Amsterdam: Editions Radopi B.V. Holmes, J. (1988). The Name and Nature of Translation Studies. In J. Holmes (ed.) Translated! Papers on Literary Translation ad Translation Studies. Amsterdam: Radopi pp 67-80. House, Juliane. (1977). A Model for Translation: Quality Assessment. T?bingen: Gunter Narr. Ivarsson, Jan. (1992). Subtitling for the Media: a Handbook of an Art. Stockholm: Transedit. Jakobson, Roman. (2000). On Linguistic Aspects of Translation. In Lawrence Venuti (ed.) The Translation Studies Reader. London and New York: Routledge pp 113-118. Karamitroglou, F. (1998). A Proposed Set of Subtitling Standards in Europe. Translation Journal, Vol.2, No.2. Retrieved 15 June 2006 from URL.http://accurapid.com/journal.htm Katan, D. (1999). Translating Cultures ? an Introduction for Translators, Interpreters and Mediators. Manchester: St. Jerome Publishing. Kovacic, I. (1996). Subtitling Strategies: a Flexible Hierarchy of Priorities. In Heiss, C. and Bollettieri Bosinelli, M. (ed.) Traduzione multimediale per il Cinema, la Televisione e la Scena. Bologna: Cooperativa Libraria Universitaria Editrice pp 297-305. Kovacic, I. (1994). Relevance as a Factor in Subtitling Reductions. In In Dollerup, C. and Loddegaard, A. (ed.) Teaching Translation and Interpreting 2. Amsterdam. Philadelphia: John Benjamins Publishing Company. Kruger, H.C. (2000). The Theory and Practice of Subtitling: The Case of Cyrano de Bergerac. Submitted in partial fulfilment of the requirements for the degree of Master of Arts. Kruger, Jan-Louis. (2006). Local is Lekker: a Seminar on Localization. http://www.translators.org.za/pretoria.pdf. Retrieved in October 2006. 134 Lewis, Jane. (2003). Design Issues. In Jane Ritchie and Jane Lewis (ed.) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London. Thousand Oaks. New Delhi: SAGE Publications. Luyken, G. et al (1991). Overcoming Language Barriers in European Television. Manchester: The European Institute for the Media. Mailhac, G.P. (1998). Optimising the Linguistic Transfer in the Case of Commercial Videos. In Y. Gambier (ed.) Translating for the Media. Turku: University of Turku, Centre for Translation and Interpreting. Mabanga, Thebe. (2002). The Stuff Soaps are Made of. In Mail and Guardian 31 May 2002. Mangena, Mosibudi. In Sunday Times, 22 September 2002. Mansour, Gerda. (1993). Multilingualism and Nation Building. Clevedon. Philadelphia and Adelaide: Multilingual Matters. Masenyama, Kurai. (2005). National Identity in Post-Apartheid South Africa: SABC TV?s Contribution. Submitted in partial fulfilment of the requirements for the degree of Master of Arts. Matseoane, Corlett. (2006). Personal interview with Collet Matseoane at his office at Generations studios in Auckland Park in January 2006. Matumo, Z. I. (1993). Setswana/ English and English/Setswana Dictionary. Gaborone: Macmillan Botswana Publishing Company (Pty) Ltd and Botswana Book Centre. Mokhoantlhe, Constance. (2006). Telephone interview with Constance Mokhoantlhe, the Generations Public Relations and Publicity Manager, on12 September 2006. Mpoyiya, Babalo. (2006). Personal interview with Babalo Mpoyiya at his office at Generations studios in Auckland Park in January 2006. Munday, Jeremy. (2001). Introducing Translation Studies: Theories and Applications. London and New York: Routledge. Newmark, P. (1991). About Translation. Clevedon: Multilingual Matters Ltd. Nida, Eugene A. and Taber, C.R. (1969 / 1982).The Theory and Practice of Translation. Leiden: E. J. Brill. Niranjana, Tejasvini. (1992). Siting Translation: History, Post-Structuralism, and the Colonial Context. Berkeley: University of California Press. 135 Nord, Christiane. (1991). Text Analysis in Translation: Theory, Methodology and Didactic Application of a Model for Translation-Oriented Text Analysis. Amsterdam: Radopi. Oder, Norman. (1991). Coloring TV. In Columbia Journalism Review (CJR) Chronicle. Retrieved 15 January 2007 from http:// archives.cjr.org Reynierse, Cecile. (1991). South African Multi-Language Dictionary and Phrase Book. Cape Town: The Reader?s Digest Association South Africa (Pty) Ltd. Ritchie, Jane. (2003). The Application of Qualitative Methods to Social Research. In Jane Ritchie and Jane Lewis (ed.) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London. Thousand Oaks. New Delhi: SAGE Publications. SABC Today 2003. SBS Style Guide. (2002). Sydney: SBS Sub-titling Department. Schwarz, Barbara. (2002). Translation in a Confined Space: Film Subtitling ? Part 1. Translation Journal, Vol. 6, No. 4. October 2002. Retrieved 15 June 2006 from http:// accurapid.com/journal/ 22subtitles.htm Schwarz, Barbara. (2003). Translation in a Confined Space: Film Subtitling ? Part 2. Translation Journal, Vol. 7, No.1 January 2003. Retrieved 15 June 2006 from URL. http:// accurapid.com/journal/23subtitles.htm Sebba, Mark and Wootton, Tony. (1998). We, They and Identity. Sequential versus Identity-Related Explanation in Code-Switching. In Peter Auer (ed.) Code-Switching in Conversation: Language, Interaction and Identity. London and New York: Routledge. Shuttleworth, Mark. and Cowrie, Moira. (1997). Dictionary of Translation Studies. Manchester: Routledge. Shota, Babalwa . (2005). Viewers Vote for the Soap that Shows the Way. In Weekend Post 15 October 2005. Smith, S. (1998). The Language of Subtitling. In Y. Gambier (ed.) Translating for the Media. Turku: University of Turku, Centre for Translation and Interpreting. Snape, Dawn and Spencer, Liz. (2003). The Foundations of Qualitative Research. In Jane Ritchie and Jane Lewis (ed.) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London. Thousand Oaks. New Delhi: SAGE Publications. 136 Speeth, Gwendolyn. (1996). Translation for the Silver Screen: a Special Case in the Subtitles for the Film ?Le moine et la sorci?r?/Sorceress. Submitted in partial fulfilment of the requirements for the degree of Master of Arts. Sperber, D. and Wilson, D. (1986). Relevance: Communication and Cognition. Oxford: Blackwell. Szarkowska, Agniesszka. (2005). The Power of Film Translation. Translation Journal, Vol.9, No.2. April 2005.Retrieved 15 June 2006 from URL. http:// accurapid.com/journal/32film.htm Teer-Tomaselli, R. and Tomaselli, K. (1994). Reconstituting Public Service Broadcasting: Media and Democracy during Transition in South Africa. In Communicare 13(2) pp 44?66. Teer-Tomaselli, R. and Tomaselli, K. (1996). Reconstituting Public Service Broadcasting: Media and Democracy during Transition in South Africa. In M.B. Anderson (ed.) Media and Democracy. Oslo: University of Oslo. Teer-Tomaselli, R. (2001). Nation-building, Social Identity Television in a Changing Media Landscape. In Zegeye, A. and Kriger, R. (ed.) Culture in the New South Africa, After Apartheid, Vol. 2, Social Identities in South Africa Series: Cape Town: Kwela Books and South African History Online pp 117-137. Teer-Tomaselli, R.Tomaselli, K. and Muller, J. (1989). Square Vision in Colour: How TV2/3 Negotiates Consent. In Ruth Tomaselli, Keyan Tomaselli and Johan Muller (ed.) Currents of Power: State Broadcasting in South Africa. Cape Town: Anthropos pp 153- 175. Tomaselli, Keyan and Tomaselli, Ruth. (1989). Between Policy and Practice in the SABC, 1970-1981. In Ruth Tomaselli, Keyan Tomaselli and Johan Muller (ed.) Currents of Power: State Broadcasting in South Africa. Cape Town: Anthropos pp 84-143. Toury, G. (1985). A Rationale for Descriptive Translation Studies. In T. Hermans (ed.) The Manipulation of Literature: Studies in Literary Translation. Beckenham: Croom Helm pp 16-41. Toury, G. (1995). Descriptive Translation Studies and Beyond. Amsterdam: John Benjamins Publishing Company. Van Aswegen, Maggi. (2006). Personal interview with Maggi Aswegen at her office at Generations studios in Auckland Park in January 2006. 137 Van Zyl, Mikki and Elion, Leandra. (1989). Trauma by Installment: Springbok Radio Operas. In Ruth Tomaselli, Keyan Tomaselli and Johan Muller (ed.) Currents of Power: State Broadcasting in South Africa. Cape Town: Anthropos pp 177-213. Venuti, L. (1995). The Translator's Invisibility: a History of Translation. London: Routledge. Wehn, Karin. (2001). About Remakes, Dubbing and Morphing: Some Comments on Visual Transformation Processes and their Relevance for Translation Theory. In: Yves Gambier and Henrik Gottlieb (ed.) (Multi) Media Translation. Amsterdam. Philadelphia: John Benjamin pp 65-72. Legislation and Official Documents Consulted Advisory Panel on Language Policy (2000) ? http://www.dac.gov.za. Retrieved on 20 November 2005. Broadcasting Act No.4 of 1999 http://www.info.gov.za/gazette/acts/1999/a4-99.pdf Retrieved on 20 November 2005. Constitution of the Republic of South Africa, Act 108 of 1996 Independent Broadcasting Act No 153 of 1993. http://www.info.gov.za/acts/1993/a153- 93.pdf. Retrieved on 20 November 2005. Independent Communications Authority of South Africa Act No 13 of 2000. http://www.internet.org.za/icasa-act.html. Retrieved on 20 November 2005. Multilingualism Action Group (2002) ? http://www.litnet.co.za/taaldebat. Retrieved on 20 November 2005. Submissions on the SABC?s Application for Licence Amendments (2004) http://www.ifj.org/doc). Retrieved on 20 November 2005. 138