Abstract
Many studies of structure in present-day Western news texts have shown that the dominant structure is the inverted pyramid, even if the use of a chronological narrative structure is acknowledged. However, the relevant literature has exclusively investigated Western news texts. In this study, we challenge the dominance of the inverted news structure by including a non-Western and less-investigated culture and ask whether textual structure of news texts can differ between cultural contexts. In total, 100 crime news texts from national Iranian and Dutch newspapers were analyzed for both the overall text and the journalists’ statements. Those texts that showed a hybrid structure (combining inverted pyramid and chronology) were subsequently studied qualitatively to understand how chronology is applied in them. News texts from the two cultures were significantly different in their structure. Qualitative analysis of the journalistic statements uncovered a quotation-based structure, a way of news writing in Iranian journalistic statements in which the journalist mainly narrates reporting eventualities that are new to Western eyes. The study offers implications for further ethnography of news and communication, suggesting that the role, pattern and effects of (journalistic) communication can be culture-specific.
Introduction
Media in general and news in particular must be considered as special ‘types of language use or text and as specific kinds of sociocultural practice’ (Van Dijk, 1988a: 2). News texts represent and share the most newsworthy events through familiar and conventionalized ways of communication in a specific society. The representation of information is accompanied by (re)imposing institutional and socio-cultural ideologies. In fact, journalists ‘do more than make an event public’ (Tuchman, 1976: 94). Regarding the power of media texts in society (Peterson, 2008; Wodak and Krzyzanowski, 2008), they re-construct ideologies and communication strategies, in turn, thus playing a ‘performative role’ in society (Broersma, 2010). In order to establish ‘authenticity and truthfulness’, journalists must know not only what to present to the public – which news events and situations – but also how to present these in the journalistic texts (Broersma, 2010; Shim, 2014). This is specifically relevant in hard news, that is, reporting articles that provide recently occurred news events involving serious issues such as politics and crime.
The constitution and construction of ideology and communicating conventions are observable in different aspects of a text. Among the textual dimensions that ‘matter the most’ in constructing meaning in media discourse is textual structure (Carvalho, 2008; Schudson, 1995; Shim, 2014; Van Dijk, 1985b, 1988b). Structure, including the ordering of news events in text, has both cognitive and social implications (Van Dijk, 1986; Van Dijk and Kintsch, 1983). Not only does it reveal ‘the interpretation of social reality that is voiced in a newspaper’ (Broersma, 2007: xxiv, 2010: 23), but such ‘organization of knowledge’ also reflects and reshapes the cognitive schemata of news producers and news consumers (Van Dijk, 1986), providing ‘an important condition for the effective cognitive processing of news texts by both journalists and readers’ (Van Dijk, 1985b: 70). The structure used to report social events influences readers’ interpretation of the reality (see Broersma, 2010: 20). Discourse structure of a news text affects readers’ comprehension, enjoyment and engagement (Knobloch et al., 2004; Van Krieken et al., 2015; Zerba, 2008). Despite their importance, news writing in general and news structure in particular are still highly ‘unexplored’ and deserve further investigation (Shim, 2014; Van Dijk, 1985b, 1986).
Discourse structure in news texts
In news-reporting articles, events are traditionally structured based on temporal and relevance criteria among other factors (Van Dijk, 1985b, 2009). Temporal criteria indicate the ordering of the events based on the temporal continuum, and relevance implies ordering of the news events based on their importance and newsworthiness. The latter is recognized by answering the main journalistic questions ‘who’, ‘what’, ‘where’, ‘when’ and sometimes ‘why’ and ‘how’. Defining the initiating or the main event is not always easy or possible since ‘a specific event can have different functions’ depending on the ‘story context’ (Knobloch et al., 2004: 261). Both journalists and researchers in the field of communication studies have discussed how events can and must be structured in a news text. In the following, we discuss three main discourse structures that can be used for presenting news: inverted (non-chronological), chronological and hybrid (mixed). In the literature, different researchers have dealt with relevance and temporal criteria differently. In this article, we include both in our definition of the structures. In the literature, sometimes the term ‘a-chronological’ is used. That usage is confusing since it can refer to both a structure that is ‘not chronological’ and one that has ‘reverse temporal ordering’. Here, we use the term non-chronological to refer to a ‘not (predominantly) sequential’ structure and ‘a-chronological’ to mean ‘totally reverse temporal ordering’.
Inverted
In the Anglo-American approach, which ‘spread across Europe and the rest of the world’ from the 1890s on (Broersma, 2010: 27) and which is still dominant in many newsrooms worldwide, journalists follow an inverted structure, which puts the most important events (typically the most recent ones) at the beginning and the least important ones at the end of a news text (e.g. Høyer and Pöttker, 2005; Ryan and Tankard, 1977). An a-chronological news text is rare. It could only occur if all of the earlier events were less important to the news story. Usually, some of the background events (that have occurred earlier) are more important than some of the more recent events, disrupting an a-chronological order (also see Van Dijk’s presentation of news schematic structure). It is also uncommon that all of the important events occur at the beginning. Usually, journalists keep a few important events to narrate in the later part of the text in order to make readers follow the whole text. News texts may not be totally inverted, but they can be predominantly inverted. In such an inverted structure, most of the main events are placed at the beginning of the text and the news events are presented non-chronologically (also see Bell, 1998; Delin, 2000; Van Dijk, 1986). Example 1 presents such a structure in a constructed but typical example news text:
Example 1
Robber Faces Arrest.
A 26-year old man was arrested in Michigan a week ago [E6] on the night that he had robbed money from people in a restaurant [E4]. A witness called the police [E5] soon after James entered the restaurant [E2] and threatened people with knife [E3]. He had run out of money [E1].
In this text, the lead presents the main information, including the main news event mentioned in the headline (i.e. the arrest). The first sentence provides enough information for the reader to understand the main news events by finding answers to the main journalistic questions: a 26-year-old man (who) was arrested (what) in Michigan (where) 1 week before the time of publishing (when) because he had robbed money (why). The text then continues by ordering the events in an inverted structure on the basis of relevance and temporality. It begins with the main event, continues with an explanation of what happened in detail and ends with the reason behind the robbery.
Such an inverted structure can be illustrated as in Figure 1. In this figure, the letter ‘E’ represents the events in the news text, the numbers indicate the actual temporal ordering of the events, and discourse time is represented as the horizontal arrow. E6 represents the main event.
Inverted structure.
An inverted structure is information-based and likely to prevent journalists from imposing their opinion into news stories (Shim, 2014). Mindich (1998) asserts that an inverted structure, together with neutrality in style, enables journalists to achieve the main journalistic principle of objectivity. The inverted structure is specifically used for writing hard-news texts and in newspapers, where the primary goal is to give information and not to entertain the audience. This structure is editor-friendly in that editors can omit the last parts of the text according to limitation of space, and it is reader-friendly since it saves the news readers’ time: people do not have to read a text fully in order to understand the main news events (Bell, 1998; Kolstrup, 2005; Pöttker, 2010). Pöttker (2010) proposes that the inverted structure ‘enhances the communicative quality of the journalistic product’ (p. 509).
Chronological
News can also follow a chronological structure, in which the most important and recent events are put at the end of the text and events follow each other chronologically (e.g. Bragg, 2000; Neveu, 2014; Vare, 2000), as in personal narratives (see Labov, 1972; Labov and Waletzky, 1967). Given the high status of objectivity and the problems of using a chronological structure for narrating complex news (Kolstrup, 2005), contemporary news may rarely be totally chronological. But news texts can still be predominantly chronological in the sense that most of the events follow each other temporally and the most important events do not appear at the beginning of the text. We adopted the discourse structure of the excerpt in Example 1 to reach such a chronological structure in Example 2:
Example 2
Robber Faces Arrest.
A 26-year-old man who had run out of money [E1] entered a restaurant [E2]. He threatened people with knife [E3] and robbed their money [E4]. The police could arrest him in Michigan a week ago on the night of robbery [E6] due to the call that a witness made [E5].
Unlike Example 1, in Example 2 the main event and answers to the main WH questions are not placed at the beginning but in the last part of the text. The text follows a more or less chronological order considering the temporal ordering and relevance of the events. Such a structure is presented in Figure 2.
Chronological structure.
Unlike an inverted structure, which manipulates the original structure of events, a chronological structure provides a natural way of narrating social events. A chronological structure is story-based, interesting and entertaining and therefore contributes to keeping the audience’s attention tuned toward newspapers – a pressing issue due to the rise of new media (Shim, 2014).
The audience-based aspect of a chronological structure has lead journalists to reconsider this structure in recent news. While some suggest that texts written in chronological structure must not be classified as (hard) news stories and that hard news is identified by non-chronological style, (e.g., Ricketson, 2004; Fulton 2005), other researchers found that ‘the label [hard news] does not specify format but content’ (e.g. Knobloch et al., 2004: 283). The results of experimental studies demonstrate that a chronological structure increases the readers’ enjoyment, engagement (e.g. Knobloch et al., 2004: Van Krieken et al., 2015) and comprehension (Zerba, 2008) while reading hard-news texts.
Hybrid
Although inverted and chronological structures have been discussed mostly in the literature, these are not the only options in contemporary news writing. News texts may have a structure that is hybrid (Van den Toren, 2010). The hybrid structure has hardly been the advised form of news writing in Western journalism during the past decades. In hybrid texts, chronology and relevance are not mutually exclusive, but instead are combined: hybrid texts have some characteristics from either of the other two types. Example 3 presents such a hybrid text:
Example 3
Robber Faces Arrest.
A 26-year-old man who had robbed people’s money in Michigan a week ago [E4] was arrested on the same night [E6]. He had run out of money [E1] so he entered a restaurant [E2] and threatened people with knife [E3]. A witness called the police after the robbing [E5].
This example has combined characteristics of the previous ones. The main event is placed at the beginning, and the text follows a more or less chronological structure. Figure 3 presents one possible type of hybrid structure, as defined here.
Hybrid structure.
Texts with a hybrid structure offer an interesting possibility for present-day news writing. Given the beneficial aspects of the inverted and chronological structures, it can be claimed that a hybrid structure, which puts the most important information at the beginning of the text and tells the news story more or less chronologically, is most suitable to fulfill readers’ needs for both immediacy and emotional engagement. This is, of course, a hypothesis that needs to be tested in further research.
Voice: Mixed voice and journalist’s voice
A journalistic text echoes different voices. Alongside their own words, journalists provide their readers with information quoted directly or indirectly from other social actors as additional news sources in order to credit their story (see Bell, 1991). A news text is consequently a story made up of the journalist’s ‘statements’ (i.e. non-quoted sentences) and indirect and direct quotations from other news actors (see Delin, 2000). Therefore, journalists are responsible for structuring two strands: they decide how to present their own wording, that is, the journalistic statements, and they make decisions about how to structure their statements in relation to the information presented by other sources in the overall text. A valuable approach to the analysis of news texts is to investigate the journalistic statements to gain insight into the structural conventions applied by the journalist, as well as to analyze the overall text which the readers are exposed to. In their diachronic analysis of Dutch news, Vis and colleagues (2012) found that news texts had become more subjective over a period of more than five decades. Equally important, they found that the subjectivity did not lie in the journalistic statements. Obviously, ignoring a separate analysis of the journalists’ statements in the news extracts and only focusing on the overall text could lead to an incomprehensive interpretation or even a misinterpretation of the reality. Although journalists can implicitly insert their personal or institutional ideology into the text by selectively using quotations from other social actors (e.g. Chen, 2009), their voice is most audible in their own narration (Sanders, 2010). Separate analyses of journalistic texts also help us understand journalists’ role in providing information to the readers and the extent to which they may interfere in reporting events by interpreting social reality.
Contextual influences on discourse structure
Communication and discourse as a communicative process are highly intertwined with culture (Carbaugh, 2007; Gumperz and Hymes, 1964; Hall, 1982; Van Dijk, 1985a, 1985b; Xu, 2005; also see Keating and Duranti, 2011). Media discourse is hardly separable from the cultural context in which it is produced since ‘[…] media producers are selecting cultural elements from the pool of public culture, transforming them and returning them through some medium, as text, to public space’ (Peterson, 2008: 59). News is a socio-cultural exchange of the most important social events; not surprisingly, it is passed from news producers to the public in structures familiar to them. Given the socio-cultural diversity, it is reasonable to expect that news discourses differ in both content and structure from one culture to another. The Anglo-American approach, for instance, ‘only became accepted when adapted to national cultural standards’ in Europe (Broersma, 2007: xi–xii). Undertaking comparative and cross-national analyses of journalistic texts and exploring non-Western societies contribute to a better understanding of the different possible journalistic conventions and communication strategies employed in other cultures (Hanitzsch, 2009; Peterson, 2008; Wasserman and de Beer, 2009).
The structure of news texts, as Van Dijk (1985b) mentions, is ‘not arbitrary but a result of social and professional routines of journalists in institutional settings’ (p. 70). Cultural context is central to the study of news structure (Shim, 2014; Van Dijk, 1985a, 1985b) for various reasons. First, the very basic journalistic principles such as objectivity are understood differently by journalists around the world (Donsbach and Klett, 1993). It is reasonable to assume that, consequently, journalistic conventions and instructions vary across countries.
Second, social practices and behavioral patterns can be culture-specific (Hall, 1982; Hofstede et al., 2010; Kaplan, 1966, 1997). For instance, contrastive rhetoric analyses demonstrate that ‘speakers of two different languages will organize the same reality in different ways’ due to different thought patterns (Kaplan, 1997: 32). As an example, in a French speaker’s writing a ‘much greater freedom to digress or to introduce extraneous material’ is observed than in an English speaker’s writing (Kaplan, 1966: 12).
Third, both institutional and socio-cultural perspectives toward news can differ from one society to another. For instance, if a society expects news to fulfill its entertainment needs while another one expects it to be exclusively informative, news in the former may be told in a chronological structure and in the latter in an inverted structure.
Finally, the socio-cultural status of journalism professionals differs between societies. While in some societies journalists are part of the in power, having high social status and professional backing, in others they are rather out of power, being less supported. It is quite conceivable that, as a result, in the former context the journalistic voice is more perceptible, whereas in the latter context the representation of eventualities is often attributed to news agencies and officials so as to apply mutual value to both the story and the authority (also see Bell, 1991).
In sum, there are several contextual factors that may affect the discourse structure of news texts. Although it is not possible and not even relevant to address all of these factors in an individual study, an awareness of such influence is required (Van Dijk, 1985b). Such awareness leads to scrutinizing various structures that journalists around the world use to tell their story, and to gaining more insight into the reasons behind preference of one structure over an other.
Research on chronology and relevance in news texts
In their analysis of news discourse structure, Van Dijk (1985b, 1988a) and Bell (1991, 1998) showed that news events are ordered non-chronologically and based on their relevance. Delin (2000) conducted a bottom-up analysis on a hard-news text by segmenting clauses and comparing the order of events presented in the text (i.e. discourse structure) to the order of the events as they actually occurred (i.e. event structure). She found that, as discussed by Van Dijk (1985b, 1988a) and Bell (1998), hard news follows not a chronological but rather an inverted structure: the most important and recent events are placed at the beginning, and the text follows a non-chronological structure.
In a diachronic analysis of old and new Dutch news, Van den Toren (2010) found that the inverted structure is dominant in contemporary Dutch news, followed by some hybrid and very few chronological texts. Following Høyer and Nonseid (2005), she used an analytical model based on temporal and relevance criteria. For our goal, however, this model has two deficiencies. First, the categories are defined as fully chronological versus inverted, which means applying more strict rules to define the chronological category that may lead to finding fewer cases of this structure. Second, a chronological structure is assessed only on the basis of temporal ordering and the inverted structure is defined only based on relevance (see Van den Toren, 2010: 36).
Following Mindich’s (1998) definition of objective news (see earlier), Thomson et al. (2008) compared English, French, Japanese and Indonesian newspapers for their inverted structure and neutrality. Their approach is novel not only in taking a cross-cultural perspective, but also in exploring non-Western news texts in their analysis. They found that the inverted structure is used in non-English-speaking cultures as well as English-speaking ones. However, they only provide the reader with a qualitative analysis of news texts; therefore it remains a question as to whether inverted is dominant cross-culturally.
Others have studied textual structure as part of narrative or inverted news writing. Johnston (2007b) studied news texts in two metropolitan Australian newspapers and found that although both chronological and hybrid structures are present in news writing, current Australian news texts mostly follow the inverted structure. In a later analysis of the same newspapers, Johnston and Graham (2012) replicated the study by Johnston (2007b) and found an even stronger preference for the inverted structure. Interestingly, they found that crime news texts tend to be written in narrative structure.
In conclusion, most studies have been dedicated to investigating Western news, which makes it difficult to compare discourse structure across languages and societies (Johnston and Graham, 2012). Although it has been proposed that inverted is the dominant structure in news writing (see Bird and Dardenne, 2009; Johnston, 2007a), none of the available studies have compared the dominance of discourse structure across non-Western cultures, and therefore this proposal remains a hypothesis (Thomson et al., 2008). Finally, all of the above-mentioned studies have analyzed the events mentioned in the overall text and none have studied the journalistic statements separately.
Present study
In this study, we aim to fill some of the existing gaps by taking a cross-cultural approach. Hard news, including crime reporting, is supposed to be non-subjective and merely informative (Delin, 2000; Pöttker, 2010). Crime news is essential news with a broad audience, occurring in all cultures, having significant societal impact; it is more socio-culturally loaded than topics such as natural disasters. Among different sources, newspapers are considered as ‘social maps’ (Broersma, 2007), and traditionally they are a major source for crime information (see Koetsenruijter and Van Hout, 2014). We compare the discourse structure of crime reporting articles published in national Iranian and Dutch newspapers.
The case of Iran and the Netherlands is relevant for their journalistic as well as socio-cultural differences. While Dutch journalists have been encouraged to write in inverted structure for a long time (e.g. Donkers et al., 2010; Donkers and Willems, 1998) and a formal instruction toward narrative writing in print media seems to be new to Dutch journalists (Blanken and De Jong, 2014; Van Krieken and Sanders, 2016), most of the Iranian reference books have traditionally advised different structures, such as chronological, inverted, surprise ending and a form of hybrid structure called ‘chronological plus lead’, as in Figure 3, which is sometimes suggested as the preferred structure for writing crime news 1 (Badiee and Ghandi, 2016; Mo’tamednezhad and Monsefi, 2013; Shokrkhah, 2004; also see Van Krieken and Sanders, 2016). In both Iranian and Dutch journalistic reference books, objectivity is valued and news is supposed to be written impartially. Also, the two countries are different in their socio-historical status of journalism (Broersma, 2011; Khiabany and Sreberny, 2009; Shahidi, 2006, 2008). In addition, the two cultures may have different preferred communication conventions and different institutional and socio-cultural ideologies and perspectives toward (crime) news. For instance, the two societies have shown diverse behavioral patterns on a number of cultural dimensions (Hofstede et al., 2010).
While Dutch news has been investigated for its structure, Farsi news is under-investigated, especially regarding its discourse structure. Most studies on the discourse structure of Iranian news (e.g. Ansary and Babaii, 2009; Bonyadi, 2010; Sabzevari and Sadeghi, 2013) have examined Iranian newspapers written in English. The problem with such studies is that first, English news texts are not aimed at an Iranian audience, which means that the news text may have lost specific features of the source culture in order to fulfill the target audience’s needs, and second, such cross-cultural studies ignore a central component of culture, namely, native language. No study has investigated the dominant structure used in Farsi news texts.
We propose the following questions:
1. What is the dominant discourse structure in Iranian and Dutch crime news texts?
Sub-question 1. Which discourse structure is dominant in Iranian and Dutch journalists’ statements?
Sub-question 2. Which discourse structure is dominant in the overall Iranian and Dutch news texts?
Regarding the different journalistic instructions and previous studies, we expect Dutch news texts to be written more in the inverted form and presume that Iranian texts use different structures. Furthermore, we will give a qualitative analysis of hybrid texts in order to understand the nature of these texts, specifically focusing on the use of chronology in them.
2. What is the chronological nature of hybrid crime news texts?
Sub-question 1. How do Iranian and Dutch hybrid texts differ in the way chronology occurs in the journalistic statements?
Sub-question 2. How do Iranian and Dutch hybrid texts differ in the way chronology occurs in the overall text?
Method
Corpus
In total, 100 crime-reporting news texts, 50 Farsi and 50 Dutch, were collected and analyzed. The corpus was collected from LexisNexis and MagIran, online databases including the archives of newspapers and magazines, by searching for five main keywords: child abuse (kindermisbruik, آودآ آزاری), murder (moord, قتل), crime (misdaad/misdrijf, جرم/جنایت), kidnapping (ontvoering, آدم ربایی) and rape (verkrachting, تجاوز). The news texts were published during a period of 1 year (from July 2013 until June 2014) in four Iranian newspapers, namely, Etemad, Jam-e-Jam, Shargh, Iran, and four Dutch newspapers, NRC Handelsblad, de Volkskrant, De Telegraaf and Algemeen Dagblad. While Etemad, Shargh, NRC Handelsblad and de Volkskrant can be grouped as quality newspapers, Iran, Jam-e-Jam, De Telegraaf and Algemeen Dagblad are considered more sensational. All of the newspapers are among the high-circulation national papers.
Model of analysis
We analyzed the structure of each text by comparing the events in the clauses (following Delin, 2000). A clause was defined as a minimal grammatical segment including a tensed verb that presents a specific action at a certain time and that can be rephrased as a separate independent clause. The tensed verb indicates the main event of the clause. This simplified definition makes the study of both languages possible and prevents the analysis from becoming complicated by too many events. Contrary to Delin’s approach, both events and states were regarded as eventualities and were not distinguished since these are interwoven in a text and construct a news story together. Both temporal ordering and relevance were set as the criteria for finding the structure of the texts. Temporal ordering clarifies the sequential relationship between eventualities, and relevance criteria specify the ordering of the eventualities based on their importance, that is, giving the answer to the most important journalistic questions. These questions are as follows: (1) What happened (the main news event)? (2) Who was involved? (3) When did it happen? (4) Where did it happen? (5) Why did it happen? (6) How did it happen? The information given in the headline was excluded in the decision about chronology and WH events.
Based on how these two criteria work in a text, the text falls into one of the following categories:
Inverted: A news text has a predominantly inverted structure if most of the WH information (answers to the six journalistic WH questions) is presented in the first third (i.e. the beginning) of the text and if it is not the case that most (more than 70% of the clauses) of the text follows a chronological order.
Chronological: A news text has a predominantly chronological structure if most of the events (more than 70%) are a subsequent event of the previous one and most of the WH information is not given in the first part of the text.
Hybrid: A news text follows a hybrid structure if most of the events (more than 70%) are ordered chronologically and most of the WH events are placed at the beginning of the text.
This model is a modified version of Van den Toren’s (2010) model of analysis. The model was applied both to journalistic statements and to the overall text. Journalistic statements were extracted from the overall text by ignoring the quotations, both direct and indirect ones. The latter excludes indirect quotations mentioned by non-verbalized structures such as ‘according to’. The main reason for following this strategy is that in the Iranian corpus, these structures were mainly used for attributing the story to the news agencies and it was not clear whether the journalist was attributing only the source-containing sentence or the whole news story to the mentioned agency.
The fact that hybrid texts are chronological as well as inverted makes them interesting cases for further qualitative analysis. In a qualitative analysis of the texts, we give an explorative inductive analysis of how chronology is applied in hybrid texts. Hybrid texts were classified into different chronological patterns, demonstrating at which layers chronology is used and for what purpose. Like the previous phase, the qualitative analysis was made separately for both journalistic statements and the overall text.
Procedure
For each text, several steps were followed. First, we determined the main event of each clause presented by the tensed verb. Then, the actual temporal ordering of each eventuality was written in front of each clause, thus explicating the event structure, after which the discourse structure was compared to the event structure (cf. Delin, 2000). Later, each text was read carefully to see whether the WH events were placed at the beginning of a text. The same procedures were followed for both the whole text and the journalist’s statements.
Below (in table 1), we provide an example analysis of a news text from Shargh. The text (and other news texts provided in this study) is not a word-by-word translation to English, but we have maintained the same clause order in order to save the discourse structure of the text. 2 The abbreviated letters may be read as S = sentence, C = clause, J (in the ‘Source’ column) = journalist’s statements and DQ = direct quotation. ‘E-All’ presents the position of the clause in the structure of the overall text. ‘E-Journalist’ does the same for the journalist’s statements. The numbers indicate temporal ordering of the eventualities as they actually happened (i.e. the event structure).
An example analysis of a news text (from Shargh, 7 May 2014).
By reading the very first sentence of the journalistic statements, we get the main news events, answers to the main journalistic questions: a woman (who) suffocated her daughter (what and how) on the street in Tehran (where) because she had become angry (why). The text is chronological, too: the clauses follow each other in a fully temporal order. The overall text is also chronological.
As to the qualitative analysis, in this news text, chronology is used by the journalist to convey information about the main news events. That is, if we look at journalistic statements only, we will be informed about the main news events. By ignoring the quoted clauses we miss some details about the story, but we can still access relevant information since much information is given in the journalist’s statements. Also, we observe chronology at three different layers: the overall text, the journalistic statements and the direct quotation are all narrated in a chronological structure. This news text can therefore be considered as a meta narrative in which micro-chronological stories, including the journalistic report of the events and the mother’s confession, are embedded.
Statistical treatments
In all, 40% of the Dutch texts and 40% of the Farsi texts were re-coded by two coders for both clause segmentation and structure of texts. Dutch data were coded by a native speaker, one of the native authors (W.S.) and the near-native author (A.R.), and the Farsi data were coded by the native-speaker author (A.R.) and a near-native speaker of Farsi. The inter-coder reliability for clause segmentation was excellent between the coders for both Farsi and Dutch texts (κ = 1). Inter-coder reliability for text structure was perfect for both Dutch (κ = 0.87, p < 0.001) and Farsi (κ = 0.82, p < 0.001). The significance of the use of discourse structure in the two countries was tested by measuring Fisher’s 2 × 3 exact test.
Results
In order to find answers to the questions, the full sample (N = 100) was analyzed. Our first question concerned the dominance of textual structure in the two countries: Which structure is dominant in the journalist’s statements in Farsi and Dutch crime news? And which structure is dominant in the overall text in the two countries? The results are presented in Table 2.
Distribution of discourse structures in the overall text and in the journalist’s statement in Farsi and Dutch news texts.
Fisher’s exact test showed that the structures used in journalists’ statements differ significantly between the two countries (p < 0.01), that is, journalists structure their statements differently. In Iranian news texts, hybrid is the dominant pattern in journalists’ statements. This indicates that chronology, as well as invertedness, is employed in Iranian news writing. As hypothesized, the dominant pattern in Dutch texts is inverted. In Iranian texts we see chronological structure, though very few, but this pattern is not observed in Dutch news texts.
What if we consider the overall text? Another Fisher’s exact test shows that the structure of Farsi and Dutch news texts is significantly different in the overall texts as well (p < 0.01). Are quotations (as personal narratives) the cause of the chronology of Iranian news texts? The answer is no: the number of hybrid texts does not increase, but it decreases when including the quotations. In sum, the analyses confirm that the hybrid use of chronology and invertedness is a frequently occurring pattern in Iranian news texts.
The outcome of the analysis of Iranian news texts does not support the claim that ‘the conventions of organizing such [i.e., news] texts include a strong tendency towards breaking the chronology of the series of events in question’ (Virtanen, 2011: 15). Rather, if we consider the use of chronology in hybrid texts, these results support the findings of previous studies that crime news tends to be written in a chronological structure (Johnston and Graham, 2012). Given that in both inverted and hybrid structures the most important events are presented at the beginning of the text, news texts from both countries are similar in this sense, that is, using a lead. This finding is in line with those from previous cross-cultural studies (Van Dijk, 1988a). Therefore it can be concluded that in news writing chronology can be culture-specific, whereas using a lead is universal.
The nature of hybrid texts
Our second research question concerned the nature of hybrid news texts, specifically, the use of chronology in them: How do Iranian and Dutch hybrid texts differ in the way chronology occurs in the journalistic statements? How do Iranian and Dutch hybrid texts differ in the way chronology occurs in the overall text?
Hybrid journalistic statements
By distinguishing between journalistic statements from the overall text and analyzing both events and states as eventualities, we were able to describe two formats of news writing regarding the reportage content. In the following, we discuss each with some examples. In four Iranian news texts, a combination of both types was observed.
In the first format, the journalist puts the most important events at the beginning and then narrates the news events following a chronological structure. All (9) of the Dutch and around half (18 out of 39) of the Iranian hybrid journalistic statements follow this pattern. The information-based hybrid structure appears in Figure 4.
Information-based structure.
As we see in an excerpt from Algemeen Dagblad in Example 4, all the necessary information is given to the news reader in a hybrid structure in the journalistic statements. That is, the most important events are placed at the beginning: Two people from Eindhoven (28 and 31 years old) were sentenced to prison stays of 8 and 9 years for the abduction and rape of a young woman from Turnhout, followed by the details and other main events presented to the reader in a more or less chronological format. Ignoring the quoted clause (printed in italics) does not cause a misunderstanding of the minimum information given on the main events. Although we will not have all the story details, we still find the most important events in most of the journalistic statements (see Table 1):
Example 4
Long Convictions for Brutal Rape
Eindhoven | Two people from Eindhoven (28 and 31 years old) were sentenced to prison stays of 8 and 9 years for the abduction and rape of a young woman from Turnhout [E7]. The older suspect also receives TBS 3 [E7]. The court in Den Bosch determined the sentences yesterday [E6]. The sentences matched those sought by the Public Prosecutor [E?]. The then 19-year-old victim was pulled into a car in her home city of Turnhout on New Year’s Eve in 2012 [E1] and blindfolded [E2]. After she was raped several times in a house in Eindhoven [E3], she was showered [E4] with the purpose of erasing evidence and in the morning she was left on a deserted path in Waalre [E5]. The judge has stated [E6] that the events were deeply humiliating and frightening for the victim [E1–5]. The criminals will pay the victim over 20,000 euro in compensation [E7]. (Algemeen Dagblad, 29 October 2013)
The other format presents a quotation-based structure, a form of news writing that up to now has not been described for Western journalism but occurs frequently and exclusively (17 out of 39) in Iranian news texts. In this format, chronology is used in journalists’ statements not to report central news events, but to state reporting events. That is, like the previous structure, the journalist gives the most important event at the beginning, but unlike the first mode, he or she does not report the events in the rest of the text but basically quotes other social actors involved in the news events. In this mode, information about the main events is mainly presented in the quotations, and the body of news text includes a chronological narration of reporting verbs (see Figure 5).
Quotation-based structure.
Although news events are quoted from different social actors, most quotations are attributed to the officials, which has implications for the socio-cultural context of Iranian society as well as the socio-political status of journalism professionals in the country. Drawing from cultural surveys by Hofstede et al. (2010), it may be presumed that Iran has a relatively high power distance. It could be the case that by using quotation-based news writing, Iranian crime journalists give more ground to social actors in power and tend to evaluate, interpret and report less from their own perspective in order to gain mutual value for both their text and the news-actors-in-power by quoting these social actors (see also Bell, 1991). The high amount of quotation-based structures indicates that this pattern is a common writing structure for Iranian journalists.
Below, we provide an example news text from Etemad, which presents a quotation-based format. If we take the quotations out from this text, what remains is the lead followed by reporting verbs, which include no major information about the main news events, put in chronological order:
Example 5
Police Chief of Sistan and Balouchestan:
7-year-old Zahedani Captive Has Been Freed
Fars| Police chief Mohsen Attar of Sistan-and-Balouchestan recently announced the arrest of four kidnappers and the release of their 7-year-old captive [E14]. Chief Attar stated [E15]: ‘on 11 Esfand a man went to the police station [E3] and reported [E4] that he had given his car to an acquaintance for safekeeping [E1] and that this person had kidnapped a seven-year-old child from school with the vehicle [E2]’. The police chief added [E16]: ‘due to the sensitivity of the issue, our department prioritized this investigation [E5] and by mobilizing all of our resources, we began a large-scale effort for recovering the kidnapped child and arresting those responsible [E6]’. Implying the identification of the hide-out of the accused kidnappers, he stated [E17]: ‘by synthesizing information, the police officers identified the location of the kidnapped child [E7] and after an investigation for verifying the presence of the hostage and the accused at this location, they took the necessary actions for the arrest and rescue [E8]’. He mentioned [E18]: ‘the police officers entered the hide-out in a covert operation, [E9] arrested four kidnappers [E10] and delivered the child uninjured back to her family’s arms [E11]’. Attar added [E19]: ‘during further investigation the car was also discovered [E12] the principal suspect mentioned having a financial dispute with the father of the kidnapped child as his motivation [E13]’. (Etemad, 6 March 2014)
Hybrid overall texts
Hybrid overall texts fall into two main categories based on the layer(s) at which chronology is applied: continuous and fragmented. In a continuous structure, the chronological pattern is observed at only one level: the overall text. No other chronological pattern is found. Two out of the eight Dutch texts and three out of the 26 Iranian texts have a continuous structure. This structure is illustrated in Figure 6 where ‘N’ stands for narrative.
Continuous structure.
Below, we provide a continuous hybrid text from de Volkskrant. As we see, the whole story is told in a more or less chronological structure. Chronology is observed only at one level (the whole text) and no other chronological story (structure) is embedded in the text:
Example 6
Woman Assaulted by Six Men
Haarlem
On Friday in Haarlem, six men assaulted a 20-year-old woman from IJmuiden [E5]. She initially entered into a discussion with two of the men [E2] who then got off the bus at the same stop as her [E3]. The woman was waited for by four men [E4] who were part of the same group as the couple from the bus [E1]. The six men (presumably of North-African descent) kicked the woman to the ground [E5]. However, she was able to escape [E6]. The police are looking for witnesses [E7]. (de Volkskrant, 7 October 2013)
Unlike the continuous pattern, in a fragmented structure more than one chronological pattern is observed. Chronology is observed in the overall text and more: in the lead, in the journalist’s statements and/or in the quotations. In other words, in fragmented texts different chronological stories are embedded into a larger story that makes up the overall news text. Fragmented structure is observed in both Dutch (6 out of 8) and Iranian (23 out of 26) hybrid texts. Figure 7 demonstrates a fragmented structure.
Fragmented structure.
As we see in Example 7 from Jam-e-Jam, chronology is observed in this news text at four different layers: the whole text is more or less chronological; moreover, the lead (L), journalist’s statements (Sa and Sb together) and the quotation (Q) are all chronological (see Figure 7). In the lead, we read that first, the child had become brain-dead and then that her body organs were donated to patients. In the following journalistic statements, we are informed about the news events from when the child went into a coma to when the doctors found no more hope for her return to life. Then the statements are interrupted by a chronological quotation from the victim’s mother, who chronologically explains the story from the beginning until when the child was transferred to the hospital, after which the statements continue to explain the official procedures and come to the present, that is, the reporting time in a chronological order. In fact, events are narrated in three micro stories, each structured in chronological order and embedded into one overall more or less chronological news text. This structure may offer a more appealing meta story to readers since it includes different chronological micro stories:
Example 7
2-Year-Old Girl, Victim of Father’s Anger
The internal organs of a two-year-old girl [E11a] who became brain dead after a beating from her father [E4] were recently donated to patients in need [E11b]. According to Tasnim, after transference of the two-year-old girl to a hospital in west Tehran, it became clear [E6] that she went into a coma caused by head trauma [E4]. After transference of the girl to Masih Daneshvari Hospital, the doctors determined [E8] that there was no possibility of recovery for the young girl [E7]. During an investigation, this child’s mother stated that [E9] her husband had become angry because of the child’s cries [E1] and beaten their daughter [E2] until she lost consciousness [E3] and she [the mother] then rushed the girl to the hospital [E5]. A criminal case being filed, a judicial order for the father’s arrest has been issued [E10]. According to the case report, following doctors’ disappointment of the child’s recovery, regarding her mother’s permission, this child’s organs were donated to patients in need [E11]. The body of the young girl was interred with deep grief by her mother and relatives [E12]. No report has yet been published regarding the father of the deceased child [E13]. (Jam-e-Jam, 20 January 2014)
Conclusion and discussion
News writing is still highly ‘unexplored’ (Shim, 2014; Van Dijk, 1985a, 1985b). While journalism practices in some cultures and languages have mostly been the topic of research, several aspects of others have remained unanalyzed. The performative role of journalism is conducted not only by the content but also through the structure applied in journalistic texts (Broersma, 2010; Shim, 2014). Textual structure requires further investigation, especially from a cross-cultural perspective (Johnston and Graham, 2012) and in non-Western societies (Thomson et al., 2008). In the present study, we compared the discourse structure of Iranian and Dutch crime-reporting articles published in national newspapers. As suggested by Van Dijk (1985b), we found that while some aspects of ordering the information in news texts ‘are fairly fixed and general’, others ‘may differ from culture to culture’ (p. 89).
Both the quantitative and the qualitative analyses confirm that at least some aspects of the news discourse structure are culture-specific. While inverted is the most used structure in Dutch news, the hybrid structure is the dominant one in Iranian news. The former confirms the finding of Van den Toren (2010), and the latter proves that inverted is not ‘the most popular’ (Johnston, 2007a, cited in Johnston, 2007b) or the ‘dominant’ structure (Bird and Dardenne, 2009) all around the world; the subordination of the inverted structure in Iranian news texts shows that this structure is not ‘the Dracula of journalism’ (DeSilva, cited in Scanlan, 2003). During the qualitative analysis of our data, we found a quotation-based structure of news writing only found in Iranian texts. This way of telling news is not recommended in Iranian reference books, but is systematically applied by Iranian journalists and is intriguing both for the use of chronology and for presenting reporting verbs instead of news events. These variations emphasize the importance and influence of cultural context and question the credibility of defining and prescribing one set of ‘professional’ journalistic practices to all cultures while degrading other possible options. Rather, journalistic conventions must be defined through a bottom-up approach based on the needs and cultural (though not necessarily socio-political or ideological) codes of the audience.
As mentioned earlier in this article, the analysis of journalistic discourse structure is not limited to finding out how events are structured in the report. It further ‘links the level of individual media to the general levels of journalism and society’ (Broersma, 2007: xxix). Regarding both production and consumption, ‘the cognitive and social implications of the nature and uses of news schemata require further attention’ (Van Dijk, 1986: 182), and this study raises inspiring questions for further research on these issues. The quotation-based structure has implications for news production that must be investigated in future studies, such as the role and the purpose of the journalist in producing such a structure. During this study we found some deviations from Western journalistic viewpoints in Iranian reference books. We suggest that a study on these different viewpoints is needed, most preferably including surveys among journalists.
Our results also have implications for future ethnographic studies of communication. Bearing in mind that news structure must be culturally accepted, it is possible that the two societies differ in their communication structures or socio-cultural ideologies toward crime news. Whether the differences between the news structures in the two countries come from different journalistic conventions, different preferred communication strategies, differences in socio-cultural ideologies toward crime news or different organizational ideologies is a topic of further study. Of course, the authors are aware of the fact that it may be difficult to differentiate between these factors since they can be highly interrelated.
Our research had strengths and weaknesses. By including news texts from a non-Western culture in our study, we came to noticeable results which broadened our view about possible journalistic as well as communication conventions and strategies. We analyzed both overall text and journalists’ statements from both quantitative and qualitative perspectives. Here, we only focused on textual structure. Whether Iranian news texts make frequent use of other narrative devices remains a question for future studies. Finally, we analyzed crime news texts only. Whether other hard-news topics, for instance political news, show the same patterns remains one of the questions raised in this study.
Footnotes
Acknowledgements
We would like to thank Chris van der Lee and Tahsin Osmani for their assistance in collecting and re-coding the data and Alexandra Carstensen for her help with translation of excerpts into English. Also, we would like to thank Ali Kafi for helping us access Iranian reference books. Finally, we would like to express our gratitude to the anonymous reviewer for her/his invaluable comments. Data are available upon request to the corresponding author by email.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.