Abstract
While numerous studies view the internet as a patron of internationalism and public empowerment, this comparative study of leading news websites in nine nations shows that online news is strongly nation-centred, and much more inclined to cite the voices of authority than those of civil society and the individual citizen. Online news is very similar, in these respects, to newspaper and television news. This convergence is due to the way in which leading media conglomerates have extended their hegemony across technologies. It also reflects the constraints exerted by the wider societal context across all media.
Keywords
In a celebrated book, Frances Cairncross (1997: xvii) predicted that the internet, and the communications revolution more generally, will ‘increase understanding, foster tolerance, and ultimately promote world peace’ because ‘people will communicate more freely and learn more about the aspirations of human beings in other parts of the world’. The internet, Bulashova and Cole concurred (1995), offers ‘a tremendous “peace dividend” resulting from improved communications with and improved knowledge of other people, countries and cultures’. ‘The Internet’, echoed the American politician, Vern Ehlers (1995), ‘will create a community of informed, interacting, and tolerant world citizens.’ Underlying these and similar statements is the belief that the internet is promoting global harmony by fostering dialogue between nations.
While this digital utopianism has subsided since the mid 1990s, there continues to be a widespread belief that the rise of the internet and other international media is fostering the flowering of a cosmopolitan sensibility and the weakening of national allegiance. This is despite the fact that understandings of cultural globalisation, 1 and of the national, 2 differ significantly. Thus, according to Jon Stratton (1997), the globalism of the internet is weakening identification with the nation (‘deterritorialisation’ ), and loosening ties to specific contexts (‘disembedding’). The globalisation of communications and the economy, agrees Saskia Sassen (2002: 13), is contributing to ‘the partial unbundling or at least weakening of the national’. Deterritorialisation fostered by the development of international communications systems, enthuses John Tomlinson (2007: 153), is ‘an exhilarating and empowering phenomenon, involving the simultaneous penetration of local worlds by distant forces, and the dislodging of everyday meanings from their “anchors” in the local environment’. Like many others, he sees the rise of the internet as contributing to an empathetic opening up to other people and places in a process that McEwan and Sobre-Denton (2011) call approvingly ‘virtual cosmopolitanism’.
The second great benefit of internet technology, it is widely argued, is that it places a powerful tool of communication in the hands of the people. The internet, according to Mark Poster (2001:175), is empowering ‘previously excluded groups’. The top-down communication between elites and the general public, afforded by traditional media, is allegedly being subverted by horizontal communication between citizens through the internet. Some see the internet as ushering in a renaissance of democracy (Agre, 1994), while others view the internet as a ‘key ingredient’ in the ‘recipe’ for democratisation that is weakening dictatorships (Howard, 2011). This view of the internet as an agency of freedom and popular empowerment has, if anything, gained even greater acceptance in the wake of the Arab Spring. Its most eloquent exponent is Manuel Castells (2012: 218, 231) who argues that ‘the technology of the Internet embodies the culture of freedom’ and is ‘changing the world in the network society’.
These two themes – the internet’s fostering of internationalism and public empowerment – have been threaded together in a seductive synthesis. One important version of this synthesis comes from the political theorist, David Held, who argues that the ‘mediated connectivity’, promoted by the internet and global communication systems, is exposing national identities to diverse solidarities, promoting acceptance of difference, and fostering an increased sense of global interconnectedness. This is contributing to the development of a ‘multilevel citizenship’ that is needed to sustain the development of multi-tiered governance (at global, continental-regional and national levels). This evolution of governance, he stresses, is already under way, and is a necessary response to globalisation and the declining efficacy of national government. New media are thus supporting the growth of cosmopolitan culture that is a necessary counterpart to the building of a ‘cosmopolitan democracy’. Held also sees the development of new global communications systems as furnishing multi-level fora of debate and public scrutiny that will make the emergent multi-level system of governance more accountable (Held, 2004, 2006; Held and McGrew, 2007).
Another version comes out of social theory, and is advanced in a seminal essay by Nancy Fraser (2007). Her starting point is that ‘the Westphalian framing of political space’ – that is, a system of national and inter-state political decision-making – is no longer adequate in a globalised world. National governments cannot manage national economies or national environments in the context of globally integrated capitalism. This means that national governments can no longer implement, even in principle, the discursively formed will of national public spheres. However, help is at hand because the ‘denationalization of communication infrastructure’ and ‘the rise of decentred internet networks’ has allegedly caused communications to break out of the ‘container of the nation’, and generate international flows of information and ideas (Fraser, 2007: 18–19). But although the international public is a legitimate locus of public opinion since all are affected by globally interconnected issues, it lacks efficacy. The paramount problem is to identify the ‘addressees’ of international public opinion ‘in the sense of new, transnational public powers that possess the administrative capacity to solve transnational problems’ (Fraser, 2007: 23).
Despite some significant differences, both Held and Fraser are clearly engaged in a common project: how to repair democracy in the age of globalisation. They both invoke an extensive literature that portrays the internet as internationalising and empowering, based on two assumptions: the internet, as a global medium, promotes internationalism; and as a low-cost, self-expressive medium, the internet amplifies the voice of ordinary citizens. These assumptions are challenged in this comparative study of nine news websites, which examines systematically the nature of online news representation.
But first we begin with a description of the methods we used, and a brief account of changes in the consumption of news.
Investigating news websites
The nine websites are based in Australia, Colombia, Greece, India, Italy, Japan, Norway, South Korea and the United Kingdom. This choice of country was determined partly by pragmatic considerations, skewing it towards affluent democracies and the exclusion of certain parts of the globe. 3 But while the sample is not fully representative, it nonetheless straddles four continents (America, Europe, Asia and Oceania). It also includes economically developing as well as developed societies, new as well as old democracies, majoritarian as well as consensual political systems, and relatively egalitarian as well as inegalitarian social systems. In other words, it provides the basis for an informed insight into how leading news websites, based in different parts of the world and in differently organised societies, report the news.
The first phase of our study took the form of a quantitative analysis of a leading news website, the two most watched TV news programmes, and two leading newspapers, in our sampled countries. 4 Our choice of news website was determined by two considerations. Each website had to be independent of other media in the sample; and it had to be among the three most popular news websites (as measured by its number of unique visitors) in each country. This last stipulation arose from the fact that much academic attention has been given to unpopular websites – a key point to which we shall return. Our focus is thus on news sites that are widely used. 5
The same, pre-determined classification scheme was applied to the relevant content of all sample media during three, non-sequential weeks (excluding weekends) in May–June 2010 by trained coders in each country. Using Huddle software, we set up a three-day ‘academy’ to train coders employed in the different countries in order to secure consistent coding.
The content analysis of news websites was confined to the first level link to which coders were directed from the home page. Coders were instructed to code links connected to actual news (that is, to avoid generic links, or links to pictures or music not accompanied by news) published on the day that they were coding. User-generated content (about which more will be said later) and certain routine features, such as weather, games, and job seeking information, were excluded, while professional blogs were included. 6
The second phase of research took the form of a survey administered to a sample of 1000 adults in each country shortly after the period covered by the content analysis. The samples were fully representative, save in Colombia and India where the survey was confined to urban areas. 7 The survey was conducted online, apart from in Greece and Colombia where interviews were conducted respectively by telephone and face-to-face. 8 In the case of online surveys, a matching procedure was used which delivered the equivalent of a conventional probability sample on the basis of specified demographic attributes from pre-established panels (for a more technical discussion of sample matching, see Iyengar and Vavreck, 2011). 9 From this estimates were derived, based on self-reporting, of frequency of exposure to different media.
Changes in news consumption
At the top of the news hierarchy in most countries is television: more people claim to attend regularly to TV news than to any other news medium in seven out of the nine countries. The exceptions are Norway and South Korea where average self-reported exposure to web news equals or slightly exceeds that to TV news. Both these two countries are in the vanguard of technological change: Norway has the highest internet penetration rate (97%) in Europe, while South Korea (83%) has the highest internet penetration rate in Asia (Internet World Statistics, 2012a, 2012b).
In sampled countries, the newspaper press and web vie for second place. The web is the second most regularly used news source in some countries, notably Greece and Australia, where newspapers are relatively weak. Newspapers are especially strong in India, and continue to have a large following in certain countries, most notably Japan (see Table 1). At the bottom of the news hierarchy is radio, used only by a minority as a news source, though it is stronger in Britain and Australia than elsewhere.
Average self-reported exposure to news media across countries (1 to 5 scale, where 1= hardly ever and 5= every day).
A MANOVA with country as the only between-participants factor reveals reliable differences across countries in the five measures of media exposure (F (45, 48730) = 105.58, p < .001, Partial Eta Squared = .09). All Univariate analyses led to reliable differences across countries (all Fs > 44.54, all p < .001).
National averages conceal large, within-country disparities of news exposure, particularly between the young and old and between those with low and high educational attainments. For example, 37 percent of 18–34-year-old adults in the sample nine nations say that they do not seek information, on a regular basis, from any news medium, compared with only 13 percent of those aged over 54 years old.
The distribution of news exposure between social groups also varies significantly between countries (see Table 2). Apart from Greece, where average newspaper consumption is very similar among older and younger adults, older adults tend to read newspapers more often than younger ones. TV news consumption is also more regular among the old than the young across all countries. A similar pattern is to be found in relation to radio (with the exception of India, where younger adults listen to radio news more often than older adults). The picture is a bit more varied when it comes to online news: in Australia, Greece, Japan, Korea and Britain, younger adults tend to use the internet as a source of news more often than do those aged over 54. But this is not the case in Norway, where there is virtually no difference between the age groups in their exposure to online news; nor is it true of Italy, where a higher proportion of older people than young people say that they use the internet regularly as a source of news (partly because older people in Italy consume more news in general).
Average exposure to news media across age groups in sampled countries.
Note: A Multivariate ANOVA showed reliable differences across countries and age groups in exposure to all media, as well as a reliable interaction between the two IVs (main effects: all Univariate Fs > 45.63, all ps < .001, all partial eta squared > .009 [main effect of age on on-line news]; Interaction: all Fs > 3.96, all ps < .001, all partial eta squared > .007 [interaction age*country for radio news]). Colombia was excluded from the analysis as it did not have data on radio attendance.
Thus, what emerges is a picture both of continuity and change. The primacy of TV persists in most countries, but a generational shift is also taking place. In most countries, though not all, young people are more oriented towards the internet and are less oriented towards the press, TV and radio as sources of news than older people. This shift suggests that, in due course, the established news media hierarchy will change (as it has done already in two countries). What are the wider implications of this cumulative shift?
Top-down perspectives
Since the internet has been widely hailed as an empowering tool of the people, we begin by examining who gets to be represented in online news compared with the news of other media. Our measure is based on who is interviewed, quoted or cited (as in ‘the government believes that …’): in other words, on explicit sources of news. For journalists to qualify as a source, they have to be interviewed or reported in a role going beyond that of reporter or presenter.
For this part of the analysis, our examination is confined to hard news stories concerned with public affairs topics (such as politics, public administration, economy and related issues). Soft news stories concerned with human interest, celebrity, gossip and entertainment are excluded. Crime news is classified as ‘hard’ if the story is contextualised and related to the public good (for example, concerned with the cause or prevention of crime in general) but judged soft if it is narrowly focused on the perpetrator, crime and victim. 10
Online news emerges as being very similar to the news of other media in being heavily reliant on state representatives (such as government ministers and public officials) and on experts. Together, these represent 70 percent of the sources of online news compared with 64 percent of press sources and 60 percent of television news sources. On average, online news actually gives less of a hearing to the political opposition, civil society 11 and individual citizen sources than either television or newspaper news in the nine countries (see Table 3).
Sources of hard news across media.
That traditional media are heavily dependent on elite sources has long been acknowledged, and has given rise to a long-running debate about whether media representation of public affairs is ‘index linked’ to elite opinion in the US (Bennett, 1990; Curran, 2011; Entman, 2004; Mermin, 1999), and framed by powerful ‘primary definers’ in the UK (Curran, 2002; Davis, 2009; Hall et al., 1978; Manning, 2001). 12 What needs to be recognised now is that the main body of online news is even more closely tied to the voices of authority than ‘legacy’ media (Table 3).
However, there are some significant differences between news websites (Table 4). Thus, although the state is the most important hard news source for all sampled websites, the most state-oriented websites are in India, Japan and Italy: the least state-oriented, relatively speaking, are in Australia, Greece and Norway. This is partly reversed when it comes to representing civil society and individual citizens, with Australia, Greece and Norway giving most attention to these sources, and Italy giving the least.
Sources of hard news across countries.
These divergences are partly attributable to national differences. This is seemingly borne out by the way in which web news in the same country is often similar to television news. Thus both the website news and TV news of Norway give more prominence to individual citizens and civil society spokespersons than their counterparts in the other eight nations. Similarly, both the web news and TV news of India give more attention to state sources than in any other sampled country. When this national effect does not operate, it is usually due to the special characteristic of the website or TV channel under consideration. For example, the reason why the sample website in Italy gives so much attention to state sources is probably due the fact that it was owned by the then Italian prime minister, Silvio Berlusconi, unlike the TV channels with which it was compared. 13
Another significant basis of variance between websites is the use of experts. Again, this tends to reflect different national conventions operating across media. Thus, in our sample, Australia makes greater use of experts than any other country in both its website and TV news.
In short, news websites do not stand out as citadels of empowerment, amplifying the voice of the people. On the contrary, web news gives greater attention to the voices of authority and expertise than to civil society and the ordinary citizen.
But perhaps the more important point is that there is an underlying affinity in the way which news is sourced across all the media. In effect the same kind of news, based on talking to the same kinds of people and reporting what they say, is being reproduced regardless of technology. And where there are differences between websites, these arise partly in response to national differences operating across media, irrespective of the technology being deployed.
National prism
Since the internet is associated with the cultivation of a cosmopolitan consciousness, we also investigated the geographical distribution of online news. Across the nine nations, 71 percent of website news is devoted, on average, to the home country. This is a lower than the proportion in the press (77%) and in television newscasts (74%) in the nine countries. Even so, web news is entirely conventional in being strongly nation-centred.
We also distinguished between stories with an exclusively foreign focus and those which make explicit reference to the home nation. This quantitative measure does not capture the subtlety of the concept of ‘domesticated foreign news’. As Lee et al. (2000) and Shoemaker and Cohen (and their colleagues) (2006) indicate, foreign news can be framed in terms of the national interest, national discourse or shared national memory – in other words within a national framework – without explicit reference to the home nation. Yet, even based on our simple measure, approaching a quarter of all foreign news in the nine websites is explicitly related to national concerns (see Table 5).
Proportion of news focusing on national, international or domesticated international news. a
Press scores exclude Colombia and Norway.
If web news apes press and TV news in being nation-centred, it is also similar in other ways. Previous research indicates that television news and newspapers are inclined to render some parts of the world relatively invisible, and to focus on great powers and neighbouring countries (Curran et al., 2010; Shoemaker and Cohen, 2006, among others). This is also true of leading news websites.
Thus the United States, the world’s superpower, accounts for 26 percent of all international hard news in the nine websites, and for higher proportions in some countries, most notably Japan where the US accounts for half of foreign web hard news and Australia where it constitutes 45 percent.
Website foreign coverage tends also to concentrate on neighbouring countries. Thus 37 percent of the foreign hard news of the Indian website centres on Pakistan; 27 percent of its equivalent in Japan focuses on China; 21 percent of its counterpart in South Korea is devoted to Japan; and the sample websites in Greece, Italy and Norway have a strong orientation towards news in European countries and the European Union.
The consequence of this dual focus on powerful countries (especially the USA) and on neighbouring countries is to neglect other parts of the globe. In particular, Africa receives minimal attention, with only two countries (Congo and South Africa) achieving respectively even 1 percent of overall foreign hard news coverage – largely due to reporting of the Congo on the Norwegian website, and of South Africa in the Japanese, Norwegian and British websites (see Table 6).
Percentage of international hard news dedicated to specific countries across sampled websites.
Note: For presentation purposes, we excluded all countries whose overall representation figure was below 1 percent. Empty cells indicate that the country is absent or accounted for less than 0.49 percent of international hard news in that country. The data presented constitute 89 percent of the overall international hard news items.
There are some differences between websites. The Norwegian website is the most outward looking of the nine websites, devoting 48 percent of its total news to foreign topics and events. The most insular are websites in South Korea, India and Italy, which allocate between 13 and 25 percent of their news to foreign stories. The websites most inclined to relate foreign news reports to domestic affairs are in Japan, United Kingdom and Greece.
Once again, the evidence suggests that these differences arise partly from national differences. Thus, Norway has not only the most internationalist news website but also the most internationalist TV news of the nine nations, reflecting the country’s strongly internationalist political culture. 14 Similarly, both website news and TV news in South Korea have a markedly insular news focus, reflecting Korea’s historical evolution as the ‘Hermit Kingdom’. The tendency to relate explicitly foreign news to domestic concerns in Japanese and British websites is also apparent in their TV news, indicating that they are following national news conventions (see Table 7).
Proportion of news focusing on national, international or domesticated international news.
Note: Press scores exclude Colombia and Norway.
In brief, leading news websites are not advancing a cosmopolitan outlook alert to other people and places. They focus on home news coverage and routinely blank out large parts of the world, just like other news media.
Conclusion
To sum up, leading websites around the world reproduce the same kind of news as legacy media. These websites favour the voices of authority and expertise over those of campaigning organisations and the ordinary citizen. They also focus on the home nation, and offer a highly selective view of the world, in ways that are familiar from watching television and reading newspapers.
Would these results have been significantly different if user-generated content had been included in our web content analysis? Clearly, the inclusion of this content would have widened the spectrum of interpretation offered by sampled websites. But our focus has been on news, and it is unlikely that user-generated content would have changed significantly the geographical distribution of foreign news reporting, or have fundamentally altered the range of sources in online news. This is seemingly corroborated, both directly and indirectly, by research into online journalism. Among other things, this concludes that user-generated input to news websites overwhelmingly takes the form of comment rather than of news reporting; this comment tends to follow the topic agenda of the website; and, in general, online news professionals are strongly resistant to ‘losing control’ to amateurs (Bakker and Pantti, 2009; Domingo, 2008; Domingo et al., 2008; Hanitzsch and Quandt, 2012; Hermida and Thurman, 2008; Thurmin, 2008, among others).
Three key influences have constrained the development of online journalism. The first is the success of established media organisations in extending their hegemony across technologies by setting up heavily subsidised news websites supported by extensive newsgathering resources, prominent brand names, and the ability to cross-promote. 15 Faced by these superior resources, and operating in a context where people expect online content to be free, independent news websites – with the potential to be different – have often struggled to survive (Downie and Schudson, 2009; Pew, 2009, among others).
Nine of the ten top news websites in the world are owned by large news organisations (Alexa, 2012b). The exception, Huffington Post, is no longer independent and is now owned by the internet giant AOL. Our sample of websites reflects this pattern: all but two – India’s Rediff.com and South Korea’s Daum.net – are owned by large media groups, which partly explains why these websites are so similar to legacy media. 16 In addition, the trend towards content convergence between national news media is being strengthened by the growing trend towards the integration of online and offline news rooms within the same organisation, and by the increasing tendency for news websites to lift stories from each other (Boczkowski, 2010; Lee-Wright et al., 2012).
The second key influence is the wider social and political context of society. As we have seen, variations between news websites tend to reflect national differences that are reproduced in the news values of other media in the same country. There are also similarities between the sample websites reflecting commonalities between nations. Expertise and official authority are privileged over the voices of ordinary citizens and civil society in online news because the former have greater prestige and power. The home nation also looms large in online news because of the enduring cultural power exerted by the nation in the era of globalisation, and also because politics and democracy is still practised primarily in the nation state.
The third key influence is the nation-centred dynamic that is shaping international online journalism. The model of networked, multilingual websites directed towards a global community has not taken off. Instead websites belonging to large news organisations based in one country – such as the BBC, Globo, Fox News, Huffington Post, New York Times, CNN, Guardian, Wall Street Journal and Times of India among others – are building international audiences among their same-language and diasporic communities. But these organisations are, to a lesser or greater degree, national news organisations, with a core national audience, and nationally inflected concerns. Their utilisation of the global technology of the internet is perpetuating, sometimes in a modified form, a national editorial legacy. 17
The reality of what is happening has been partly obscured by the way in which scholars have tended to concentrate attention on a small number of alternative, international websites such as International.OhMynews (Dencik, 2012), openDemocracy (Curran and Witschge, 2010) and the exhaustively studied global activist network, Indymedia (Atton, 2004; Coyer et al, 2007; Jankowski and Jansen, 2003; Hanke, 2005; Miloni, 2009; Morris, 2004; Pickard, 2006; Platon and Deuze, 2003; Salter, 2006, among others). These websites are both cosmopolitan and politically progressive, seemingly fulfilling widely shared perceptions of the impact of the internet. However, these websites have also attracted relatively small audiences. In terms of global user traffic, they come lower – indeed very much lower – in the international pecking order than every one of the websites studied here (Alexa, 2012 a). 18
Thus, the central conclusions of this study are that popular online news is much like other forms of news, and that this is largely the consequence of the incumbent economic power of leading media conglomerates, and the constraints exerted by the wider social context across all media. This runs counter to the received wisdom outlined at the beginning of this article. Popular online news emerges as not being in the vanguard of globalisation but as being tenaciously nation-centred. It is also a vehicle of authority rather than a voice of the people. This does not detract from the innovatory theorising of Held and Fraser, who correctly identify a central problem in the functioning of liberal democracy, and outline a route map of reform. But it does strongly suggest that their solutions, based in part on the belief that the internet is an agency of globalisation and popular power, will be more difficult to achieve than they anticipate.
While this survey report goes against the grain of received wisdom, it is not in fact unsupported by other work. One group of researchers questions the extent of the empowering nature of the internet principally by arguing that the net is constrained by state, corporate and elite influences (e.g. Curran et al., 2012; McChesney, 2013; Morozov, 2011, 2013). Another group emphasises – more as qualification than as a rebuttal – the limits of globalisation by stressing the continuing importance and relevance of the nation and national state institutions (Calhoun, 2007), the importance of national cultures in shaping the content of the web and online interactions (Miller and Slater, 2000), the weakness of global media (Hafez, 2007), and the incomplete, region-centred nature of global economic integration (Weiss, 1998). The conclusions of this study are thus part of a dissenting tradition saying similar things.
But in siding with this dissenting tradition, we are very definitely not subscribing to a doctrine of technological irrelevance. The internet has wrought important changes in some areas of life but not in others, and more in some countries than others. Thus, if popular online journalism has not broken the mould, there is compelling evidence that the internet has increased the effectiveness of political activists, whether on the left or the right, and whether operating locally, nationally or transnationally (e.g. Alexander, 2011; Juris, 2005; Rohlinger and Brown, 2009, among others). But ultimately activists’ effectiveness has depended on whether they connected to a powerful groundswell of popular feeling. Thus in the Middle East and North Africa region, insurgencies took place in countries where internet penetration was relatively low (with the exception of Bahrain) but where resentment against the regime was strong, not in countries where internet penetration was high but the local regime was strongly embedded. 19
A cumulative picture is beginning to emerge. We have extensive knowledge about the impact of the internet in some areas (most notably its profound influence on the economy) but also large tracts of ignorance. What is also coming into focus is an impression of uneven, internet-facilitated change, strongly constrained or extended by the external environment. However, much more work is needed in the form of empirically based case studies – such as this one – if we are to make an informed judgement about the existence or otherwise of the ‘internet revolution’.
Footnotes
Funding
This work is supported by the UK Economic and Social Research Council: Japan’s Society for the Promotion of Science; the Korea Research Foundation and Korean Government; the Ministry of Information, Greek Government; Research Council of Norway; Sky Italia; University of Madison-Wisconsin; Australian Research Council (DPO665146) and University of Hyderabad/MICORE.