Evaluation in media texts: A cross-cultural linguistic investigation

A quantitative0interpretative approach to the comparative linguistic analysis of media texts is proposed and applied to a contrastive analysis of texts from the English-language China Daily and the UK Times to look for evidence of differences in what Labov calls “evaluation.” These differences are then correlated to differences in the roles played by the media in Britain and China in their respective societies. The aim is to demonstrate that, despite reservations related to the Chinese texts not being written in the journalists’ native language, a direct linguistic comparison of British media texts with Chinese media texts written in English can yield valuable insights into the workings of the Chinese media that supplement nonlinguistic studies. (Media, China Daily, Times, Labov, evaluation.) I N T R O D U C T I O N : T H E C A S E F O R C R O S S C U L T U R A L C O N T R A S T I V E A N A L Y S I S O F B R I T I S H A N D C H I N E S E M E D I A T E X T S Sophisticated linguistic tools such as those used in critical discourse analysis (CDA) have long been available for the precise analysis of media texts in English. In the UK alone, much work has been done on debunking, for example, the idea that news coverage of events is the unbiased reporting of “hard facts” (e.g., Fowler 1991; Glasgow University Media Group 1976, 1980; Hall et al. 1978, 1980). The UK literature is also good on analyzing what Fowler 1991 calls the “social construction of news” through processes of selection (i.e., choices about what has “news value”) and transformation (i.e., choices about how news is presented – often linguistic choices, but also involving elements such as layout, headline size, and position in a newspaper) (e.g., Glasgow University Media Group 1976, 1980; Hall et al. 1978; Philo 1983). Similarly, work has been done on news stereotypes and the way these reinforce themselves (Fowler 1991) and on the social and economic factors involved in news selection (Fowler 1991, Philo 1983, Cohen & Young 1973, Hall et al. 1980). There have also been forays into contrastive functional linguistics, including attempts (e.g., Ghadessy 1988, Carter 1988) to match “situational Language in Society 33, 673–702. Printed in the United States of America DOI: 10.10170S0047404504045026 © 2004 Cambridge University Press 0047-4045004 $12.00 673 factors” with “linguistic features” in written media texts – that is, to look at how the situational constraints (commercial pressures, expectations of target audience) acting upon a medium influence the particular linguistic choices made in presenting the news. These studies have combined to give linguists, social scientists, and media commentators a sophisticated understanding of the complex functional role the media plays in UK society. The Western literature also contains a number of studies of the media in China. Various commentators (e.g., Li 1999, Zhao 1998, Lee 1990) have written of the comparative lack of freedom enjoyed by the Chinese media compared with the media in so-called Western societies, even in the present period, when Chinese journalists perhaps enjoy more freedom than they have for some time. Zhao, a former Chinese journalist himself, concedes that over the past decade or so there have been significant changes in the news media in China. Nevertheless, he insists that the Chinese Communist Party still retains “overt political control” of the news media, one of whose roles is to serve as a government mouthpiece. Zhao writes: It is certainly true that the rise of mass communication, especially television, has brought profound changes to the ideological landscape. The increasing variety and liveliness of cultural entertainment forms, together with a reduced, explicitly propagandist content, has resulted in a proliferation of new symbolic forms. This does not mean, however, that the media are no longer doing ideological work or politically dominating. Indeed, the media’s promotion of consumerism is no less ideological than their promotion of class struggle during the Mao era. (1998:6) On the evidence of the literature, the roles played by the media in the two societies appear different. The mainstream government-controlled media in China – newspapers such as the People’s Daily, Guangming Daily, and Economic Daily, which, according to Zhao (1998:129), are not available to buy on the streets but are subscribed to with public money and circulated in offices, classrooms, and other workplaces – are largely expected to report positive events and0or put a positive spin on other events that are reported (Zhang 1997, Conley & Tripoli 1992). When the “correct” positive line to be adopted is not apparent – for example, at times of uncertainty such as the Tiananmen crisis of 1989, when for a week disarray within the Chinese Communist Party was so complete that Propaganda Department officials did not hold their regular meetings with top Beijing editors – newspapers may be left floundering, unsure of the line to take. In the case mentioned, many ultimately adopted an ultra-cautious line (Conley & Tripoli 1992). By contrast, the British and British-influenced press industries are commercial at heart, relying on newspaper sales to generate advertising revenue. Fowler goes so far as to say: L I LY C H E N 674 Language in Society 33:5 (2004) The main economic purpose of newspapers appears to be to sell advertising space. . . . Consumer advertising is based on the representation of ideal fictional worlds, i.e. sets of beliefs about desirable personal and social behaviour in relation to such products as cars, deodorants, coffee, hair care, washing powders and sweets. The texts of newspapers themselves also offer fictional model worlds, for example the obsessive discussion in the tabloids of television soap operas as if real, of actors, personalities and stars, the escapism of the travel pages in the middle class papers. (1991:121) Where mainstream Chinese media accentuate the positive, British and Britishinfluenced newspapers appear to thrive on conflict and on negative news stories. Thus, the then editor of the British-influenced South China Morning News – based in Hong Kong, not a million miles away from Beijing – is quoted by Knight & Nakano (1999:173) as saying; “News is conflict. News is where there is disagreement. . . . News is where someone puts up something and someone else on the other side says where they have problems with it.” Unfortunately, although much useful work has been done on the role of the media both in Britain and in China, a direct linguistic comparison (at least at the level of CDA) between the media in a Western English-language society such as the UK and the media in China, though likely to yield fascinating insights into role differences between the two country’s media and how these affect language choices made by journalists, has always been difficult because journalists in the two societies work in different languages. I believe, however, that a contrastive critical discourse analysis of the English-language China Daily and Englishlanguage newspapers from a Western society such as the UK may yield interesting insights into how the differing roles of the media in the two societies affect the ways in which journalists in the two countries approach the writing of texts, as evidenced by the linguistic choices they make. Obviously, just as no single UK newspaper can be taken as representative of the UK print media in general, no single Chinese newspaper can be taken as typical of the Chinese print media generally – especially when that newspaper is not even written in Chinese. Nevertheless, I would argue that a substantial portion of the ideological and political constraints operating upon the Chinese-language media generally operate also upon the China Daily, making contrastive analysis interesting and fruitful – a claim that will be discussed further below, along with my reasons for choosing the UK Times to represent the Western print media. The aim of this article, then, is to demonstrate that a linguistic analytic system that focuses on one specific aspect of language – what Labov has called “evaluation” – can be fruitfully applied to a comparative analysis of these particular British and Chinese print media to reveal differences between the patterns of language choices made. This, in turn, can deepen our understanding of the differences in the roles played by the media in the two societies, and it can supplement existing nonlinguistic comparisons of the two countries’ media. E VA L U AT I O N I N M E D I A T E X T S Language in Society 33:5 (2004) 675 E VA L U A T I O N : A T O O L F O R C O N T R A S T I V E L I N G U I S T I C A N A L Y S I S My general field of interest is contrastive functional linguistics. My particular aim in this study is to sketch a method for unpicking one aspect of what Carter (1988:8) describes as “the degrees of neutrality or bias which are inscribed in the choice of words which reporters make,” and to apply it to a contrastive analysis of British and Chinese media texts taken from the Times and the English-language China Daily. The tool I have chosen for doing this is a particular aspect of language which Labov 1972 terms “evaluation.” Evaluation is an aspect of the narrative structure of a text. The term “evaluation,” Labov says, refers to “the means used by the narrator to indicate the point of the narrative, its raison d’être: why it was told and what the narrator was getting at” (1972:207). The effect of evaluative elements is to enrich the narrative. Evaluative devices say to us: “This was terrifying, dangerous, weird, wild, crazy; or amusing, hilarious, wonderful; more generally that it was strange, uncommon or unusual – that is, worth reporting. It was not ordinary, plain, humdrum, everyday or run-of-the-mill” (Labov 1972:209). Labov developed the concept of evaluation while studying the speech patterns of African Americans in south central Harlem, New York. In an attempt to study the way they used their verbal skills to evaluate their own experience, he asked a range of subjects, including pre-adolescents, adolescents, and adults, to record oral narratives in which they talked spontaneously, with the help of a little prompting from the interviewer, about personal experiences or events from their past. When analyzing these narratives, he identified an important element of discourse that he labeled the “evaluation of the narrative,” which broadly had to do with the way in which the speaker embellished his or her narrative to make it more interesting. Labov identified and labeled a number of linguistic devices that narrators employed – consciously or unconsciously – to do this. They fall into four main categories, which Labov labeled intensifiers, comparators, correlatives and explicatives. Intensifiers select one of the events that are organized in a linear series in a narrative discourse and strengthen or intensify it. Intensifier devices include repetition, as well as what Labov calls “quantifiers,” such as all, some, a lot, much, many, few, all of which add emphasis. Comparators provide a way of evaluating events by placing them against the background of other events that might have happened but did not; thus, they enrich the telling of what did indeed happen. Correlatives bring together two events that actually occurred so that they are conjoined in a single independent clause. Explicatives are used to explain why or how an event happens; they are often carried out using conjunctions such as while and though, or connectives such as since or because (Labov 1972). L I LY C H E N 676 Language in Society 33:5 (2004) By analyzing the use of such elements, we can learn much about the cultural context within which a linguistic exchange takes place. They help, in a sense, to fix the immediate linguistic exchange within a much richer, deeper cultural context. They also reveal a good deal about the relationship between speaker or writer and listener or reader, and about their expectations and perceptions of each other. Evaluation was developed in the context of analyzing African American spoken English vernacular in 1970s Harlem. It could be argued that it is a big step from that to the kind of written media discourse being analyzed here. I believe, however, that there are two reasons why the use of evaluation is justified in a study of this nature. First, the difference between the written discourse being analyzed here and the spoken discourse on which Labov was working when he developed the concept of evaluation is not as great as it at first appears. The oral narratives on which Labov worked in the early 1970s were not themselves examples of spontaneous conversation. He did try to record such spontaneous conversations in the form of face-to-face interviews, but he found that the resulting discourse was frequently interrupted and broken. Even worse, he found that because of the formal nature of the interviews, the conversation he was obtaining was over-monitored by the speakers themselves, and therefore atypical of unmonitored African American vernacular. To overcome this, and to obtain large bodies of uninterrupted, unmonitored speech, he resorted to suggesting topics to his subjects, encouraging them to talk about important events from their own past in an uninterrupted way. In this way, he wrote, he was able to produce “narratives of personal experience, in which the speaker becomes deeply involved in rehearsing or even reliving events of his past” (Labov 1972: 207). The point here is that the examples of discourse Labov obtained were atypical of the fragmented, broken nature of ordinary unmonitored speech, containing as they did a strong, uninterrupted narrative structure. In that sense, they are not unlike newspaper texts. Labov himself developed a model to represent the structural elements that make up the kind of narrative he was analyzing, identifying six elements: the abstract, which sets out what a narrative is about; the orientation, which establishes the who, where, when, why, and what; the complicating action (Then what happened?); the evaluation (So what?); the resolution (What was the outcome?); and the coda, which signals a return to the present. I would argue that a typical news report conforms very closely to this model of narrative: indeed, with the possible exception of the coda, it could arguably serve as a good model for journalism schools to use when teaching the writing of a news report. Second, evaluation was developed as a way of understanding vernacular narratives where the audience could not be taken for granted. The whole point about evaluative devices is that they are a way of enriching a narrative, of grabbing and holding attention. In what circumstances might a newspaper need to do this? It seems plausible that a newspaper would need to do this if its very survival E VA L U AT I O N I N M E D I A T E X T S Language in Society 33:5 (2004) 677 depended on attracting and holding a large readership. This is certainly the case with Western newspapers such as the Times, commercial operations that rely for their economic survival on attracting advertising, which in turn is directly related to the size of the newspaper’s readership. It is not the case with the China Daily, which, as will become clear, is not even available to buy on the street and is instead circulated by public subscription. This fundamental difference between the two newspapers is precisely the point of using evaluation in this study. Given that the point of evaluation is to grab attention and make a narrative interesting, we would expect a commercial newspaper like the Times to be rich in evaluation, while a publicly subscribed newspaper like the China Daily, essentially a publicly funded government mouthpiece, would not need to bother with the extra textual work that evaluation involves. The results do indeed seem to bear this out – a sufficient justification in itself for using the technique. For reasons largely of economy of scale, this study will look at only one category of evaluation when comparing the two sets of texts being analyzed: comparators. To conduct an analysis of all four types of evaluators in an essay of this scope would be unrealistic, and an initial analysis revealed that the most interesting differences between the two sets of texts being examined occurred in the use of comparators. Comparators add richness and complexity to a narrative by placing the events described against the background of events that might have happened but in fact did not, or events that have not happened yet. For the purposes of this study, and again for reasons of scope, I will furthermore be concerned with only three main types of comparators, those that Labov calls negative evaluators, future evaluators, and modal evaluators. Negative evaluators – such as the element has no plans in the sentence The headmaster at William Straw’s school says he has no plans to suspend or discipline the teenager – specifically place the events described in the context of what could have been the case, but in fact is not. In this sentence, William Straw’s headmaster could have suspended the teenager, but in fact he has not. Future evaluators – for example, will make in The mainland will make greater effort towards furthering cross-straits ties – tend to be used when reporting on future events or developments, things that will be the case in the future but are not at present. Modal evaluators – like could be taken away in the sentence Inquiries could be taken away from (police) forces and carried out by independent investigators – make possible speculation about events that could or might or should come to be, but that are not the case at present. Negative evaluation, as I hope will become apparent from the examples analyzed below, is a very rich and effective linguistic device for heightening and enriching the drama of a narrative; at times, it even makes it possible to construct a reportable narrative out of very little. Future evaluation is used when talking about what will be. Modal evaluation can be used in a number of ways, including speculation and also exhortation. L I LY C H E N 678 Language in Society 33:5 (2004) This study will examine the use of evaluation in a corpus of 50 texts from the Times and 50 from the China Daily. Of particular interest will be not only the difference in the frequency of use of evaluation by journalists working for the two newspapers, but also the difference in the ways they use evaluation. These differences, I believe, can be interpreted so as to throw interesting light on the different roles played by the media in China and Britain. T H E TIMES A N D T H E CHINA DAILY As already acknowledged, no single newspaper can be held to be fully representational of the print media in either the UK or China. Nevertheless, it is necessary to restrict the analysis to just a single newspaper from each culture in order to keep the study manageable, and also to reduce the problem of variation among different newspapers from the same country and hence to produce a clearer set of data for comparison. The use of a single newspaper from each culture to, in a sense, represent the media of that culture, while admittedly less than ideal, is not without precedent. Kathleen Jamieson and Karlyn Kohrs Campbell argue that it is acceptable to use “elite” newspapers to stand in for others in samples because they shape opinion in ways that others do not (Jamieson & Kohrs Campbell 1992:18–19). The China Daily was chosen because it is the principal English-language newspaper in China. Being written in English, it is targeted mainly at non-Chinese people with an interest in China. Jean Conley and Stephen Tripoli, two American journalists who worked as sub-editors on the China Daily from 1986 to 1987, point out that “it is one of the major voices to the outside world of the People’s Republic of China” (Conley & Tripoli 1992:27). Nevertheless, I would argue that it operates under many of the same cultural, political, and social constraints as other mainstream national Chinese newspapers. It is, in fact, owned by China’s principal Chinese-language daily newspaper, the People’s Daily, and is produced in the same building in Beijing. Zhao 1998 distinguishes between two main categories of newspaper in China: those, mainly local evening newspapers, that are sold on the streets; and those – the major Party organs such as the People’s Daily and specialized newspapers published by government departments – that are rarely sold on the streets but rather are subscribed to with public money and are circulated for consumption in offices, classrooms, and other places of work. The former, Zhao says, are largely under the direct control of the municipal Party propaganda committee, but they at least have a more diversified content than Communist Party organs and are more entertainment-oriented. The latter – to which group the China Daily, effectively an English-language sister paper of the People’s Daily, certainly belongs – are much more directly under the control of the Party. Being so closely connected with the People’s Daily, the China Daily draws on many of the same sources of information as its sister paper and adopts the E VA L U AT I O N I N M E D I A T E X T S Language in Society 33:5 (2004) 679 same line on major news items. It is written in English by Chinese journalists, whose work is then “polished” by foreign journalists working as sub-editors. These “polishers,” say Conley & Tripoli, regularly participate in daily editorial meetings but have little influence over editorial line: “Polishers had virtually no influence on, and often little knowledge of, decisions involving the most sensitive matters” (Conley & Tripoli 1992:27). Conley & Tripoli leave no doubt about the extent of Party control over the content of China Daily: “The Party’s control over the news media is exercised through the Propaganda Department, which answers directly to the Central Committee” (1992: 36). Moreover, “Propaganda officials, according to China Daily staff members, meet with media leaders regularly (usually every two weeks) to discuss news events and to deliver the Party line on how they should be covered.” Given that the China Daily is so closely associated with the People’s Daily – and is so clearly subject to the same processes of Party control as its sister paper – I would argue that, even though it is written in English, its use for the purposes of this contrastive analysis is meaningful and can reasonably be expected to shed some light on the role and workings of China’s national, paid-forby-subscription Party newspapers Similarly, I would not attempt to argue that the Times can be held to be fully representational of the full breadth and diversity of the British print media. The differences in editorial policy, target readership, style, and criteria for selection of items found among the various British print media are well documented in the literature. It is not, however, these differences in which I am interested. For the purposes of this analysis, what British newspapers have in common is more important. They are all commercial operations, which rely on selling as many newspapers as possible to generate the advertising revenue necessary for economic survival; they are independent of direct political (though not commercial) control; they appear to thrive on conflict and negative reporting; and they all generally make similar claims about wanting to inform and entertain readers. In these respects at least, I held the Times to be fairly representative of British newspapers – sufficiently representative, at least, to make a direct linguistic comparison between it and the China Daily interesting. The Times also has the advantage, from the point of view of this study, of being a broadsheet, as is the China Daily. It is, in fact, one of the longest established broadsheet national daily newspapers published in the UK, and arguably one of the country’s most influential newspapers. Moreover, at the time of gathering materials (1998), it was the British newspaper most readily available on the Internet. D A T A A N D M E T H O D S A total of 50 reports each was analyzed from the Times and from the China Daily. Certain criteria were used when selecting particular newspaper texts for analysis, so as to maximize points of comparison between the two sets of texts. L I LY C H E N 680 Language in Society 33:5 (2004) These criteria were as follows: articles were to be between 200 and 600 words in length, about “home” news (i.e., British news for the Times, Chinese news for the China Daily), and published over the Internet in the period January to March 1998. The first 50 texts from each newspaper that I found on the Internet that met these criteria were chosen for the analysis. The approach to analysis and interpretation of data texts adopted here is essentially what Grotjahn 1987 describes as a “Paradigm 7” approach to research in applied linguistics – that is, an exploratory or nonexperimental, quantitative, interpretative approach. The 100 newspaper texts were analyzed and instances of negative, future, and modal comparator evaluation recorded and tabulated. The data for the two sets of texts were then compared for evidence of quantitative differences in the use of evaluation. To complement this quantitative approach, texts displaying particularly interesting usages of certain types of evaluation were also analyzed qualitatively, and the results of the combined quantitative and qualitative analyses were used to shed further light on role and cultural differences between the two media.