Introduction

In research on intercultural rhetoric, the use of linguistic conventions to structure a text is of paramount importance. Equally important is the way authorial stance is constructed to achieve the objectives of writing (Pho, 2013; Peng, 2019). Intercultural rhetoric is concerned with the influence of the writer’s first language on the production of a text in the second language. Such an influence is explored by comparing writings in two languages. Of the most under-researched types of writing is the writing of newspaper editorials, which is used to express opinions and argue for certain positions. It is no accident that editors and members of the editorial board opt for certain stance markers in their editorials to unveil the mystery behind their own judgements and evaluations. That decision to pick a certain stance feature in a language is not a random one but rather a deliberate and thoughtful one based on the setting in which it exists. Such acts of evaluation or expression of opinion merely draw on a variety of different factors, including the newspaper’s ideology and the editorialist’s stance.

Editorials are the institutionally constructed voice of the newspaper. Therefore, a balanced, fair, and factual manner of reporting events is seen as something that most journalists and editors aspire to do in journalism. However, it could be argued that editorial writers do not have the option of excluding their views and their organisational cultures in the writing of their editorials. In fact, editorials can be less restricted with the absence of biases and prejudices as long as they completely concur with the stipulations of the institution. Therefore, editorials may influence one’s eyesight or sentiments of an issue over time. Richardson and Lancendorfer (2004), in their discussion of the framing theory and affirmative action in newspaper editorials, stated that ‘the opinion page is just that- a place where editors can bluntly state exactly what they think’ (p. 75). In broad terms, editorials have been proved to be a stance-rich environment where the ideology of the newspaper can be represented in journalism circles and to the public. For Hyland (2008, p. 5), stance means “the writer’s textual voice or community recognised personality”.

A straightforward reflection of opinion in an editorial targeted at the newspaper readership is not as simple as it seems. The editorial stance is generally seen to have a strong power in the process of agenda-building and political agenda-setting in their respective countries. Jaffe (2009, p. 8) pointed out that ‘stance attributions are tools of control and ideological domination’. Thus, editorial stance is believed to have a considerable influence on the ideologies, social attitudes and values of the community. Resulting of this, many studies have suggested a number of functions that stance performs in a written text (e.g. Biber and Finegan, 1988; Hunston and Thompson, 2000; Hyland, 2005; Dafouz-Milne, 2008). For instance, Biber and Finegan (1988, p. 2) stated that ‘a broad range of functions falls under the umbrella of ‘stance’ including expression of certainty, generalisation, and actuality’. It is worth noting in this respect that stance as a linguistic term was referred to in the literature using different labels, such as evaluation, appraisal, voice, and writer’s commitment. Hunston and Thompson (2000, p. 6), in their discussion of why evaluation has attracted broad interest, also identified three functions that evaluation is used to perform: expressing opinion, maintaining relations, and organising the discourse.

In this study, we compare the use of stance features in English and Arabic editorials to identify differences (if any) in how editorialists construct their stance in the two languages. The aim is to highlight the similarities and/or differences between both languages in the use of stance features, with a view to determining the factors affecting the use of stance in editorials. To achieve this objective, the study analyses editorials published in two well-known newspapers: The first in English (The Guardian, the UK), and the second in Arabic (Addustour, Jordan). These newspapers were selected based on their popularity and wide readership in their respective contexts. This study seeks to answer the following search questions:

  1. 1.

    How do newspaper editorialists construct their stance in English and Arabic newspaper editorials?

  2. 2.

    What are the similarities and/or differences (if any) in the use of stance markers in English and Arabic newspaper editorials?

Literature review

The literature on metadiscourse use abounds with studies that analysed different text types and genres (e.g. Abusalim et al., 2022; Alghazo et al., 2023a; Alghazo et al., 2021a; Rabab’ah et al., 2022; Rabab’ah et al., 2024). However, little research has been found on the use of metadiscourse in media discourse genres. For example, Chen and Li (2023) examine how two newspaper agencies, namely China Daily (CD) and The New York Times (NYT), use interactional metadiscourse to engage the readers and the similarities and differences in their usage. Specifically, the research focuses on identifying the rhetorical tools used by each agency and how/why they are used. The researchers used corpus analysis where the data was collected from 60 commentaries, 30 from each newspaper agency, which were published over the course of 6 months and the analytical framework used to analyse the data was derived from Hyland’s works on metadiscourse. The results show that both newspaper agencies used interactional metadiscourse, but their usage differed in terms of total frequency; the NYT used interactional discourse more. Secondly, in terms of the macro level, the NYT uses of stance and engagement were balanced, whereas the CD heavily focused on stance. Thirdly, in terms of the subcategories, the NYT used ‘hedges, self-mention, and engagement markers’ more frequently compared to the CD. As for ‘boosters and attitude markers,’ the differences were insignificant.

McCambridge (2022) examined how commenters on YouTube use stance to construe their voice while reacting to a video by compiling a list of 2000 comments on a viral video of Greta Thunberg made by 1949 users and analysing them using ATLAS.ti. The study states that from the data collected, the voice that seems to be construed is that of bullying. In terms of the frequency of the codified categories, attitude markers came first, followed by boosters, reader addresses, self-mentions, and hedging, respectively. The overall stance of the comments was negative and can easily be ‘[characterised] as bullying.’ The comments as a whole nurtured hate towards Thunberg through the use of sarcasm and insults, amongst other methods which fall under the above-mentioned categories. Finally, McCambridge (2022) extrapolates these results to the ‘wider social scale’ where group commentaries such as the one covered in this paper can be used to create an intimidating force against a given entity or figure.

Hyland and Zou (2021) studied the usage and function of stance in the genre of ‘Three Minute Thesis presentation (3MT)’. More specifically, the research focuses on analysing the usage of stance by the presenters, the differences that exist in said stances between the different fields, and the reasoning behind the differences using a corpus analysis of 140 presentations, which were transcribed from the video format. The presentation material covered multiple fields, including education, applied linguistics, history, and sociology. The corpus was categorised under hard sciences and soft sciences. Hyland’s model was used as the analytical framework, and AntConc was used to analyse the data. The results showed that in total, there were 4616 stance markers, where 2086 belonged to the social sciences, and 2529 belonged to the hard sciences. This difference was proven to be statistically significant. Moreover, when it comes to the subcategories, the hard sciences had more uses of boosters and hedges, whereas ‘attitude markers and self-mention were only slightly more frequent in the social [sciences]’. Finally, the hard science presenters used ‘more epistemic devices’ to support their claims, whereas soft science presenters used ‘a more affective and visible stance’.

Droz-dit-Busset (2022) explored the representation of Social Media Influencers (SMIs) by English news agencies by analysing the usage of the celebration and derision stances through the usage of ‘legitimation and delegitimation’ approaches. Droz-dit-Busset (2022) used a corpus analysis of 143 pieces, which included editorials, features, and opinion pieces extracted from a number of international English news agencies through LexisNexis, as well as pieces from the Guardian and the New York Times. The results show that for ‘celebration,’ techniques used included discussing positive metrics (view count, reach, etc.) as well as ‘narratives of [societal] upward mobility,’ where the SMIs are celebrated as individuals who went from zero to hero. As for the ‘derision’ stance, approaches to delegitimise SMIs include ‘negative moral evaluations’ regarding ‘their work [ethics]’ and referencing their ‘lack of institutional legitimation’.

Yazdani et al. (2014) investigated how articles from Persian and English news agencies utilise interactional metadiscourse, using a corpus analysis tool to analyse 30 articles (15 from Persian and 15 from English) and adopting Hyland’s (2005) system of metadiscourse. The results show that, in total, the English articles had a much higher frequency of interactional metadiscourse compared to the Persian articles, with a value of 70% compared to 29% respectively. In terms of the subcategories, English articles’ most frequently found markers were hedges (38%), followed by attitude markers (26%), boosters (18%), self-mentions (12%), and engagement markers (6%), respectively. As for the Persian articles, the most frequently used markers were the attitude markers (48%), followed by hedges (32%) and boosters (20%), respectively. Interestingly, the engagement markers and self-mentions were not present in the Persian articles, which led the researchers to conclude that these differences are attributed to the different writing styles of English and Persian writers, the writer-reader relationship dynamics, and the cultural differences between the two languages.

Fu (2012) examined how interactional metadiscourse is utilised in job postings, adopting a corpus analysis method to analyse a corpus consisting of 220 different job postings obtained from five sub-corpora of job postings, including that of ‘The Daily Telegraph’ and ‘The Guardian.’ These postings were further divided into postings made for college students and another for non-college students. Hyland’s (2005) model was used as the analytical framework, and the instances of interactional metadiscourse were categorised. Moreover, a questionnaire was given to 30 randomly selected students. The results show that, in total, the number of engagement features and stance features are similar, 1757 to 1804, respectively. Among these features, self-mentions (1145) and reader-inclusive pronouns (1497) are the most frequent. On the other hand, micro-level features (hedges/boosters) are not frequent. Personalisation is also a distinct feature, as displayed in the use of ‘we’ and ‘you’ when referring to the writer and the reader, respectively. As for the difference between postings made by college students and non-college students, interactional metadiscourse was more frequent in the postings by college students.

Al-Subhi (2023) investigated the use of interactional metadiscourse in the USA and UAE-based leading newspaper editorials, using a corpus analysis on a corpus consisting of 24 editorials with 12 from UAE-based news agencies ‘namely, Khaleej Times, Gulf Today, and The National Gulf,’ and 12 from USA-based news agencies namely, ‘The New York Times, Los Angeles Times, and Washington Times’. The analytical framework used was Hyland’s (2005) model of interactional meta-discourse. The results showed that both corpora had similar frequencies when using interactional metadiscourse. Moreover, both corpora used stance markers more frequently compared to engagement markers. As for the subcategories of stance, both corpora used hedges most frequently, followed by attitude markers, boosters, and self-mentions, respectively. Engagement markers were few, with only ‘5%–8%’ present in the corpora. Al-Subhi (2023) concludes by addressing how editorials ‘constitute a persuasive genre par excellence’ given their usage of ‘rhetorical and argumentative structures’ which ‘[aims] to influence public opinion and [express] the institutional voice of a newspaper’.

Shen and Tao (2021) examined the usage of stance markers in scientific medical research articles and newspaper opinion columns, using a corpus analysis on two corpora: 52 articles and 175 opinion articles from ‘The New York Times’. Hyland’s (2005) framework of stance was used as the analytical framework of the study. The results showed that stance markers were frequently used in both corpora; however, the number of markers in the opinion columns was nearly twice as many compared to the medical corpus, which in turn ‘[reflects] a high level of interaction between writers and readers’ in that genre. As for the frequency of the categories, hedges were the most frequent, followed by boosters and attitude markers; all these categories were more frequent in opinion columns compared to medical articles. Self-mentions were also used in both genres; however, their usage differed, with opinion columns using ‘first person singular forms’ and medical articles using ‘the first person plural forms’. The results also showed that there was some overlapping in the usage of markers and that even though stance markers were more frequent in opinion articles, there were some subcategories under the difference between the two genres.

In a recent study on the use of stance features in the editorial section of newspapers, Al-Anbar et al. (2023) compared the use of interactional features of metadiscourse in two groups of editorials: the first was written by native English authors, and the second by nonnative English authors, using Hyland’s (2019) framework. The findings showed that non-native editorialists used fewer hedges and more boosters than native editorialists, who were found to also use more engagement markers in writing. In a related investigation, Alghazo et al. (2023b) explored the interactive features of metadiscourse in 80 newspaper editorials written in first language (L1) English and second language (L2) English. The analysis showed that, in the main, there was no significant difference in the use of interactive features. However, a slight variation in the use of frame markers and evidentials was observed in the data analysed.

Methodology

The data were gathered from two broadsheet daily newspapers in Jordan and the UK, namely Addustour newspaper and the Guardian newspaper. This study is focused primarily on editorials which were published on the websites of the two newspapers between 2020 and 2021. Simply put, the present study gives an account of how stance is expressed in Arabic and English editorials. Therefore, it was decided that the best method to adopt for this investigation was to analyse the data both quantitively and qualitatively. A mixed-methods approach, combining both the quantitative and qualitative analysis of the data, was used to allow for a deeper insight into the use of stance markers by editors of both languages. Eighty editorials were divided into two sets; each set consisted of 40 editorials in each language. The data was collected from the two newspapers’ websites (Addustour and the Guardian) and then transferred to a Microsoft Word document. This study utilises Hyland’s (2005, 2019) model of interaction, which includes stance and features of writer positioning. As for Hyland’s (2019) typology of stance, features of writer positioning are of four kinds: hedges, boosters, attitude markers and self-mentions.

In order to identify stance markers in the two sets of editorials, a functional analysis was carried out to capture these markers within contexts. Studies of stance have traditionally relied upon the contextual analysis of texts in stance detection (e.g. Dobbs, 2014; Aull and Lancaster, 2014; Wu and Paltridge, 2021). We prepared a list of potential stance markers in English and Arabic after reviewing many relevant studies in the literature (e.g. Biber and Finegan, 1988, 1989; Fitzmaurice, 2004; Jaffe, 2009; Johnstone, 2009; Kiesling, 2009; Grey and Biber, 2012; Alghazo et al., 2021b). After that, each potential stance marker was double-checked within context to verify its analysis. Each stance marker was then highlighted to lay down the ground for the calculation of the number of stance markers in each set of editorials. It should be noted here that Arabic stance examples were translated carefully after ensuring that they are considered stance features. In the quantitative part, a statistical analysis was performed using SPSS and the Mann–Whitney U test to compare the differences between the two independent groups. And for the purpose of qualitative analysis, numerous examples have been discussed with reference to earlier studies of stance. An explication of these instances has been made to uncover the interpretations and justifications of stance-taking options in both languages.

Findings

The findings of the study show that the most frequently used stance markers in the Arabic set of data were the attitude markers, followed by boosters, hedges, and self-mentions, respectively. Attitude markers, interestingly enough, were far more frequent than any other kind of stance features in the Arabic group of editorials. While boosters were the second most employed feature of stance across the Arabic editorials (23.4%), hedges were used less and less frequently (7.3%). There was only one instance of self-mentions in the Arabic collection of editorials (see Table 1). Self-mentions are considered to be the least recurring feature of stance in the Arabic editorials (0.1%), which was exactly the same as that of the English ones (0.1%).

Table 1 Stance features frequencies, percentages and frequencies per 1000 words in Addustour set of editorials.
Full size table

Table 2 below shows the frequencies, percentages, and frequencies per 1000 words of stance features in the Guardian collection of editorials. By far, the most frequent stance feature in the English set of editorials was hedges (37.7%), followed by attitude markers (37%) (see Table 2). It is also worth noting that the gap between the usage of hedges and attitude markers in English editorials was relatively small (only 0.7%). Boosters ranked third among the most frequent stance features in the Guardian English editorials (25.2%). At the very least, self-mentions were used only once in the English editorials (0.1%), as can be seen in the table below.

Table 2 Stance features frequencies, percentages and frequencies per 1000 words in the Guardian set of editorials.
Full size table

The Mann–Whitney U Test, also known as the Wilcoxon Rank Sum Test, was adopted to capture further in-depth information on the similarities and/or differences between both languages in the practices of stance. This non-parametric statistical test was selected for its reliability and validity when comparing two independent groups. The results of the correlational analysis are summarised in Table 3, and all types of stance markers were evidently present in the editorials of both languages. There is also evidence of a statistically significant difference between both sets of editorials in the total usage of stance markers, where the Z value reached (−7.713). Most strikingly, the test revealed a significant difference between the two groups of editorials in the usage of each type of stance markers. The two sets of editorials differed in the use of every category of stance markers (namely hedges, boosters, attitude markers, and self-mentions). This means that the expression of stance in editorials has significantly diverged from the Jordanian versus the British setting.

Table 3 The Mann–Whitney U test results.
Full size table

We turn now to consider the different types of stance with some examples from both sets of editorials. Stance, with its roots in metadiscourse and Hyland’s interpersonal model, is generally broken down into four elements. Hyland (2005) categorised stance elements as being hedges, boosters, attitude markers, and self-mentions. Hyland’s (2005) typology of stance features is very widely used in the studies of interaction and stance in written discourse.

Hedges

Hedges have basically been used to refer to the linguistic items used by writers to distance themselves from their content. Hedging can be loosely described as ‘the softening’ of the writer’s statements or comments (Housen et al., 2012). Kaltenböck et al. (2012, p. 1) defined hedging as ‘a discourse strategy that reduces the force or truth of an utterance and thus reduces the risk a speaker runs when uttering a strong or firm assertion or other speech act’. Van Dijk (1995, p. 16) has also emphasised that ‘opinions may not only be expressed implicitly but also be implied indirectly’. Examples of hedges from the editorials of both languages are presented below. Hedges are used in the below-mentioned examples to moderate and tone down the rhetoric in these utterances.

  1. (1)

    1 In defense of a nation that is almost overwhelmed by despair.

  2. (2)

    This would also promote our national economy.

  3. (3)

    Perhaps we could cultivate such attachments (The Guardian view on second-hand clothes, The Guardian, 2021).

  4. (4)

    But Prince Philip was also, perhaps paradoxically, the trailblazer for the idea of royalty as a profession. (The Guardian view on Prince Phillip, The Guardian, 2021).

Although Arabic editorials contained a relatively small amount of hedges (7.3%), they performed a variety of vital functions in these editorials. Editors may comment on a topic of interest using an adverb of degree, which is placed before the verb it modifies (see Example (1)). Badawi et al. (2013) also mentioned that words such as kada كاد ‘almost’ or ‘nearly’ are regarded among the most commonly used words in Arabic discourse. A possible explanation of this use of the stance adverbial ‘almost’ in Example (1) as a hedge may be related to the employment of hedges as face-saving strategies (Hübler, 1983). In this example, the writer has the ability to say غلبها or ‘that is overwhelmed’, but the writer preferred using كاد ‘almost’ to avoid any face-threatening act (FTA). Crompton (1997, p. 278), in a discussion about hedging, argued that “the writer is not displaying a lack of confidence in his own proposition but politeness towards the discourse community”.

In the second example, an expression of stance was delivered through the use of the modal verb ‘would’. Almeida and Vazquez (2009, p. 1171) mentioned that ‘modal verbs grammaticalize speaker’s subjectivity’. The category of modals or modal verbs can be considered a grammatical category or a pragmatic one (Kreutz and Harres, 1997). The epistemic modals or modals of probability, like ‘would’ in Example (2) and could in Example (3), are used to soothe the firmness and intensity of these narratives. Since hedges were the most commonly employed stance feature in the English corpus, the editors often utilised the hedging word to express a sense of ambiguity. Thus, it is obvious that the adverbial hedging word was an interpretation devised and presented by the writer, as shown in Examples (3) to (4). Biber et al. (2021, p. 861) claimed that ‘news and academic prose also use probably and perhaps, with predictions, suppositions, explanations, and interpretations that have not been clearly proven’. As for Biber et al. (2021), ‘perhaps’ is one of the most common stance adverbials across all registers for marking doubt about what you are saying. Hyland (2019, p.143), in a discussion about Milne’s (2003) comparison between Spanish and English editorials, highlighted that ‘similarities in the use of hedges and attitude markers, for instance, reflect the combination of mitigation and opinion needed to persuade newspaper readers’.

Boosters

In the literature, the term ‘boosters’ tends to be used to refer to the amount of affirmation or emphasis a writer is putting on a certain claim. Multiple terms were used to describe the notion of placing a level of ‘assertion’ on a proposition, such as certainty markers, emphatics, and modality markers. Hyland (1998) reminds us that boosters ‘allow writers to negotiate information, helping to establish its perceived truth by strategically presenting it as consensually given’. In fact, the degree of certainty voiced by boosters in editorials is not equal in all cases. Therefore, previous research has established a scale of certainty to describe the level of assertion that a writer asserts (e.g. Holmes, 1982; and Rubin, 2007). To illustrate, the continuum of certainty in these scales ranges from absolute certainty to low certainty with varying degrees in between. A point to note is that we observed the use of boosters with high levels of certainty or with what is called ‘absolute certainty’ in the editorials of both languages, as shown in Examples (5) to (8). A possible explanation for this result is that editors are trying to build authority and trust between themselves and their readership. Perhaps stance-taking from a strong foundation exerts a hidden power on the public. For example, Hyland (2008) suggested that boosters might help bring readers to the text or involve them in building some kind of solidarity. Since editorials are meant to be the newspapers’ voice, the journalistic stance of the newspaper about issues of the day is assumed to be firmly present by means of pragmatic boosters.

In Examples (5) and (6), writers of the editorials used boosters in their discussion of political issues in the Jordanian context. Namely, the writers handled the partnership between the public and private sectors in Jordan and the prime minister’s visits to the cities of the country. The same applies to their English counterparts from the Guardian newspaper, where boosters were utilised to discuss some politically oriented topics in the British context (see Examples (7) and (8)). In particular, the writers addressed topics such as Biden’s 100 days after the presidency and the view on the BBC.

  1. (5)

    The relationship between the sectors and its sustainability will necessarily improve the citizens’ standards of living.

  2. (6)

    The approach of communication and fieldwork, that aims to serve Jordanians, has been consistently the focus of the leader’s attention and care.

  3. (7)

    No one doubts the sincerity of the Biden team. (The Guardian view on Biden’s 100 days, The Guardian, 2021).

  4. (8)

    It is often said that the BBC has never faced such towering challenges. In this occasion, that is certainly true. (The Guardian view on the BBC, 2021).

Attitude markers

The use of ‘emotive markers’ or ‘frames of mind’ in speech, originating from Aristotle’s attitude to Austin’s illocutionary force, is one of the deeply entrenched topics in linguistics and has its deep roots within the field of sociolinguistics. In Searle’s (1975) taxonomy of illocutionary acts, he defined what is meant by an illocutionary force and argues that ‘the illocutionary force indicating device in the sentence operates on the propositional content to indicate, among other things, the direction to fit between the propositional content and reality’ (p. 359). In speech, the illocutionary force of a sentence might be used for asserting, demanding, promising, exclaiming, or questioning. The idea of attitude markers in writing is similar to that of emotive markers or illocutionary acts in speech. Attitude markers are set forth by the writers to describe their own faith and feelings about the given content. Hyland (2019, p. 36) points out that ‘almost any linguistic choice conveys an attitude of some kind, expressing our likes and dislikes, our approval and disapproval’. Since editorials are packed with opinions and attitudes, it is essential to analyse the linguistic aspects used to express these sentiments. Below are some examples of attitude markers in their context of use.

  1. (9)

    The Hashemites were wise and capable of being with their people

  2. (10)

    These meetings, where citizens and officials present both their vision and their hopes for their officials, are responsible for achieving the desired development, which is the most useful and beneficial.

  3. (11)

    It is disappointing but not surprising that the PM appears uninterested in plans to make people healthier. (The Guardian view on the way we eat, The Guardian, 2021)

  4. (12)

    The situation is extraordinarily difficult, with cases at about 40 universities so far and further outbreaks expected. (The Guardian view on universities, The Guardian, 2020).

Examples (9) to (12) illustrate the use of attitude markers in the editorials of the two languages. In every instance of these attitude markers, the writers used an adjective to describe their own stance. This is exactly what Dafouz-Milne (2008), in her discussion of attitude markers, referred to as the use of ‘attitudinal adjectives’. An example of an attitudinal adjective would be the usage of the word ‘disappointing’ in Example (11).

Self-mentions

Drawing on features of interactional metadiscourse, self-mention markers can be considered as the link between authors of the texts and their communities. Hyland (2001, p. 208) demonstrated that self-mention is ‘a strategy that maximises the credibility of the writer to elicit credence from the reader’. Self-mention markers in the editorials of both languages were indicated using first-person plural pronouns rather than using self-reference words such as the author(s) or the name of the institution. In only one instance was the existence of self-mention markers in English and Arabic editorials. Numerous studies attributed the low frequency of self-mentions in certain texts to the writer’s absence of awareness about their metadiscoursal and pragmatic role. However, we find it unreasonable that the editorial authors are unaware of the self-mentions’ pragmatic role but are rather constrained by the genre style and conventions.

As far as we are concerned, there are two likely causes for this result, as it seems possible that the paucity of self-mentions in the editorial genre will be made for a good reason. Firstly, the number one reason for this might be the idea that editorials are responsible for representing the institution’s voice rather than the individuals’ viewpoints. Secondly, this result can also be attributed to the intricacies of authorial identity construction in the genre of newspaper editorials. It might be the case that editorial writers struggle with separating their own sense of identity or personal beliefs from their institutional identity (personal voice versus institutional voice). Wu and Zhu (2014, p. 137) argued that ‘everyone has a ‘core identity’ connected to internal states, but all people have multiple identities connected to their performances in society’.

Wu and Zhu (2014) also mentioned three aspects of the authorial identity: the detached self, the individual self and the collective self. Though this classification has been proposed in the context of academic discourse, it seems that it also applies to the editorial genre of texts. In this sense, it can be noticed that ‘the collective self’, where the writer is showing a strong affiliation with the discourse community, is the dominating aspect of identity within newspaper editorials (Wu and Zhu, 2014). A prominent explanation about self-mentions’ scarcity in certain genres was proposed by Hyland (2002), who argued that the little existence of self-mentions may be due to ‘the culturally specific views of authority’. The use of self-mentions in newspaper editorials of both languages is shown in Examples (13) and (14).

  1. (13)

    As a result, we are facing a promising new phase full of vital projects.

  2. (14)

    The pandemic has opened our eyes to the predicament of the poor. (The Guardian view on food banks, The Guardian, 2020).

In Example (13), the usage of the first-person plural pronoun ‘we’ may partly be explained by the intention of involving a broader audience in the discourse. An alternative explanation is that the writers tend to capitalise on the collective identity of their target audience (mainly Arabs and Jordanians in this context). Furthermore, Arabs lean toward a shared sense of belonging to a group or a collective identity, which is formulated by the sociopolitical environment (Smooha, 1992, as cited in Amara, 2016). Direct self-mention using a first-person pronoun can be regarded as a method to invite readers to collective bargaining. The use of the first-person pronoun ‘our’ in a discussion of a current issue (i.e. the pandemic in this respect) was the only instance of self-mentions in the Guardian (see Example (14)). The authors’ use of the self-mention marker ‘our’ is likely to be related to the assumption and promotion of a level of societal responsibility. In this regard, ‘our’ is more than a sign of visibility or author presence. Finally, here are a few more instances of stance features from the editorials of both languages Table 4.

Table 4 Stance features in English and Arabic Editorials.
Full size table

Discussion

The aim of the present study was to examine the stance-taking strategies that editors employ when defending the newspapers’ standpoint. It has then investigated how the use of stance markers varied across two different newspapers using two separate languages with two totally different cultures. Editorials, however, are a remarkably distinct genre with respect to persuasion, evaluation, and appraisal. The discrepancy between editorials and the other text genres in the articulation of stance arises from the power of editorials on the political decisions in their respective geographic areas. For instance, Van Dijk (1995) observed that editorials might have an impact on the presidential elections or the “formulating” of people’s ideologies. Van Dijk, in this sense, defined the term ‘ideologies’ as ‘the basic ‘axioms’ of socially shared representation of groups about themselves and their relations to other groups, including such categories as membership criteria, activities, goals, values, and crucial group resources’ (1995, p. 30). This brings us to the conclusion that there is a consensus among researchers about the idea that public ideologies and beliefs are being framed or at least influenced by the stance of newspapers offered in editorials.

The present study contributes to our understanding of intercultural rhetoric and the use of rhetorical devices such as stance markers across two languages (i.e. Arabic and English). It also lays the groundwork for future research into stance markers in the editorials of other languages. The results of this study add to the rapidly expanding field of rhetoric in editorial journalism. In this regard, the findings of this study can be utilised to train editors, editors-in-chief, and members of the editorial board about stance and positioning in their early or mid-career phases. Introductory webinars for editors can take advantage of the findings of this research to help editors pursue a professional usage of stance devices in editorials. This paper is of interest to the society of editorial freelancers to assist them in understanding the tactics through which editors-in-chief optimise their strategies in showing stance.

Stance-taking and writer positioning in media discourse, plus the idea of finding the appropriate policy to present your perspective and the counterargument, would indeed be a daunting task. Kärkkäinen (2006, p. 699) argued that ‘stance in discourse is not the transparent linguistic packaging of ‘internal states’ of knowledge, but rather emerges from dialogic interaction between interlocutors’. Therefore, argumentative writing in editorials is quite complex and sometimes misleading and requires a great amount of observation to be fully understood by the average reader. A contrastive analysis of the rhetorical strategies, such as stance devices and how they are utilised in editorials from diverse cultures, is needed for better comprehension of them. For that being the case, this study compared the use of stance devices in English and Arabic editorials from two national broadsheets in Jordan and the UK. This study has found that generally, Arabic editorials in Addustour newspaper employed stance markers in a different way than English editorials in the Guardian newspaper.

Some statistically significant differences were detected between the two languages in the total usage of stance markers. Significant differences were also identified between the two languages in the use of each category of stance features (i.e. hedges, boosters, attitude markers, and self-mentions). Although all stance types were present in the English and Arabic editorials, self-mentions were the least-employed stance features with the same value in both languages (0.1%). Perhaps this resemblance in the little use of self-mentions is due to the genre conventions and style of the editorials. To illustrate, it seems that the authorial presence via self-mentioning is a privilege in editorials’ argumentative writing rather than a necessity. By and large, the editorial section of the newspaper is designed to display the newspaper’s voice and not the author’s voice.

A thorough understanding of stance features in editorials, which are concerned with the most pressing current political issues, would foster the public audience’s comprehension of these columns. Liu and Hood (2019, p. 589) pitched the idea that ‘the non-neutral construal of people and events in the media is an issue of considerable significance’. Editorials are now considered as the organisation’s or media institution’s portrayal of the issues of the day, and wherefore this genre has come to be referred to as ‘leading articles’ or ‘leaders’. Vigilance in observing stance features within the scope of metadiscourse in editorials is not only fruitful for an avid reader of politics but also for foreign language learners. Hashemi and Golparvar (2011, p. 122) claimed that ‘metadiscourse can play a more influencing role on reading comprehension if the consciousness of the EFL learners’ is raised by their teachers, especially at the intermediate level’. In consequence, students who are learning English or Arabic as a foreign language can take advantage of the findings of this study in their comprehension of such kinds of opinion articles.

Additionally, a persona in editorials is generally affected by the dominating culture in that language. To elaborate, the term ‘persona’ here refers to ‘the mask served to express the speaker’s own opinion, only by a tactful indirection’ (Mayer, 2003, p. 60). Broadly speaking, editorials are designed by the editor-in-chief and writers from the editorial board about socially or politically sensitive topics with careful attention paid to the culture and community convenience. This idea is positively related to what Strauss (2004) has reported about ‘cultural standing’ in the expression of opinion. She demonstrated that opinion display, argumentation, hedges, and modality are heavily impacted by the ‘cultural standing’. According to a definition by Strauss (2004, p. 161), cultural standing is ‘the location of a view on a continuum that ranges from highly controversial to completely taken for granted in the relevant opinion community’. Taken together, this suggests that the examination of how stance features are used in two different languages will have significant implications for the understanding of language within a cultural and institutional frame of community.

Conclusion and implications

The findings of this study have implications for the teaching of argumentative writing in both languages (i.e. English and Arabic). These findings reveal something about the nature of discourse markers used in opinion articles. This research may provide insights to editorial training and editorial training fellowships because it might enhance the skills of argumentative writing. In these training and fellowship programmes, trainees can emulate some of the examples of stance given in this study. This simulation would further help them fine-tune their strategies for displaying stance.

Further research might explore the use of stance features in editorials in other languages to gain a greater understanding of stance in the editorial genre. That is to say, the use of stance markers as a compelling force in editorials is known to be affected by a range of factors. To start with, the differences in the use of stance markers between the two sets of editorials can be linked to the language and register that these editorials belong to. Matthiessen (2019, p. 207) initially stated that ‘registers are functional varieties of language that have evolved as adaptations to different institutional settings’. Then, he argued in his discussion of language as an aggregate of registers that ‘a key aspect of the adaptive nature of language is that it adapts to its contexts of use’ (Matthiessen, 2019, p. 212). Subsequently, it is possible to assume that register and language are the reasons behind some of the significant differences in stance markers between the two groups of editorials.

Secondly, the sociocultural dimensions can play a significant role in the choice of stance markers in editorials. Du Bois (2007, p. 139) claimed that ‘stance can be approached as a linguistically articulated form of social action whose meaning is to be construed within the broader scope of language, interaction, and sociocultural value’. Thirdly, routines can be regarded among the factors affecting the use of stance markers in editorials. Reese (2016, ‘levels of analysis’) said that ‘if journalism is primarily a social practice, routines are the ways of working that constitute that practice’. Along these lines, routines and institutional policies of the newspaper can determine the approach and means of displaying stance. To conclude, factors affecting the use of stance markers in English and Arabic editorials can be discussed under four headings: language, register, routines and institutional policies, and sociocultural systems and values.

As a matter of fact, there is no alternative to adopting a position and taking a stance in editorial journalism. Even when the editorial holds no position on the issue of discussion, this is believed to be a stance on its own, as Jaffe (2009, p. 3) suggested that ‘neutrality is itself a stance’. Hyland (2019, p. 63) also reported that ‘writers cannot avoid projecting an impression of themselves and how they stand in relation to their arguments, their community and their readers’. The grasp of this idea makes it extremely significant to pinpoint these positionings in editorials and to understand them (i.e. stance markers). As far as objectivity is meritorious in news coverage, it is thought of as blameworthy in editorials. Firmstone (2019, p. 6) observed that ‘in direct contrast to most other forms of journalism, subjectivity and opinion is not only permitted in editorials- it is expected’. And there is no better way to express opinion in editorials than using stance markers. The evidence from this study suggests that the analysis and use of stance features should be carefully observed by both lay readers and editorial specialists.