New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions

Discourse, Context & Media xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Discourse, Context & Media journal homepage: www.elsevier.com/locate/dcm

New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions Tayyiba Bruce Newman University, Genners Lane, Birmingham B32 3NT, UK

a r t i c l e

i n f o

Article history: Received 5 June 2017 Received in revised form 14 October 2017 Accepted 24 October 2017 Available online xxxx Keywords: Reader comments News Online discourse Ideology Religion Islam Catholicism

a b s t r a c t The recent growth of populism has resulted in dramatic political change in British, European and American politics. Given the role of right-wing media in legitimising and encouraging traditionalist, nationalist or sometimes far-right populist discourse, the linguistic and semantic relationship between news and readers’ discourse in the context of right-leaning news websites merits investigation in the current climate. This paper examines the relationship in terms of representations of Islam and Catholicism in Daily Mail and Telegraph websites. Empirical keyword analysis of news articles and linked comments is presented. Key semantic categories revealed evidence for stereotypical representation of these religions, as applicable to new media contexts and enacted in the websites of traditional print media. Results highlighted that long-term themes such as ‘Islamic terrorism, war and extreme belief’ and ‘the Pope, Vatican and scandals in the Catholic Church’ continue to dominate news stories online. 18 out of 19 categories in Islam news repeated in comments, and 9 out of 13 in Catholicism news repeated in comments. This strong overlap indicated close connections between news and comments and shared ideologies. Higher numbers of semantic categories in comments showed greater variation of topics than news, with 15 and 23 additional categories in Islam and Catholicism comments respectively. However, key multiwords reflecting collectivisation of communities, such as Muslim country, were identified in news and repeated in comments. Investigation of Muslim country revealed nationalistic and polarising statements by readers, while investigation of penitents revealed portrayal of Catholics as an ‘out-group’ through lexis which denoted violence, shock and crime. Closer analysis of some keywords revealed polarity between readers displaying differing views in relation to various aspects of religion. Taken together, however, findings suggested that age old stereotypes of minority religions were perpetuated in new media contexts. Crown Copyright Ó 2017 Published by Elsevier Ltd. All rights reserved.

1. Introduction An array of recent studies provide evidence for hostile coverage of Islam in the British press (e.g. Baker et al., 2013; Taira et al., 2012; Saeed, 2007; Moore et al., 2008). News articles indicate first, Islam and Muslims as a threat and ‘‘suspect community” (Nickels et al., 2012) and second, as a cultural ‘Other’. The first of these themes stems from stories about violence, terrorism, extremism and criminality as associated with Islamic belief (Baker et al., 2013; Poole, 2002). Muslims have been discursively constructed as a threat to global peace (Baker et al., 2013), and to Britain or British culture (Saeed, 2007). They are portrayed as clashing with British values, such as equality and human rights, which form

Abbreviations: INC, Islam News Corpus; ICC, Islam Comments Corpus; CNC, Catholicism News Corpus; CCC, Catholicism Comments Corpus. E-mail address: [email protected]

the cornerstones of democratic society (Ratcliffe, 2004) and the established British religion, Christianity (Taira et al., 2012). The second theme involves representation of Muslims as foreign, with ill-placed loyalties or as immigrants and asylum seekers (Richardson, 2009), thus presenting unresolvable cultural differences typified by negative characteristics such as barbarism, irrationality, primitivism and sexism (Benn and Jawad, 2004). Many of these studies were post-9/11 representations of Islam but earlier studies also report deviance, violence and mental imbalance as key topics (van Dijk, 1991). Studies of Catholicism in the media also report the dominance of similar themes. Catholics have been presented as an ongoing threat in British and American media and as antagonistic toward the Church of England (Woolley, 2012). Most recently, this has been centered around child abuse scandals (Pepinster, 2012; Woolley, 2012; Pierre, 2011; Jenkins, 2003). As with Islam, portrayal of Catholicism as a cultural ‘Other’, or ‘‘social pariah”

https://doi.org/10.1016/j.dcm.2017.10.001 2211-6958/Crown Copyright Ó 2017 Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

2

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

(Pepinster, 2012:171) has been typical of media coverage recently, leading to accusations of anti-Catholic bias and hostility to the Catholic Church (Royle, 2012; Cozens, 2004; Watts, 2004), with Catholics depicted as intolerant and anti-reason (Woolley, 2012). This tendency can also be identified in British literature and political discourse about Catholicism (O’Malley, 2006; Colley, 2005). Furthermore, association between Catholicism and terrorism was a key theme in news coverage until relatively recently (Nickels et al., 2012). Although some level of positivity and inclusivity in representation was recently identified in BBC, Times and Telegraph news about Popes and Papal visits to Britain (Taira et al., 2012), this has been explained in terms of the need to portray a more foreign Other, namely Islam, as the enemy (Lund, 2011). Therefore, Otherness may be placed on a gradient by language users rather than an absolute characteristic applied to a particular group.

(Bivens, 2014;). However, others describe UGC as fundamental to contemporary journalism, (Manosevitch and Tenenboim, 2017; Batsell, 2015). Reader comments sections, which form the main type of UGC in news websites (Hermida, 2011), allow unprecedented opportunities for readers to express their opinions and beliefs, including religious and political orientations and more private views (Loke, 2012). Most national UK newspapers and many local papers now offer readers this opportunity, free of charge (Collins and Nerlich, 2015) thus empowering citizens in a participatory culture (Jenkins, 2006, Leung, 2009). From the journalistic perspective, opportunity to comment generates website traffic and improves brand loyalty (Batsell, 2015). In fact, there is some evidence for UGC as correlated to offline political participation (Kaufhold et al., 2010). 1.2. Reader comments: participation, dialogue or incivility

1.1. Shifts in media and participatory practices Traditionally, media communication with the public was largely a one-way process, with limited opportunities for readers to respond. For example, letters to the Editor were selected by newspapers to represent the institutional voice and more challenging points of view may not have been published (Wahl-Jorgensen, 2002). The ‘web 2.0’ era changed communication conventions and boundaries in journalism, with the advent and growth of social networking sites, blogs and news websites online, while televised, radio and print news continued offline. For instance, free access to publishing technology brought a dramatic increase in ‘citizen’ or ‘amateur’ journalism and a plethora of small news outlets, which provided a ‘horizontal’ alternative to the tradition vertical (or top-down) structure of media communication and platforms for dissemination of alternative worldviews (Bivens, 2014). News websites with participatory features for audiences, such as reader comments sections, have become increasingly popular in recent years (Santana, 2014; Milioni et al., 2012; Neuberger and Nuernbergk, 2010). The initial perception of these innovations as a threat to professional journalism (Lewis, 2012) has been debated, with continuation of traditional norms and values cited in recent research (Domingo et al., 2014). Bivens (2014: 205) argues that audience desire for interactivity, transparency and immediacy in televised reporting, are best met by established institutions which can meet the financial investments, such as in satellite technologies, which make live reporting or ‘‘breaking news” broadcasts possible. The ability to satisfy audiences in this way, is thought to increase loyalty and trust while the variability of consumption patterns, and audience need for continuity and routine has maintained the place of established or elite journalism (Bivens, 2014). This could mean that traditional production processes, brought about by top-down directives continue, with emphasis on simplification, ‘‘human interest”, ‘‘memorableness”, drama and conflict rather than on providing context and history to readers (Bivens, 2014). Nevertheless, social media technologies, including usergenerated content (UGC) have been shown to deviate from these constraints (Bivens, 2014): ‘‘Many see them as functioning in the opposite direction by confusing, blurring, and outwardly, unapologetically pushing the boundaries of journalism to make room for new models such as amateur and participatory journalism.” UGC is considered by some as peripheral to news production (Karlsson, 2011) and unlikely to bring societal change (Christensen, 2011; Morozov, 2009), partially because the potential of tools to understand audiences (such as numbers of likes, shares and comments) is not yet fully being realised by journalists

In research on comments as ratings systems, debates correlated with levels of interest, while increased commenting was considered as success of a given article by journalists (Nagar, 2011; Reich, 2011). Research has examined user comments in terms of linguistic features such as rhetoric, focus and register (Freund, 2011; Galily, 2008; Kohn and Neiger, 2007; Manosevitch and Walker, 2009). According to Gurak, 1997, comments tend to take an oral or casual style and anonymity in such contexts encourages open dialogue between people of all ages and backgrounds, including opportunity to explore self and gender identities. This in turn provides a large amount of data for linguists to explore, for instance in terms of public perceptions of minority religions that would otherwise be difficult to access. Reader comments may encourage expression of hostile emotions (Gurak, 1997) due to lack of moral restrictions and norms (Virilio,1997). According to Hecht (2004), linguistic aggressions, such as talk-backs, slander and expletives indicate personal protests about social issues and signal tension within the discourse. These have been likened to graffiti in the offline world (Rosenthal, 2004). Santana (2014) revealed causation between anonymity and ‘‘incivility” on news websites. Conversely, users who must identify themselves tend to contribute more deliberative comments, with diverse perspectives (Aharona, 2012; Nagar, 2011). Collins and Nerlich (2015) argued that incivility in the form of vulgarity, peroration, name-calling and stereotyping is not as prominent as the demand for rational argumentation. However, other studies highlighted that hard news, controversial issues or high-profile figures with a particular political stance attracted incivility (Coe et al., 2014; Holliman, 2011) more comments (Diakopoulos and Naaman, 2011; Tenenboim and Cohen, 2015) and more one-time rather than reciprocal comments, resulting in polarisation rather than two-way debating (Manosevitch and Tenenboim, 2017). Tenenboim and Cohen’s (2015:207) recent study revealed that that the main focus of comments was likely to be politics/government followed by crime, military/defence and society/welfare (such as in hard news topics). They noted that comments can be ‘‘extremely blatant”, which has occasionally led to comments sections being disabled, for instance in the case of Christian Science Monitor in 2012. The editor of the magazine based his decision on the large numbers of ‘‘unproductive comments” posted over a period of two years (see Batsell, 2015). In his work on the British press, Richardson (Richardson, 2009:373-6; Atkin and Richardson, 2007) argued that comments which incite hatred and promote extreme views (e.g. ‘‘GET SHOT OF THE LOT OF THEM”, ‘‘ban Islam”) are the responsibility of news institutions because they choose to publish and therefore endorse them; comments sections enable institutions to distance themselves from controversial statements while at the same time circulating them. Ben-

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

David and Folkman (2010) also presented the contradictory nature of comments, in that language allowed may be inexcusable in printed version. 1.3. Religious discourse online Beckett (2012) claimed that religious hegemonies are challenged as people are exposed to alternative ideologies, and express non-conventional attitudes with greater freedom. However, this ‘freedom’ may be counter-acted due to increased influence of online moderators and frequent posters (see Campbell, 2007). Thus, for those still reading online versions of their traditional choice of newspaper, perhaps the extent of change (of previous conventions) is limited. Studies of news audiences have highlighted the importance of a reader’s individual engagement with texts (Branum and Charteris-Black, 2015; Fowler, 1991), while those of internet contexts revealed discourse which is fluid and evolving rather than easily definable and fixed (Karaflogka, 2002; Pihlaja, 2011). However, recently, Pihlaja (2014) identified a tendency for users on YouTube who hold similar religious worldviews to group together. This suggested parallels between online and offline discourse. For instance, readers may actively source news that generalises, supports or matches their own experiences and ideologies in an ‘echo chamber’ (Colleoni et al., 2014). This study adopts van Dijk’s (2006: 115) approach that ideologies are relatively stable systems of ideas based on fundamental beliefs rather than personal opinions, with discourse, at various levels of text and talk, providing a means for ideologies to be ‘‘acquired, confirmed, changed and perpetuated”. One of the main features of ideological structures, which can be found in discourse, is referred to as ‘‘ideological polarisation between in-groups and outgroups” (van Dijk, 2006:115); whereby members of in-groups tend to emphasize their own positive aspects, and minimise or deny negative ones, while at the same time doing the reverse when describing members of the out-group. Ideological discourse involves a general strategy, which may function at any linguistic level, of positive self-presentation and negative otherpresentation. Ideology affects discourse forms (formatting, choice of imagery, placing of story in newspaper, syntactic choices – active/passive) and meanings (van Dijk, 2006; Moore et al., 2008). The focus of a corpus approach on lexical patterns, for instance through identification of (statistical) keywords, is useful to develop understanding of discursive construction of social or religious groups (Mahlberg, 2013). Keywords, as repetitive but distinct linguistic behaviour of a discourse community, indicate ‘‘labelling” of people, events and institutions (Mahlberg, 2013). 1.4. Contribution of the study The recent growth of populism has resulted in dramatic political change in British, European and American politics. Given the role of right-wing media in legitimising and encouraging traditionalist, nationalist or sometimes far-right populist discourse, the linguistic and semantic relationship between news and readers’ discourse in the context of right-leaning news websites merits investigation in the current climate. In this paper, I present an empirical analysis of key themes in discourse about Islam and Catholicism in Daily Mail and Telegraph websites. The main aim was to investigate how the religions were portrayed or labelled in online right-leaning British news articles and whether and how themes were replicated, changed or rejected in reader comments. More broadly, this provided insight into the relationship between news and comments as discursive genres. Most previous studies in this area investigated representations of individual religions in the British media rather than comparison between religions (e.g. Baker et al, 2013; Pepinster, 2012). In depth, linguistic

3

comparison of Islam and Catholicism, two ‘newsworthy’ religions (Knott et al., 2013), enabled better evaluation of the particular features of right-wing news discourse around these particular social groups (Partington et al., 2013:12). Mitchell and Gower’s (2012) book provided contributions from journalists and religious leaders, with reflections of their experiences of British news coverage of religion. The focus was on main religions reported in ‘serious’ news (Islam, Church of England and Catholicism). However, the chapters therein did not incorporate research of online news, which due to its centrality in current news consumption trends, is essential. Therefore, comparison of news articles with reader comments, was the focus of this study. The corpus methodology deployed here allowed analysis of large numbers of comments, which was a challenge in previous studies (Tenenboim and Cohen, 2015). This shed some light on the nature of comments, and how readers discursively constructed social or group identity. Mainstream news plays a key role in citizenship, democracy and political decision-making, and therefore it is important to continue to investigate representations of minority religions in popular contexts. Previous studies had already contrasted left and right, or broadsheet and tabloid representation of Islam (e.g. Taira et al., 2012; Baker et al., 2013). Two right-leaning British news providers were chosen in order to gain insight into the relationship between potentially hostile coverage of minority religions (Knott et al., 2013) and ensuing comments. Given that reader comments indicate public perceptions of political and social issues (Tenenboim and Cohen, 2015), research of the register is crucial, especially in terms of its connections to news articles. Finally, the popularity of traditional news providers has continued (National Readership Survey (NRS), 2014; Audit Bureau of Circulation (ABC), 2014), indicating wide readership of Daily Mail and Telegraph news (with first and third highest combined print and online UK readership from January to December 2014, as well as increase in overall readership due to online news) and consequently current importance of rightleaning perspectives (Inglehart and Norris, 2016).

2. Methodology 2.1. Data and corpus design News articles relating to Islam and Catholicism were identified and collected from Daily Mail Online and Telegraph Online, using their own search engines. Data published between 2010 and 2015 were selected for analysis. Search terms used to find articles were: Islam, Islamic, Muslim and Muslims for the Islam corpora, and Catholicism, Catholic and Catholics for the Catholicism corpora. Articles were selected according to relevance. The relevancy ranking provided by the websites’ search functions sorted results so that articles in which search terms occurred more frequently appeared higher in the list. Muslims received more news exposure than Catholics. I therefore attempted to balance my dataset by collecting similar amounts of articles and comments for Islam and Catholicism corpora (Evans, 2007). It was crucial that articles were collected with their corresponding comments. This avoided loss of contextual evidence and enabled direct comparison. This criterion for selection took precedence over issues of perfect balance and diachronically even distribution. Four sample, comparative corpora (Xiao, 2007) were compiled: ‘Islam News Corpus’ (INC), ‘Islam Comments Corpus’ (ICC), ‘Catholicism News Corpus’ (CNC) and ‘Catholicism Comments Corpus’ (CCC). INC and CNC consisted of 176 and 166 news articles respectively, while ICC and CCC contained the related comments threads. In terms of word tokens, this was approximately 203,000 in INC, 178,000 in CNC, 2.4 million in ICC and 1.6 million in CCC (full details in Table 1, p.27).

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

4

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Table 1 Data collected for Macro-Analysis.

Number of Daily Mail articles Number of Telegraph articles Number of Daily Mail comments Number of Telegraph comments Total number of word tokens

Islam News Corpus (INC)

Islam Comments Corpus (ICC)

Catholicism News Corpus (CNC)

Catholicism Comments Corpus (CCC)

90 86 – – 202,875

– – 34,247 30,809 2,361,870

85 81 – – 177,640

– – 24,352 12,737 1,564,320

News articles with less than 10 comments were discarded. Variation in comments thread size (10–2000 comments) was a challenge and meant that more popular articles were emphasised. However, no attempt was made to apply an upper limit on comments as this would have involved removing parts of interactions and restricted the range of perspectives considered. Furthermore, discourses receiving more attention from readers were perhaps more important to investigate (Nagar, 2011). Finally, lexical ‘closure’ was analysed while building each corpus, by counting numbers of new word types (a measure of lexical growth) after each addition of ten new articles or threads (see Tables 3–6, and Figs. 1–4, pp. 29–32 for lexical growth statistics and curves). Data collection stopped at the stage when lexical growth slowed and stabilised (Miller and Biber, 2015; McEnery et al., 2006). This process systematised decisions about levels of representativeness of data collected. 2.2. Preparation for corpus analysis

2.3. Keyword analysis

New types in sub-corpus Mail

Telegraph

1–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90

2937 1612 1363 1346 627 844 489 542 501

3317 1518 1031 827 807 557 508 480 358

Table 4 Number of new word types added to ICC.

1–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90

New types in sub-corpus Daily Mail

Telegraph

7386 12058 7378 3043 3088 2875 1105 1034 1124

13,499 5437 4094 5644 1283 2515 3876 1306 1742

Table 5 Number of new types added to CNC.

2.3.1. Reference corpora Study corpora were compared with a 3-million-word reference corpus of general news articles (NC14-15). This was built using a web-crawler site called ‘Webhose.io’, with open-access data from news sites, blogs and discussion forums (Table 2, p.28). 2353 articles published online between 2014 and 2015 were obtained from a range of popular UK national news providers (NRS, 2014). The corpus included a short section of comments (approximately 300,000 tokens).

Table 2 Components of NC14-15.

Article range

1–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80

New types in sub-corpus Daily Mail

Telegraph

2038 4199 5886 7058 8386 9193 9740 10145

2352 4516 5878 7172 8328 8837 9537 9950

Table 6 Number of new types added to CCC.

News source

Number of online articles

The The The The The The The The

400 400 400 400 392 190 142 29

Total number of articles Total token count (AntConc)

Article range

Thread range

Each article was checked manually to ensure that search terms appeared in its title or main text and it had received at least 10 reader comments. Duplicate articles and comments were deleted. Unrelated items such as side-panels and unrelated URLs were then deleted, and metadata (e.g. user IDs, journalist names, publication dates, shares) were removed. However, links posted by readers as part of comments were kept. Finally, a spelling variant detector (VARD, see Baron and Rayson, 2008) was used to normalise spelling in the comments corpora before analysis.

Guardian Times Daily Mail Telegraph Scotsman Daily Mirror Daily Express Independent

Table 3 Number of new word types added to INC.

2353 3,066,604

Thread range

1–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80

New types in subcorpus Daily Mail

Telegraph

7359 10,872 14,267 16,105 20,693 21,824 23,347 24,077

13,586 16,635 18,905 20,834 23,479 25,699 27,029 30,567

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

5

Fig. 1. Lexical growth for INC.

Fig. 2. Lexical growth for ICC.

Fig. 3. Lexical growth for CNC.

NC14-15 comprised diverse political orientations (left/right) and news formats (tabloid/mid-market/broadsheet), and English and Scottish news articles. This provided a relatively varied and balanced collection of views. Comparison with NC14-15 revealed how reli-

gion data in INC, ICC, CNC and CCC contrasted with online news in general. Comparisons of certain words were also made with normalized frequencies in English Web 2013 (enTenTen13) corpus to determine how they compared with more general usage.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

6

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Fig. 4. Lexical growth curve for CCC.

2.3.2. Keyness statistics ‘‘Keyness” of words was measured using the LogDice Coefficient using Sketch Engine. This measure was suitable for heterogeneous datasets and avoided overemphasis of outliers (McCune et al., 2002). After preliminary tests using a variety of cut-off points, a minimum frequency threshold for inclusion of keywords was set at 0.002% in normalised frequency (NF). In absolute frequency (AF), this translated as 6 in INC, 58 in ICC, 5 in CNC and 38 in CCC. Finally, keywords were subjected to a check of frequency distribution. Only items which occurred in more than two articles or threads were included in subsequent categorization. 2.3.3. Categorisation into semantic sets 150 keywords and 50 key multiword items (KMWI) were grouped into keyword categories (Baker et al., 2013). First, keywords were automatically grouped using 3 semantic categories in Wmatrix3 (Archer et al., 2002). Second, to facilitate further analysis and interpretation, those categories attracting large numbers of keywords were divided into data-driven subcategories. For example, the largest keyword category in all four study corpora was ‘religion and the supernatural’, which was divided into subcategories to provide clearer indication of topics (e.g. ‘doctrine’). The category was also renamed ‘religious belief, identity and community’ to better reflect the keywords included. Next, keywords (mostly proper names) placed in a list of ‘unmatched’ terms by automatic categorisation were manually grouped. Rather than creating a large list of ‘personal names’, these were placed according to typical use of words in the data. For instance, USAS placed terrorism in ‘crime’ but left ISIS uncategorised. Therefore, ‘crime’ was extended to incorporate ‘terrorism/groups and affiliation’ and ‘terrorism/criminals’ as subcategories, where proper names could fit. Categorisation was a subjective process, with some keywords able to fit in more than one category and much overlapping between categories. To minimise subjectivity, all terms in ‘unmatched’ were checked in word sketches (in Sketch Engine) which indicated meanings in the context of use. Finally, interpretation and contextualisation of keyword results was supported through bigram, collocational and concordance analyses. The collocation span was set as 5 words either side of the node (Jackson, 2014). Hypotheses formed by quantitative results were further tested by analysis of small sets of concordance lines (50 randomly selected lines where available). Although with the larger comments corpora this would limit the range of views included, it was beyond the scope of this paper to increase the amount of data analysed qualitatively. The insights gained provided supporting evidence for quantitative findings as well as counter-arguments

to earlier hypotheses. Note that spelling errors within news or comments were not corrected in the examples below. Normalised frequencies are provided per thousand words. 3. Results 3.1. Islam news: violence, extremism, muslim identity and beliefs Nineteen semantic categories were identified in INC (see Table 7, pp.33–34). The two main ones were ‘religious belief, identity and community’ and ‘geographical names’. Others were ‘crime’, ‘thought/belief’, ‘dead’, ‘politics’, ‘war’, ‘clothes’, ‘education’, ‘media’, ‘law and order’, ‘violent/angry’, ‘dislike’, ‘unethical’, ‘time’, ‘no religion’, ‘no respect’, ‘exclusion’ and ‘foolish’. ‘Religious identity, belief and community’ included 36/150 (approximately one quarter) of the above keywords and 16/50 (almost one third) KMWIs. As well as Islam, Christianity was key and religions were also referred to under umbrella terms such as faiths and religions. Sometimes use of these keywords had a bridging effect: 1. ‘‘Reason is surely a principle that good people of all faiths, or none at all, can agree upon.” but sometimes they were used by journalists to attribute negativities in a blanket manner, as part of an argument seeking to undermine religion: 2. ‘‘There are fanatics in all religions. Government officials have always refused to admit that this is a religious war.” Example 2 was part of an argument made by the journalist that politicians should be treating war against terrorists as a religious war. INC keywords and collectivizing terms in the ‘people’ and ‘world’ subcategories (e.g. Muslim community) indicated that collectivised portrayal of Muslims was an important tool in Islam news, as reported in Richardson (2009). Only 8/153 (5.2%) occurrences of Christian, formed part of a collectivizing term (Christian town/community/minority), compared to 130/538 (24.2%) occurrences of Muslim (Muslim community/world/population/countries/ nation/land). This supported the idea that generalisations were more likely to be made about Muslims than Christians. In CNC, only 3/84 (3.6%) occurrences of Christian were similar collectivisations (Christian community) and even fewer (14/719, or 0.3%) instances of Catholic (Catholic population/community). However, the main collocation pattern of Catholic was the Catholic Church, which may also be used in blanket statements. Therefore,

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

7

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx Table 7 Keyword Categories in INC. KMWI = Key Multiword Item. Keywords category

Subcategory

Keywords

KMWI

Religious Belief, Identity and Community

Role/title

IMAM, PREACHERS, CLERIC, PROPHET, PREACHER, PREACHING, IMAMS, CALIPH, WORSHIPPERS MCB

FORMER ARCHBISHOP

Groups and affiliation

MUSLIM COMMUNITY CENTRE, ISLAMIC CENTRE, MUSLIM CHARITY, ISLAMIC SOCIETY, ISLAMIC GROUP

Follower of religion/ religious sect Doctrine Name of religion People

MUSLIM, MUSLIMS, ISLAMIC, SUNNI CHRISTIANS, CATHOLIC

God/Prophet Place of Worship Practice/Belief Cleric The World Attributes

MOHAMMAD, MOHAMMED, ALLAH MOSQUE, MOSQUES RAMADAN, HALAL RAUF, FEISAL

Places

KOBANE, BENGHAZI, SINJAR, WOOLWICH, CALIPHATE, MOSUL, LEVANT, LAHORE, KABUL, ISLAMABAD, PAKISTAN, DABIQ, WOOTTON, BASSETT, NINEVEH, HAMLETS KURDISH, YAZIDIS, YAZIDI, IRAQIS ASSYRIAN, PAKISTANI, KURDS, SOMALI, IRAQI, OTTOMAN, IRANIAN, MULTICULTURALISM

SYRIAN CITY, SYRIAN TOWN,

Terrorism/groups and affiliation Terrorism/criminals

HAMAS, ISIL, QAEDA, ISLAM4UK, INTERPAL, TALIBAN, HIZBOLLAH,

ISLAMIC STATE, TERRORIST ORGANIZATION, TERROR GROUP, TERRORIST GROUP,

Thought/Belief

Extremism

ISLAMIST, EXTREMISM, RADICALISATION, ISLAMISTS, RADICALS, FUNDAMENTALISM, FANATICS, EXTREMIST, ISLAMISM, EXTREMISTS, HARDLINERS, FUNDAMENTALISTS, ZIONISTS,

ISLAMIC EXTREMISM, EXTREMIST GROUP, EXTREMIST PREACHER, RADICAL ISLAM, RELIGIOUS EXTREMISM, EXTREMIST IDEOLOGY,

Dead

Victims

AID WORKER

General

HENNING, FUSILIER, HEBDO, FOLEY, HAINES, RIGBY, KASSIG, SOTLOFF BEHEADING, MARTYR, KILLERS,

Foreign politician/ office Anti-Islam

AHMADINEJAD, ERDOGAN, WILDERS, TAYYIP, RECEP, CONSULATE, PEGIDA, EDL

AMERICAN EMBASSY

JIHAD, JIHADIST, JIHADISM, YPG, PESHMERGA, MUJAHIDEEN

CAR BOMB, HOLY WAR, BESIEGED SYRIAN CITY

Geographical Names

Nationality/ ethnicity Crime

Politics

War, Defence and the Army Clothes

KORAN, QURAN, VERSES, FATWA, DOCTRINE, WRITINGS ISLAM, CHRISTIANITY, FAITHS, RELIGIONS

MUSLIM WORLD, ISLAMIC WORLD RELIGIOUS

BRITISH SOCIETY, ARAB WORLD

CHOUDARY, OSAMA, ANJEM, JIHADI, LADEN, EMWAZI, ADEBOLAJO, HAITHAM, BAKRI, BAGHDADI, AYMAN, KOUACHI

SUICIDE BOMBER, BRUTAL MURDER, SUICIDE BOMBING, SUSTAINED ACTION

HEADSCARF, NIQAB, UNMASKED, BURKAS, BURKA, VEILS, HIJAB,

Education

Schools

OLDKNOW, HILLOCK, NANSEN, GOVERNORS, ASSEMBLIES,

Media

Television, radio and film Arts and craft

EMERSON

Law and Order Violent/Angry Dislike Unethical Time No religion No respect Exclusion Foolish

ISLAMIC FAITH, MUSLIM FAITH MUSLIM COMMUNITY, MUSLIM POPULATION, ISLAMIC COMMUNITY, MUSLIM CONVERT, MUSLIM COUPLE, MUSLIM WOMAN,

ANTI-ISLAM FILM, PROPAGANDA VIDEO

CARTOONS, CARTOONISTS, STATUES SHARIA MILITANT, MILITANTS, SAVAGERY ISLAMOPHOBIA, ISLAMOPHOBIC TROJAN

SHARIA LAW, BLASPHEMY LAW, ISLAMIC LAW

PURE EVIL 11TH ANNIVERSARY, NINTH ANNIVERSARY

SECULAR BLASPHEMY SEGREGATED

differences between followers of a religion, in particular Muslims, were sometimes minimized in the context of right-leaning news. This suggested that Muslims were not represented as a diverse community of individuals and therefore could be blamed or expected to apologise for the actions of small groups of criminals: ‘‘Lord Carlile, the former Government anti-terror adviser, who urged the Muslim community to publicly speak out against Mr Choudary’s views and actions” In the example above, there was an expectation that moderate Muslims take a more public and active role against extremists. However, the term Muslim community could be perceived as ideo-

COMPLETE IDIOT

logically polarising. When recycled in the news, it could separate out Muslims from other groups and, hold them responsible as a group for the actions of individuals (Petley, 2010). Furthermore, combined with extensive references to foreign places and people, such language may help to distance problems such as terrorist attacks and radicalisation of young people from British society. Keywords also revealed that Islamic leaders and dissemination of Islam (e.g. imam, preaching), were important topics in INC, and focal points for media scrutiny. Islam was discussed in terms of its doctrine, practices, beliefs, and British Muslim organisations, indicated by keywords such as Koran, Ramadhan, halal and MCB (Muslim Council of Britain). However, these are likely to have been

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

8

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

evaluated in terms of other predominant areas of discussion, such as crime and extremism, rather than in isolation from them. Keyword categories reflected the wider discourse when considered together. Categories ‘crime’, ‘dead’, ‘war, defence and the army’ and ‘violent/angry’, which overlapped considerably, consisted of over a quarter of keywords (39/150) and almost a third of KMWIs (14/50). This reflected discourse about terrorism and war. Names of victims of terrorism were key, for instance, American journalists, French cartoonists, British soldiers and aid workers killed by members of IS (e.g. Sotloff), as were various terms for types of killings (e.g. beheading, suicide bombing). Key places in ‘geographical names’, also indicated terrorist attacks (Benghazi) and warzones (Kobane) in the Middle east, with the only exception being (Tower) Hamlets. There was a focus on terrorism abroad and on home soil, for instance in the case of British militant, Jihadi John: ‘‘the Prime Minister had raised a second video showing a British ISIL fighter unmasked delivering a further message.” In this example, the journalist referred to events surrounding the capture and killing of British volunteer aid worker, Alan Henning, by militant group ISIL. Nine of the 19 keywords in the ‘crime’ category were also British individuals or groups. The most key person was Anjem Choudary, who was accused for many years of ‘hate preaching’ culminating in his arrest in September 2016 on terrorism charges. The subcategory ‘nationality’ and smaller categories such as ‘time’ also connected to stories about conflict (e.g. Sinjar/ Yazidis, Mosul/Iraqis, 9/11, 11th anniversary). Therefore, keyword categories indicated that Islam news continues to be about violence and conflict, with stories about perpetrators and victims of terrorism being recycled (Moore et al., 2008; Poole, 2002). One in eight key terms indicated extreme belief types (e.g. fundamentalism), rather than strong or moderate (see Baker et al., 2013: 151). The seriousness of these terms and the concerns they reflected was apparent in suffixes -ist and -ism (e.g. Islamist, jihadism). In INC, -ist occurred with an NF of 4.8 per thousand words compared to 1.6 in enTenTen13 and its most common node forms were terrorist

(NF = 1.1) and extremist (NF = 0.9), compared to assist (NF = 0.09) and list (NF = 0.028) in enTenTen13. Although terrorist was the fifth most common node form in enTenTen13, its NF was just 0.02. In INC, most terms ending -ist denoted ‘crime’ or ‘extreme thought or belief’, which was not the case in enTenTen13, where only 3/49 terms could be placed in these categories (terrorist, activist, activists). These results indicated unusual and excessive use of the suffix in Islam news, typically with negative connotations. The remaining categories pointed to subsidiary topics in INC discourse; far right politics, Iran and Turkey, radicalization in British Education, Sharia law, the hijab, freedom of speech and blasphemy, secularity, segregation and intolerance. A number of smaller categories, for instance ‘dislike’, ‘unethical’, ‘no respect’ and ‘foolish’, consisted of words which conveyed varying levels of negativity. Finally, as with collectivisation of Muslims, KMWIs British society and Arab world indicated simplification (Bivens, 2014), hinting at cultural and ideological distinctions made in INC discourse. British society occurred 13 times in INC (NF = 0.06) and usually in one of three patterns (see Fig. 5, p.34). The first pattern reflected separation from British society, for example when preceded by constructions like marginalized from or isolated from (concordance lines 1–5), while the second reflected integration into British society (6 1 0). The third pattern alluded to problems within British society (e.g. discrimination, bigoted views) (11–13). Arab world occurred 7 times (NF = 0.03), and in two patterns (Fig. 6, p.34). The first pattern included descriptions of the Arab world as a large group of people who were angry (1 3) denoted by terms fury and boiling over: ‘‘The U.S. braces itself for fury of Arab world as Osama Bin Laden is killed in Pakistan” In the above example, the KMWI was surrounded by words denoting violence/anger and its use implied that all ‘‘Arabs” or ‘‘Arab” countries were sympathisers of Bin Laden and that the whole of the ‘‘Arab world” could be angry (humanization of a geographical label) at once. The second pattern constructed descriptions of the Arab world as a collection of countries or large

Fig. 5. British society in INC. AF = 13, NF = 0.06.

Fig. 6. Arab world in INC. AF = 7, NF = 0.03.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

geographical area where there was considerable political instability and terrorism (4–7) reflected in lexis such as destabilise, uprising, overthrow and Islamic State. This contrasted representation of British society, through ‘‘values” such as democracy, tolerance: ‘‘how to integrate immigrants into British society, and to ensure that they adopt values that are not actively hostile to the central ideals of our society – secular democracy, freedom of conscience, tolerance and the equality of everyone before the law.” The use of the verb ‘‘adopt” in the statement above (underlined) implied that immigrants could not have accepted or believed in ‘‘values” or ‘‘ideals” mentioned prior to their arrival in Britain. This in turn implied moral superiority of British Society. Overall, these themes indicated that Daily Mail and Telegraph news articles were preoccupied with the threat posed by Islam or Muslims to global peace, British security (through institutions such as the education system), and Western ideals. This sort of lexis, when repeated regularly, could serve to make differences between Western countries and Islam appear irreconcilable. 3.2. Islam comments: identity, extremism and collectivised portrayal Thirty-four keyword categories were identified in ICC (Table 8, p.35-36). Results implied high repetition of lexis and semantics in news articles. The ten categories attracting most keywords were ‘religious belief, identity and community’, ‘thought/belief’, ‘politics’, ‘crime’, ‘geographical names’, ‘no religion’, ‘law and order’, ‘foolish/ lack understanding’, ‘inclusion’ and ‘media’. All ten, except ‘inclusion’, were also key in INC and six other categories repeated INC categories. Keywords denoting ‘dead’, ‘dislike’, ‘violent/angry’ and ‘exclusion’ were again unusually frequent. These types of negativities were extended by additional categories of ‘fear’, ‘egoism’ and ‘evaluation: false’. 61/150 ICC keywords repeated INC keywords and 7/50 KMWIs, which indicated that, although readers worded comments in their own way more often than repeating the same multiword expressions, concepts introduced in news articles provided the foundation for reader comments. For instance, the same religious concepts, beliefs and identity labels were key as well as references to terrorism and extremist groups and individuals. The ‘religious identity, beliefs and practices’ category consisted of more than a quarter (41/150) of ICC keywords and almost one fifth (9/50) KMWIs, which was similar to the equivalent category in INC. The subcategories in ICC were all also key in INC. The two religions discussed most were the same as in the news (Islam and Christianity) and there was evidence of discussion of religions in general again (faiths and religions). Furthermore, one in ten KMWIs continued collectivised portrayal of Muslims as identified in news data (e.g. Muslim community, Muslim country). Approximately one third (3227/9308) of occurrences of Muslim in ICC were as an adjective. The ten most frequent nouns modified denoted large groups of people rather than individuals; Muslim country, Muslim community, Muslim women, Muslim population, Muslim world, Muslim men, Muslim area, Muslim leaders, Muslim state, Muslim nation. For example, there were 127 occurrences of Muslim women compared to just 23 of Muslim woman. This tendency implied descriptions which limited differences between Muslims, as in the example below: ‘‘Men are trying to control their women and make them cover in the name of Islam, I hope there will come a day the Muslim women will object.” In the first clause, the readers’ use of ‘‘their women” evokes a sense of Muslim women as the property of men, while the second clause presents this problem of ‘‘men [..] trying to control their women” as something which is the norm.

9

Keyword Muslim country indicated that readers may emphasise the concept of homogeneity at the geographical level, more so than journalists (NF = 0.15 in ICC, compared to 0.056 in INC). One potential reason for this was more explicit expression of nationalistic beliefs in comments. The main collocate of Muslim COUNTRY was live, which occurred 52 times. The most typical pattern (21/52 instances) was as follows: go (and) live in (a) Muslim COUNTRY Concordance analysis revealed use of this KMWI to reinforce national differences between Muslims and Christians (see Fig. 7, pp. 36–37). The category of words denoting ‘no religion’ (e.g. Atheism, unbelievers) was bigger in comments than news, indicating more variation in beliefs discussed. Keywords infidels, infidel and kuffar conveyed the presence of concerns about Islamic beliefs around non-Muslims and derogatory terms, such as scum and barbarians, expressed a level of anger based on news events reported. Whether such evaluations were most frequently directed at criminals or religion in general would require further investigation, which was beyond the scope of this paper. However, collocates of atheist (Buddhist, Hindu, Christian, Jewish, agnostic, nihilist, USSR, glad, secularist, Sikh, Hitler) also highlighted association or comparison with controversial ideologies or concepts, such as nihilism, communism and Nazism. This suggested that comments linked to ‘hard news’ topics tend toward ‘incivility’, regardless of the specific subject discussed (Coe et al., 2014; Holliman, 2011). The fact that readers could post anonymously, may also have led to more frequent expression of stronger personal views (Santana, 2014). Some keywords in ‘religious identity and belief’ also implied a potential to cause offence, for instance Taqiyya; a practice not commonly discussed or practiced by Muslims, whereby fearing persecution they may opt to hide belief, whereas lying is forbidden in most circumstances (Walker, 2009). Closer investigation of Taqiyya showed that there was difference in opinion about whether Taqiyya was a Shia or mainstream Islamic practice. The main narrative was; Taqiyya as an Islamic concept which allowed lying and deception (e.g. to deceive non-Muslims), making Muslims untrustworthy (21/50 instances): ‘‘Heard of ‘Taqiyya’? It’s a ’secret deception’ to fool all Kafirs (unbelievers) of the heinous and ugly truths of Islam” Other concordances revealed belief that Taqiyya was practised by Muslims to hide association with terrorism (5/50) (Fig. 8, pp.38–39); ‘‘I don’t expect a truthful or even logical response since I understand how Taqiyya works. They are terrorist sympathisers at best.” The examples above showed how something which originated in avoiding persecution was misrepresented and denigrated in reader comments. The keyness of lemma PREACH in INC and ICC indicated that it was an important activity referred to in this context. Further analysis revealed imam as the main subject of PREACH and hate and hatred as main objects. This connected the ‘religious identity, belief and community category’ to other categories such as ‘dislike’, showing that religion or religious people were sometimes the cause of hatred. This also connected to the second main category in ICC, ‘thought/belief’. The ‘thought/belief’ category, consisted of one in ten key terms, which was again dominated by keywords denoting extreme belief (e.g. Islamism, extremism, fundamentalism). As in the news, unusually frequent occurrence of suffixes -ist and -ism in this category was evident. As well as extremism, ICC was underpinned by talk

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

10

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Table 8 Keyword Categories in ICC. Keywords category

Subcategory

Keywords

Religious Belief, Identity and Community

Follower of religion/ religious sect

MUSLIM, MUSLIMS, CATHOLIC, CHRISTIANS, MOSLEMS, CATHOLICS, MOSLEM, SUNNI, BUDDHISTS, HINDUS, ORTHODOX CALIPHATE

The World

Thought/Belief

MUSLIM COUNTRY, CHRISTIAN COUNTRY, MUSLIM WORLD, ISLAMIC COUNTRY, ISLAMIC WORLD

Role/title Doctrine God/Prophet Name of religion General term for religion Practice/Belief Place of worship Attribute People Groups and affiliation

PREACHING, PREACHERS, PREACH, IMAMS, IMAM, KORAN, QURAN, VERSES, DOCTRINE, HADITH, PASSAGES MOHAMMED, PROPHET, ALLAH, PROPHETS, GOD ISLAM, CHRISTIANITY, RELIGIONS, RELIGION, FAITHS HALAL, RAMADAN, TAQIYYA MOSQUE, MOSQUES ISLAMIC, RELIGIOUS MUSLIM POPULATION

Extremism

ISLAMISTS, ISLAMIST, ISLAMISM, FUNDAMENTALISTS FUNDAMENTALISM RADICALS, EXTREMISM, FANATICS, EXTREMISTS TOTALITARIAN, BRAINWASHED, SATANIC BELIEFS, BELIEVERS, UNBELIEVERS

ISLAMIC EXTREMISM

LIBLABCON, LEFTIES, LEFTY, APOLOGISTS, CORRECTNESS, APOLOGIST, LIBERALS, BLIAR

General Politics

KMWI

OTHER RELIGION, PEACEFUL RELIGION

MUSLIM COMMUNITY

Anti-Islam

EDL, BNP, PEGIDA

POLITICAL CORRECTNESS, RIGHT WING, FOREIGN POLICY, POLITICAL CLASS, BRITISH GOVERNMENT, LEFT WING, PC BRIGADE VOTE UKIP

Crime

Terrorism/groups and affiliation Terrorism/criminals

HAMAS, TALIBAN, QAEDA, MB, INTERPAL

ISLAMIC STATE, ISLAMIC TERRORISM

Geographical Names

Nationality/ethnicity Places

ARABS, WESTERNERS, KURDS, PAKISTANIS, ISRAELIS, KURDISH PAKISTAN, HAMLETS

Atheism/Secularism Non-believer

ATHEIST, ATHEISTS, SECULARISM, SECULAR, DAWKINS KUFFAR, INFIDEL, INFIDELS

No Religion

General

OLD TESTAMENT

CHOUDARY, ANJEM, STALIN, OSAMA, EMWAZI

Law and Order

SHARIA, TREASON

Foolish/Lack understanding Inclusion

NUTTERS, IDIOTS, SPOUTING DELUSIONAL, BACKWARD TOLERANT, INTEGRATE, MULTICULTURALISM, CIVILIZED, INTEGRATING DM, DT, MSM, HODGES, OBORNE SAS, INVADE, INVADING, CRUSADES, DRONES BURKA, NIQAB, HIJAB, VEIL BEHEADING, BEHEADINGS, STONING, HENNING

Media War, Defence and the Army Clothes Dead People

MIDDLE EAST

BRITISH LAW, SHARIA LAW, FREE SPEECH, DEATH PENALTY

Dislike Time

HATRED, ISLAMOPHOBIA, SCUM

Violent/Angry Exclusion Fear Egoism Evaluation: false Relationships; intimacy, sex Discourse markers Money: poverty The World Culture No respect Affect: cause Moving, coming and going Weak Happy Possession Lack of attention Understand Chance, luck Evaluation: good Different

BARBARISM, BARBARIANS, BARBARIC SEGREGATED, INTOLERANCE, INTOLERANT COWARDS, COWARD BIGOTS, BIGOT, BIGOTRY HYPOCRITES, DRIVEL GAYS, VIRGINS EH, BTW GHETTOS

ARAB SPRING

HUMAN RACE, INDIGENOUS POPULATION, WESTERN SOCIETY, BRITISH SOCIETY GOOD RIDDANCE 7TH CENTURY, 21ST CENTURY, NEXT TIME, FIRST PLACE

NO SUCH THING

THIRD WORLD WHOLE WORLD, WESTERN WORLD BRITISH CULTURE, BRITISH WAY

BLASPHEMY INCITING MASS IMMIGRATION GULLIBLE LOL

about violence and crime. Keywords in ‘war, defence and the army’ category (e.g. invading) connected with ‘geographical names’ relating global conflict and terrorist attacks. Furthermore, the cate-

OWN COUNTRY BLIND EYE COMMON SENSE GOOD LUCK GOOD IDEA OTHER WAY

gories ‘crime’, ‘dead’, ‘violent/angry’ also mirrored INC keywords. These overlapped and together, included 26 key terms. Comments focused on a smaller number terrorist groups, individual terrorists

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

11

Fig. 7. Muslim countr(y)(ies) in ICC. Query: muslim countr*, AF = 387, NF = 0.15.

and victims of terrorism than news. Some keywords in the ‘dead’ category were distinctive from those in INC, (treason, death penalty), and overlapped ‘law and order’. These indicated that comments were about Islamic or Sharia law as well as British law, and that readers discussed punishments for crimes. Some instances of treason indicated that readers raised the question of treason laws in retain to terrorism, nationalism and solutions for dealing with British citizens involved in terrorist acts: 1. ‘‘Isn’t this called treason? Perhaps if Britain were to return to executing traitors a few of these ingrates biting the hand which fed them might get the Message.” 2. Time to pull out of international laws and the EU that are not in the best interest of the British people. What happened to OUR law of treason?

In the first example, the reader calls for a stronger response to British terrorists, while the second expresses dissatisfaction with current laws to deal with them. Other distinct keywords showed readers’ interest in Middle Eastern or North African unrest (Arab spring), references to oppressive forms of dictatorship (totalitarian, Stalin), types of punishment associated with Islam (stoning), a focus on the British context in relation to the army (SAS) and interest in historical rivalry between Christians and Muslims (crusades). Specifics about locations of attacks and conflict in articles were replaced by more long-term or historical causes for ideological separation. The word crusades varied in use and provided access to strong opinions in the data; 1. ‘‘the crusades against the blasphemy of Islam, murder and rape authorised by Pope Urban II”

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

12

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Fig. 8. Taqiyya in ICC. Query: taqiyya, AF = 66, NF = 0.025.

2. ‘‘It is one religion versus the rest. Let’s end it! The crusades only slowed them down its now the time to finalise the situation” In example one, the reader argued the crusades were inspired by material gains, while in example two, the crusades were presented as a necessary undertaking against the enemy – Islam.

These examples showed that the crusades were interpreted via different worldviews and readers invoked the same events with different purpose, reinforcing historical causes of polarisation in the current context. In comments, British victims were emphasised more than others (Henning rather than Foley, Haines, Kassig or Sotloff), and the more varied KMWI forms used to denote terrorism

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

and extremism in INC were encapsulated in the terms Islamic terrorism and Islamic extremism in ICC. In ‘politics’, many of the keywords reflected comments written by readers dissatisfied with mainstream British political parties; Conservative, Labour and Liberal Democrats (lefties, liblabcon, Bliar) combined with discussion about anti-Islamic political groups and parties (EDL and UKIP). Keyword UKIP typically occurred in the bigram vote UKIP (AF = 334, NF = 0.12), conveying readers’ association of Islamic extremism and terrorism, with the need for far-right political solutions. Ideological distinctions made in the discourse were also indicated by other categories of words, in particular, ‘people, groups and affiliation’, ‘the world’ and ‘culture’, where terms such as Western society and British way provided a different set of homogenising labels for people to those mentioned earlier (e.g. Muslim community). Finally, a combination of 16 smaller categories reflected minor themes; the media, Muslim women and the hijab, homosexuality and virgins in Islam, ‘ghettoization’ in Britain, immigration and blasphemy. In the following section, keyword categories for Catholicism news are presented and discussed. 3.3. Catholicism news: the vatican, foreignness, crime and scandal 13 keyword categories were identified in CNC (Table 9, pp. 40– 41). The dominant one was ‘Religious identity, belief and community’, which consisted of 125/150 keywords and 38/50 KMWIs. This was approximately three times the size of the equivalent category in INC. The three main subcategories in this category were ‘Pope and Vatican’, ‘other role/title’ and ‘practice/belief’. These indicated that the main religious aspects of Catholicism provided the topics of news articles. ‘Pope and Vatican’, consisted of 21/150 keywords and 14/50 KWMIs. Keywords included current, past and future Popes (e.g. Pius, next Pope), references to the Vatican (e.g. Vatican) and to papal activities or elections (papal visit, white smoke). Together keywords in ‘Pope and Vatican’ suggested that Catholicism was evaluated in terms of the actions, statements and beliefs of the Pope. Keywords such as new leader and Buenos Aires in ‘geographical names’ was explained by articles on the new Pope, his background and capability or legitimacy for the role. Concordances of keyword Pope Francis indicated the following narratives (Fig. 9, pp. 41–42): Pope Francis as; popular, friendly or loving (18/50 instances), associated with South America (8/50), new or different (8/50), associated with Far East Asia/Asian politics (5/50), praying or humble (4/50), advising or urging change (3/50), or unpopular in US (2/50). This suggested that Pope Francis was presented as foreign but generally in a 3 light in Daily Mail and Telegraph news. The ‘other role/title’ subcategory included 26/150 keywords and four KMWIs, reflecting the role or status of Catholics within a hierarchy. Other than the Pope, much of the discourse was about others representing church leadership (20/30 key terms in this subcategory) (e.g. cardinals, bishops) and included leaders of other Christian denominations (Anglican clergy). Occurrences of Archbishop were in Archbishop of Canterbury and (Roman Catholic) Archbishop of Westminster, who were interviewed in the news. Therefore, to some extent at least, evaluation of the Catholic Church was based on Church of England statements and perspectives as shown previously (Woolley, 2012). Keywords, penitents and devotees signalled discourse about ordinary Catholics rather than clergy members. However, further investigation of penitent (s) revealed negative and violent connotations, with reference to foreign Catholics participating in rituals involving self-inflicted injury and children abused by paedophiles (Fig. 10, p. 43). Closer inspection of the most frequent of keyword in ‘practice/ belief’, confession, also provided insight into how a key element of the Catholic faith was associated with crime. Concordances of confession (AF = 48, NF = 0.25) reflected that it was described as an oppressive practice and one that had been imposed on the Catholic

13

Church, an imposition which should now be lifted (Fig. 11, pp. 44– 45). The following were two examples of this: 1. Childhood confession, and the ideas it put into young heads, was oppressive.” 2. ‘‘Backed by the convictions of many hundreds of lay Catholics, I have sent an open letter to Pope Francis and top cardinals in Rome, seeking a ban on childhood confession.” The statements above were made in an article drawing on investigative journalism carried out for a book about child sex abuse cases involving priests. The journalist portrayed confession as an oppressive practice which provided opportunities for abuse to take place (example 1). This was used as a precursor to invalidate the need for childhood confession and call for reform in the Catholic Church (example 2). Furthermore, keyword crucifixion indicated stories about violence (9/35 instances), rather than the historical significance of Christ’s crucifixion to Christianity and Christians. Most of the discourse around this was found in Daily Mail and involved reporting of foreign ‘re-enactments’ of the crucifixion from around the world (San Fernando, Gauhati, Spain, Philippines, Argentina, Indonesia). In this way, Catholicism was distanced from Britain, and linked to controversial practices, involving extreme forms of religious devotion: ‘‘About two dozen Filipinos were nailed to crosses in an extreme display of devotion” Sometimes, the journalist admitted that the ritual described was an ‘‘extreme” example, however, this was rarely balanced by more common rituals or practices, a type of sensationalism which is typical of tabloid press (Boykoff, 2008). Furthermore, analysis of the most key attribute in CNC indicated that descriptions of Catholics were often negative. The term holy may be presented as a divine or revered quality in religious texts or contexts, but its use varied in internet English, as indicated by nouns modified by the adjective in enTenTen13 (grail, crap, shit, cow, scripture, shrine, trinity, relic, spirit, hell). In CNC, examples of the adjective in use showed association of the keyword with crime, therefore connecting the categories ‘religious identity and belief’ and ‘crime’: ‘‘an investigation into multiple counts of abuse by Fr Maciel Degollado, founder of a holy order called The Legionaries of Christ.” In the above example, the negative association formed between the words abuse and holy questioned not only the trustworthiness of the individual, but also the Catholic order mentioned, the Catholic Church more generally and even the value of being holy. This sort of sensationalism could be seen to be intrinsically newsworthy, in the Daily Mail in particular, which, as a mid-market tabloid newspaper (National Readership Survey, 2014), caters to readers seeking entertainment as well as information about major events nationally and globally. News values prioritised in tabloid news are personalisation, scandal and conflict, domestic stories, simplicity, entertainment and sensationalism (see Boykoff, 2008: 550), which serve to deemphasise objectivity (van Zoonen, 1991) and restrict detailed political, economic and social analyses (Djupsund and Carlsson, 1998). Other key terms showed discourse about other religions or Christian denominations (Islam, Protestantism) and their followers (Muslim(s), Protestant(s)), as well as a category for ‘no religion’ (atheism, secular). The Catholic order of which Pope Francis is a member was also key (Jesuits), while keywords Protestant and Reformation signalled discourse reinforcing historical divisions between Britain and Rome.:

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

14

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Table 9 Keyword categories in CNC. Keywords category

Subcategory

Keywords

KMWI

Religious belief, community and identity

Pope and Vatican

VATICAN, PONTIFF, POPE, PAPAL, BENEDICT, BERGOGLIO, XVI, PAPACY, FRANCIS, POPES, RATZINGER, POPEMOBILE, APOSTOLIC, OSSERVATORE, PIUS, HOLINESS, XII, LOMBARDI, PONTIFICATE, PONTIFICAL, TURKSON

Role/title

CARDINALS, BISHOPS, PRIESTS, CONCLAVE, CARDINAL, ARCHBISHOP, NUNS, NUN, PRIEST, MONSIGNOR, CLERICAL, FR, PRIESTHOOD, MISSIONARY, CLERIC, CLERGY, CLERICS, CONGREGATION, CONGREGATIONS, PARISHIONERS, PENITENT, PENITENTS, WORSHIPPERS, PILGRIMS, WORSHIPPER, PILGRIM, DEVOTEES CONFESSION, BEATIFICATION, PROCESSION, COMMUNION, CRUCIFIXION, CELIBACY, ORDINATION, HOMILY, CANONISATION, PROCESSIONS, BEATIFY, EXORCISM, ROSARY, LITURGICAL, PRAYER, PENANCE, CRUCIFIXIONS, RE-ENACTMENT ALTAR, PARISHES, CONVENT, SYNOD, SQUIRES, SISTINE, CHURCH, CHAPEL, ORATORY, CATHEDRAL, CHURCHES HOLY, ORDAINED, RELIGIOUS, DIVINE, BEATIFIED, CANONISED, PASTORAL, SAINTHOOD, EPISCOPAL, PRIESTLY, SPIRITUAL CATHOLIC, CATHOLICS, CHRISTIANS, ANGLICANS, PROTESTANT, ANGLICAN, PROTESTANTS, MUSLIM, MUSLIMS, JESUIT, JESUITS, FRANCISCAN CATHOLICISM, CHRISTIANITY, ISLAM, FAITHS, PROTESTANTISM, ANGLICANISM

NEW POPE, STATE VISIT, PAPAL VISIT, WHITE SMOKE, FIRST PONTIFF, FIRST POPE, NEXT POPE, NEW PONTIFF, NEXT PONTIFF, BLACK SMOKE, PAPAL CONCLAVE, RELIGIOUS LEADER, SPIRITUAL LEADER, CATHOLIC PRIEST, SENIOR CARDINAL, PARISH PRIEST, ANGLICAN CLERGY

Practice/Belief

Place of worship Attributes Follower of religion/ religious sect or order Name of religion

Geographical names No religion Media

Time

Doctrine/Education

DOCTRINE, THEOLOGY, GOSPELS, THEOLOGICAL, SEMINARY

Administration Cleric God/Prophet Groups and affiliation People Life

ORDINARIATE, DIOCESE, ARCHDIOCESE, ENCYCLICAL, CURIA SODANO, NICHOLS, NEWMAN, BROADHURST CHRIST, JESUS

Places Atheism/Secularism Television, radio and film News New

ASSISI, BUENOS, AIRES, PAMPANGA, CUTUD, COFTON ATHEISM, ATHEISTS, ATHEIST, SECULARISM, SECULAR, DAWKINS PHILOMENA, JUDI, SIXSMITH, DENCH, COOGAN

PARISH CHURCH

FIRST JESUIT, RELIGIOUS ORDER CATHOLIC CHURCH, CATHOLIC FAITH, CHRISTIAN FAITH, ROMAN CATHOLICISM, ANGLICAN CHURCH CATHOLIC SCHOOL, CATHOLIC DOCTRINE, LIBERATION THEOLOGY VATICAN BUREAUCRACY

CATHOLIC COMMUNITY CATHOLIC POPULATION RELIGIOUS LIFE

ROMANO REFORMATION

CELIBATE, HOMOSEXUAL

NEW LEADER, 70TH ANNIVERSARY, NEW STRUCTURE, MODERN WORLD ANNUAL EVENT GAY MARRIAGE

CONDOMS

BIRTH CONTROL

Frequency Relationship: Intimate/sexual Medicines and medical treatment Crime Kin Fear/shock Hiding Dead Politics

HOLY COMMUNION, FIRST CONFESSION, INAUGURAL MASS, ANGLICAN COMMUNION,

CHILD ABUSE, ISLAMIC STATE TRADITIONAL FAMILY SHOCK RESIGNATION VATILEAKS MORTAL PAISLEY

‘‘The greatest Victorian historian of Tudor England was James Anthony Froude, who eagerly explored the archives, but read them through inherited spectacles. A Protestant to his fingertips, he hated clergy, doctrine, religious mystery and, above all, Catholicism. He saw the break with Rome as the beginning of Britain’s rise to imperial greatness, and the Reformation as a confrontation between two incompatible civilisations.” (Telegraph) Ultimately, statements such as this portrayed Catholicism as inherently and historically incompatible with Britishness. Smaller categories reflected discussion relating to St Francis of Assisi, the adoption scandal and film ‘Philomena’, celibacy, relationships and the family, child abuse, secrecy within the RCC, beliefs about sins, Irish politics and Marxist ideology. KMWI Marxist ideology indicated associations made between Pope Francis and socialism in some articles. By linking the Catholic Church to socialism, right-

MARXIST IDEOLOGY

leaning news institutions could distance themselves ideologically and politically. Associations with religious extremism were even made, through references to the group Islamic State. 3.4. Catholicism comments: the pope, religious debate and nationalism Twenty-nine keyword categories were identified in CCC (Table 10, p. 46–47). The ten categories attracting most keywords were ‘religious identity, belief and community’, ‘no religion’, ‘ethics’, ‘politics’, ‘relationship: intimate/sexual’, ‘thought/belief’, ‘people’, ‘egoism’, ‘foolish’, ‘life’. As, with CNC, the dominant category was ‘religious identity, belief and community’, with 101/150 keywords and 21/50 KMWIs. Six of the smaller categories were not key in CNC; ‘ethics’, ‘thought/belief’, ‘people’, ‘egoism’, ‘foolish’ and ‘life’, but some of these were key in ICC. For instance, keywords relating to bigotry in ‘egoism’ were key in both sets of comments data, reflecting it as a topic of interest to readers in particular.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

15

Fig. 9. Pope Francis in CNC. Query: pope francis, AF = 294, NF = 1.49.

Over one third of CCC keywords (59/150) and nearly half KMWIs (17/50) were in the top CNC key terms, showing close connection between news articles and subsequent comments. Many subcategories also repeated; ‘Pope and Vatican’, ‘name of religion,

‘doctrine and education’, ‘practice/belief’, ‘role/title’, ‘follower of religion/religious sect or order’, ‘God/Prophet’, ‘general terms for religion’, ‘attribute’, ‘place of worship’, ‘the world’. However, compared to the news, ‘places of worship’ was much smaller and

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

16

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Fig. 10. Penitent in CNC. Query: penitent, AF = 25, NF = 0.13, A = self-inflicted injury during a Catholic ritual, B = child abuse.

‘church administration’ was no longer key. New subcategories were ‘doctrine/bible quotation’, ‘power/organising’ and ‘different’. ‘Pope and Vatican’ again attracted most key terms, denoting past, present and future Popes, the Vatican and papal elections. Although not as frequent in comments (NF = 0.35) as in the news (NF = 2.07), Francis was still the strongest collocate of Pope, showing readers’ interest in the current or new Catholic leadership. The main collocates of Pope Francis in CCC were; bless, i., francis, luck, 1, wish, congratulations, ii, welcome, guide, new. These implied at least some level of popularity of the new Pope with readers, as shown by the main pattern identified: God bless (you) Pope Francis However, concordance analysis revealed a range of descriptions (Fig. 12, p. 47–48); Pope Francis as supported, respected or liked (12/50 instances), the object of hope or expectation for change in RCC (7/50), frail, old or lacking ability or sincerity (7/50), Argentinian or Latin American (5/50), humble or compassionate (5/50), infallible or powerful (3/50), under investigation (3/50), educated (1/50), of European origin (1/50), not on friendly terms with US (1/50). The following is an example of the second and third narratives: ‘‘Good luck Francis the 1st you’ll need every ounce of luck you can get. If you can give way on Contraception and get rid of celibacy in the priesthood you will the greatest Pope of recent times.” In this case, the readers’ expression of encouragement is overshadowed by pessimism toward the new Pope’s ability to bring about change, which s/he deems essential to the Pope’s success in the role. Other readers countered this pessimism:

‘‘Pope Francis is a lovely guy- actually looking towards the poor in our community instead of the privileged. He leads a life of example to future popes- a life of modesty and compassion.” This comment supported the Pope through lexis that legitimised his position in society and countered previous comments (underlined). However, as mentioned above, aligning the Pope with Argentina may have been a nationalistic, polarising mechanism used by the media to reinforce historic rivalry between Britain and Catholicism. This was reinforced in comments where Falkland was the strongest collocate of Argentina, whereas in enTenTen13 it was 26th highest collocate. Comments which made these close links between Pope Francis, Argentina and the Falklands voiced concerns about the Pope’s stance, and claims about who the islands ‘‘belonged” to (Fig. 13, p.49). This was evident in the lexis referring to possession (give, belong, keep, claim, return), polarisation (side with) and conflict (invade) (concordance lines 1–11). Some readers stated that the Pope could or should dissuade the Argentinian government from a perceived invasion (12–13) and occasionally, readers expressed the opposing viewpoint, i.e. that the islands belonged to Argentina (13–14). Despite assurances given by the Vatican of its neutrality in the Falklands issue (quoted in the news), many readers still referred to the appointment of the new Pope as a threat to British interests. In CCC, Catholicism and its beliefs were portrayed and discussed in the context of other religions and beliefs. Some readers grouped non-established religions or denominations in Britain to reject them: ‘‘Catholicism is as damning a religion as Islam, Mormonism, and Jehovah’s Witness. Stay away from religion!”

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

17

Fig. 11. Confession in CNC. Query: confession, AF = 48, NF = 0.25.

Which was then countered by others: ‘‘Religion from Catholicism, Judaism, Islam etc is the one thing that gives people hope and inspiration when they have no one or nowhere to turn to especially when Secular government turn their backs.” These examples placed boundaries between religion and the secular and presented simplifications which reinforced polarisation between readers. Comparisons made by readers between dif-

ferent Christian branches were also varied. For example, Catholicism was grouped with the Orthodox churches to emphasise positive aspects by some readers but negative aspects by others. This also connected to the smaller category ‘evaluation: true’, which included keywords such as true Church and implied discourse supporting one Christian denomination over another. ‘Doctrine and education’ was a larger category in CCC comments, with keywords in ‘doctrine/biblical quotation’ indicating references to the bible by readers as support for arguments.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

18

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Table 10 Keyword Categories in CCC. Keywords category

Subcategory

Keywords

KMWI

Religious Belief, Community and Identity

Pope and Vatican

VATICAN, POPE, POPES, PAPAL, PAPACY, RATZINGER, HOLINESS, FRANCIS, BERGOGLIO, BENEDICT, PONTIFF, CONCLAVE CATHOLICISM, CHRISTIANITY, RC, RCC, ISLAM, COFE, PROTESTANTISM, SCIENTOLOGY

NEW POPE, BLACK POPE, NEXT POPE, BLACK SMOKE, WHITE SMOKE CATHOLIC CHURCH, CATHOLIC FAITH, CHRISTIAN FAITH, CHRISTIAN CHURCH, CATHOLIC RELIGION, ROMAN CATHOLICISM, ANGLICAN CHURCH CATHOLIC SCHOOL, OLD TESTAMENT, CATHOLIC DOCTRINE

Name of religion

Doctrine and Education Doctrine/Bible quotation Practice/Belief

Role/title

Follower of religion/ religious sect or order

God/Prophet General terms for religion Attribute Place of worship Power/ Organising Different The World

DOCTRINE, BIBLE, SCRIPTURE, THEOLOGY, GOSPELS, COMMANDMENTS, SCRIPTURES, DOGMA, THEOLOGICAL, DOCTRINES, BIBLICAL, WRITINGS THOU, THY, SHALT, AMEN, UNTO BAPTISM, CELIBACY, CONFESSION, COMMUNION, RESURRECTION, BLESS, PRAYED, WORSHIP, MIRACLES, CRUCIFIXION, ORDINATION, WORSHIPPING, TRANSUBSTANTIATION, HEAVEN PRIESTS, BISHOPS, CARDINALS, NUN, NUNS, PRIEST, CARDINAL, CLERGY, ARCHBISHOP, CLERICAL, VICAR, PRIESTHOOD, PREACH, PREACHING, CONGREGATION CATHOLIC, CATHOLICS, CHRISTIANS, BELIEVERS, PROTESTANT, MUSLIM, MUSLIMS, PROTESTANTS, ANGLICAN, ANGLICANS, ORTHODOX, JESUIT, JESUITS GOD, CHRIST, JESUS, APOSTLES RELIGIONS, RELIGION, FAITHS, FAITH, DENOMINATION, DENOMINATIONS RELIGIOUS, DIVINE, HOLY, DEVOUT, BAPTISED, SPIRITUAL, PRACTISING CHURCH, CHURCHES, CONVENT INFALLIBILITY HERETICS

CATHOLIC PRIEST

RELIGIOUS BELIEF RELIGIOUS PERSON,

ORGANIZED RELIGION OTHER RELIGION CHRISTIAN COUNTRY

No religion

Atheism/Secularism

ATHEISTS, ATHEIST, ATHEISM, DAWKINS, SECULAR, SECULARISM, ATHEISTIC, AGNOSTIC

SECULAR SOCIETY

Ethics

Unethical

SIN, DEVIL, SINS, SATAN, SINNERS, SINFUL

ORIGINAL SIN, MORTAL SIN, (TAR EVERYONE WITH THE) SAME BRUSH

General

MORALS

Politics Relationship: Intimate/sexual Thought/belief People Egoism Foolish Life Medicines and Medical treatment Money: Poverty

PAISLEY, BLIAR, LEFTY, STALIN CELIBATE, GAYS, WEDLOCK, HOMOSEXUAL

STATE VISIT, RIGHT WING

BELIEFS, SUPERSTITIOUS, SUPERSTITION, BRAINWASHED MANKIND BIGOTRY, BIGOTS, BIGOT FOOL, TWADDLE AFTERLIFE CONDOMS, AIDS

BELIEF SYSTEM HUMAN BEING, FELLOW MAN, HUMAN RACE

Evaluation: true Crime Time: Early/New Evaluation: false Media Autonomy Dead Education Humility Avarice Unimportant Inclusion Exclusion Weak Possession Evaluation: Good Dislike

PAEDOPHILES REFORMATION HYPOCRITES DM

OLD FOOL ETERNAL LIFE, OTHER LIFE BIRTH CONTROL THIRD WORLD, THIRD WORLD COUNTRY, 3RD WORLD TRUE CHURCH, TRUE RELIGION HISTORICAL FACT CHILD ABUSE EARLY CHURCH, MODERN WORLD (SORT YOUR) OWN HOUSE RED ARROW FREE WILL

MORTAL TEACHINGS HUMILITY GREED MEANINGLESS TOLERANT EXCOMMUNICATED GULLIBLE

Some key practices and beliefs were the same as in CNC, the main one being confession (NF = 0.065). Closer investigation showed that the practice received mixed evaluation in comments. In February 2014, an article published by Daily Mail made explicit links between the Catholic practice of confession and increased rates of child abuse. The headline read:

OWN CHURCH GOOD PERSON MUCH HATRED

‘‘How a Pope called Pius turned the confessional box into a paradise for paedophiles: From a leading Catholic writer, a devastating exposé of a Vatican ruling” In a selection of 50 (whole) comments which followed the article, 28 were found to be ‘against confession and/or Catholicism’, 17 ‘in favour of confession and/or Catholicism’, three ‘neutral or difficult

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

to categorise’, one ‘against Christianity’ and one ‘against religion in general’. This showed how discourse around confession connected to evaluation of Catholicism, because the practice of confession distinguished Catholicism from other Christian denominations. Most negative comments about confession were also negative about Catholicism, portraying it as a religion which encouraged unnecessary guilt, facilitated child abuse, and accommodated criminals:

19

‘‘Guess he can now go to confession as a Catholic and have the deaths of thousands of innocent people wiped from his conscience (if he has one that is, which I doubt).‘‘ In the example above, the reader defined former Prime Minister Tony Blair’s conversion to Catholicism as a way out of accountability for his mistakes. Therefore, media cynicism around confession

Fig. 12. Pope Francis in CCC. Query: pope francis, AF = 249, NF = 0.19. Narratives in discourse about Pope Francis in CCC (number of lines identified): d supported/respected/ popular/liked (12); d expectation/hope he will address RCC problems (7); d frail/old/ pessimism about ability/sincerity (7); d Argentinian (5); d humble/compassionate (5); d infallible/powerful/superior/correct (3); d under investigation (3); d educated (1); d European origin (1); d not on friendly terms with US (1).

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

20

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

Fig. 12 (continued)

was shared by some readers and may have guided opinions about the legitimacy of Catholic practices and Catholicism itself. However, CCC keywords were also different in some ways. Rather than ritual or ceremony which was emphasised in CNC practices and attributes (e.g. beatification, beatified, procession, ordained), CCC keywords highlighted afterlife and spiritual belief (e.g. miracles, eternal life). Keyword atheists, in the ‘no religion’ category, was ranked fourth in CCC. It had a keyness figure over five times higher than in CNC (398.7 compared to 71.8) and an NF about 10 times as high (0.62 compared to 0.07). This indicated that discussion in CCC widened to include readers’ views about religion compared to atheism. Similarly, keywords in ‘thought/belief’, for instance superstition signalled discourse about religion versus logic or reason: ‘‘Religions are all superstitions. This particular brand of superstition treats all women like dirt. The Roman Catholic church deserves to die.”

In the above comment, the reader used the word superstition to describe all religions, to explain gender discrimination in Catholicism and finally to express strong dislike toward the Catholic Church, potentially encouraging violence. This keyword, combined with other terms such as fiction and fairy tale in both comments datasets, pointed to discourse in which some readers rejected religion and expressed hatred in the discourse. However, the keyword, Stalin, led to comments made by some readers as counter-criticism to those claiming that atheism was peaceful and religion violent. Ironic comments such as the following one aimed at atheists, were equally simplistic and divisive as those aimed at religion above: ‘‘Good thing I’m atheist. so was Stalin” The ambiguity of this controversial remark highlighted problems faced by news institutions in moderating insensitive comments which sometimes appeared to promote extremism. Smaller categories also revealed the following themes; ‘morals,

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

21

Fig. 13. Falklands as a collocate of Argentina in CCC. AF = 15, NF = 0.013.

sins, and sinfulness’, ‘British and Irish politics’, ‘relationships and the family’, ‘the human race’, ‘bigotry’, ‘poverty and the third world’, ‘child abuse’ and ‘the Reformation’. Many conveyed some level of negativity in the discourse (‘egoism’, ‘foolish’, ‘crime’, ‘evaluation; false’, ‘avarice’, ‘greed’, ‘dead’, ‘exclusion’, ‘weak’ and ‘dislike’), although some indicated discussion about positive aspects (‘autonomy’, ‘humility’, ‘evaluation: good’). Overall, results showed a larger range of perspectives in comments compared to news, through the greater number of keyword categories and the examples identified, in which some readers countered the news and other readers’ negative opinions about Catholicism. However, as with Islam data, Catholicism comments were prone to strong and negative opinions. Despite evidence for optimism around Pope Francis, comments sections in Daily Mail and Telegraph served as platforms where a right-wing, nationalistic political and ideological stance prevailed and anti-Catholic, or sometimes anti-religion, perspectives were frequently posted.

4. Discussion Keyword analysis highlighted the contexts in which two minority religions in Britain, Islam and Catholicism, were portrayed in Daily Mail and Telegraph websites, as distinct from online UK news in general. The six dominant keyword categories in Islam news corpus (INC) were; ‘religious identity, community and belief’, ‘geographical names’, ‘crime’, ‘thought/belief’, ‘dead’ and ‘politics’. These encompassed aspects of religion discussed, as well as news stories focused on extremism, terrorism, places of war or terrorist attacks, murder victims, killings and interest in Turkish, Iranian and Dutch politics as well as far right political and anti-Islamic groups. These categories were repeated in Islam comments corpus (ICC). However, ICC keywords included fewer proper names referring to terrorists and places of terrorist attacks or war and more keywords in ‘politics’ which indicated criticism of mainstream politics and government, through terms such as political correctness, apologist, liblabcon and lefties. The ‘no religion’ category also attracted more keywords in ICC, suggesting some shift in the discourse from specific events in news articles to religion in general and comparison to atheism in comments. The two main keyword categories in Catholicism news corpus (CNC) mirrored those in INC; ‘religious belief, identity and community’ and ‘geographical names’. These indicated focus on the Pope,

Vatican and various practices, beliefs and attributes in news articles. Keywords in ‘media’ implied a focus on film Philomena which depicted the Irish adoption scandal, while those in ‘time’ reflected discourse about newness or change in the Catholic Church but also about the Reformation; the movement which defined separation between England and Catholicism. The ‘no religion’ and ‘relationship: intimate/sexual’ categories represented talk about atheism, secularity, celibacy and homosexuality. The main keyword categories in Catholicism comments corpus (CCC) overlapped those in CNC and were ‘religious belief, identity and community’, ‘no religion’, ‘ethics’, ‘politics’, ‘relationship: intimate/sexual’ and ‘thought/belief’. The Pope remained a central focus in reader comments. As with Islam comments ‘no religion’ was a larger category than news data, although the category of ‘politics’ attracted less keywords. This indicated that Islam was perceived more strongly as a political issue or problem compared to Catholicism by journalists and readers. However, there was evidence for ideological distancing, reflected in discourse around Pope Francis’ background, Argentinian politics and the Falklands in both Catholicism corpora. The results for news data confirmed previous studies which argued that Islam is portrayed as a religion of violence and extreme belief (Baker et al., 2013; Richardson, 2009) and Catholicism as a foreign institution defined by separation from Church of England, a problem due to practices such as confession and scandals (Taira et al., 2012). Thus perhaps, in spite of the change in technology used to report the news, i.e. the shift from print to online news, portrayal of Islam and Catholicism was based within the same dominant ideologies as previously reported. Furthermore, comments analysed showed strong overlap with news representation of these religions. 18 out of 19 keyword categories in Islam news repeated in comments, and 9 out of 13 of those in Catholicism news repeated in comments. As long-term phenomena, the generally negative discursive representations of Islam and Catholicism may point to historic rivalry between Islam and the West and Catholicism and Britain. This can be traced back to orientalism and neo-oriental theories (Halliday, 1996; Said, 1997) and 18th century Protestant literature (Brewer, 2000). In the latter, correlations were made between Catholicism and negative traits or behaviours, for instance primitivism, lower moral level, violence or riotousness, whereas Protestants were portrayed as culturally superior (Romani, 1998). Discursive repetition of beliefs and values by important social institutions reveal

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

22

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx

dominant and therefore difficult to challenge ideologies (Abercrombie and Turner, 1978) to which news audiences are exposed. Portrayal of Islam and Catholicism is perhaps less crude in contemporary British media contexts, including the rightleaning websites studied here, yet corpus analysis techniques adopted here provided evidence for subtle and explicit links made in news articles between the religions and violence, crime and oppression. In some news articles, the importance of ‘moderate’ Muslims taking a more public and active role against extremists was emphasized. However, at the same time, separation between Muslims and mainstream British society was reinforced, which was somewhat contradictory. This separation was constructed using collectivized terms, and repeated associations made with foreign names and places, confirming Taira et al. (2012) who identified ‘distancing’ in British media representations of Islam. Comments relayed concerns reported in the news, which were then framed in terms of discontent with the Government, left-wing parties and laws currently in place to deal with terrorism. This was coupled with support for far-right politics in ICC. Interestingly, keyword results indicated a relationship between the nature of news reporting about events and people, and the highly negative evaluation combined with nationalism and partisanship in the discourse. Close analysis, shed further light on whether and how portrayal of religion in comments overlaps with beliefs about race, ethnicity; i.e. whether intersectionality was at play. Finally, there was evidence that ideological divisions between Christianity and Islam were reinforced in comments, based on reminders of historical conflict. Readers expressed condemnation of particular militants described in the news, but there was also evidence for negativity toward Islam in general based on the actions of those criminals. By overrepresentation of terrorists in news relating to Islam, more mainstream Islam was not fairly represented. There was also evidence of extreme responses from some readers, calling for violence against whole communities as a solution, similarly to Richardson’s study (2009). There was a somewhat surprising similarity between these world-views and those of the terrorists that are being criticized. The greater number of keywords in ‘religious belief, identity and community’ in CNC than INC indicated that Catholicism was discussed in terms of its beliefs and practices in the news, moreso than Islam. Catholicism was represented as an institution rather than a community (the Catholic Church compared to Muslim community) and evaluated in terms of the Pope and the Vatican. The co-textual environment of some keywords representing ordinary Catholics (penitents), Catholic practices (confession) and religious attributes (holy) revealed associations made with violence, crime and oppression. At the same time, there was evidence for relatively positive evaluation of Church leadership (Pope Francis) in news and comments. Potential reasons for this may have been Catholic readership of the Daily Mail or Telegraph, or the phenomenon noted by Lund (2011) whereby Islam is represented as more ‘other’ than Catholicism. Negativity toward confession was countered in CCC, by readers who were more favourable toward Catholicism and occasionally, religions in general. However, an ironic tone in some complimentary posts aimed at the new Pope was identified, where the reader emphasised hopelessness of his mission. This showed the difficulty of analysing this register on a macro-scale. Although it was beyond the scope of this study to measure levels of incivility, many ICC and CCC keywords could be considered derogatory terms, which confirmed Coe et al. (2014) and Holliman (2011) findings that correlated incivility in reader comments with hard news and controversial topics. Examples of use of keyword Stalin also confirmed that ‘unproductive’ counter-arguments signalled conflict in the discourse. Ultimately, in comments, Catholicism and Catho-

lics were portrayed in a wider range of ways, compared to news articles, where institutional ideology tended to frame opinions projected. However, some readers evaluated religions, or atheism, more explicitly and more negatively than the news. By underrepresenting the views of ordinary Muslims and Catholics, and focusing on negativities, the media may perpetuate difference and hostility between mainstream and minority communities, causing increased Islamophobia (Garrido Rebolledo and Marquina, 2000; Benn and Jawad, 2004) or anti-Catholicism and legitimisation of anti-religious prejudice, racial abuse, vandalism and other violence (Fekete, 2005). However, exclusion constructed through discourse can be based on religious or other differences. For instance, racism may also occur in press portrayal (Ballibar and Wallerstein, 1991:60; Khosravinik, 2010). Some studies of online space and new media had foreseen a shift from traditional to new religious hegemonies (Beckett, 2012; Campbell, 2007). However, the present study revealed evidence for continuation of traditional ideologies regarding religion in websites of rightleaning news providers. Such institutions may have adapted to and benefitted in online media contexts, by offering audiences the opportunity to play a role in maintaining the shared and dominant ideology but potentially increasing audience separation from alternative worldviews, through filter bubble (Pariser, 2011) and echo-chamber effects (Jamieson and Cappella, 2008). In spite of debate and difference between readers, closer analysis of some key people, practices and beliefs highlighted strategies used to reject the place of minority religions, such Islam and Catholicism, in Britain today. Further, qualitative work studying the relationship between news and comments from a chronological perspective will enhance the work carried out here, for instance to shed light on how far-right political beliefs or nationalism become a focal point for discussion in news websites and whether deliberative communication also occurs in such contexts. Funding This work was part of a PhD study supported by Newman University, Birmingham, UK. References Abercrombie, N., Turner, B.S., 1978. ‘The Dominant Ideology Thesis’. Br. J. Sociol. (The London School of Economics and Political Science: Wiley-Blackwell) 29(2), pp. 149–70. Aharony, N., 2012. WikiLeaks comments: a study of responses to articles. Online Inform. Rev. 36 (6), 828–845. Archer, D., Wilson, A., Rayson, P., 2002. Introduction to the USAS category system. Benedict project report, October 2002. Atkin, A., Richardson, J.E., 2007. Arguing about Muslims: (un)reasonable argumentation in letters to the editor. Text Talk 27 (1), 1–26. Audit Bureau of Circulation, 2014. http://www.abc.org.uk. (accessed May 2015). Baker, P., Gabrielatos, C., McEnery, T., 2013. Discourse Analysis and Media Attitudes: The Representation of Islam in the British Press. CUP, Cambridge. Ballibar, E., Wallerstein, I., 1991. Race, Nation, Class: Ambiguous Identities. Verso, London, p. 60. Baron, A., Rayson, P., 2008. VARD2: A tool for dealing with spelling variation in historical corpora. In: Postgraduate conference in corpus linguistics. Batsell, J., 2015. Engaged Journalism: Connecting with Digitally Empowered News Audiences. Columbia University Press. Beckett, C., 2012. ‘Networked Religion’. In: Mitchell, J., Gower, O. (Eds.), Religion and the News. Ashgate, Farnhame, pp. 154–164. Ben-David, B., Folkman, A., 2010, ‘‘Readers’ comments (talk-backs) on the IsraeliPalestinian conflict in Canadian news-websites: a lighting-rod for extreme views vs a ban on free discussions. http://www.boazbendavid.org/uploads/4/5/ 7/5/4575323/ben-david__folkman_2010.pdf. Benn, T., Jawad, H.A. (Eds.), 2004. Muslim women in the United Kingdom and beyond: Experiences and images (Vol. 2). Brill Academic Pub. Bivens, R., 2014. Digital Currents: How Technology and the Public are Shaping TV News. University of Toronto Press. Boykoff, M.T., 2008. The cultural politics of climate change discourse in UK tabloids. Polit. Geogr. 27 (5), 549–569. Branum, J., Charteris-Black, J., 2015. The Edward snowden affair: a corpus study of the British press. Discourse Commun. 9 (2), 199–220.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001

T. Bruce / Discourse, Context & Media xxx (2017) xxx–xxx Brewer, J.D., 2000. Understanding anti-catholicism: a summary of the arguments. Stud.: Irish Quart. Rev. 89 (355), 200–204. Campbell, H., 2007. Who’s got the power? Religious authority and the Internet. J. Comput.-Med. Commun. 12 (3), 1043–1062. Christensen, H.S., 2011. Political activities on the Internet: Slacktivism or political participation by other means? First Monday 16 (2). Coe, K., Kenski, K., Rains, S.A., 2014. Online and uncivil? Patterns and determinants of incivility in newspaper website comments. J. Commun. 64 (4), 658–679. Colleoni, E., Rozza, A., Arvidsson, A., 2014. Echo chamber or public sphere? Predicting political orientation and measuring political homophily in Twitter using big data. J. Commun. 64 (2), 317–332. Colley, L., 2005. Britons: Forging the Nation, 1707–1837. Yale University Press. Collins, L., Nerlich, B., 2015. Examining user comments for deliberative democracy: A corpus-driven analysis of the climate change debate online. Environ. Commun. 9(2), PP. 189-207. Cozens, C., 2004. ‘BBC accused of ’gross insensitivity’ to Catholics’. The Guardian. (4 February 2004). Diakopoulos, N., Naaman, M., 2011. Towards quality discourse in online news comments. In Proceedings of the ACM 2011 conference on Computer supported cooperative work. ACM, pp. 133–142. Djupsund, G., Carlson, T., 1998. Trivial stories and fancy pictures. Nordicom Rev. 19 (1), 101–113. Domingo, M., Jewitt, C., Kress, G., 2014. Multimodal social semiotics: Writing in online contexts. Evans, D., 2007. Corpus building and investigation for the Humanities. University of Nottingham http://www.corpus.bham.ac.uk/corpus-building.shtml. Fekete, L., 2005. ‘Protecting ethnic minorities. Eur. Race Bull. (Institute of Race Relations) 53, pp. 15–30 (21-6). Fowler, R., 1991. Language in the News: Discourse and Ideology in the Press. London: Routledge. Harcup T and O’Neill D (2001) What is news? Galtung and Ruge revisited. Journalism Studies 2(2) Freund, E., 2011. ‘‘‘Discuss this article!’. Participatory uses of comment sections on Spiegel Online: a content analysis”, available at: http://www2.lse.ac.uk/media@ lse/research/mediaWorkingPapers/MScDissertationSeries/2010/Freund.pdf (accessed 24.01.11). Galily, Y., 2008. The (re)shaping of the Israeli sport media: the case of talk-back. Int. J. Sport Commun. 1 (3), 273–285. Garrido Rebolledo, V., Marquina, A., 2000. The dialogue between the European Union and the Islamic World. Annals of the European Academy of Sciences and Arts (Vol. 24). Gurak, L., 1997. Persuasion and Privacy in Cyberspace. Yale, New Haven. Halliday, F., 1996. Islam and the Myth of Confrontation: Religion and Politics in the Middle East. IB Tauris, London. Hecht, J., 2004. Online discourse and a social medium. Internet Association Magazine. Hermida, A., 2011. Participatory journalism: Guarding open gates at online newspapers. Mechanisms of participation, pp. 11–33. Holliman, R., 2011. Advocacy in the tail: Exploring the implications of ‘climategate’ for science journalism and public debate in the digital age. Journalism 12 (7), 832–846. Inglehart, R., Norris, P., 2016. Trump, Brexit, and the rise of populism: Economic have-nots and cultural backlash. Jackson, H., 2014. Words and their Meaning. Routledge. Jamieson, K.H., Cappella, J.N., 2008. Echo Chamber: Rush Limbaugh and the Conservative Media Establishment. Oxford University Press. Jenkins, H., 2006. Convergence Culture: Where Old and New Media Collide. MIT Press, Cambridge, MA. Jenkins, P., 2003. The New Anti-Catholicism: The Last Acceptable Prejudice. OUP, Oxford. Karaflogka, A., 2002. Religious discourse and cyberspace. Religion 32(4), pp. 279– 291. Karlsson, M., 2011. The Immediacy of Online News, the Visibility of Journalistic Processes and a Restructuring of Journalistic Authority. Journalism 12 (3), 279– 295. Kaufhold, K., Valenzuela, S., De Zúñiga, H.G., 2010. Citizen journalism and democracy: How user-generated news use relates to political knowledge and participation. J. Mass Commun. Quart. 87 (3–4), 515–529. KhosraviNik, M., 2010. The representation of refugees, asylum seekers and immigrants in British newspapers: A critical discourse analysis. J. Lang. Polit. 9(1), pp. 1–28. Knott, K., Poole, E., Taira, T., 2013. Media Portrayals of Religion and the Secular Sacred: Representation and Change. Ashgate Publishing Company. Kohn, A., Neiger, M., 2007. ‘‘To talk and talkback: analyzing the rhetoric of talkbacks in online journalism”, paper presented at the Israel Communication Association Conference, 19 February, Tel Aviv. Leung, L., 2009. User-generated content on the internet: an examination of gratifications, civic engagement and psychological empowerment. New Media Soc. 11 (8), 1327–1347. Lewis, S.C., 2012. The tension between professional control and open participation: Journalism and its boundaries. Inform., Commun. Soc. 15 (6), 836–866. Loke, J., 2012. Public expressions of private sentiments: unveiling the pulse of racial tolerance through online news readers’ comments. Howard J. Commun. 23 (3), 235–252. Lund, C.C., 2011. New victims of the old anti-catholicism. The. Conn. L. Rev. 44, 1001.

23

Mahlberg, M., 2013. Corpus Stylistics and Dickens’s Fiction. Routledge. Manosevitch, E., Walker, D., 2009. Reader comments to online opinion journalism: A space of public deliberation. In: International Symposium on Online Journalism, vol. 10, pp. 1–30. Manosevitch, I., Tenenboim, O., 2017. The multifaceted role of user-generated content in news websites: an analytical framework. Dig. J. 5 (6), 731–752. McCune, B., Grace, J.B., Urban, D.L., 2002. Analysis of ecological communities, vol. 28. MjM software design, Gleneden Beach, OR. McEnery, T., Xiao, R., Tono, Y., 2006. Corpus-Based Language Studies: An Advanced Resource Book. Taylor & Francis. Mitchell, J., Gower, O. (Eds.), 2012. Religion and the News. Ashgate Publishing, Ltd.. Milioni, D.L., Vadratsikas, K., Papa, V., 2012. ‘Their two cents worth’: Exploring user agency in readers’ comments in online news media. Observatorio (OBS) 6 (3). Miller, D., Biber, D., 2015. Evaluating reliability in quantitative vocabulary studies: The influence of corpus design and composition. Int. J. Corpus Linguistics 20 (1), 30–53. Moore, K., Mason, P., Lewis, J.M.W., 2008. Images of Islam in the UK - The representation of British Muslims in the national print news media 2000-2008. Morozov, 2009. Morozov, E., 2009. From slacktivism to activism. Foreign Policy 5 (September). Nagar, N., 2011. The Loud Public: The Case of User Comments in Online News Media. State University of New York, Albany, NY. PhD dissertation. Neuberger, C., Nuernbergk, C., 2010. Competition, complementarity or integration? The relationship between professional and participatory media. J. Pract. 4 (3), 319–332. Nickels, H.C., Thomas, L., Hickman, M.J., Silvestri, S., 2012. Constructing ‘‘suspect” communities and Britishness: Mapping British press coverage of Irish and Muslim communities, 1974–2007. Eur. J. Commun. 27(2), pp. 135–151. O’Malley, P.R., 2006. Catholicism, sexual deviance, and Victorian Gothic culture. CUP, Cambridge. National Readership Survey, 2014. http://www.nrs.co.uk (accessed May 2015). Pariser, E., 2011. The filter bubble: How the new personalized web is changing what we read and how we think. Penguin. Partington, A., Duguid, A., Taylor, C., 2013. Patterns and meanings in discourse: theory and practice in corpus-assisted discourse studies (CADS) (Vol. 55). John Benjamins Publishing. Pepinster, C., 2012. ‘Religion and the Specialist Press’. In: Mitchell, J., Gower, O. (Eds) Religion. Petley, J., 2010. Rules, recycling, filters and conspiracies. J., Sources, Credibility: New Perspect. Pierre, D.F., 2011. Double Standard: Abuse Scandals and the Attack on the Catholic Church. eBookIt. com. Pihlaja, S., 2011. Cops, Popes, and Garbage Collectors: Metaphor and Antagonism in an Atheist/Christian YouTube Video Thread. Language@internet, 8. Pihlaja, S., 2014. Antagonism on YouTube: Metaphor in Online Discourse. Bloomsbury, London. Poole, E., 2002. Reporting Islam: Media Representations of British Muslims. IB Tauris, London. Ratcliffe, P., 2004. Race, Ethnicity and Difference: Imagining the Inclusive Society. Open University Press, Maidenhead. Reich, Z., 2011. User Comments. Participatory journalism: Guarding open gates at online newspapers, pp. 96–117. Richardson, J.E., 2009. ‘‘Get shot of the lot of them”: election reporting of Muslims in British newspapers. Patterns Prejudice 43 (3–4), 355–377. Romani, R., 1998. British views on Irish national character, 1800–1846. an intellectual history. History Eur. Ideas 23 (5), 193–219. Rosenthal, R., 2004. The Joy of the Hebrew Language. Am Oved, Tel Aviv. Royle, R., 2012. ‘Popular Media, News and Religion’, in J, Mitchell, & O, Gower. (Eds) Religion and the news. Ashgate, Farnhame, pp. 229–241 Saeed, A., 2007. Media, racism and Islamophobia: the representation of Islam and Muslims in the media. Sociol. Compass 1 (2), 443–462. Said, E.W., 1997. Covering Islam: How the Media and the Experts Determine How We See the Rest of the World. Vintage, London. Santana, A.D., 2014. Virtuous or vitriolic: The effect of anonymity on civility in online newspaper reader comment boards. J. Pract. 8 (1), 18–33. Taira, T., Poole, E., Knott, K., 2012. ‘Religion in the British Media Today’. In: Mitchell, J., Gower, O. (Eds) Religion and the News. Ashgate, Farnhame, pp. 60–76. Tenenboim, O., Cohen, A.A., 2015. What prompts users to click and comment: A longitudinal study of online news. Journalism 16 (2), 198–217. van Dijk, T.A., 1991. Racism and the Press. Routledge, London. van Dijk, T.A., 2006. Ideology and discourse analysis. J. Polit. Ideol. 11 (2), 115–140. van Zoonen, L., 1991. A tyranny of intimacy? Women, femininity and television news. Commun. Citizenship, 217–235. Virilio, P., 1997. Cybernetics. Anyone, New York, NY. Wahl-Jorgensen, K., 2002. The construction of the public in letters to the editor: Deliberative democracy and the idiom of insanity. Journalism 3 (2), 183–204. Walker, P.E., 2009. ‘‘Taqıyah”. The Oxford Encyclopedia of the Islamic World. Oxford University Press, Oxford. Watts, G., 2004. Bishops Accuse BBC of Anti-Catholic Bias. National Catholic Register (28 March 2004). Woolley, P., 2012. ‘A Relationship Worth Getting Right’. In: Mitchell, J., Gower, O. (Eds) Religion and the News. Ashgate, Farnhame, pp. 99–121. Xiao, R., 2007. What can SLA learn from contrastive corpus linguistics? The case of passive constructions in Chinese learner English. Indonesian J. English Lang. Teaching 3 (2), 1–19.

Please cite this article in press as: Bruce, T. New technologies, continuing ideologies: Online reader comments as a support for media perspectives of minority religions. Discourse Context Media (2017), https://doi.org/10.1016/j.dcm.2017.10.001