Abstract
This paper examines the relationship between unofficial party blogs and official party sources in the UK using a mixed-method approach. Specifically we combine interview data with content analysis, user surveys and usage data, and finally hyperlink analysis to profile the emergence, popularity, audience and online prominence of four major party blogs since 2005. The core question posed is how far the blogs are challenging parties as the focal point for member activism and offering an alternative public ‘voice’. The findings show blogs occupy an important alternative critical space for party debate, particularly outside elections. They are not mobilizing tools, however, being used by the grassroots largely for information-gathering and discussion purposes.
Introduction: organizational change in political parties and ICTs
To date, examinations of new information and communication technologies (ICTs) and political parties have centered on the public arena of e-campaigns and electioneering (Bimber and Davis, 2003; Foot and Schneider, 2006; Ward et al., 2008). By contrast, the impact of ICTs on intra-party affairs has been considerably less explored. This article examines the implications of new ICTs for party organizational change through the lens of new partisan blog networks that have grown up in the UK. In particular we explore the significance of four unofficial party blogs that have developed since 2005 – Conservative Home (CH), Labour Home (LH), Labour List (LL) and Liberal Democrat Voice (LDV). Our central question is what role is played by these outlets in party affairs and specifically whether they provide a new quasi-official voice for members and a basis for grassroots activism. We use a multi-method approach to address these questions and qualitative and quantitative data detailed below.
Academic attention on the internal uses of new ICTs, while limited, has been increasing. Against a backdrop of organizational problems (declining membership; falling levels of activism; shortages in funding; and weakening links into civil society through falls in partisan attachment) (Needham, 2005; Norris, 2002; Seyd and Whiteley, 2002; Webb, 2006), parties across liberal democracies have shown an increasing interest in the use of new technologies for organizational renewal.
Much of the work on parties’ internal and organizational uses of ICTs has been speculative and non-empirical. Early writers, such as Rheingold (1994), Negroponte (1995), and Budge (1996) offered predictions about parties’ declining role in the Internet era as part of the wider erosion of representative institutions and the increased use of direct methods of citizen input. The continuance of the state and collapse of the dot.com bubble challenged these more radical interpretations of the impact of new ICTs, and more moderate and party-driven accounts followed, with authors coalescing around two alternative scenarios: one being the emergence of a more decentralized, transparent and grassroots-led ‘cyber-democratic’ party model; the other, a trajectory of an increasingly centralized and ‘techno-elite’-dominated organization that would erode members’ collective power to hold their leaders accountable (Lipow and Seyd, 1996; Lofgren and Smith, 2003; Smith, 2000). Subsequent work by Margetts (2006) and Saglie and Heider (2005) reinforced these dual possibilities, but as two sides of a single new party model − the ‘cyberparty’ or ‘network party’ – in which members were afforded more flexible channels for participation and opportunities to form and join looser issue- and policy-based networks. The levels and timing of individuals’ involvement could vary, ranging from simply receiving regular news updates to donating funds and contributing feedback on an individual policy or issue basis. Such easy exit and entry barriers were also seen to create the possibility for a more passive and floating support base less able to hold leaders accountable, and the emergence of a narrower digital elite that displaces the more traditional activist base.
Empirical investigation of these projections has been limited to date. Smith and Webster (1995) provided a brief account of the uptake of computing and information systems within the three main parties in the UK from the 1980s onward. Their analysis revealed that usage was confined to an already active IT-literate minority and that the technology was regarded as a labor-saving device rather than a means for promoting grassroots activity. Gibson and Ward’s (1999) study of national party elites’ attitudes toward internal computerized communication systems confirmed this apparent lack of interest in more participatory uses of the technology. Subsequent analysis of Labour and Liberal Democrat party members presented a more positive scenario, however, with new ICTs appealing more strongly to the younger cohort. Online joiners, however, were found to be more passive than those joining through traditional means, while IT also served to deepen activism among the already active, creating an internal digital divide (Lusoli and Ward, 2003, 2004).
In general then, it seems that parties’ internal adaptation to new ICTs may be promoting a more centralized mode of operation and a narrowing of their support base. However, the limited and rather dated nature of the evidence supporting this argument raises questions about how robust such conclusions are. Additionally, more recent online initiatives with social media appear to challenge earlier assumptions. The development of the party blogs investigated here is one prominent example of how new ICTs may be helping to democratize and revitalize internal party politics.
Research questions, data and methods
In light of the above discussion, we sought to carry out a broad-based exploration of the role of these new grassroots spaces and their relationship with more formal party websites. In particular we focus, first, on inter-party comparisons to discover whether the three major UK parties and their associated bloggers and audiences take a similar approach to online communication and networking. Second, we are interested in an intra-party perspective that is capable of understanding the relation between the parties and their grassroots and blogger communities. The development of these online communities raises important questions in terms of the quality and inclusiveness of political discussion around party structures, which we operationalized with five sets of research questions:
What are the origins of the partisan blog sites and why were they created?
Who contributes to the sites and what type of content is present? In particular, how much dissent and criticism of the party is featured?
How wide is the reach of these sites on the Web and who forms the audience for such sites?
To what extent do the parties’ official websites and the partisan and independent blogs form a cohesive online issue network? Or do they form a separate alternative and independent network for party political discussion?
How does the wider political context, understood in terms of position in the electoral cycle and party incumbency, affect the contents and structure of party and blog websites?
Given the range of questions and the exploratory nature of the research, a multi-method strategy was considered appropriate. Four modes of data collection and analysis were used to address our questions:
(1) Seventeen semi-structured interviews were conducted between November 2008 and January 2009 with party officials, bloggers and blog editors. The interview data was used to provide background information on the context and origins of the blog sites and the function, purpose and audience for the sites.
(2) Qualitative content analyses of three blog sites (CH, LL and LDV) were used to address question 2. 1 Specifically, we examined the number and type of postings across two time points: (i) during the last week of the election campaign (1−7 May 2010) and (ii) after the election (10−16 October 2010). Capturing data from these two time points meant that we could assess both the effect of the election campaign on the sites and the new (Conservative−Liberal coalition) government. Given the decline of LH in terms of updated content by 2010, we replaced it with the LL site which had become more prominent by this point. As well as the number and frequency of postings, we identified the source of the posts (editors, politicians/official party personnel or private individuals), and the tone of postings. The latter categorization used five basic codes − ‘supportive’, ‘neutral’, ‘critical/challenging of the party’, ‘opponent-focused’ and ‘other’. This was done to assess how far the sites really offered an alternative voice to, or challenged, their party leadership.
(3) Audience/user statistics: two data sources were utilized to measure the sites’ reach (question 3). The first was a range of site statistics produced by http://Alexa.com, a Web information and traffic-ranking service. The data used by the service are compiled primarily from its large body of registered Alexa Tool Bar users, although they also incorporate other ‘unspecified’ sources. The data reported here included traffic rankings, the percentage change in traffic to the site over a three-month period (calculated in terms of overall global Internet users), and the number of in-links a site receives. A second source of information on these sites’ audience was provided via an online survey of users of the LDV blog fielded in September−October 2009. The survey gathered a range of demographic and attitudinal data on site visitors as well as more specific information about their motives in using the site and perceptions of its importance. 2
(4) Hyperlink network analysis was conducted using two publicly available social network analysis software applications. These tools allowed analysis of the structure and content of online networks occupied by blogs in relation to other party-related sites (question 4). Specifically, Issue Crawler was used to collect the hyperlink data and map the basic parameters of the relevant networks. More in-depth analysis and visualization of the contents and properties of the blog networks was undertaken with a Microsoft Excel template called NodeXL. 3 Two stages of link harvesting with Issue Crawler were then performed. The first stage focused on collecting the links contained in the home pages of the official party website and the relevant party blog, so a total of three ‘seed sets’ of links were created − one each for Labour and LH/LL, the Conservatives and CH, and the Liberal Democrats and LDV. In a second stage, we created larger seed sets of links that were taken from the official profiles of parties and blogs on the most popular social media websites (see Appendix 1 for full details). Both stages were performed across our two time points of May and October.
(5) The final question on the impact of political context was addressed by cumulating and comparing the findings of the content analysis, usage statistics and hyperlink analysis over the two time points (May 2010 and the general election campaign, compared to the October−November 2010 ‘peace-time’ political context).
Origins: the rise of unofficial party blogs
CH pioneered the unofficial collective partisan blog format just before the 2005 election, with LH and LDV emerging during the following 18 months, while Labour List (LL) is a more recent addition, having been created in early 2009. Indeed, the initial success of CH acted as a catalyst motivating other parties’ responses. As the editors’ welcome post on LH admitted: ‘Last year Tim Montgomerie set up www.conservativehome.com and it seems to have done a good job at focusing grassroots Tory opinion – so we thought, “why can’t we do something like that?”’ (Labour Home, 20 June 2006).
All the sites claim that representing the grassroots or providing a space for supporters’ voices to be heard is one of their primary goals. The LDV editor’s initial post explains the site as a ‘place for activists and party members to have a conversation … This place is intended to be by activist for activists … It is not an official blog’ (LDV, 9 September 2006). The tag lines of the sites are also indicative, with LDV described as ‘our place to talk’, LH as ‘back to the grassroots’, and LL as ‘Labour’s biggest independent grassroots e-network’. Similarly, Montgomerie’s reflections on the fourth birthday of CH listed 10 broad aims and achievements, including representing the grassroots, protecting party democracy, campaigning for a balanced Conservative message, and initiating campaigns (CH, April 2009).
Despite claims to represent the grassroots, none of the sites were created by party outsiders or ordinary supporters. CH was launched by Tim Montgomerie an ex-senior aide to former Conservative leader Iain Duncan-Smith and was funded by Stephan Shakespeare, director of the polling organization YouGov. LDV was started by Rob Fenwick, who was the party Internet Campaigns Officer at the 2005 election. Fenwick was joined subsequently by Mark Pack, previously the party’s Head of Innovations. The remaining editors were largely a mixture of local councilors or activists. At LH, founder and editor Alex Hilton had been PPC for Labour in 2005, and his co-founder Jag Singh had worked on John Kerry’s 2004 US presidential campaign. Although initially unsponsored, LH was bought by Mike Danson, publisher of the left-of-centre New Statesman magazine in 2008, although Hilton remained the editor. 4 LL is comparable to LH in terms of its quasi-official status (unlike the Left Foot Forward blog, for instance, which is promoted as a non-aligned source for left-wing progressives). In its early days LL was enmeshed in a political scandal over leaked emails designed to discredit leading Conservatives. Despite its rocky start, LL has increasingly strengthened its position and profile, eclipsing LH as the most prominent independent party blog of the centre left.
Over time, CH has developed into the most popular and influential of the sites under investigation. It has developed the largest audience of the three sites (see below for details) and regularly gains coverage in the mainstream media. Tim Montgomerie appears frequently on radio and television news shows to offer the voice of the Conservative grassroots, and he contributes articles to the broadsheet press. In part, the media coverage and attention is not surprising since CH has three full-time staff dedicated to the site (Montgomerie, ex-Daily Telegraph journalist Jonathan Isaby and former MP Paul Goodman). However, it is not simply the mechanics or the professionalism of the site that explain the development of CH. It established a prominent voice and its grassroots credentials from the start. It quickly became an influential voice in internal party debates and an agenda-setter. Within months of its creation, CH led opposition to then-leader Michael Howard’s attempt to abolish the one-member one-vote rule for the election of the party leader. It also built its audience through its coverage of the subsequent 2005 leadership battle. Nor has current leader David Cameron always had a smooth ride from CH, especially in the initial stages of his leadership. CH has been critical of the so-called ‘A list’ of parliamentary candidates drawn up by Central Office to widen the social base of party representatives. More recently, it has also raised questions about future Conservative spending plans and argued for a reduced role for the state.
Compared with CH, the other sites occupy a lower profile both within and outside their respective parties and have yet to exercise as strong an influence on policy. In autumn 2008, LH did make headlines in the mainstream media (and was criticized by the party) for running a poll of its members that called for Gordon Brown to resign as party leader. LL clearly has strong aspirations in this direction, identifying itself as ‘an independent progressive blog providing a platform for open debate about centre-left issues and the future of the Labour movement’.
Contributors and content: the emergence of grassroots communities?
All three sites in our content analysis (CH, LDV and LL) were regularly updated with multiple posts each day. The volume of posts was, unsurprisingly, more intense in the last week of the election campaign for all the sites. As Table 1 shows, CH clearly dwarfs the other two, and particularly LL, in terms of the quantity, regularity and range of postings (averaging 21 and 16 per day in the election and peace-time periods, respectively). That is nearly three times as many posts as LL in the election-week period. CH effectively offers a comprehensive, professionalized news service with large numbers of links to the mainstream media, and blog stories relating to key issues and Conservative affairs and personalities. There are extensive links to a video site with clips of the latest political interviews and speeches. The other two sites have more of a feel of traditional blogs, although they do mirror the CH format on a more limited scale. They both have regular daily links to news stories and audio-video coverage. LDV also provides a members-only discussion forum in addition to the public blog site.
Range and volume of posts and authors.
In both periods, editorial teams dominated the postings on all three sites but especially CH and LDV, and in particular, in the election period. During the campaign, over 90 percent of posts on LDV came from the editorial team and 83 percent on CH. While this figure is considerably smaller in October, editors still account for over half the postings on all the sites. Whilst all three have grassroots input, this is arguably greater for LDV and especially for LL. LL postings contained contributions from both ordinary party members and well-known figures such as Alistair Campbell and Ken Livingstone. While elected party representatives did post on all three sites, they did so in relatively limited numbers and hardly at all during the election campaign. CH tends to have most regular postings by elected officials and other quasi-official party voices and groups, but again these differences partly reflect the structure of the sites. For instance, CH has regularized and formalized features, sections and posters. Besides the editors, there are regular posts from right-of-centre think-tanks and internal pressure groups such as Policy Exchange, the Taxpayers Alliance and the Social Market Foundation.
We also looked at the tone of postings and the level of criticism of the official party. Here a number of trends stand out (see Table 2). First, the level of questioning/challenging posts during election time is very low. Of the four posts which might be considered to have mildly challenging comments three came the day after the election, raising questions about the campaign or warning party leaders not to do deals with other parties. The blogs seem to have taken on a more supportive role during the campaign, with the vast majority of posts offering either partisan support for the party or focusing on attacking opponents. In the post-election period we see a rise in challenging posts on CH and LDV. In part, this can be accounted for by tensions in maintaining the coalition between the two parties. The biggest issue of contention, especially for LDV, was the controversial issue of university tuition fees that surfaced as the main political issue during this week. Ten of the twelve critical posts were on this issue alone on LDV. In the case of CH, some of the challenging posts relate to the issue of the European budget, where posters wanted to push the leadership towards a more Euroskeptic position. Third, and somewhat surprisingly, given the problems of the Labour election campaign and the subsequent leadership debate, critical posts on LL were almost entirely absent. Indeed, in the post-election period, there appeared to be a considerable shift towards attacking the new government’s policies. Since our October study was conducted in the immediate aftermath of Ed Miliband’s election as party leader, it could be that the findings indicate a coming-together of the party to support the new leader after a period of internal party debate. Finally, there was little outright hostility or open criticism in any of the postings. The tone of challenging posts remained polite, although the comments sections on all three sites allowed a significant amount of vitriol to be aimed at all the parties.
Tone of posts.
Although the extent of commentary fell after the election, there remains a healthy discussion in response to posts on all the sites. Most notably, the level of debate on LDV had risen since the controversial decision of the party to back proposals for increases to university tuition fees. This was an issue that had drawn more interest and feedback than any previous debate, according to LDV editors.
Overall, the content study indicates that these sites represent active communities for partisan supporters. They allow a growing diversity of voices to be heard and could certainly be seen as good indicators of issue importance to the party grassroots. They all follow similar patterns in terms of broad content and structure, although CH is the most professionalized and has a stronger focus on news coverage. The differences in campaign mode and peace-time study suggest that such sites can function in a more supportive manner at election time, focusing more on internal debate outside that competitive context.
Audiences: reaching out or reaching in?
Beyond the content of these sites, we were also interested to examine their significance and potential impact in the wider Web sphere. To do this, we utilized two data sources – Alexa statistics on site rankings and an online survey of LDV users.
The Alexa ranking results displayed in Table 3 demonstrate that the Conservatives are the most prominent political force online. The party’s official homepage and CH are the most frequently visited of all the sites investigated and enjoy the highest number of in-links, i.e. links directed to them from other sites. The only exception here is during the May 2010 election period when the Liberal Democrats rose slightly higher in the rankings than either site: a surge undoubtedly related to the higher profile of Liberal Democrat leader Nick Clegg after he performed beyond expectations in televised leadership debates.
Alexa basic site statistics (May 2010 and Nov 2010) a .
Dates of data collection: 2 May 2010; 1−2 November 2010.
Figures are calculated from registered Alexa Toolbar users (volume of unique URL visits and page view requests), plus other sources (unspecified) over a rolling three-month period and are updated weekly. Rankings are reported for top-level domains only. Rankings below 100,000 are seen as unreliable and differences are not statistically meaningful. Scarcity of data for these sites means their ranking may be subject to large swings.
Figures are Alexa statistics for unique visits to home pages, standardized as a percentage of global Internet (as reported 15 August 2009).
In-links are crawled and indexed by Alexa generally two to three times per month. They do not count links from ‘link farms’ or ‘exchanges’ in their calculations or from sites that have no traffic.
Comparing the performance of the blogs to that of the party sites overall, we can see the latter proving more of a popular venue during the election and the former asserting their dominance in the post-election period. The traffic rankings (where lower scores mean higher ranking) show that the major parties’ sites were ranked within the top 60,000 of all websites analyzed by Alexa and within the top 1000 UK sites during the election in May. The only blog to join the parties in this level of prominence was CH. Post-election, while CH and LH both fell in overall popularity, the latter spectacularly so, both LL and LDV by contrast appeared to maintain and even increase their audiences: LL on the UK rankings and LDV on the global rankings. Such findings are interesting in that they indicate that party sites are of value during an election period, but quickly fall off in popularity afterward. Blogs, on the other hand, would appear to have a more stable audience.
Alexa also provides statistics on the origins of the traffic arriving at the sites. The evidence here points very strongly to the fact that individuals are specifically looking for the sites, rather than stumbling on them accidentally. The search engine Google.co.uk is the most commonly visited site prior to arrival and the most commonly searched terms preceding visits are the party or blog names. One notable exception here is LH where it is Conservative blogger Iain Dale’s blog spot that proves to be the key ‘upstream’ site. This overlap between Labour blog readers and Iain Dale could reflect Dale’s highly dominant presence within the party/blogosphere. However, given that this crossover does not appear to occur among LDV or CH blog readers it suggests that the site has lost its core left-wing audience and now consists mainly of Tory supporters monitoring the opposition.
Closer analysis of blog audiences was possible through a survey fielded by the authors (in conjunction with the Hansard Society) of LDV users. This survey was posted on the LDV site between 19 September and 20 October 2009, attracting 129 respondents. Participants were self-selected and recruited through a link posted on the blog by contributing editor Mark Pack. Given the self-selecting nature of the sample, we anticipated that they would constitute a core group of participants in the site rather than casual users. The findings support this, with 88 percent of respondents reporting that they visited the site either daily or a few times a week. Demographic data show respondents were overwhelmingly male (85.3 percent), younger (68.2 percent were 44 years of age or younger) and educated to degree level (70.5 percent with a university diploma, degree or postgraduate qualification). Respondents also displayed a strong similarity in levels of political interest and active support for the Liberal Democrats. All respondents reported their interest in politics as seven or higher on a scale of 0 to 10 and nine out of ten (n = 113) were strong or moderate supporters of the Liberal Democrats. Three-quarters of the sample (n = 94) had voted Liberal Democrat in 2005. Party membership was also high, with 73 percent (n = 94) of respondents reporting that they were currently members of the Liberal Democrats. Nine percent (n = 12) reported being members of other parties, and less than one in five (n = 23) had no party membership.
As well as profiling the social and political composition of LDV users, we were also interested in learning more about the motives of these individuals in accessing the site. A range of alternatives were offered (See Table 4) about the value of the site and individuals were asked to rate the importance of each on a scale of 0 to 10. 5
Importance of reasons for using Lib Dem Voice.
Source: LDV User Survey ‘How important are the following reasons for visiting the site?’ on a scale of 0–10.
The results suggest that a primary function of LDV, at least for its regular users, is accessing party-related news and information and discussion rather than taking action to promote any particular issues. Respondents valued the site mostly as a source of news (mean score 7.42) and for engaging with the wider views within the party (mean score 7). The least important reason was to find out more about how to get involved with offline political actions (mean score 3.52), and finding out more about possible online political action and resources scored only slightly higher. Of the remaining reasons, being part of a wider community and getting involved in discussion and debate were seen as of moderate importance. Given that we only have user data from one of the blogs being analyzed, it is important not to over-generalize these conclusions. Tentatively, they suggest that the core audience for these sites is a group of younger male politicos who share the ideological outlook of the blog authors and who use the site to stay ‘in the loop’ on party affairs, rather than providing a space to take political action.
Hyperlink networks: online integration?
Our final methodological tool to investigate the relationship between party blogs and parties was a network analysis of the hyperlinks connecting them. Our first step in understanding the linking practices of the main political parties and associated blogs was to construct a snowball network from three simple seed sets made up of pairs of the home pages of official party websites and the main unofficial blogs. A snowball network simply highlights all the websites that have been linked to from the seed set. We do not report the details in full here but one finding was important for refining our method for a more sophisticated network analysis. We found a heavy presence of links between the parties and ostensibly non-political but well-known Web 2.0 services, e.g. Flickr, Twitter and Facebook. These sets of links primarily constituted official parties highlighting their own presence in these social networks. The findings suggest that major parties at least are making strong efforts to extend their visibility within more informal online arenas and away from their own ‘shop windows’.
For a more detailed analysis of hyperlink networks we utilized larger seed sets, which included the profiles of both official parties and blogs on the most popular social media platforms (e.g. Twitter, Facebook, YouTube). The goal is to identify a cohesive ‘issue network’ that is marked by reciprocity of linking, rather than simply a list of linked websites. Crawls were launched in May 2010 and repeated in October 2010. Following Issue Crawler’s co-link technique, the seed sites are crawled and those sites that do not receive two or more links from the resulting network are removed. The websites identified in the resulting networks were coded manually according to the type of actor they represented, using categories as defined in Table 5.
Website type coding definitions.
Some basic structural properties of the three networks are presented in Table 6. Here we see that parties produced networks of broadly similar sizes in the pre-election period. The Conservatives’ issue network is a little smaller and more densely connected, while the low number of site-to-site links and the low network density score for the Liberal Democrats is suggestive of a rather diffuse network lacking internal coherence. The changes in network size and density in our post-election crawls are explored further in Tables 6 and 7.
Three issue networks, overview data.
Types of website in issue networks (number of sites and group density).
Turning to the substance of the issue networks in terms of actor types (Table 7) we can see that they generally contain more official party sites than any other category, with post-election Labour being the exception here. Party and independent blogs constitute significant components of the networks. The figures in parentheses in Table 7 report the densities of the links among the different categories of website (i.e. the proportion of all possible site-to-site links in a network that actually do exist). From this, it is clear that Labour and Conservative party blogs and independent blogs are more densely interconnected than official websites. The blog networks are also diverse in that they frequently feature ‘opposition’ sites. Labour’s pre-election network, for example, includes both CH and Iain Dale’s blog, both of which are more central than LL and LH (see Figure 3). So, in the build-up to the election, blog authors’ linking practices created a dense network of news and opinion resources that crossed party lines. 6 While most people may well have got their information about parties from the official sites (as indicated in the pre-election audience statistics above), those who did visit the blogs would have been presented with many more enticements to consider a wider variety of information sources, including those of political opponents.
Figures 1 to 3 present the data from Table 7 in visual form and particularly allow for comparison over our two time periods; i.e. pre- and post-election. 7 The results show that for the Conservatives and the Liberal Democrats there are significant changes between the two time points, while the Labour network remained more static. The figures present several kinds of information. Specifically, nodes are scaled by in-degree centrality and represented by symbols that indicate their category (see the key in Table 5). Ties are weighted by the number of links between each pair of nodes. Finally, while the network statistics in Tables 6 and 7 include peripheral websites that make links to key network actors but receive none, the figures include only websites that receive links (i.e. in-degree centrality > 0).
Conservative issue networks (seed set 1); Top: May 2010; Bottom: Oct 2010.
Liberal Democrat issue networks (seed set 3); Top: May 2010; Bottom: Oct 2010.
Labour issue networks (seed set 2); Top: May 2010; Bottom: Oct 2010.
In the top half of Figure 1, while the official site (http://conservatives.com) is prominent, the presence of large square nodes in the Conservatives network near the centre of the graph indicates the importance of blogs (both party and independent) in structuring the conservative issue network. The blogs are further connected to a periphery of smaller nodes representing news media and NGO websites. The post-election picture, however, is rather different: although a few blogs stand out (notably CH, Iain Dale and Guido Fawkes’ order-order.com) the official party site has become central to the network and there is an increase in the overall volume of official party sites. In terms of in-degree, the two blog categories together were the targets of 52.4 percent of all in-links in the network in May, but only 29.1 percent in October. Conversely, official party sites counted for only 6.2 percent of network in-links in May but 37.0 percent in October.
In Figure 2 we see some interesting changes in the nature of the Liberal Democrats network before and after the election. In the immediate lead-up to the election Central Official party sources dominate (in the form of libdems.org.uk and nickclegg.com), although a number of Web 2.0 sites and independent and party blogs give the network some diversity. Post-election there appears to be a clear parting of the ways, with a densely connected mass of official party sites occupying almost a separate network to that of the bloggers. In fact, while the number of official party sites receiving in-links in the network rose from 21 to 35, the proportion of network in-links they account for rose from 43.2 percent to 83.4 percent, and the density of links within this group increased from 0.18 to 0.52. These trends are interesting in terms of what they indicate about the impact of being in government for each parties’ online presence. Essentially, within the Conservative issue network, the official party site appears to assert a stronger authority post-election but retains a connection with the blogosphere. For the Liberal Democrats, a separation appears to have emerged between the two.
Finally, in Figure 3 we see less pronounced changes in the Labour network over time. Table 7 had shown a drop in the number of official party sites and a rise in the number of NGOs. The latter, mostly representing trade unions and left-leaning think-tanks, have increased from four to 13 sites that receive in-links (or two to eight if we only include those that receive as well as emit links). The proportion of network links that NGOs receive has grown from 3.5 percent to 20.5 percent of the total network links. The increasing importance of this group, along with an increase in the centrality of independent blogs at the expense of party ones, suggests that contestation over the future policy direction of the Labour Party has brought a greater diversity of voices into online debate.
Discussion and conclusions
Drawing these findings together we can provide some answers to our original research questions about the purpose and significance of these new party blogs within British politics. In terms of their origins and purpose these sites are clearly designed to provide an alternative to party politics ‘as usual’. While their producers are sympathetic to the parties with which they are linked, and populate their sites with issues and content generated by party affairs, they are ultimately motivated by a desire to see parties ‘doing things differently’, particularly with regard to bringing new voices into party debates, beyond the traditional members and leaders.
The findings from our content analysis and survey suggest that the blogs are succeeding in providing alternative spaces for grassroots voices and sources of internal debate and criticism within the parties, although varying degrees of editorial control are exercised. Overall, the sites do not appear to be serving as rival platforms for organizing activism among ordinary supporters. Whether this continues in the longer term is clearly an open question. One possible longer-term effect of the arrival of these blogs in party politics is the decline of geographically based localized activism. This was certainly a theme arising from the interviewees consulted. As one put it, ‘Is having a membership base for a party the most important thing these days? No it’s not actually. It’s more important that a party knows who its supporters are.’ Another similarly queried the value of traditional members, pointing instead to: the concept of being a friend of the party, where you can have things without having to pay … The more you’re open as a party, the more you move away from being a membership-based party and become kind of much more fluid, a party based on a coalition of interests. (Interviews with author, 2 and 4 December 2008)
While they may not as yet be replacing parties’ mobilizing and organizing roles, it does appear that the blogs may be siphoning attention away from the parties within the wider electorate during ‘normal’ i.e. non-election periods. Audience figures indicate that the blogs are more popular than their respective party sites in ‘peace-time’ although during elections, the parties’ appeal increases quite sharply. The network analysis enriches this picture in that it shows that parties and blogs largely interconnect with one another during the election period to form fairly coherent and dense partisan issue networks. Thus, one might expect a degree of mixing of the audiences between the sites. However, post-election, the issue networks of the governing parties reduced in size, with the blogs becoming somewhat more separate and less central. This shrinkage may reflect a more (small c) conservative approach by the governing parties toward their linkage partners, and a desire to avoid the controversy and adversarial politics that blogs are seen to represent (Ceren, 2006).
Overall, therefore, we would argue that our findings show that CH and its Labour and Liberal Democrat counterparts are not weakening established party politics by offering an alternative basis for coordinating activism, nor do they act as hotbeds of internal dissent toward party leaders and policies. The best-known and popular of the sites, CH exhibits a strong professionalized ethos and acts more as a news source than as a catalyst for activism. Since assuming government, however, it is notable that the profile of both Conservative and Liberal Democrat blogs have dropped within their respective issue networks, while official party sources have become more central and self-referential. The extent to which this continues through the life of the current government is one that is worth returning to for future research.
Footnotes
Appendix
Seed sets for party network analysis
Funding
This research, while not directly funded by, is linked to and is part of, research output under Rachel Gibson’s ESRC grant – ESRC Professorial Fellowship Award ‘The Internet, Electoral Politics and Citizen Participation in Global Perspective’, 2010−11.
Notes
Rachel K. Gibson is a Professor of Political Science at the Institute for Social Change, University of Manchester. She has worked in the area of digital politics and political parties since completing her PhD at Texas A&M in 1996. She is currently holder of an ESRC Fellowship to examine online democracy and elections in comparative perspective.
Kevin Gillan is a Lecturer in Sociology at the University of Manchester. His publications have focused on the generation and communication of ideas in social and political movements and on their relationships with technology and globalization. His latest research examines the spectrum of critiques of corporate power found in contemporary movement and business circles.
Fabienne Greffet is a Lecturer in Political Science at the University of Nancy 2 and Research Associate to the Pacte-CNRS, Institute of Political Studies in Grenoble. Her research focuses on uses of the Internet by political parties, mainly in France and Britain, and the transformations of political parties.
Benjamin J. Lee is a PhD candidate at the Institute for Social Change based at the University of Manchester. His work focuses on the adoption of new online campaign techniques by constituency-level campaigns in the UK, but he is also interested in the wider links between the Web and politics.
Stephen Ward is a Reader in Politics at the University of Salford. He specializes in the impact of the Internet on political campaigning and participation and has published widely in this area.