Cultural knowledge in News Comprehension 
Roskilde University, Denmark
This paper presents the main lines of the design and the findings of a reception study on news comprehension. This empirical study is a comparison of the comprehension processes of Danes and French Canadians over a set of news texts from both countries. Comprehension is explored from a cultural perspective, through the lens of cognition and pragmatics, revealing the role played by cultural knowledge in comprehension and the underlying relationship between a text and its intended audience. It is argued that recipients ‘problematise’ the news texts, a process by which the texts answer questions that take their roots in the cultural knowledge of the recipients.
Keywords: Reception study, News Comprehension, Cognition, Pragmatics, Cultural models, Problematology, Comparative Study, Canada, Denmark.
Texts (…) are constructed in anticipation of experienced and informed readers, and readers approach texts as structures awaiting their appropriate contribution
- Sonia Livingstone, 1998: 172
This reception study is a comparison of the comprehension process of Danes and French Canadians (or Quebecer ) over a set of news texts from both countries. The project focuses on the process of inferential comprehension, revealing the role played by the cultural knowledge of the recipients and the underlying cultural relationship between a text and its intended audience that effectively results in comprehension. Specifically, it emphasises the implicit processes underlying mass-mediated communication, and hence develops a view of news comprehension as a communicative phenomenon, rather than a matter of textual representation or perception of a stimulus.
Although pragmatics and cognition form the main theoretical framework, and the emphasis is on news comprehension, the project is truly meant as a contribution to communication studies and in particular reception studies, for reasons that I hope will appear obvious after reading this paper. Indeed, this project is best understood as a communication study with news as its case, rather than a research study in journalism. Introducing considerations related to the production of news and the insights produced by the sociology of journalism would certainly add to an understanding of the whole circuit of communication. The usefulness of such a holistic approach has been exemplified in the work of Deacon et al. (1999) concerning the news, or Schrøder et al. (2003), concerning advertisements. Yet, the specific contribution of this reception study is to explore the news text from the perspective of recipients.
The news text is in turn understood within the field of pragmatics. Broadly conceived, this project is an attempt to integrate the linguistic, cognitive and cultural dimensions of the reception of news into a coherent theory of comprehension, or to put it simply, to provide a pragmatic perspective on news comprehension. Pragmatics has been characterised successively as the study of ‘language in use’, ‘intentional meaning’ or ‘utterances in context’ (Verschueren, 1999; Yule, 1996). These three designations are, to different degrees, all acknowledged in this project, but I wish to stress the importance of context, which has become programmatic in recent developments of the field. Indeed, in recent understandings of its project, pragmatics is basically synonymous with the study of context . The approach adopted here is clearly contextual – and hence shares similarities with the cultural studies tradition of reception studies – but as it relates specifically to pragmatics, which acknowledges a cognitive and linguistic dimension often absent in cultural studies. There is no space here to develop a consideration of context theory, but many of the ideas presented in this paper are indebted to this relatively new concern. One ambition of this study is indeed to explore, empirically, the contexts of knowledge used by recipients to comprehend news discourse.
The first section of this paper serves different purposes. It briefly introduces the reader to the literature, theories and conceptualisations on which this project is based, while presenting the main problems and perspectives that the project is set to address. Following a brief look at the ‘golden age’ of news comprehension studies in communication, I present a pragmatic conception of comprehension that is based on the notion of inference. This conceptualisation is also the result of several considerations in connection to previous work in reception studies. Inferential comprehension describes the attempt to articulate the theme of ‘diversity of interpretations’ in a view of reception as an on-line process , which acknowledges the point of view of the recipients. The last part of this first section develops the objectives of the empirical study, essentially a cross-cultural comparison, and stresses some of the theoretical background that is crucial for the subsequent analysis of the comprehension process.
The second section of this paper presents the design of the empirical study. It explains how the interview method was adapted to the study of on-line comprehension and the rationale behind the selection of both the participants and the texts. It also clarifies the nature of the cross-cultural comparative approach, where the relationship between the knowledge of the recipients and the text is a central element.
Finally, the core of this paper appears in the third section: the analysis. Describing briefly the comprehension process in general terms, I then concentrate on the empirical process of comprehension, here conceptualised as a problematisation, of the Canadian news text by the recipients of both countries.
1. Theoretical framework and problem orientation
The comprehension of news was extensively studied in the eighties, mainly through recall studies (for example: van Dijk, 1987, 1988; Findhal and Höijer, 1982; Graber, 1988; Gunter, 1981, 1987). In these studies, comprehension was mainly associated with the ability of the recipients to reproduce a given text. Many studies have shown a positive correlation between, on the one hand, the amount and sophistication of background knowledge and, on the other hand, recall and comprehension. Despite these findings, the influence of background knowledge on comprehension remains largely under-theorised and unexplored (van Dijk, 2001, 2003). One possible explanation is that comprehension has been largely approached as being representational, where the text is seen as the source of the process and the goal of comprehension is the reproduction of this text in memory.
The conception of comprehension that governs this project owes much to the Cognitive Pragmatic theory introduced by Sperber and Wilson (1989/1995). Relevance theory assumes that incoming information is comprehended against a background of knowledge - a context of interpretation - that makes this information relevant. In other words, recipients process information that is relevant to them, and this relevance is relational to a context of interpretation elaborated on-line by the recipients. This process of contextualisation is an inferential process by which the combination of incoming information with the context yields cognitive effects or conclusions. I refer to this kind of comprehension as inferential comprehension.
1.1 Inferential comprehension
Inferential comprehension is concerned with how incoming information is incorporated to and affects our existing representations of the world. To put it more simply, what difference does it make to know this information? What conclusions are reached on the basis of this newly acquired information? What does this new information change in the context of what I already know? Consider this example, an extract taken from the Canadian news texts used in the empirical study, which illustrates what is meant by inferential comprehension:
(…) the Canadian parliament has passed with an overwhelming majority of 256 against 16 a resolution that recognises (…) Quebec as a nation within a united Canada.
The information an overwhelming majority of 256 against 16 can be comprehended differently by different recipients. For most of the Danish recipients who read this extract, it meant that the resolution was ‘uncontroversial’, that there was a “wide political consensus’ or even a ‘consensus within the Canadian population’. The picture was quite different for Quebecers, who did not consider the information at face value, but placed it in a large web of knowledge about the political context in Canada. For some, this majority was a sign of the ‘resignation of the politicians’, an indication that they could no longer ‘deny this reality’, that they were ‘forced to vote in favour’ of this resolution, although ‘they did not really agree with it’. For another, it was simply ‘easy for all the politicians to agree on something that had no practical consequence’, as the resolution was, in fact, only a symbolic gesture.
Of course, the comprehension of this information necessitates that the word majority is well understood and indeed part of the representation of the events that recipients elaborate on the basis of the text. In that sense, representational comprehension is to be associated with the semantic meaning of a word, and is therefore a pre-requisite to inferential comprehension, which then expresses the meaning that a word takes in particular contexts.
1.2 Some aspects of doing comprehension
The differences in the conclusions drawn by the Danes and Quebecers are of course explained by their respective background knowledge, used to contextualise the same piece of information. One of the premises on which this project is based comes from the bulk of reception studies showing diversity in interpretation of media texts by an active audience.
Much of the effort invested in understanding the active audience and questions of interpretation has been motivated by concerns for ideological influence. For example, Hall’s Encoding/Decoding model (1980), regarded as the instigator of the reception studies tradition (Gray, 1999: 26), is an attempt “to study decoding in both semiotic and ideological terms’ (Schrøder, 2000: 238). Acknowledging a distinction between the study of recipients’ sense-making processes and the broader study of ideology, Schrøder (2000) suggests an empirically based model of reception that distinguishes the reading experience of the recipients (motivation, comprehension, discrimination and position) from the researcher’s evaluation of it (evaluation and implementation). Limiting my inquiry to the dimension of comprehension, I wish to investigate the reading experience of news recipients, as something independent from its evaluation, by focusing on the micro, empirical and cognitive processes that lead to and explain the activity of news recipients and the diverse interpretations they produce.
Thus, the explanatory framework that I am trying to develop in this project concerns the possibility of arriving at different interpretations of a same text. It is well known that recipients comprehend media products, and in particular news, in different ways. But as Livingstone (1998) remarks, research will not be able to answer the question of how recipients make sense of media outputs simply ‘by offering up a vast diversity of individual readings made under different circumstances’ (1998: 152). This project is therefore not meant to show that recipients comprehend news differently, but instead to explain such an outcome in a theoretical framework that stresses the importance of cultural tacit knowledge in comprehension and its relationship with the language use in news discourse.
This project also aims to investigate on-line processes, as opposed to off-line processes such as remembering, discussing or reflecting on the news (Findahl, 1997). On-line processes are not very well known by communication researchers, as the method of investigation commonly used in reception studies – the interview – is most often used to stimulate off-line responses. In this project, the interview method has been developed as an on-line method that allows the researcher to follow the unfolding of comprehension during the reading of the text as a dynamic process, providing new insights about what it means to comprehend a text.
The project also addresses the perspective of the recipients in comprehending news discourse (Schaap, 2001). It focuses on what recipients actually do with the news, rather than what they do not do or what they have to do, according to a model of the coherence of the text (like in linguistics), to a scholarly analysis (like in discourse analysis) or a model based on an ideal reader, either real (like an ‘expert’) or imagined (like an inscribed reader).
Taken together, previous findings on news comprehension (van Dijk, 1987, 1988; Findhal and Höijer, 1982; Graber, 1988; Gunter, 1981, 1987) seem to indicate that recipients learn, remember and comprehend very little from the news. These findings, however, stemmed mainly from the use of recall and reproduction methods. As pointed out by Schaap (2001), some scholars have raised doubts concerning the appropriateness of such methods to study comprehension. First, making sense of the news involves certainly more than remembering and reproducing its content. Second, Schaap observed that correct reproductions are assessed by the researchers themselves, and that processes such as elaboration or alteration of news content have sometimes been classified as wrong retention, even if they may be legitimately considered a part of what it means to comprehend news. Thus, the search for an objective measurement might have led to the discarding of some important aspects of news comprehension and, not least, to ignoring other aspects that are not captured by the methods. Consequently, previous findings are tied to the methods used, and unless we are content with the conclusion that recipients do almost nothing with the news, alternative methods need to be explored. In that respect, qualitative methods seem more appropriate to investigate what recipients actually do with the news.
The empirical findings that show what recipients do not do with news are complemented by an earlier theoretical interest in what recipients have to do to comprehend or make sense of a text. This line of thinking can be found in various research traditions, varying between linguistics, literary studies, semiotics, and even cognitive sciences. These traditions have in common that they all emphasise the text as the source of meaning . Each of these traditions is, of course, motivated by the particular theoretical assumptions that they bring in their analysis (language is something to comprehend, discourses have coherence, texts have [intended] meaning, comprehension rests upon cognitive processes and structures that need to be discovered, etc.). However, all these approaches, as justified as they may be, are not concerned with meaning as a product of actual recipients, who may not have the resources presupposed by these theoretical assumptions (goals, interest, depth, knowledge, cognitive skills, attention, and even the time or the will to penetrate a text, etc.).
All in all, this emphasis on the perspective of the recipients and on their on-line processes is also a way to stress that comprehension of news discourse must be assessed in relation to the actual recipients who receive this discourse, which includes an assessment of their sense-making resources, not only in the light of a long and thorough analysis conducted in front of a computer desk. In other words, even if textual meaning can be assessed by a scholarly analysis, this kind of analysis should not be considered in lieu of the quick and gist analysis that actual recipients make of news discourse (O’Halloran, 2003; Shore, 1998).
In a nutshell, the conceptualisation of comprehension that is advocated in this project moves away from an interest in the output, where the aim is to measure comprehension or assess its quality, towards an investigation of its process. The main question at the heart of this project might well be formulated in these terms: What do recipients do when they are doing comprehension, that is, what are the processes involved that lead a recipient to acquire an understanding of a news story?
1.3 Background knowledge and cultural models
Background or world knowledge consists in tacit assumptions about the way the world is or ought to be. Such knowledge is thought to be embedded in schemas (Bartlett, 1932) or mental models (Johnson-Laird, 1983). World knowledge is also believed to be highly social and cultural. Some authors use the term cultural knowledge to stress that world knowledge is shared within a given community (Holland and Quinn, 1987: viii; van Dijk, 2003, 2001). Shore (1998), in the context of cognition, and Gee (1999), in the context of discourse analysis, talk instead of cultural models.
Gee (1999) explains that cultural models are cognitive representations associated with the use of words or concepts in context. They develop from recognising similar patterns in our experience of the world that transcend the different contexts in which concepts are used. Thus, the meanings associated with a concept are always connected to their contexts of use. These contexts do not only refer to direct experience, but can be mediated by other people, in which case we can talk of discursive contexts.
Not only do people recognise patterns in their experience, but they also develop explanations as to why these patterns hang together. These implicit theories consist in taken-for-granted assumptions about what is normal, typical or appropriate.
A cultural model is usually a totally or partially unconscious explanatory theory or ‘storyline’ connected to a word – bits and pieces of which are distributed across different people in a social group – that helps to explain why the word has the different situated meanings and possibilities for the specific social and cultural groups of people that it does (Ibid: 44).
From this rather brief sketch of the theory of cultural models, I wish to draw certain implications for the current study. Words are loaded with implicit meaning. When authors use words, they ‘mean’ the cultural models associated with these words as part of what is communicated, although these meanings remain unsaid and implicit. This needs not be a conscious choice, but an aspect of the common ground (Clark, 1992) on which communication becomes possible. Conversely, the implicit patterns and theories associated with the words used play a crucial role in the elaboration of a context of interpretation. In their attempt to comprehend a text, recipients might look for patterns usually associated with a word. Similarly, the theory implied in a cultural model allows making predictions and inferences about the textual storyline under consideration. That is, cultural models create expectations that may guide (or hinder) the comprehension process.
Gee stresses that cultural models are not integrally in the mind of each individual. Rather, bits of them are distributed in a group or even through institutions and practices. This observation is rather important because it implies that no particular ‘piece of knowledge’ is presupposed in the use of a word. The presupposition concerns the patterns implied by the cultural model, which can be retrieved by calling in any of the contexts associated to the cultural model (what is called episodic knowledge in cognitive psychology) or general knowledge developed on the basis of these contexts, that is, the patterns themselves (what is called semantic knowledge).
As final remark, Gee treats the concepts of context, situated meaning and cultural model ‘as existing in the mind and in the world’ (1999: 53). This position is at the root of this empirical study, which has the ambition to investigate the cultural dimension of reception and comprehension in a realist fashion. But at the same time, Gee reminds us that these terms are still primarily ‘tools of inquiry’, tools used by researchers to answer certain questions and not others.
Although, in this project, knowledge is assumed to be cultural from a methodological and theoretical point of view, this issue is still a debate that rages between, and even within, different traditions of research. Generally speaking, the cultural dimension of knowledge is absent from most, if not all, systematic theoretical accounts of comprehension (see for example the recent account of Zwaan and Singer, 2003). Secondly, the cultural dimension of knowledge is not being investigated in contemporary empirical studies of discourse processing. This is to be expected, given that knowledge is conceived as mostly a-cultural and a-social, a position that the cognitive psychologist Arthur C. Graesser makes explicit in a recent, but short paper published in Discourse Studies (2006).
In contrast, there is a wide agreement that media reception is cultural within the qualitative tradition of reception studies (Schrøder et al, 2003). However, there is no consensus as to a definition of culture, its nature, structure and agency. There are instead many different positions that conceptualise culture in different ways: in relation to knowledge, social interaction, institutions, identity, language, discourse, and so on. No matter how culture is conceptualised, however, it is often posited solely in theoretical terms in the humanistic-interpretative tradition. Culture is something that explains reception, rather than something that needs an explanation. Culture remains an abstract category used to interpret the data, rather than something to be analysed empirically.
This project is an attempt to explicitly articulate, in an empirical study, the cultural dimension of knowledge (and by extension comprehension); knowledge that is expressed by individuals in their reception of media products. In other words, the project stresses the role of culture in the mind of individual recipients, without adopting a private conception of the mind, and is concerned with the ‘application’ or ‘use’ of culture, as a ‘cognitive’ resource, rather than the acquisition of culture which has traditionally been the main concern of scholars.
It is evident that the cultural dimension of knowledge in comprehension is not problematised adequately, if at all, in cognitive psychology. First and foremost, this is due to the relative absence of cross-cultural comparative studies. But the choice and conceptualisation of the material used to test comprehension, combined with the selection of participants, as well as the limitations of the methods used to assess comprehension, also explain this situation.
I want to argue that using material and participants that belong to the ‘same culture’, in other words, using material that is intended to be received by a particular audience and not by another one, makes the cultural dimension of comprehension invisible to methods and researchers alike, leading to universalist claims. Moreover, text material is often purely conceived as a stimulus and is elaborated and tested as such. Traditionally, cognitive psychology has laid major emphasis on perceptual processes (or perception-based processes) – its main contribution is indeed to have uncovered the mysteries of perception – and has conceived responses to text as originating in the perception of a stimulus.
1.4 The cross-cultural comparison
This project articulates comprehension as a knowledge-based process rather than a perception-based process, that is, as taking its roots in the tacit knowledge of the recipients. It has been showed that news discourse presupposes a lot of knowledge (van Dijk, 2001; Bekalu, 2006). This knowledge is in great part cultural, in the sense that the news text addresses a specific audience whose knowledge is being presupposed.
The cross-cultural comparison is therefore not simply a matter of comparing the cultural knowledge of two national groups. It is more accurate to conceptualise this comparison in terms of presupposed cultural knowledge. The comprehension of a group for which its tacit knowledge is presupposed by a news text, is compared to the comprehension of another group, for which its tacit knowledge is not presupposed by the same news text. Thus, the cultural dimension of comprehension is contrasted in the comparison by considering both the origin of knowledge (or their recipients) and of the text.
As implied previously, the role of culture in comprehension is largely invisible. One of the challenges facing this study was therefore to make culture ‘visible’. The strategy adopted was to find ‘traces of culture’ through diversity of interpretation. Therefore, a cross-cultural comparison was designed in order to contrast processes of comprehension that rely on different cultural knowledge. To do that, two national groups were chosen as the basis of the comparison. Yet, there is a clear danger, both methodological and theoretical, in using diversity as a way to compare two national groups. As Jandt remarks in relation to large-scale groups, ‘the diversity within cultures probably exceeds the differences between cultures’ (2004: 7).
In comparing recipients from two national groups, I do not want to imply that these two nations form homogeneous groups – something that this empirical study easily contradicts – nor am I trying to stress the importance of national or territorial boundaries in the formation of communities and cultures. I am simply comparing recipients from two national groups in relation – and this is important – to specific texts for which the distinction is relevant and empirically confirmed. All in all, the choice of these two groups should be regarded as a methodological strategy, not as a theoretical assumption about the importance of national identities.
The cultural dimension that is being investigated here is not a static conception of culture that involves looking at inactive and ossified groups of people, out of any consideration, but a complex relationship involving both texts and recipients. As a matter of fact, it hardly makes sense to talk about two distinct cultural groups in any absolute way, outside of any specific situations, i.e. the reception of concrete texts. In some respects, these groups can be said to form one culture, while in other respects they can be distinguished. In other words, the distinction that is created between these two groups is as much a product of the knowledge of these recipients than of the choice of the texts. With other texts, such distinction would not have been possible. In this respect, the design of the study articulates the ‘situatedness’ of meaning and culture (Gee, 1999).
It was therefore assumed, from a methodological point of view, that the chosen recipients form two distinct cultural groups and that the text will elicit this distinction. These assumptions needed to be realised in the actual empirical study, so that it became possible to contrast the comprehension of these two groups. The texts were therefore deliberately chosen, so that it became relevant to talk in terms of two distinct cultural groups. The distinction that is created is an empirical one, which then serves as a heuristic to study the process of comprehension. That is, the methodological strategy needed to be actualised in the empirical study in order to proceed with the analysis of the process of (cultural) comprehension, which represents the real interest in this study.
From the perspective of each group reading the foreign text (for example Danes reading the Canadian text), the study was not meant to be a reflection of the comprehension of foreign news. Foreign news is usually selected for and minimally adapted to the needs, preferences, concerns and resources of a local audience . The news texts that ‘played the role’ of foreign news in this study were not adapted and were clearly not aimed at a foreign audience. Indeed, most participants have been painfully aware that the foreign text they were reading was not meant for them.
1.5 An analysis of the problematisation of the text
So far, I have argued that inferential comprehension consists in reaching conclusions via inferential processes that combine textual information and background knowledge. We lay emphasis on the information that is relevant to us and this information is relevant in the context of our background knowledge. Relevance is therefore an implicit strategy to read and comprehend text. What is comprehended and how it is comprehended depends to a large extent on our background knowledge, which operates an interpretative work on the text: knowledge is not simply applied to the text or activated by it, but is actively used to investigate the text. In this process, the text answers questions that take their roots in the background knowledge of the recipients. In other words, recipients problematise the text with the help of their background knowledge.
The notion of the problematisation of discourses by recipients derives from a conception of meaning developed by Michel Meyer (1995) in his book Of Problematology, which sheds light on the nature of the interpretative work of knowledge in the comprehension of discourse. The philosophical position that Meyer calls ‘problematology’ is an attack to the propositionalist conception of meaning (that the meaning of a sentence resides in its truth values), and as such is truly a pragmatic position. For Meyer, the meaning of a sentence depends on the question to which this sentence is an answer for. ‘In a general sense, to understand discourse is to see it as answer, that is, as an answer to a question of knowing what it answers or what is the question within what is said’ (Meyer, 1995: 250-251).
In a style worthy of philosophers, Meyer explains his ‘Fourth Meditation: The question of meaning, or meaning as question’:
A listener is potentially an explicit questioner, as is a reader, although rarely will a reader be able to question an author directly. Whether he agrees or not with an answer to a question, a listener will make his own assessments about the adequacy of that answer. This process is called comprehension, or, if we want to refine the expression, it can be called the hermeneutic process. An answer does not state its meaning, since it doesn’t say, “This is the question which I am resolving,’ nor does it identify itself as an answer. It refers to what it says without saying that it does; it suggests its meaning without stating it. Meaning here remains implicit; it corresponds to the implicit character of understanding. The meaning of a discourse is provided by the problematicity it deals with. When someone speaks of the meaning of a sentence, for example, he is referring to what is in question in the sentence (Meyer, 1995: 223; his emphasis).
A text presents itself as an answer or result, but it is the questioning process of the recipient that ultimately defines what the result stands for or what it is an answer to. ‘Even if it is a result in regard to some originating problem, it can as well be a result for other questions’ says Meyer (1995: 213). Thus, he conceives any text as being inherently problematological, that is, as offering the possibility to be turned into a subjective understanding by virtue of the context that is used to problematise the text. ‘Context (…) transforms what is problematological in the answer into something which is problematic for someone’ (Meyer, 1995: 222; his emphasis).
I would like to suggest that the cultural knowledge of the recipients forms problematics that are used to comprehend the news text. It is the presence of a problematic that leads to comprehension, while the absence of any problematic leaves the text inert. More than that, I would like to show that comprehension rests upon the presence of presupposed problematics. Cultural differences in comprehension derive from the different problematics that are carried by our cultural knowledge. When this knowledge is being presupposed by the news discourse, comprehension will be successful to the extent that the problematics carried by this knowledge facilitate comprehension and enhance the inferential capacities of the recipient. When recipients use problematics that are different than the ones presupposed by the text, they become a hindrance to comprehension.
Recipients make relevant and exploit the textual information in a way that is compatible with the problematics that are carried by their cultural models. In fact, recipients elaborate contexts precisely to be able to problematise the news text and to derive their own comprehension of what the news text is about. Therefore, the analysis that is carried out in this study can be characterised essentially as an analysis of the problematisation of the texts by recipients.
As a result of these considerations, an empirical study was set up to investigate the role of knowledge, especially cultural models, in inferential comprehension and the dynamic process of on-line comprehension. The design of this empirical study is presented in the next section.
2. The design of the empirical study
The cross-cultural comparison resulted in 2 hour in-depth and semi-structured interviews with 10 Danes and 10 Quebecers, for a total of 40 hours of interview material. Each recipient was asked to read 2 news stories, one from Quebec (Canada) and one from Denmark, in either a short or a long version . The foreign text was translated in the local language of each group and the interviews were also conducted in the local language (French and Danish) . The interview with Quebecers took place in January 2007 between the 15th and 29th, at Laval University in Canada . The interviews in Denmark were spread between the 11th of April and the 16th of May 2007  and were conducted at Roskilde University or in Copenhagen.
2.1 The method
The in-depth interview was designed as an adaptation of the think-aloud protocol (see Schaap  for an application to reception studies of news), a method that consists in asking people to think out loud while performing a task, such as for example solving a problem. This adaptation consisted in asking participants to read the news text, paragraph by paragraph, starting with only the title, and to ask non-leading probes after each extract in order to elicit verbal reports concerning different aspects of the comprehension process.
The use of probes was primarily a tool to go in-depth in the comprehension process of each recipient. The probes were non-leading to make sure that comprehension was investigated from the perspective of the recipients. In other words, the interviewer paid special attention not to suggest any interpretative path that the recipients did not themselves bring up. In effect, most of the probes were ways to follow up on what was verbally reported by the recipients.
The aim of the method was mainly to investigate the contexts of interpretation elaborated by recipients to comprehend the news text. More precisely, the method consisted in collecting verbal reports about 1) what was retained from the text and what appeared to be the important information for each recipient; and 2) the inferences made by recipients, the conclusions reached and the assumptions used to contextualise the news text. Additionally, questions were asked to tap into different aspects of the dynamic process of comprehension, such as the formation of expectations concerning the rest of the text or the formation and development of the macro-structure of the text (van Dijk, 1987) .
The investigation of background knowledge is by no means an easy task. It implies bringing to the foreground what is usually in the back of the mind. As these assumptions are taken-for-granted and rather obvious for the people who hold them, they are the most difficult to speak about. Moreover, as an assumption probably hides further underlying assumptions, there is an obvious limit to how much both the interviewer and interviewee can ‘keep digging’ in the course of an interview. And finally, all this is interactively produced: the interview is based on a collaborative work between the interviewer and the interviewee, which itself relies on their reflexivity and the ability of the participants to verbally report on their comprehension. The method was indeed not familiar, nor natural to the participants, and around 20 minutes of the interview was dedicated to presenting, explaining and training  the method with the participants.
2.2 Selection of the participants
10 Danish students and 10 students from Quebec were recruited for an interview . The initial objective of the recruitment was to form more or less homogenous groups along various socio-demographic factors, like age, sex and especially level of education, without any pretense for standardisation . All participants were first year university students and recruited from courses related to communication .
University students are not only a convenient group to get access to, but they were also preferred over other recipients because the method was so demanding. I am not sure the method would have been successful with other groups who have not been and are not currently and constantly asked to articulate their thoughts verbally, such as it is done in a university. Yet, choosing first-year students would prevent the interview from becoming an exercise in academic analysis or media literacy. It was however expected that the use of students, the nature of the method, as well as the setting of the study (most interviews were conducted at the university) all contributed into shaping the news reading experience into an intellectual achievement, which it isn’t necessarily. This research product might be remote from the more emotional or social, from the less attentive or reflexive experience of daily news reading, but it still does not appear completely misplaced given the object under study: comprehension.
2.3 Choice and elaboration of the texts
It was a premise of this empirical study that the news stories should relate to real issues (instead of being fictional) and that these issues should not be too distant in the mind of the recipients, given that recency defines in large part, at least in common-sense, what news discourse consists of. But because the texts were manipulated and translated into a foreign language, there was a long gap of time between the moment the story broke the news and the moment the interviews were conducted.
Although it may have been preferable to study the reception of these news texts in more pristine conditions, recipients can be said to always read news texts under the influence of other ‘texts’, be those previous news on the same topic or interpersonal conversations. Such understanding of meaning-making can go as far as to suggest that a whole culture amounts to a complex web of intertextuality (Jensen, 2002: 186), so that any cultural reading is an instance of intertextuality. Yet, the objective of this empirical study is to articulate the role of tacit knowledge as it is ‘mobilised’ (Vershueren, 1999) and made ‘relevant’ (Sperber and Wilson, 1995) by actual recipients in order to form an on-line context of interpretation, rather than an understanding of its history or genealogy. Considering also that the investigation involves critically the world knowledge of the recipients, it appeared more important to rely on real news items rather than on asceptic reading conditions typical of the use of artificial texts in laboratory experiments. Thus, the limitation of the fieldwork was seen as a trade-off in naturality between recency and the reliance on real news items.
The news texts used for the empirical study were elaborated by the researcher as accounts or recapitulations of domestic events that unfolded over approximately a week period in the press. That is, the material that was presented to the informants was pieced out by the researcher on the basis of actual news extracts taken from various newspapers. The idea was first to select and elaborate a coherent, consistent and inviting news story to be read by both a local and foreign audience, rather than using actual news reports that account for the day-to-day events as they unfold, and which may appear detached to a foreign audience, or even to a domestic audience unaware of current affairs.
As a way to reflect natural uses of language appearing in news discourses, the actual news extracts were incorporated in a ‘cut and paste’ fashion, with no or minimal alterations if only to insure local and global coherence to the material. Of particular importance in the selection of these extracts were the ways journalists make use of the language – their own or the language of their sources – to refer or index actions, processes, events, as well as protagonists, their roles, and intentions. The theoretical rationale of this selection is that although news is commonly understood as a referential discourse (a representation of reality), the ways this reality is engaged in language use hide an important cultural pre-understanding that is crucial for the comprehension of what is said. In that respect, the texts, together with the cross-cultural comparison, were conceived in order to address the contribution of cultural models in making what is said either ‘transparent’ or ‘opaque’ to its readers (Hanks, 1996). This line of reasoning comes directly from an understanding of language use in pragmatics as being ‘interadaptable’ (Verschueren, 1999), ‘indexical’ (Widdowson, 2004), ‘indeterminate’ and ‘saturated by context’ (Hanks, 1996).
An additional challenge was to translate the ‘foreign’ news texts in the local language of the informant . Again, special attention was paid to preserve the natural ways the original language indexed events and their protagonists. Original formulations or expressions were kept intact when this was directly possible and grammatically correct. This means that the ‘foreign’ texts were not adapted to the tastes or needs of a domestic audience (see also the endnote 20).
The Canadian text is about the recognition of Quebec (a predominantly French-speaking province in Canada) as a nation . The news story from Denmark concerns the making and coming of a new national smoking policy. The analysis that follows in the next section covers only the comprehension of the Canadian text. Therefore, this text is introduced in more details and only methodological information is given about the Danish text. In the following, the title and lead of the Canadian text is presented, which was common to both the short and long versions, as well as a synopsis of the rest of the story.
‘The time has come for national reconciliation’, but costs government a minister
Motion recognising Quebec as a nation passed yesterday
Although it costs Canada’s PM, Stephen Harper, a minister from his cabinet, the Canadian parliament has passed with an overwhelming majority of 256 against 16 a resolution that recognises people in the francophone province of Quebec as “a nation within a united Canada’. With the adoption of this “historical’ resolution, Canada became last night, according to independentit  party Bloc Québécois, the first country to recognise Quebec as a nation.
After the lead, the article presents the rationale behind the decision of Stephen Harper, the Canadian Prime Minister, to pass this resolution, which was done in reaction to a similar resolution presented by Gilles Duceppe, the leader of the Bloc Québécois, a party that promotes the independence of Quebec at the Canadian parliament. The difference between the two resolutions is that the first one, the one originally presented by Duceppe, makes no mention of the clause ‘within a united Canada’, which was intentionally added by the prime minister of Canada.
The article then develops the argumentation of both Harper and Duceppe, the former seeing this recognition as a way to maintain the Canadian unity, and the latter as the first step towards independence. The article moves on to explain the reasons for the resignation of the minister mentioned in the title and the lead. Finally, it presents the point of view of the Liberal party, which was hesitant to support the motion.
2.4 Experience and knowledge
The rationale behind the design of this empirical study is in the way each text relates to the experience of each national group. First, each group knows a priori very little about the other. It was a condition of participation to this study that each recipient knew very little about the foreign country . Second, the texts were chosen on the ground that they would relate differently to the experience of each national group.
The Danes could not relate directly to the kind of experience dealt with in the Canadian story, which can be said to be domain-specific. Denmark is not in the same situation; it is a nation as well as a nation-state. There are perhaps a few similarities involved in the historical relationship between Denmark and Sweden, as well as between Denmark and its former colonies (Iceland, Greenland, Faeroe Islands; the two last are now integral parts of Denmark). These were experiences that some Danes made use of in their comprehension of the Canadian text, but the limits of such similarity appeared clearly, also to the recipients themselves, in their verbal reports. In other words, Danes had no or little firm premises from which to start their inferential comprehension of the Canadian text.
On the other hand, Quebecers could relate to the kind of experience involved in the Danish story - the making of a smoking policy - as there is also a restrictive smoking policy in Quebec, but this experience is however different in relation to the considerations and objectives of the law, its strictness and the firmness of its application. In other words, Quebecers could possibly relate to the Danish issue, but they would probably do so assuming a different set of premises or assumptions.
The texts used for the comparison are therefore not ‘equal’ and were not meant to be equal, if that were possible. The rationale behind their selection was to see the extent to which each group would bring their local set of assumptions into their understanding of both the local and the foreign texts, and how this ‘pre-comprehension’ would affect their selection of information, processes of inference and their overall problematisation of the text.
3. The analysis
The following analysis is a work in progress. It only presents the main findings and is based exclusively on the comprehension of the Canadian text, by both Danes and Quebecers. It focuses on group tendencies rather than on the analysis of each individual comprehension. Such qualitative analysis is difficult to present succinctly. As much as it is possible, examples taken from the interviews will be used to illustrate the main points, but it is obvious that the analytical work rests upon much more than what is presented here.
The analysis first provides a general description of the comprehension process. But the core of the analysis consists in presenting the main problematics used by recipients and their subsequent problematisation of the news text.
3.1 Comprehension: A dynamic process
The process of comprehension is a dynamic process of formation and confirmation (or disconfirmation) of hypotheses and expectations (see Reboul and Moeschler, 1998, for a theoretical account of this conceptualisation). At the very beginning of the text, after having read the title, recipients form many hypotheses, sometimes alternative or competing hypotheses about the states of affairs talked about in the news text or about expectations of possible outcomes. The subsequent process is one of verification, abandonment or adjustment of hypotheses and expectations.
Genre is also crucial to comprehension, as it creates expectations and reflects reading conventions that both guide and place limitations upon the reading process. For example, Pateman (1983) showed that expectations associated with the genre ‘advertisement’ (such as talking positively about the advertised product) are responsible for the identification of the connotative meanings of visual ads. In a similar vein, Schrøder et al. (ch. 7, 2003) showed that an emerging sub-genre (ethical or SCR ads) that broke with existing genre conventions (such as not requesting a clear ‘action’, i.e. buying) confused recipients as to the intended meaning of the ad. Zwaan (1994) demonstrated the phenomenon experimentally, showing that inappropriate genre expectations slow down and confuse text comprehension. Also relevant is the analysis of Livingstone (1998), which shows that the divergent interpretations of recipients regarding plots in soap opera never stray aside the borders of the genre.
The question of genre was not explicitely considered in the empirical study as recipients are all positioned within the same genre, which in itself does not contribute to explain differences in news comprehension, and because an emphasis on individual differences regarding knowledge of and familiarity with the news genre, typical of the analyses of performance in the psychology of discourse processing, would distract us from the more basic, shared cultural competence underlying news comprehension that was aimed at in this study, i.e. problematisation. Some basic information on the informants‘ news habits were nevertheless collected, and they indicate that, albeit with some variation, none of the recipients were ignorant of the news genre in a way that could be detrimental to their comprehension, and which could have compromised the conclusions suggested in this paper. Although my sample does not allow firm generalisation, it confirms previous quantitative findings by Findhal and Höijer (1982, 1985) and van Dijk (1987) to the effect that experienced news readers have a better comprehension of news texts.
Seen from the dynamic perspective of the unfolding of comprehension, the more confirmation, the better is comprehension. Conversely, less confirmation (or the more disconfirmation) leads to an impairment in comprehension. Consequently, the beginning of the comprehension process is crucial as it frames the rest of the process. This original framing of the text depends on the textual features of the title as much as on the background knowledge of the recipients.
Many Danes started their reading of this text with the assumption that the recognition of Quebec as a nation meant that Quebec had become a country or an independent nation. The news story also dealt with the relationship between Quebec and Canada. Some Danes understood the recognition by the Canadian government as a good gesture and an indication of collaboration between Canada and Quebec. However, the rest of the article could not confirm these two hypotheses. As a result, Danes who started the text with these inadequate assumptions had difficulties to comprehend the rest of the story. If they pursued their comprehension with these assumptions in mind:
1) The text did not confirm them and even refuted them;
2) They could hardly draw conclusions on the basis of these assumptions;
3) They did not pay attention to the textual information that was incoherent with or unrelated to these assumptions.
In other words, the information that the Danish recipients read was not relevant in the context of the assumptions they held, and as such, could not be exploited in inferential processes to draw conclusions. Or if this information was exploited, on the basis of inadequate assumptions, it would lead to misleading conclusions, generating even more confusion.
That is to say that these recipients had difficulties orienting themselves throughout the text. To say it metaphorically, they were visiting Canada with a map of Denmark and could not recognise anything their map was telling them to expect. Some Danes persisted in their use of unconfirmed assumptions, and they are also those who comprehended very little of the Canadian text. Others abandoned their initial assumptions and therefore needed to pay much more attention to the textual context. The process of comprehension then turned into a process of forming assumptions, hypotheses and problematics out of the textual context, which then appeared more difficult and less constrained by existing knowledge.
Generally speaking, recipients grabbed any opportunities to confirm their hypotheses. This has resulted in very different assessments of the text, as recipients pay attention to the textual information that confirms their hypotheses and very often ignore the rest. Each paragraph was systematically probed to assess what information each recipient retained, and there is a clear link between the textual information that is retained and the answers this retained information provides to the hypotheses, questions and expectations of the recipients. That is, the information that was not retained was not relevant because it did not correspond to any assumptions held by the recipient.
The following analysis of the problematics and the problematisation of the Canadian news text will illustrate, with the help of quotations taken from the interviews, this dynamic process of comprehension.
3.2 The main problematisations of the text
Many different problematics were used by both Quebecers and Danes in their comprehension of the Canadian text. But I want to argue that there were some central problematics that played a critical role in the comprehension of this text. Not only could these problematics be used to comprehend different aspects of the text, but they also brought these different aspects into a coherent whole. The following analysis will limit itself to the presentation of two main problematisations: the problematisation of the nation and the problematisation of the relation between Canada and Quebec.
For each problematisation, I will first show how the background knowledge held by the recipients from Quebec implies problematics that were presupposed by the text, facilitating comprehension, and how the background knowledge of the Danes was incompatible with these pressuposed problematics, impairing their comprehension. Second, I will provide examples of how the problematics used by Quebecers and by Danes enhanced or impaired their selection of information and their inferential comprehension.
The concept of nation is, of course, at the heart of the Canadian story. And the cultural model held by recipients about the concept of nation, prior to reading, affects their comprehension of the text, at least as a starting point. What does nation mean for Quebecers and Danes? What does it mean that Quebec has become a nation? And what does it mean when it is something said by the Prime minister of Canada or by the leader of the independentist party?
To begin with, the translation of the word nation in the Danish version of the news text needs some clarifications. Some native Danish speakers have pointed to me that the meaning of nation in Danish is that of a nation-state. This is attested by the definition provided in the Danish dictionary Politikens. Yet, the definition of nation as a nation-state is only the most common, and there exists also at least two other senses to the word nation in Danish: cultural and ethnic. These senses can be found in the Danish lexicon (Gyldendals Leksikon) or in the academic literature, thus attesting of their existence: ‘Even the word nation has (at least) three different meanings, both in Danish and many other European languages. It is a widespread use in daily language to use the word nation as a synonym of state (…) But the word nation is also used to designate a group of people with common cultural characteristics, especially a language, a history and a culture, and sometimes also an ethnic origin or religion’ (My translation from Danish; Adriansen, 2003: p. 27-28) .
Without surprise, nearly all Danish recipients conceived nation as a country or as an independent nation. One extract taken from the verbal report of one Danish recipient illustrates the presence of this cultural model in the Danish group:
What do you think it means that Quebec has become a nation?
Well, I don’t know. I suppose that they have become an independent nation. I don’t know if it’s possible to be a nation without being an independent nation, eh. So, I, I don’t know. I’m thinking that they will have their own, their own nation, their own… yes, their own parliament or their own structure, and their own dumps and hospitals… (…) Well, when I think about nation, then I think about a country (DK3).
This recipient wonders if it is possible to be a nation without being a nation-state. Indeed, Denmark constitutes both a nation and a nation-state, surrounded by other European nation-states. Moreover, the status of Denmark as a nation outside the definition of a nation-state is, for obvious reasons, not a matter of controversy, as another Danish recipient made explicit in her reading of the title of the article:
I have a hard time to remember the title, but yes, the government has passed a resolution about… and Quebec becomes a nation… a national state it’s called. (…) It says that the government has passed a resolution that Quebec will be a national state. So I guess that it’s not just a small town, but… but precisely a big area. (…) Yes, yes… eh… But really… nation, it’s, you know… eh… Yes, it doesn’t quite fit… But eh… but Denmark is a national state, isn’t? (DK8).
For many of the Danish recipients, it is as if the text had said that Quebec had become a nation-state. This is the answer that the text provided to the Danish recipients, given the familiar patterns implied by their cultural model of the word nation.
Of course, the expectations, created from the assumption that Quebec had become a country or an independent nation, were not going to be confirmed by the subsequent text. As a result, the comprehension of these recipients would be impaired, as the context they held could not make the textual information relevant. In these extracts, the assumption that Quebec has become a country clashes with the textual context presented in the subsequent paragraphs, but also with some implicit theory regarding the concept of separation as held by a Danish recipient (marked in bold in the extract):
Verbal report after the title:
But really, I don’t understand at all, because I can also see, that it’s probably not… that Quebec is not going to become its own country, or… or I don’t know, but… Yeah, and that it has been passed as a resolution, it’s probably that… It has surprised me a lot that Canada, if it… you know, who the hell give away willingly a part of its country. But, there is nobody who does that. So, so, yeah… (DK3).
Verbal report after the lead:
Yeah. Well now, I think I am really as confused or just like so uninformed in a way, because now I find out that there was an overwhelming majority that has voted for Quebec to be a nation. And there was also this wording ‘a nation within a united Canada’. (DK3)
Verbal report after the 2nd paragraph:
Yeah. But I am even more confused… (DK3).
One Danish recipient stands as an exception that somehow confirms the rule. She understood that Canada was the nation and that Quebec wanted to be part of this nation, that is, of Canada. Of course this initial representation had dramatic implication for the course of her comprehension. She only fully realised the inaccuracy of her assumptions after reading the 5th paragraph of the long version:
Eh… Well, it looks like they rather want to be on their own. Which is the complete opposite of what I understood this far. I thought that they were going to be under a bigger power, but now it looks like that they have become their own… not a part of a big power (DK6).
The absence of relevant knowledge, combined with an inaccurate representation of the text (the title did say ‘as a nation’ and not ‘in the nation’) led this recipient to hold mistaken expectations that were kept until the 5th paragraph, to the expense of her comprehension of the 4 previous paragraphs, as her macro-structure of the text, collected after the 4th paragraph, indicates:
‘Eh… Well, I think that, it’s about the idea of a nation, what it is… And then, also something about being part of a big power, but I don’t know if it is a part of it or if it separates from it. I’m not really sure about that’ (DK6).
Some Danes insisted in keeping their initial assumptions and could not properly contextualise the textual information. The information they were reading was simply not relevant in the context of these assumptions, and as a result, many pieces of information were simply ignored and did not take part in the representation of the text that these recipients elaborated. Generally speaking, Danes did not retain very much from the first few paragraphs of the text. But the majority of the Danish group progressively realigned their assumptions to accommodate the newly acquired textual information.
At the end of reading the short version, one Danish recipient, who realised that the recognition was only symbolic, reinterpreted the gist of the text in light of this new information, as it appears in her macro-structure:
It was symbolic to recognise Quebec as a nation. Yeah. Well, then it’s about national feelings, and about, eh, the right to be your own people (…) So it’s not really in connection to the economy, and to… and to all the things that follow when one becomes a nation. So it’s… it’s in relation to feelings, will I say. Yes (DK2).
Compared to an earlier macro-structure:
It’s about, eh, Quebec as an independent… or perhaps not independent, but as a declared nation, and eh… and it’s about old disputes, maybe, and maybe about traditions (DK2).
Information that did not make sense before now did, but many of the details of the story were lost. As a general rule, the information that is comprehended under an inappropriate problematic is not being processed again, even when the recipient forms a new problematic in which previous information could usefully be reinterpreted. In other words, to be able to lead to any conclusions, the textual information needs to be actively comprehended under the light of an appropriate problematic. Otherwise, the text remains inert and does not contribute to the inferential comprehension of the recipient.
It should come as no surprise that recipients from Quebec started their reading of the text with different assumptions than the Danish group. For all Quebecers that were interviewed, it is taken-for-granted that Quebec is a nation: ‘Well, it is something obvious, but at the same time, I was happy [that it was recognised]’ (QC1). As Quebec has always been regarded as a nation, the text could not be about the making of a new nation. Recipients from Quebec perceive themselves as a nation mainly because of their distinction, on the basis of their ‘culture, language and institutions’, traits common to the conception of nation offered by all recipients from Quebec.
In the following extract, not only the conception of nation provided by this recipient excludes any considerations about national state, but it also denies the status of nation to Canada:
Well nation, it’s eh... it’s eh... it’s a contentious definition. There can be many different definitions you know. (...) According to me, Quebec is a nation, because for me... OK, if we push it, we could admit that Canada is a nation. Eh, everything is a question of definition, but for me, I identify myself more to the nation of Quebec. To say that it’s a Canadian nation, well, I don’t really agree with that because most of the people in Quebec don’t want to be... Well, perhaps not most, but a large portion of the population in Quebec does not want to be part of the Canadian nation. So for this reason, we cannot admit out of any doubts that there can be found a Canadian nation (QC6).
For this recipient, Canada is a country without a nation. The problematic related to his conception of nation - or to put it differently, what is in question in his conception of nation - is more a matter of identification than of definition. Who cares what a nation means, I mean we are a nation!
This example also shows that the implications of a cultural model go deep and can radically contrast with the understanding of the same cultural model in another community. I greatly doubt that it would have crossed the mind of any Danish recipients, given their cultural model of nation, to call into question that Canada is a nation.
On the contrary, the meaning of the concept nation puzzled many Danes when applied to an entity that did not seem to have become a nation-state. While for them, the meaning of nation was a question of knowing what it means in this context (does it mean that Quebec has become an independent country or does it mean something else?), for the recipients from Quebec, it was more a question of knowing who says it and for what purpose. The use of nation by the different politicians in the news story was there conceived as a site of struggle. In other words, the recognition that Quebec forms a nation, and the underlying definition of this nation, were seen as contentious by Quebecers.
The assumptions underlying the cultural model of nation by Quebecers provided them with a problematic, which we could call the problematic of the contention, from which to comprehend the information in the text, like ‘national reconciliation’ or ‘united Canada’, ‘historical resolution’, etc., as these extracts from a recipient in Quebec illustrate:
When you recognise the nation of Quebec, just like that... well, you recognise that they are able… that they could have their own country to themselves. So finally, they could manage their country and they could be independent from Canada. But when you specify ‘within a united Canada’, it means ‘you are still with us, so don’t forget that’ (QC9).
And after that they say that it’s a historical vote because Canada is the first country to recognise Quebec as a nation. (…) It’s a political move, more than a representative or historical move, as it said in the other paragraph. (…) And they say: ‘time has come for reconciliation with Quebec’. Well I think it’s even more a political move (QC9).
And I think that Harper probably thought that if he gives us a candy, there might be a rapprochement between Canadians, to really form a united Canada… in which he wants to put us. It’s really… I think that’s what the reconciliation is about. Has it worked? I don’t know (QC9).
Compared with a Danish recipient who could not problematise ‘national reconciliation’:
And I don’t really know what he means with reconciliation. I couldn’t really… yeah… I think ‘reconciliation’, it has to be… if you say reconciliation and you talk about Canada, then you could almost think that they mean the opposite (DK1).
Eh… because by saying that Quebec is separate… some sort of state in its own, and is something special, then they part from Canada. So when I think about reconciliation, I am rather led to think that the whole thing should be like… together and integrated, so eh… it’s a bit mystical. I can’t see the logic of it at all… I thought it was very confusing (DK1).
The reading of the Danish recipient illustrates the distinction between semantic meaning and pragmatic meaning, that is, the difference between meaning out of context or in context. The Danish recipient knows what reconciliation means, but it does not make sense in the context of the story, or what she understands of it. In the representation of the recipient from Quebec, reconciliation is understood in the context of being uttered by Stephen Harper, the prime minister of Canada. The problematic used by this recipient allows him to incorporate this important information, which makes a difference, in his representation of the events.
For this recipient, as well as for other recipients from Quebec, it is meaningful that it is the prime minister of Canada who wants to reconcile the country or who recognises Quebec as a nation. In other words, for Quebecers, the news story is not simply that Quebec is being recognised as a nation, which is what many Danes comprehended, but that it is Harper who recognises Quebec as a nation and that this nation is recognised ‘within a united Canada’. This is an aspect of the story that was not problematised by the Danish recipients.
This importance given to the intentions of Harper is stressed in the macro-structure of nearly all recipients from Quebec, while it is absent in the macro-structure of the Danes:
- The resignation of the minister in relation to the fact that Harper has recognised that Quebec is a nation (QC1).
- Mister Harper (…) adopts a motion about Quebec, about the nation of Quebec in a united Canada (QC2).
- I would say that the news is about the decision that Stephen Harper has taken (QC3).
- It’s the news that concern the recognition of the nation of Quebec by Ottawa (QC6).
- OK, the news, in one sentence, it’s that Canada has recognised the nation of Quebec (QC9).
As stressed earlier, the question of the meaning of nation is more a question of identity than a matter of definition for the recipients from Quebec. There is an underlying conception of ‘us’ and ‘them’ implied in their problematic, which makes relevant the question of knowing who says what. Moreover, it allows recipients from Quebec to reach conclusions on the basis of the textual evidences. Notice here all the intentions that this recipient attributes to Harper after he had read the 3rd paragraph:
Well, now the question is being defined. It defines how Harper sees the question of the nation of Quebec. He says that he accepts that Quebec forms a nation, but within a united Canada. So by saying this, he pleases Quebecers, but he also doesn’t want to displease people who don’t agree with that, so he says ‘No wait! I’m not saying that they are independent, that they are an independent nation’, so at the same time he protects himself (QC1).
This problematisation, this emphasis on the intentions of Stephen Harper, allows this recipient to draw many inferences. It is interesting to note that the textual information ‘united Canada’ had not been problematised before by this recipient, as it had been attributed to the journalist. Only when it was attributed to Harper could the recipient infer meaning out of it:
I think that before, when it was mentioned… it was the journalist. I can’t remember exactly how it was put. But now, it’s really his discourse [Stephen Harper], so I think it has more impact. He really said that. It was said with quotations marks (QC1).
So what was the goal of this paragraph? You have said it a little bit but…
It was to define the position of Harper. To say what he means by nation (QC1).
This is not an isolated phenomenon. Another recipient from Quebec was puzzled by the use of “national reconciliation’ as it appeared in the title of the article. He first attributed this use to a pro-conservative bias of the journalist. Only when it appeared clear that it came from Harper’s mouth could the recipient make sense of it as an intention of Harper to ‘put an end to the conflict between Quebec and Canada’ (QC2).
All things considered, the comprehension of Quebecers seems to have been tainted by the implications that their knowledge has for their identity, and that they therefore attributed intentions in order to rationalise the propositions that clashed with their taken-for-granted assumptions.
Rightly or not, it is often a premise in Canadian politics that the province of Quebec, in one way or another, is in conflict with the central government in Ottawa, the capital of Canada. One of the main episodes of the news story concerns the political warfare between Stephen Harper, the prime minister of Canada, and Gilles Duceppe, the leader of the independentist party Bloc Québécois. This opposition was not explicitly stated in the text, nor did it need to, at least for recipients familiar with it. The conflict problematic was therefore a map that could orient the comprehension of the recipients. Consider the reading of the title by a recipient from Quebec who makes this problematic apparent:
I think it’s reductive. It says: the reconciliation between – as far as I understand – between Canada and Quebec is achieved by the recognition. (…) It’s because there is a resolution that has been passed. It solves the conflict. I think it is reductive of all the conflict that has been going on for more than 200 years. And I feel like it reduces… the news reduces this issue in relation to what it really is (QC2)
What is the conflict?
I would say it’s the nation of Quebec amongst Canadians (QC2)
The same recipient, using the conflict problematic, provides a clear picture of the role of each politician in the story:
So, Duceppe, in fact, he’s part of the Bloc Québécois. So he wants to separate from Canada. Well, he wants that Quebec be separated from Canada (…) He acts… he plays his role of a politician of the Bloc Québécois. (…) You know, it’s an issue that they will always have, anyway in this context, they will always camp on their position. One will want to separate, the other will want to reconcile (QC2).
This problematic of the conflict also takes its roots in the assumptions held by Quebecers about being a nation, in that Quebec forms a nation that is conceived, perhaps not in complete opposition, but as different from the rest of Canada:
According to me, Quebec society, in itself, represents a nation because of its distinct characters. Eh, because of its language and so on. (...) A nation, it’s... it’s a population that distinguishes itself, in fact, we have our language, our heritage (...) It’s also a question of our values that distinguish ourselves. (...) And especially in relation to our language, I think we are different. And it makes us, you know, we’re from French origin, in relation to the other who have English origins. So in relation to that, you know, it’s two societies that have had a lot... that have distinguished themselves a lot. I’m not necessarily saying that they are against each other. But you know, there is a distinction to make. So I think it’s normal to consider ourselves as a nation (QC2, my emphasis)
In contrast, many Danes did not start their reading with any expectations concerning the relationship between Canada and Quebec. This is something they had to figure out as they were reading the text. Some Danes integrated indications from the text that pointed to this conflict problematic, but others did not, which led them to understand that the recognition was a sort of collaborative effort between Canada and Quebec. For example:
One could say that Canada wishes that Quebec becomes an independent state (DK1)
Notice, in the following extract, how this assumption has influenced the comprehension of the information that Canada was the first country to recognise Quebec as a nation, as it appears in the lead of the article:
Well, eh… Really. If, if eh… well if Quebec was before a part of Canada, so I think that it’s important that they are the first to come up and say it (…) because it seems like Quebec has not separated itself against the will of Canada. It is a common decision, and in fact it is something that Canada gave them permission to do (DK5).
One Danish recipient had minimal knowledge about Quebec and she profited from this in her comprehension (or problematisation) of the information ‘national reconciliation’. Her knowledge about Quebec helped her to elaborate a conflict problematic, which she used to comprehend the textual information:
I would like to hear what you know about Quebec?
They speak French. And there has been a little movement that wishes to free Quebec from Canada, as far as I know. But not the majority. That’s what I think (DK4)
You have noticed that the text says that the time has come for reconciliation. What does that mean for you?
Well, it means, that there has been, eh, I think, a separatist movement in Quebec that has fought against their inclusion in Canada. And, what is it called?… as far as I know, it was not a Quebec-terror… eh, separatist movement, but who knows, eh, I can’t really remember, but there has probably been an extremely harsh tone, or an extremely irreconcilable… eh, and an unpleasant political environment, maybe. Eh, yeah, maybe even some violent assaults. Yeah, something irreconcilable and now they say that it’s time for reconciliation, by giving them some rights that they have wanted for a long time (DK4).
This other Danish recipient could not count on any previous knowledge, but he managed to take advantage of some textual clues, from which appropriate expectations could be formed:
What do you think it means that there is a minister who left?
Well, eh… it’s… it’s like an escalated conflict. That it’s not everybody who claps their hands and say: Hey let’s get a new flag and a new national song. There is indeed a minister from the government that has left. And it is a conflict, a conflict about which I expect to learn more (DK8).
It is interesting to compare what this recipient (DK8) retained from his reading of the second paragraph, to another Danish recipient (DK6), for whom there was no trace of a conflict problematic.
Well, then, eh… Yeah, there has happened some sort of conflict that has been there before… eh, the recognition. (…) It presented the two speakers, who fought from each their side. Well, there is him, the prime minister, this Harper or what’s his name… He says something. And then there is the other one, him from… eh, the party for… that wants separation. (…) We get to know who they are and what opinions they have. (…) And then there is this problematic. Eh… and some quotations, eh… from the two guys, which say what it is they want (DK8).
Eh… Something like the prime minister… I have to think about it a little bit… No, it was hard, I think. Anyway, there was something like that Quebec is or is not a nation, and then the prime minister was a little bit… He didn’t really like to call them a nation. He would rather say that they were a part of a united Canada. But in order to do that, they had to… they obviously needed to make a compromise… in order to call them that, they had to become a nation, or something like that. Yeah, it was a bit complicated. But eh… it could be that I need to read it again (…) It’s also this thing that the prime minister has changed his mind suddenly. Eh, yeah, I paid attention to that because he definitely… I didn’t quite get why he changed his mind. It was something with this nation and with the United States, or united Canada (DK6).
Not only the reading of the second recipient (DK6), who lacked a problematic, clearly indicates that he was unable to find what was in question in the text, but it also shows how little information was retained from the text, which, as a result, remained inert, without meaning.
The next example is an interesting case. This Danish recipient developed a conflict problematic, based on some schematic knowledge of the French people, but she also understood nation to mean independence, as her reading of the lead indicates:
Well, I got out of it that obviously, there are a lot of French people in Quebec. That it… that maybe it makes them feel like… like a minority and that they would like to be independent. Eh… Yes, and obviously that they have their own political party (DK9).
Why do you think that there is a minority that has a wish to be independent?
Because they are French. Yes. (…) Well, eh, it’s also that… I don’t know… I don’t know what happened before. But I’m thinking of the French as people who hold on very much to being French. Yes. (…) Now I don’t even know why there are so many… why they speak both French and American in Canada. I don’t know if there came a lot of French immigrants (…) Yes, I don’t know, of course, if people speak French all over Canada, or mostly in Quebec. But if there has been a mass of French immigrants, then I could imagine that they think that they are so distinct from the rest of the country (DK9).
In the first extract, this recipient tries to find evidence, in the text, of the relevance of her background knowledge about the French people (from France), which she provides a glimpse of in the second extract. She notes that there are many French people, but that they are in the minority, and that they have their own political party. Although these implications are accurate, they are a very subjective selection of the information presented in the lead.
Indeed, her reading of the lead indicates that she was absorbed in developing an adequate problematic rather than applying an ‘already known problematic’, which therefore explains her partial selection of information from the lead. By developing a conflict problematic, which she based on the existence of a French minority that wants independence, she enhanced her inferential power, as can be showed by her problematisation of two textual evidences that appear later in the text: that Harper was warmly applauded by the members of his cabinet and the information ‘united Canada’:
I think it [the warm applause] means that they [the member of the cabinets] don’t think that they [Quebec] should be independent, but… At least, that they [Quebec] will still be under Canada. So it therefore explains why they think it’s fine (DK9)
These preceding examples illustrate that it is not the presence (or absence) of specific prior knowledge that leads to comprehension (or lack of), since some of the Danish recipients who lacked knowledge of the situation in Quebec could still comprehend the text. It is instead the presence of problematics that orients the reading of the text, the selection and integration of the textual information into a coherent whole.
In this analysis, comprehension was analysed as a dynamic cultural phenomenon involving the relationship between the text and the recipients’ knowledge. By contrasting the comprehension process of Danes and Quebecers, it was shown that background knowledge, and in particular cultural models, are presupposed in news discourse and account for the implicit part of what is communicated, but not said. This implicit relationship between a text and its intended audience reminds us that comprehension is not self-evident, nor a matter of pure perception. It is neither completely a textual phenomenon, nor a private expression of cognitive skills, but a deep-seated cultural phenomenon that often goes without notice in the context of ordinary reception. Only by looking at those circumstances where there is a misfit between the cultural models implied in a text and those used by recipients, which this cross-cultural comparison was meant to articulate explicitly, does this cultural phenomenon appear unmistakably, because it is conspicuous by its absence.
Cultural models contribute to the process of comprehension in various ways. First, they motivate the search for relevance. Recipients search for a textual context in which they can put their cultural models to use. For the most part, the Danish recipients found a textual context that contradicted or that did not meet their expectations. In other words, they had troubles making the textual context relevant, given the assumptions they held.
Additionally, recipients select and retain parts of the textual context that are relevant to their cultural models. This is to say, recipients ignore information that is irrelevant under the assumptions implied by their cultural models. This could be obviously seen when Danish recipients generally disregarded the information ‘within a united Canada’, while Quebecers made it a central element of their situation model.
Thirdly, recipients make inferences based on their cultural models. Given the adequacy and the richness of their cultural models, recipients from Quebec drew more inferences. Their conclusions were not only in phase with the textual context, but they also made the whole text coherent. Danish recipients, whose cultural models turned out to be inappropriate, had more difficulties drawing inferences. Their inferences were more general, and were sometimes explicitly refuted by the subsequent textual context.
Finally, and perhaps most importantly, cultural models provide expectations about what is in question in news discourse, about the problems that the news is seen to address, about the questions that the news is seen to answer. In the course of the comprehension process, these cultural models form problematics that are used to investigate the text. Conceived in this way, comprehension is an active search process for answers to questions that take their roots in the cultural models of the recipients. Inferential comprehension, as I have put it, is not pure recognition of what is already known, nor does it stray too far from what is already known.
As I hope to have shown, it is the recipients without problematics, irrespective of the group they belong to, who have the most difficulty comprehending the text. Of course, this was the case for more Danes than for Quebecers, as the text presupposed the cultural models of the latter. It is interesting to note that, through their verbal reports, many Danes showed that they were busy developing adequate problematics, rather than applying a familiar problematic. As is reflected in their many hesitations and doubts, Danish recipients often questioned the adequacy of their cultural models, without completely letting go of them. They seemed unable to avoid making the textual context relevant using the same cultural models that they judged inadequate.
It is as if recipients from Quebec used the textual context as evidence of the adequacy of their cultural models, while Danes were condemned to use the textual context as clues from which both questions and answers could be provided, or to miscomprehend the text given the inadequacy of their cultural models. Both processes reflect an effort to problematise the text, although they clearly differ in outcome.
Additionally, the information that the Danish recipients retained from each extract indicates that they used the textual context to answer basic questions that their lack of knowledge made pressing to answer, such as geo-political information about Quebec. It is not only that Danes produce conclusions that are less helpful to understand what is going on in the news story. It is also that their inferences stray aside from what is expected by the news genre. Indeed, comprehension is perhaps not the most appropriate term to describe the use that the Danish recipients made of the Canadian text, as they used the text for learning basic facts about Quebec . This prompts the question whether the news, especially foreign news, is an appropriate medium for such learning to take place? In any case, the difference between Quebecers and Danes seem to indicate that the role of the news genre depends crucially on its problematisation and requires a wider cultural competence that is grounded in tacit knowledge, what other scholars would readily call ideological knowledge.
This study shows that news – its language – can be quite opaque for those who do not share the cultural models that are tacitly presupposed. For those who do share these cultural models, and who use them as a source of problematisation, the language appears as transparent, as leading directly to meaning, and the world behind it becomes easily intelligible. What is said, yet in many ways ‘sketchy, partial, incomplete, masked, elliptical, concealed, ambiguous, or misleading’ – to use a formulation by Harold Garfinkel (1972: 317; quoted in Widdowson, 2004: 11) – becomes nothing less than a straightforward representation of reality as understood in a specific community. That is, the news text becomes a transparent representation of reality by virtue of what it serves to problematise. In a sense, news does not represent reality. Rather, news problematises reality.
As one underlying goal of this project was to combine and integrate insights from two traditions that are too often put in opposition, I wish to conclude this paper with an example that not only encapsulates the main argumentation of this analysis, but that also exposes the way comprehension, as understood in cognitive psychology, does, can and should converge with concerns within the interpretative tradition. In cognitive studies of discourse processing, it is well established that the primary goal of recipients in comprehension is not the formation of a textual representation – what can be associated with what I have called representational comprehension - but the understanding of what the text is about (van Dijk, 1988; van Oostendorp and Bonebakker, 1999) – what I have referred to as inferential comprehension. That is, recipients are believed to elaborate a model of the situation, not simply a model of the text. In this last example, we can see how a recipient from Quebec readily elaborates a situation model out of her sole reading of the title, as if the rest of the article was as superfluous as predictable:
From what I understand, indeed, it’s… if you want… Stephen Harper has chosen to, if you want, to reunify Canada, precisely by his declaration that Quebec is distinct within a united Canada, even though for that he has lost a minister (QC3).
Her assessment of the title does not only incorporate the main aspects of the story, but also such elements that make up what we call situations. He did X by Y to achieve Z, even though A, etc. Guided by her cultural knowledge, this recipient could instantly assess what was in question in the text.
Such assessment of the situation is obviously subjective and evaluative. Winners and losers are identified; goals are seen to be reached or missed; actions are justified or criticised; consequences and obstacles are assessed, etc. So that ‘the resolution of Harper was an attack to Duceppe, who counter-attacked’ (QC1) or it was ‘an end in itself [a way to put an end to the conflict between Quebec and Canada], while Duceppe’s resolution was a first step towards independence’ (QC3), ‘Harper wanted to gain the heart of Quebecers, and in that sense, it was a good strategy, although I doubted it would work’ (QC6), etc.
These situation models, as elaborated by actual recipients on the basis of their cultural knowledge, will normally exhibit what can be called misrepresentations, elaborations, losses of information and biases. Rather than considering these as errors of computation, it is more useful to regard them as the expression of the way our cultural knowledge allows us (or limits us) to understand and organise the world. In that sense, cultural knowledge does not incapacitate recipients in perceiving the situation presented in the text. Instead, it contributes in defining that very situation, in that the text, through its problematisation, answers questions that our cultural knowledge makes possible to ask.
The notion of problematisation gives space for a conception of diversity and cultural specificity in matters of comprehension. It stresses that recipients connect news discourse with world knowledge and that it is in this interaction that comprehension arises and that news become meaningful for their recipients. It does not conceive of differences in comprehension as errors or biases, but as inherently cultural. Seen under this light, comprehension is not a derived product of perception, prone to errors of computation.
Rather, the notion of problematisation stresses the cultural role that texts, and in particular news, play for their readers. Texts are written for their capacity to provide answers to underlying questions. As these questions take their roots in the cultural knowledge of the recipients, it seems only natural that news presuppose cultural knowledge that increases this capacity. In turn, as news may appeal to problematics that are familiar to their recipients, they contribute to reproducing culturally-specific conceptions of the world and provide these recipients with a meaningful understanding of that world.
The importance of studying comprehension empirically as an activity which actual recipients engage in should be clear at this point. It was argued that cultural models are implied in news discourse, something that became empirically visible by contrasting the comprehension of two cultural groups. Indeed, these cultural models cannot be identified by textual analysis alone, as they are not to be found in the text . But that does not mean that cultural models are detached from the text at all.
It has been said that cognitive psychology only contributes to promote a conception of an isolated text processed by a detached recipient (Livingstone, 1998), and hence obscures the broader cultural context in which texts and recipients are embedded. It is not surprising that cognitive psychologists start their empirical investigation with a conception of a detached recipient. For it can be argued that recipients discover the text precisely by processing or interpreting it. What is more surprising is that the same conception of a detached recipient still prevails after the investigation of cognitive processes is undergone. In other words, cognitive processes as studied by cognitive psychology may be said to contribute very little to affirming the mutual dependence of texts and recipients.
While maintaining an empirical stance, it can be argued that it is precisely by interpreting the text that actual recipients realise (or fail to do so) the relationship that unite them with texts. This can act as a guiding premise in the conduct of both reception studies of actual recipients and studies of cognitive processes. Cognitive processes related to the reception of media products are complex, and this complexity does not only follow from some internal logic of cognition, but needs to be related to the broader cultural context in which it is naturally embedded. Yet, close attention to the processes involved in reception can provide a more precise picture of how culture plays a role in communication. Here, I am advocating for a fruitful integration of media studies and cognition.
I regard this cross-cultural comparison as an example of such integration and of the methodological and theoretical compromises that necessarily follow from multi-disciplinary research. Although impaired by many limitations, this project was an occasion to revisit our understanding of comprehension in light of the cultural relationship that ordinarily unite texts and recipients, and to offer some empirical insights into the ways in which ‘texts (…) are constructed in anticipation of experienced and informed readers, and readers approach texts as structures awaiting their appropriate contribution’ (Livingstone, 1998: 172).
List of references
Adriansen, I., Nationale Symboler i det Danske Rige, 1830-2000, Copenhagen: Museum Tusculanum, 2003.
Bartlett, F. C., Remembering: a Study in Experimental and Social Psychology, Cambridge: Cambridge University Press, 1932.
Bekalu, M. A. ‘Presupposition in News Discourse’. Discourse & Society, vol. 17, no. 2, 2006, pp. 147-172.
Bell, A., The Language of News Media, Oxford: Blackwell, 1991.
Clark, H. H., Arenas of Language Use, London: The University of Chicago Press, 1992.
Deacon, D. et al., ‘From Inception to Reception: The Natural History of a News Item’, Media, Culture and Society, 21, 1999, pp. 5-31.
Duranti, A. and C. Goodwin, Rethinking Context: Language as an Interactive Phenomenon, Cambridge: Cambridge University Press, 1992.
Fairclough, N., Media Discourse, London: Edward Arnold, 1995.
Findahl, O. and B. Höijer, ‘The Problem of Comprehension and Recall of Broadcast News’, in J-F LeNy and W. Kintsch (Eds.), Language and Comprehension, Amsterdam: North-Holland Publishing Company, 1982, pp. 261-272.
Findahl, O., ‘Go Cognitive and Go On-line. New Directions in Media Research’, in Carlsson, U. (ed.), Beyond Media Uses and Effects, Göteborg: Nordicom, 1997.
Gee, J. P., An Introduction to Discourse Analysis. Theory and Method, London: Routledge, 1999.
Graber, D. A., Processing the News: How People Tame the Information Tide, New York: Longman, 1988.
Graesser, A. C. and G. H. Bower, Inferences and Text Comprehension, San Diego: Academic Press, 1990.
Graesser, A. C, ‘Views from a Cognitive Scientist: Cognitive Representations Underlying Discourse Are Sometimes Social’, Discourse Studies Vol. 8 (1), 2006, pp. 59-66.
Gray, A., ‘Audience and Reception Research in Retrospect: The Trouble with Audiences’, in P. Alasuutari, Rethinking the Media Audience, London: Sage Publications, 1999, pp. 22-37.
Gunter, B., ‘Forgetting the News’, Intermedia 9(5), 1981, pp. 41-43.
Gunter, B., Poor Reception: Misunderstanding and Forgetting Broadcast News, Hillsdale: L. Erlbaum Associates, 1987.
Hall, S., ‘Encoding and Decoding in the Television Discourse’, in S. Hall et al., Culture, Media, Language, London: Hutchinson, 1980.
Hanks, W. F., Language and Communicative Practices, Oxford: WestviewPress, 1996.
Holland, D. and N. Quinn, Cultural Models in Language and Thought, Cambridge: Cambridge University Press, 1987.
Jandt, F. E., An Introduction to Intercultural Communication: Identities in a Global Community, Thousand Oaks: Sage Publications, 2004.
Jensen, K.B. and K.E. Rosengren, ‘Five Traditions in Search of the Audience’, European Journal of Communication 5(2-3), 1990, pp. 207-238.
Jensen, K.B., A Handbook of Media and Communication Research. Qualitative and Quantitative Methodologies, London: Routledge, 2002.
Jensen, K.B., News of the World: World Cultures Look at Television News, London: Routledge, 1998.
Johnson-Laird, P. N., Mental Models: Towards a Cognitive Science of Language, Inference, and Consciousness, Cambridge: Harvard University Press, 1983.
Katz, E. and T. Liebes, The Export of Meaning: Cross-Cultural Readings of Dallas, Cambridge: Polity Press, 1993.
Levinson, S. C., Pragmatics, Cambridge: Cambridge University Press, 1983.
Livingstone, S., Making Sense of Television. The Psychology of Audience Interpretation, London: Routledge, 1998.
Lorch, R. F. and E. J. O’Brien, Sources of Coherence in Reading, Hillsdale: L. Erlbaum Associates, 1995.
Meyer, M., Of Problematology. Philosophy, Science and Language, London: The University of Chicago Press, 1995.
Mucchielli, A., Étude des Communications: Approche par la Contextualisation, Paris: Armand Colin, 2005.
O’Halloran, K., Critical Discourse Analysis and Language Cognition, Edinburgh: Edinburgh University Press, 2003.
Pateman, T., ‘How is understanding an advertisement possible?’, in Davis, H. et Walton, P. (Eds.), Language, image, media, Oxford : Basil Blackwell, 1983, pp. 187-204.
Reboul, A. and J. Moeschler, Pragmatique du Discours : de l’Interprétation de l’Énoncé à l’Interprétation du Discours, Paris: A. Colin, 1998.
Schaap, G., ‘Using Protocol Analysis in Television News Research: Proposal and First Tests’, Communications 26 (4), 2001, pp. 443-463.
Schrøder, K. C., ‘Making Sense of Audience Discourses. Towards a Multidimensional Model of Mass Media Reception’, European Journal of Cultural Studies 3(2), 2000: 233-258.
Schrøder et al., Researching Audiences, London: Arnold, 2003.
Sperber, D. and D. Wilson, Relevance: Communication and Cognition, Cambridge: Harvard University Press, 1995.
Shore, B., ‘Models Theory as a Framework for Media Studies’, in B. Höijer and A. Werner (Eds.), Cultural Cognition. New Perspectives in Audience Theory, Göteborg: Nordicom, 1998, pp. 7-40.
Trabasso, T. et al., ‘Explanatory Inferences and Other Strategies During Comprehension and their Effect on Recall’, in R. F. Lorch and E. J. O’Brien (eds.), Sources of Coherence in Reading, Hillsdale: L. Erlbaum Associates, 1995, pp. 219-39.
van den Broek, P., ‘The Causal Inference Maker: Towards a Process Model of Inference Generation in Text Comprehension’, in D.A. Balota et al., Comprehension Processes in Reading, Hillsdale: Lawrence Erlbaum Associates, 1990, pp. 423-446.
van Dijk, T. A., News as Discourse, Hillsdale: L. Erlbaum Associates, 1987.
van Dijk, T. A., News Analysis: Case Studies of International and National News in the Press, Hillsdale: L. Erlbaum Associates, 1988.
van Dijk, T. A., (2001) ‘Knowledge and News’, [WWW document] URL http://www.discourses.org/Teun.html. [visited 29/10/2006].
van Dijk, T. A., ‘The Discourse-Knowledge Interface’, in G. Weiss & R. Wodak (Eds.), Critical Discourse Analysis. Theory and Interdisciplinarity, Houndsmills: Palgrave-MacMillan, 2003, pp. 85-109.
van Dijk, T. A., ‘Discourse, Context and Cognition’, Discourse Studies, 8 (1), 2006, pp. 159-177.
van Dijk, T. A., Discourse and Context: A Sociocognitive Approach, Cambridge: Cambridge University Press, 2008.
van Dijk, T. A., Society and Discourse: How Social Contexts Influence Text and Talk, Cambridge: Cambridge University Press, 2009.
van Oostendorp, H. and C. Bonebakker, ‘Difficulties in Updating Mental Representations During reading News Reports’, in H. van Oostendorp and S. R. Goldman (eds.), The Construction of Mental Representations During Reading, Mahwah: L. Erlbaum Associates, 1999, pp. 319-339.
Verschueren, J., Understanding Pragmatics, London: Arnold, 1999.
Widdowson, H.G., Text, Context, Pretext. Critical Issues in Discourse Analysis, Oxford: Blackwell Publishing, 2004.
Yule, G., Pragmatics, Oxford: Oxford University Press, 1996.
Zwaan, R. A. and M Singer, ‘Text Comprehension’, in A.C. Graesser and al. (eds.), Handbook of Dicourse Processes, London: L. Erlbaum Associates, 2003, pp. 83-121.
Zwaan, R.A., ‘Effect of genre expectations on text comprehension’. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20 (4), 1994, pp. 920-933.
 I wish to thank all the persons that have commented on this paper or its earlier drafts, and in particular Kim Schrøder and Teun van Dijk for their respective insights into my project.
 In Canada, a distinction is made between francophones (who have French as a mother tongue), anglophones (English) or allophones (another language). The participants in my study were all francophones from Quebec, who form the majority of the population of this province. In the rest of the article, I refer to these participants only as Quebecers, keeping in mind that they are French Canadians living in Quebec.
 For useful accounts of context theory from a variety of perspectives, see van Dijk, 2006, 2008, 2009; Mucchielli, 2005; Verschueren, 1999; Hanks, 1996; Sperber and Wilson, 1995; Duranti and Goodwin, 1992.
 The use of the term on-line refers to the application of the interview as an on-line method, as opposed to an off-line method (see further in section 1.2), in order to investigate comprehension in its tight connection to the text and as it unfolds throughout the reading process, hence the terms on-line processes or on-line comprehension that will be used throughout this paper. Although the term on-line does not here refer to ‘on-line media’ or ‘on-line news’ (this study only relies on print news) and is meant to qualify a point of method, a quick search on Google may convince of their common roots, as on-line is associated with ‘on-goingness’ and ‘in progress’ and indicates ‘a state of connectivity’ (while off-line expresses a ‘disconnected state’), qualifications that can equally apply to the internet and the method of investigation used in this study.
 See Jensen and Rosengren, 1990, for an overview of five traditions relevant to communication studies.
 The selection and adaptation that is operated on a foreign news story in order to fit a local audience could reveal much about the cultural character of comprehension. The fact that there are alterations made to foreign news is an indication that a text is not an isolated entity, which exists in a vacuum, and that its comprehension is not self-evident or a matter of pure perception, but rather is meant to rely upon something other than itself if it has to make any sense at all to the audience for which it is intended to.
 As a way to keep the duration of the interview within reasonable limits (maximum 2 hours) and, by the same token, explore the contribution of the textual context (more precisely, its granularity or levels of precision) in comprehension, each recipient read one short (6 paragraphs) as well as one long text (circa 14 paragraphs), the origin of which was arbitrarily but evenly distributed within each group. As an example, half the Danish recipients would read the short story from Quebec and the long story from Denmark, and vice-versa for the other half. The main idea behind this strategy was to see if “culturally-proximate’ recipients could fill-in the gaps left in the short version of the story and, conversely, if the addition of various details is of any help to the culturally-distant recipients. As the findings seem to suggest, the granularity of a text does play a role, but one that is so close to each individual response or textual proposition that a very large group of recipients would have been needed to see if any clear group tendency would emerge out of this condition. For this reason, the presentation of the short vs. long condition is left out in this paper, and is not considered in the analysis.
 The author, a native of Quebec, has conducted the interviews in Canada, while a Danish research assistant was enrolled to interview the Danish students.
 In this connection, I wish to thank especially Jean de Bonville and Jean Charron, directors of the research group GRMJ at Laval University, and everybody else in the group, for the help and the facilities they have provided me.
 It was harder and longer than expected to recruit the Danish students at Roskilde University, who, at that period, are normally busy writing their project report.
 The macro-structure is a sort of summary of the main propositions (called macro-propositions) of a story.
 The method was briefly trialled with the informants at the beginning of the each interview, using a short story of their own liking from the newspaper of the day. This rehearsal did not take part in the analysis.
 In fact, 13 interviews were conducted in Quebec, but only the 10 interviews that produced the most relevant verbal reports in relation to the research objectives were chosen for analysis and compared to the 10 Danish interviews.
 The Danish group was composed of three men and seven women between the age of 21 and 28 years old, but 8 of them were 21 or 22 years old. The students from Quebec were equally distributed between men and women through an age range of 19 to 23 years old, with a median of 21. That is, the group from Quebec is slightly younger and more equally distributed. This is explained by the fact that more students in Quebec were interested than what was needed for the study, so that a more restrictive selection could be made.
 The students from Quebec were all enrolled in the Communication programme offered at Laval University. Danish students were recruited from the Humanities Basic programme, as the Communication programme is only offered from the third year of a 5 year higher education degree. The intention was to exploit the interest of the students to participate in the study.
 The translation was realised with the help of a professional translator, Mette Huuse, who I wish to thank for her precious help. It also rested on a more ‘pragmatic’ (in the common sense of the term) research about actual language uses in each community, with help of dictionaries, encyclopedia, newspapers, literature and also the Internet (the latter revealed itself an invaluable resource in this particular context). The Canadian story also appeared in the form of short news stories in different Danish newspapers, and those were used as guidance for translation. Comparisons between the Canadian story in francophone and anglophone newspapers in Canada were also used as a reference point to think about issues of translation. Despite my effort, I take the entire responsibility if the linguistics choices I made appear inappropriate in relation to real editorial decisions. Yet, it is important to remember that the texts were translated for specific research purposes.
 The events related by this material took place between the 22nd and 28th of November 2006. News extracts were selected from the consultation of 22 articles from two omnibus newspapers, La Presse, the main francophone newspaper edited in Montreal, and The Globe and Mail, a major anglophone newspaper edited in Toronto with national distribution. The story also appeared, in the form of a brief foreign news, in various Danish newspaper (Jyllands-Posten, Politiken, Dato and Berlingske Tidende), the day of or the day after the voting of the resolution.
 The opponents to Quebec’s independence refer to the Bloc Québécois as a ‘separatist’ movement, while its advocates speak of a ‘sovereignist’ movement. This terminology colours the daily language of Canadian politics and is not without echoing the kind of distinction made between ‘terrorists’ and ‘freedom fighters’. The expression ‘independentist’ is more politically neutral and used in both Canadian French and English. For the rest, the translations into English that are offered here are only indicative and for the sole benefit of the reader.
 The knowledge of the participants towards the foreign country was briefly assessed in an oral questionnaire at the beginning of the interview. As expected, all participants had no or very minimal knowledge about the foreign country.
 In the various Danish newspapers in which the Canadian story appeared, both ‘nation’ and ‘folk’ (meaning ‘people’ or ‘peuple’ in French) were used indiscriminately. This is, however, misleading with respect to the Canadian political context. For the first, it is widely recognised that the Canadian federation results from the union of the two founding people of Canada: The French and the British. Second, precisely because the word nation relates to nation-state has it become such an object of controversies in Canada. It is the ambition of the independentist party in Quebec to become a nation-state, and its recognition as a (cultural) nation by the Canadian parliament has been understood as opening the door to this possibility. This is only one example of the cultural context in which the word nation is controversial in Canada. In that respect, the word ‘people’ is not the subject of a controversy that could motivate the passing of a resolution in parliament. Additionally, Danes are a nation in the sovereign, cultural, and ethnic senses of the term. While there is no relevant distinction to be made between these senses in the Danish context, the gaps they create in Canada is an important source of problematisation to understand what the news text is about. Using the word ‘folk’ in Danish does not only misrepresents the local context of its use, but it also greatly diminishes the point of the news story. A word like nation does not simply point to a specific cultural meaning; it points to a large web of implicit theories and assumptions about the world, which some scholars would readily identify with ideological knowledge (Hanks, 1996; Gee, 1999), that are used to problematise a text.
All in all, the challenging issues that this project raised with regards to translation do not compromise the claims suggested in this article; on the contrary. Pragmatics does not draw a straight line between language and culture, as language is always something that people use in a cultural context. Rather, language and culture permeate each other (Hanks, 1996). When nation is readily associated with nation-state, it may seem as if the Danish recipients were ‘linguistically limited’ in their comprehension, rather than ‘culturally constrained’. Yet, along a pragmatic understanding, this amounts to making the same argument from a different side of the coin. Danes associate such meaning to the word nation because of the use they make of that word in their cultural context.
 I wish to thank Kim Schrøder for having pointed this out to me.
 Here, I do not want to imply that textual analysis denies the possibility of identifying a cultural dimension to texts, but that it avoids doing this from the perspective of the recipients, and hence denies the significance of this perspective for the objectives of an analysis of cultural mediation. Textual analysis, such as critical discourse analysis (Fairclough, 1995), may only serve to express the reading grid of the analyst, not the ones used by actual recipients; a point raised also by Schrøder (2007) and Widdowson (2004); the latter directly inscribing his work in a contextual approach to discourse.
David Mathieu is a Ph.D. student in the Department of Communication, Business and Information Technologies, Roskilde University, Denmark.