Dynamics of Intra-Party Organisation in the Digital Age: A Grassroots Analysis of Digital Adoption

Abstract

Over recent decades, scholars have explored political parties’ adoption of digital technology. Tracing successive eras of change, scholarship has examined the degree to which digital disrupts or embeds traditional power structures—with many studies finding evidence of ‘controlled-interactivity’. In this article, we revisit debates around the adoption of digital tools from a bottom-up perspective. Moving beyond attempts to categorise elite strategies for digital adoption, we consider practices on the ground to document how, in practice, digital technology is being taken up and used. Using a case study of the UK Labour Party, we categorise a range of different practices, highlighting and theorising the presence of digital adherents, laggards, entrepreneurs, renegades and refuseniks. Discussing the drivers of these practices, we offer new insight into variations in digital adoption and consider the significance of these trends for our understanding of party organisation.

1. Introduction

Parties’ accommodation to successive changes in communication technology has been mapped and theorised by many scholars of party politics (Lusoli and Ward, 2004). Most recently, this has resulted in studies of the way in which social media, databases and intra-party software have been mobilised for electoral and organisational purposes (Pedersen and Saglie, 2005; Williamson et al., 2010; Jungherr, 2016). Recurrent within this literature is attention to the question of organisational control. Building on an established literature that has traced successive ‘models’ of party organisation (Duverger, 1959; Kircheimer, 1966; Katz and Mair, 1995), scholars have asked whether digital technology disrupts traditional models of party organisation or embeds the power of elites.

In this article, we engage with debates around digital adoption, party organisation and control, but unlike much existing literature, we do not focus on the ideas and experiences of elites. Instead, we conduct a bottom-up analysis exploring the practices and ideas of the grassroots activists who are adopting digital technology. This approach recognises that while elite strategies have an important impact on parties’ digital activities, elite intentions do not automatically translate into grassroots practices (reflecting an acknowledged structure-agency problem; Enos and Hersh, 2015). Noting this, we ask: in what ways do grassroots activists adopt digital tools? And what factors explain these trends? These questions are then used to consider the significance of our findings for understandings of party organisation and control. Within this article, we adopt an expansive conception of digital technology, looking at the varied digital systems used for party campaigning, organisation and communication. This includes activists’ use of party canvassing databases, online platforms, social media and ‘mundane’ tools (Nielsen, 2013) such as email or spreadsheets. Taking this approach, we explore the case of the UK Labour Party, looking beyond the official, elite story of digital adoption, to examine ‘the unofficial story about how these parties campaign’ and organise (Kefford, 2018, p. 658). This article therefore uses a bottom-up analytical focus to cast new light on our understanding of digital adoption and party organisation.

2. Digital technology and party organisation

Studies of parties (and, indeed political organisations; Dennis, 2019) adoption of digital technology have proliferated in recent years. From early studies, monitoring the use of Web 1.0 and 2.0 (Römmele, 2003; Lilleker and Jackson, 2010; Baxter et al., 2011; Lee, 2014), more recent studies have traced parties activities in a platform society (Van Dijck et al., 2018) and context of data-driven campaigning (Roemmele and Gibson, 2020). This work has explored the patterns and significance of digital adoption trends, with a recurrent focus on organisational control. Building upon an established tradition of organisational modelling, scholars have sought to test the degree to which the tenets of Michel’s iron law of oligarchy (1962) (which states that…in all organisations power tends to fall into the hands of a small number of leaders regardless of the formal constitution of the organisation) are mirrored in contemporary practices. Studies have explored the extent to which digital media facilitates elite domination or allows more decentralised forms of control, often being framed around the ideas of ‘normalisation’ and ‘equalisation’ (Gibson and McAllister, 2015).

While it is important not to draw simplistic conclusions from diverse literature, it is notable that numerous studies have concluded that ‘democratic intermediaries are mostly interested in the technology as a means to continue performing their existing functions, only to a better level’ (Gibson et al., 2004, p. 198). While some evidence exists that parties (Gerbaudo, 2018, or their activists; Penney, 2017) have used digital tools in innovative ways, there appears to be a recurring tendency for elite control and limited grassroots power. Lioy et al. (2019, p. 44) therefore found that while ‘parties might claim that their online presence makes them closer to the electorate… attempts to disintermediate decision-making often fall short of creating real responsiveness and participation’. Elsewhere Gerbaudo concludes that within participatory digital parties ‘the reality of online democracy to date paints a rather pessimistic picture’ (2018, p. 127). Although variation therefore exists, studies have demonstrated a tendency for what Stromer-Galley has called ‘controlled-interactivity’ in the adoption of digital tools. This idea suggests that while digital affordances have given ‘a greater role for and visibility of citizens in the daily work of campaigning’, it has also enabled elites to control these activities ‘to ultimately advance the objectives of the campaign’ (2014, p. 104). As such, digital technology is not seen to have revolutionised party organisation, but has led to new forms of elite dominance and control.

In reacting to this literature, in this article, we draw attention to the tendency within much of this scholarship to explore the impact of digital media from an elite perspective. We argue that by tending to study national party websites (Chen and Smith, 2010), social media accounts (Enli and Skogerbø, 2013) or to conduct interviews with central party staff (Dommett, 2018), scholars have tended to neglect the experiences and practices of activists on the ground (c.f. Nielsen, 2013). This means we have a good understanding of how elites are intending to adopt (and indeed are adopting) digital technology, but have less understanding of what local activists ‘and volunteers actually do as they work together’ (Nielsen, 2010, p. 763). This makes it unclear whether grassroots practices conform to elite ideals or whether there are variations in digital adoption practice.

We argue that existing research in two often-distinct subfields indicates the importance of this form of analysis. First, scholarship on party activity, and specifically on the use and adoption of different tools, has suggested that elite intentions do not always translate into grassroots practices. Indeed, work by Enos and Hersh (2015) has shown activists to be unreliable mediums for the translation of elite messages. Similarly, Kreiss has found that ‘there were often disconnects between the desires of staffers and the at times competing expectations, wills, and even demands of volunteers’ (Zittel, 2009, p. 299; Kreiss, 2014, p. 543). Secondly, work on party organisation has indicated that elite power is not as uniform as might be supposed. Indeed, recent scholarship has traced the emergence of ‘franchise’ or ‘stratarchical’ party structures that suggest ‘organizational power and authority does not finally rest in any single place, or with any single set of individuals’ (Carty, 2004, p. 9). Both of these literatures indicate that party elites do not have ultimate control, and that grassroots can have autonomy. While some scholars have begun to explore these dynamics (Kefford, 2018), in this article, we provide a new categorisation of grassroots practice and explain the drivers of these trends. This analysis allows us to consider why variations exist and whether alternative grassroots practices have implications for our understanding of central party control.

3. Methodology

In asking: ‘in what ways do grassroots activists adopt digital tools, and what factors explain these trends’, this article engages with one case in detail, the UK Labour Party. The Labour Party is a major electoral force in the UK and possesses the single largest membership in the UK (reported to be 580,000 in January 2020 (Perry, 2020)). While the party’s internal organisation has evolved over the years, in recent history, scholarship has highlighted changes that have strengthened elite control (Gould, 1999; Minkin, 2014).

To structure our analysis, we used interviews, documentary analysis and secondary analysis of the Election Agent Survey. The purpose of these methods was two-fold. First, to provide a baseline for the analysis, we undertook a preliminary study of Labour’s digital adoption practices around the 2017 General Election. These interviews sought to establish elite intentions for digital technology at an elite and local level, providing a benchmark against which we could assess grassroots practices. In total, we conducted eight interviews with party elites and strategists as part of a wider project looking at party renewal. These interviews were supplemented with an analysis of publicly available documents and webpages pertaining to the party's use of digital technology.

Secondly, to address our primary research question, we conducted a second and more extensive round of interviews with party activists. In total, 18 interviews were conducted with activists connected to digital campaigns in Labour Party branches, constituencies and regional parties in England.¹ Interviewees were individuals who were active in organising the local party and in almost all cases had a formal role on the executive of either the branch, constituency or region. We selected local parties for analysis that displayed ranging proficiency with digital tools, reviewing online content for parties around the country to identify those with extensive or minimal online presence. Interviews were conducted either in person or over the phone and lasted on average 50 min. Participants were asked about their local activity and use of digital tools, their relations with the national party, and their local party demographics and skillbase using a semi-structured interview format. Interviews were audio recorded and transcribed before being analysed using NVivo.² To ensure anonymity, names and geographical locations of interviewees are not disclosed. However, it should be noted that we surveyed campaigning activity in 11 geographic locations. To reflect on the wider applicability of our findings, we supplemented our analysis with data from the Agent Survey that gathers information on local campaigning activity around the country.

4. Characterising Labour’s elite digital adoption strategy

Looking first at elite intentions for digital adoption, interviews revealed that historically ‘there was not really that much expertise’ in the Labour Party around digital (Interview 1) and that digital generally has often been ‘under-resourced and not influential enough internally’ (Interview 2). However, since 2015, interviews revealed that the Labour Party have invested heavily in digital—creating a Digital Transformation Team (Interview 3; see also Dommett, 2018) and a raft of new digital tools. One particularly notable aspect of this activity is that rather than relying on external digital expertise, party staff emphasised their scepticism about relying on ‘existing commercial solution(s)’, arguing that there was instead a need ‘to build our own’ systems and expertise (Interview 4). In line with these ideas, in October 2018, Labour launched a new bespoke digital organising system—Achieve. This is a suite of centrally designed digital software that local activists can use, comprising:

• Organise—A volunteer management tool that replicates many of the functions previously provided by Nationbuilder;

• Doorstep App—An app that allows canvassers to enter canvassing data in real time on mobile devices;

• Promote—A platform for enabling targeted messaging on Facebook;

• Insight—A platform for data analysis; and

• WordPress Network—A website creation tool.

These complement existing—and continually evolving—platforms such as Contact Creator, Campaign Creator and MemberCentre. Achieve provides an extensive programme of online training courses and the party emphasises sharing best practices—including an annual ‘Best Digital Campaign Award’ presented at the national party conference (Labour Party, no date).

Thinking about the significance of these practices for central control, these trends suggest that the national party is seeking to structure and guide local activity. Indeed, looking in more detail at these resources, it appears that Doorstep prompts activists to gather certain kinds of voter information when canvassing, while Promote includes a sign-off procedure for the approval of any locally developed Facebook advertising content.³ From an elite perspective, therefore, Labour’s strategy aligns with the notion of ‘controlled-interactivity’ as it seeks to guide and monitor local digital activity to ensure alignment with elite objectives.

4.1 Categorising local activists’ digital adoption practices

Looking beyond elite intentions, our primary interest lies in asking in what ways do grassroots activists adopt digital tools, and what factors explain these trends? To consider these questions, we began by exploring available data about grassroots digital adoption practices among Labour activists across the country as a whole.

Data from the Election Agent Survey in 2017⁴ suggest that of the 333 Constituency Labour Parties (CLPs) who responded the adoption of public-facing digital tools is by no means universal. Three-quarters of CLPs reported having a local website for either the candidate or local branch, suggesting that a considerable amount did not have a ‘landing page’ for their branch outside of social media. And, while nine out of ten CLPs claimed to use social media in their campaigns, 20% were willing to admit they put little to no effort into Facebook, while 38% said the same about other social media (such as Twitter). Only 38% claimed to have someone in a role that might be described as a ‘computer officer’ and less than a quarter of CLPs claimed to have any kind of website design or content assistance from party headquarters. This initial analysis indicates significant variation in uptake, despite central party efforts to provide activists with appropriate tools. A somewhat unified central-party narrative does not necessarily equate to uniform practice on the ground.

To generate new insight about the kind of variations suggested by this data, we used our interview data to identify five types of digital adoption practice. Across the local parties we studied, there was significant variation in the type of tool that grassroots parties utilised, with some parties relying on centrally provided systems and others looking elsewhere for high or low tech alternatives. In addition, we found variation in the extent of usage, with some local parties relying heavily on technology, and others hardly using digital tools at all. Classifying these differences, we differentiate between practices we categorise as digital adherence, entrepreneurship, laggard, renegade or refusenik (Table 1).

Introducing each classification in turn, we first identified ‘adherence’, a practice that demonstrated the high use of digital tools offered by the central party. Here activists adopted the digital tools provided by elites, utilising campaign tools such as Contact Creator and Organise and/or intraparty digital communication tools. Those presenting this behaviour reflected: ‘Doorstep, Turnout… I can certainly see the benefit of them. They are really well thought out resources’ (Interview 6), ‘Doorstep is absolutely brilliant’ (Interview 7) and ‘[i]n Organise itself, we’ve been given a really powerful tool not just in terms of communicating with members, but also organising with members’ (Interview 8). These views accorded with the belief of one party elite that ‘[local parties are] very receptive normally. When it’s things that they can see are going to make their life easier, I think they’re generally like, “Oh, that’s great”’(Interview 9; also 5). Such accounts provide evidence of local branches capitalising on affordances offered by the central party and suggest that, through these mechanisms, party elites can guide local party activities .

Our second category, ‘entrepreneurship’, was different. Again we saw high usage of digital tools, in line with a belief in the ‘valuable role of technology in managing and mediating’ (Interview 16) political participation. However, here the tools used were often not provided by the central party. Practice here was by no means uniform, with some drawing on ‘mundane’ tools and others using more specialist software or expertise. In the former category, interviewees described frequently using Google Docs, explaining that: ‘I haven’t seen any Labour way of doing this better’ (Interview 10). Yet, others were more innovative, with one describing how a skilled volunteer has ‘started messing around making his own mini animations, we made a couple of those and they went viral’ (Interview 17). Another noted that they had created a MediaWiki to log agendas, meetings and minutes of local branch meetings, all hosted on their own servers.

In our analysis, it was rare that local parties exclusively used non-official tools, rather, we found local actors to be supplementing official party tools in ways that central party elites did not devise (or seem to foresee). For instance, one activist described how they:

‘…wrote a computer programme that strips through an Excel spread sheet that has been exported from Contact Creator and it figures out how many people live in the house, and if it is five different surnames then it just lists their first names, and if there are two people who have different surnames it will say “hello Jenny Baker and Frank Goodwin” and it will just figure out how many characters it can fit…in an address field’ (Interview 7).

In such instances, official party tools were not neglected, but they were often only a part of the digital arsenal used by local activists, meaning that a proportion of local party activity was beyond the control of central party structures and procedures.

The third category, ‘laggard’ behaviour, describes practice in which there was use of official digital tools, but it was limited and often inefficient. Activists appeared reluctant to use the official tools and tended to describe their experience in negative terms, for instance reporting central party software to be ‘horrible, clunky impenetrable systems’ (Interview 8). These attitudes resulted in limited uptake of digital technology, with activists instead relying on non-digital techniques. While some of these processes did conform to party objectives (such as conducting canvassing), the lack of digital feedback to the central party limited opportunities for oversight.

‘Renegade’ practice again describes low levels of digital usage, but in this instance, any digital tools that were mobilised tended to be unofficial rather than provided by the party. Activists preferred to rely on familiar services and platforms, using generic options such as email to organise their activities. In one example, we found activists using Excel spread sheets to manage and process voter data rather than the party system, Contact Creator. In such instances, the central party’s capacity to influence and oversee local activity was therefore limited, suggesting minimal elite control.

Finally, we also gathered data that some local parties exhibited ‘refusenik’ behaviour whereby local activists appeared not to demonstrate any form of engagement with official or unofficial digital tools. Although the branches and constituencies we studied did not describe themselves in this way, our interviewees recounted stories of other local parties who did not use digital tools at all. They also related examples of refusenik attitudes among some local actors who were sceptical about the advantages of digital. One interviewee, therefore, described how they had faced ‘resentment’ and were ‘told councillors…ought to be knocking on doors and talking to people’ rather than engaging in digital activity (Interview 14). Evidence of such practices suggests, as above, that central elites have minimal opportunities for control.

In offering these five classifications, it is important to clarify that our interest is in identifying different practices regarding the use of digital tools. It is important to make it clear that most local parties will contain a range of practices. This occurs because individuals themselves can change their attitude over time (moving from laggard practices to adherence through training), and because local parties are comprised a range of different individuals with different attitudes and competencies. It is, therefore, possible, for example, for a single local party to exhibit refusenik and entrepreneurial behaviour among different activists. When attempting to categorise grassroots practices, it is therefore important to be aware of variations, and to reflect on how local parties manage competing digital practices. Rather than focusing on such questions in this article, we instead turn to consider explanations for these different practices. Then, in the discussion, we reflect on why these variations may be significant for our understanding of party organisation.

5. Explaining digital adoption practices

In seeking to explain the variations identified above, we asked interviewees to reflect more broadly on the dynamics of digital adoption. Coding interview transcripts, we identified four recurring themes that highlighted alternative explanations for the variations seen above. These relate to: local party organisation, perceptions of digital, digital skills and central party relations.

5.1 Local party organisation

While the Labour Party boasts a large membership relative to other UK parties (Audikas et al., 2019), it is by no means the case that this membership is distributed evenly across the country. While some of the local parties we spoke to had a membership of many hundreds, others had only a handful. This meant that some parties struggled to operate functionally at any level, not just when it came to digital. The reasons for membership distribution are somewhat random—reflecting factors such as geography and marginality—but they have significant implications for practice on the ground. In many places, it was simply the case that there were not members available to utilise official or unofficial digital tools.

These variations were compounded by differences in the degree of activist involvement. While some local parties have a history of engaged membership—often underpinned by strong social structures—others did not, or were highly dysfunctional. One interviewee described how despite having ‘around 1,500 members in constituency’ they had about 200 people ‘active in some sense’ and only about ‘60 people’ actively involved at election time (Interview 8). Another similarly argued that ‘I don’t think we’ve got an activist base’, despite having one of the largest constituency memberships in the country (Interview 14). These variations in activism were significant for the uptake of digital tools we witnessed. While those with a large, active membership tended to have the capacity to devote specific attention to digital technology, resulting in adherent or entrepreneurial behaviour, those with smaller or more dysfunctional structures faced difficulties in conducting offline, let alone online activities—resulting in laggard, renegade or refusenik practices. Indeed, one interviewee reflected that their local party was so ‘moribund’ it had ‘no networks to disseminate or coordinate activity amongst activists’ (Interview 6). The dynamics of the local party, therefore, shape the uptake of digital tools in multiple ways.

5.2 Perceptions of digital

In addition to party capacity, our analysis also showed variations in practice to be linked to perceptions of digital. Indeed, among our interviewees, we found diverse views of the capacities and utility of a range of official and unofficial digital tools that informed uptake. While many of our interviewees—especially those exhibiting entrepreneurship and adherence—recognised the value of technology for engaging members and voters, many others were sceptical or downright hostile towards technology.

These different attitudes were reflected in the perceived importance of digital within local parties’ executive structure. While some local parties saw digital technology as core to the activities of the party, many others saw digital adoption and activity as an optional extra that supplemented parties’ day-to-day activities. This was reflected in parties’ executive committees, with interviewees often describing that their local party did not have ‘social media or communications officers’ and reflecting that many branches did not devote attention or resource to digital activity (Interview 6). Among these activists, there was reported to be ‘a kind of inertia issue because local people who have done it for years think they’re doing it right’ and therefore have no interest in digital tools (Interview 14). These trends were reflected in the aforementioned Election Agent Survey data which showed that while parties across the country reported having ‘core’ executive positions filled, only 38% claimed to have a ‘computer officer’.

Our analysis, therefore, suggests that the perception of digital within local parties can be influential for the uptake of digital tools, as it affects the degree to which digital is integrated into the activity of the local party as a whole. In our cases, local parties exhibiting high levels of adherence or entrepreneurship displayed a commitment to digital technology, often devoting activists and budget to this resource. In contrast, laggards, renegades or refuseniks often showed little interest in the potential of digital.

5.3 Digital skills

Related to these factors, our analysis also suggested that variations in activist skill and attitude affected practices. Activists were by no means always equipped with digital skills, and even large parties were not guaranteed to have members who possessed such expertise, leading to laggard, renegade or refusenik behaviour. This point was forcefully underlined by one interviewee who expressed frustration at the lack of local skills, reflecting:

‘I don’t understand why I couldn’t [find someone with the necessary digital skills]. I mean we’re bloody 650 people in a middle-class area, an academic area… There must be somebody who can do, who’s got the software… I couldn’t find anybody who had the software to do anything more than the most basic leaflets’ (Interview 13).

Some interviewees described this lack of skill as a result of demographics—pointing to an absence of young ‘digitally native’ members familiar with, and enthusiastic about, using digital tools (a point rendered particularly poignant by the tendency of some local parties to pass responsibility for digital to young members). Others noted a basic level of skills, with one interviewee noting that most local activists ‘still struggle with email, let alone Facebook and WhatsApp’ (Interview 6).

Within parties exhibiting adherence or entrepreneurial practices, we found evidence of a skilled membership. Among entrepreneurs, the use of unofficial tools was often driven by those with pre-existing digital skills. The activists we spoke to therefore often had backgrounds in digital businesses or computer science, Masters degrees in political communication or design skills. These individuals were highly competent and had ‘skills in making videos and graphics design, web design that perhaps not everybody who becomes an organiser would necessarily have’ (Interview 7). This meant that they were able to create bespoke pieces of software and databases (Interview 11), acting in an entrepreneurial way to not only capitalise on tools offered by Labour, but to complement them with other software. Among adherents, we also saw evidence of familiarity with digital technology, with those responsible reporting to be users of a range of platforms.

We also found evidence that local parties were being supported by external campaign organisations who promoted digital skills and technology. In particular Momentum, an organisation created as part of Jeremy Corbyn’s leadership election campaign in 2015 (Dennis, 2019), supported digital skills training and activity. For instance, a Momentum training session at a fringe event at the 2018 national Labour Party Conference targeted and sought to push resources into constituencies in Cornwall and Devon, in contrast to the Regional Office that was argued to have ‘written off’ much of this area. Interestingly, we generally found evidence that it was more entrepreneurial behaviour being facilitated through Momentum and other non-party networks (Interviews 18, 16). These developments created alternative systems of digital adoption, but were often implemented alongside official party tools rather than supplanting them. These examples suggest that adoption practices are driven by the availability of skills among membership, and are affected by the presence of external support and expertise around digital technology.

5.4 Relationship with central party

A fourth factor we identified was variation in local parties’ relationship with elites. Our analysis revealed considerable variation in knowledge of centrally provided services, and attitudes towards central systems. First, in regard to knowledge, this explained significant differences in practice. Those exhibiting adherent or laggard practices were often aware of support from the central party. Activists informed us of a national party mailing list for campaign organisers, webinars and conference training sessions designed to promote the adoption of digital tools. One local interviewee reflected that their own digital activism had followed:

‘…a lot of training with the party nationally…it opened my eyes as to how we could do things. They were training us to how to write good emails, trained us on how to use Organise. They did training in a kind of workshop fashion. I was sat with people from all over the country and we shared ideas and best practice from people all over the country…I took a lot of ideas from that…I told our conference debrief meeting, this is what we’re doing wrong and right’ (Interview 8; see also 6; 13).

Another reflected that ‘the training campaign bulletin….[if] you are on that, you are not going to miss anything, you’re not going to miss the national campaign day that they want you to be doing, and you’re not going to be missing the training either – I can imagine if you are not on that then you could miss stuff, a lot of the stuff, and the training’ (Interview 18). For those not aware of this infrastructure, however, practice tended to be laggard or entrepreneurial in nature. Many local actors were not aware of central party support, as evident in comments such as: ‘I was expecting guidance with how to run the social media campaign… [but] they left me to my own devices’ (Interview 10), and: ‘[i]f the party provide support, I have no idea how’ (Interview 17). One particularly vocal respondent reflected at length on the lack of support, asserting that ‘[w]hen it came to social media we had nothing from region, no content, no protocols’. They lamented this failing, asserting:

‘The software is only as good as the person that is using it. The Labour Party have a responsibility to get everyone up to scratch, because it is their software. I would look at it how a business would do it – I’ve seen it happen in my industry and you have champions, and they go to national training and filter that down by giving training – I’ve never seen that happen’ (Interview 6).

These perceptions led to low levels of adoption for official party tools. They may also have been responsible for some entrepreneurial practice, as those displaying this practice often voiced surprise at the existence of Labour Party systems for problems they had sought to rectify themselves.

In addition to awareness, we also observed variations in attitudes towards central systems. Within our interviews, we found evidence that activists were more or less willing to comply with often-complex oversight systems. This became particularly apparent when discussing activists use of Facebook advertising—a tool that has become prominent in recent campaigns (Dommett and Temple, 2018; Dommett and Power, 2019). In recognition of the importance of Facebook, the Labour Party—as outlined above—created the Promote system to allow local activists to design and pay for adverts that could be targeted on the basis of Labour Party canvassing data and Facebook data. This system was designed to facilitate campaigns whilst also maintaining control, as users had to submit adverts for approval to the Regional Office before they were published online. In practice, however, the Promote system was exceedingly complex, with activists presented with a 100-page manual—leading one activist to reflect that they ‘didn’t get the impression from the national party that this is something we should be doing’ (Interview 10).

The complexity of these systems prompted different responses from activists. While those willing to defer to central authority devoted many hours to learning the system (adherence), others challenged the logic of the party’s approach and developed alternative mechanisms. Indeed, those displaying entrepreneurial practices reported by-passing Promote by using Facebook’s advert manager system direct. Indeed, one interviewee recounted how ‘we couldn’t work with region, we couldn’t use Promote – so we got people to put it on their own pages’ (Interview 8). This approach to adoption was evident elsewhere, and was often explained by the perception that centrally provided tools were ‘clunky’ and ‘impenetrable’ (Interview 8)—inspiring entrepreneurial, laggard or refusenik practices. Interviews, therefore, revealed that variations in practice were affected by knowledge of central systems and training, but also perceptions of the value and effectiveness of those centrally provided mechanisms.

6. Implications for party organisation

Our analysis of grassroots practices has revealed previously hidden variations in the adoption of digital tools. Despite the presence of a clear elite strategy to promote controlled-interactivity, we found evidence of five different digital adoption practices. We categorised ‘adherence’, ‘entrepreneurship’, ‘laggard’, ‘renegade’ and ‘refusenik’ practices, and identified the explanatory significance of local party organisation, perceptions of digital, digital skills and central party relations. These findings enhance our knowledge of grassroots practice, but we argue that they also have implications for our understanding of party organisation, specifically about conceptions of elite control. Reviewing our findings, we argue that variations in the extent to which official and unofficial tools are adopted have different consequences for questions of elite control and controlled interactivity.

When it comes to differences in the extent to which grassroots parties were using official party tools, the variations we identified did not appear to have implications for central party control. Although elites would ideally like to have local parties adopting digital tools in a uniform way, our elite interviewees reflected that ‘[e]very CLP is different’ (Interview 11) and they ‘wouldn’t have thought that all branches use…[party provided] channels’ (Interview 9). While these differences meant that elites could not treat all local parties in the same way, laggard or refusenik practices were not seen to challenge elite power, but rather to require a different form of response. In this way, elites spoke about needing to work with local parties to launch membership drives or training initiatives, or to create more direct means of elite control. These alternative responses suggest that the low adoption of official digital tools does not challenge the idea of controlled-interactivity, but does require elites to make an alternative response.

Turning to variations in the uptake of unofficial tools, however, we argue that questions emerge about the extent to which elites are able to exercise control. In discussing explanations for different practices above, our analysis showed how the adoption of unofficial tools was shaped by factors including the availability of activists with digital skills, complex relations with central party and support from external actors. While not uniformly leading to the uptake of unofficial tools, where this did occur, we argue that elites’ capacity to exercise control was affected.

To demonstrate this point and the kind of implications it may have for our understanding of party organisation, we consider two trends we observed within our cases. First, we identified instances where, dissatisfied with (or unaware of) official party tools, activists created or adopted alternative systems for party activity (such as using Excel spreadsheets to facilitate canvassing). Returning to the example of Promote, our interviews showed that activists circumvented official sign-off procedures and elite control by using the unofficial Facebook advert manager tool. This removed elites’ ability to exercise oversight and control. Such behaviours are significant and need to be understood because our interviews showed that such behaviour can be problematic. Indeed, we uncovered evidence of local digital activists intentionally creating controversial and emotive content in order to provoke a reaction to their online campaigns. This type of campaign messaging and strategy is highly controversial and unlikely to be approved through official party procedures. This shows how unofficial mechanisms can circumvent elite oversight and message control, behaviours that are likely to be of interest to scholars of party organisation.

In addition, our cases also showed that external actors can play a role in promoting unofficial digital adoption practice. Specifically, in our interviews, we heard stories of Momentum providing digital resources and campaign messages that were developed without elite party oversight. This meant that some local parties eschewed official digital content in favour of materials and messages that had not been subject to oversight. While we did not observe any problematic practices in our cases, the presence of alternative sources of expertise reveals that party elites are not the only actor able to inform the uptake of digital tool and that different actors may be able to exercise influence and control. Given the potential for factionalism and opposition between party grassroots and elites, this example shows the potential for local parties to develop alternative procedures that elites have little control over. Returning to the idea of stratarchy outlined at the beginning of this article, this example, therefore, suggests that party organisation and power can be less hierarchical and more networked in nature, indicating that a wider range of actors beyond party elites need to be considered to understand how parties work (Dommett et al., 2020).

These trends suggest that party elites may not have as much power as is sometimes suggested, and that digital adoption cannot be characterised simply by looking at elite intentions. This is significant given that party elites are responsible for exercising oversight and ensuring compliance with strict legal and financial regulations on parties.⁵ With power potentially held not only by elites, it also becomes necessary to look at the dynamics of local party activity, considering how and why local parties are adopting digital technology differently, and whether their practices are compliant with systems of oversight and accountability.

Such analysis is important because there are a number of signs from recent scholarship on parties and digital technology that the distribution of power within parties is by no means fixed. To take just two examples, recent scholarship on party membership has shown the growing significance of supporters and new forms of activist affiliation (Fisher et al., 2014). Such resources may provide new forms of grassroots power in regards to digital (and other aspects of party activity), developments that have the potential to lead to the adoption of unofficial tools and circumvent elite oversight and control. An alternative body of literature also suggests the need for such analysis, as work on data-driven campaigning has indicated that digital technology may enhance the power of party elites (Chadwick and Stromer-Galley, 2016). However, it is as yet unclear how parties’ dependence on data will affect the way that local party activities are organised. These examples indicate the need to consider where power lies within parties, specifically exploring whether grassroots activities match elite intentions, and whether elites are able to exercise control.

7. Conclusion

In this article, we have sought to extend our understanding of party organisation using a study of digital adoption within the UK Labour Party. In contrast to previous studies, we focussed on grassroots practices to show that, despite a desire from elites for controlled-interactivity, many different practices exist. We presented five categorise to capture these differences, categorising ‘adherence’, ‘entrepreneurship’, ‘laggard’, ‘renegade’ and ‘refusenik’ practices. While the picture we produce is perhaps messier and more complex than that shown when focusing on elite desires alone, we argue that a focus on grassroots practices helps to capture the complex dynamics of party organisation. Explaining these variations, we highlighted the significance of local party organisation, perceptions of digital, digital skills and central party relations, and reflected on the significance of these differences for party control. While arguing that variations in the extent of adoption around official tools have limited implications for our understanding of control, we argue that differences in the adoption of unofficial tools have more consequence. Specifically, we suggest that this form of grassroot practice raises questions about the distribution of power and elites’ ability to exercise oversight and control.

In distilling these insights, it is important to reflect on the limitations of this study. Our analysis focuses solely on the UK Labour Party and hence is not able to comment on the transferability of these findings to other cases. However, they can be used to pose questions for future studies of this type. These include asking whether the adoption practices we identify are encountered elsewhere, whether alternative explanations for variation occur, and how elites have responded (if at all) to these dynamics? In calling for further study, we argue that there is a need to examine parties that exhibit an elite intention for controlled interactivity , but also those that favour other forms of elite strategy to enable comparison. Such inquiry is required in order to ask whether our categories resonate in other contexts.

Looking beyond the Labour case, our findings suggest that any party seeking to promote the use of digital tools—or indeed to exercise central control—faces a range of challenges about how they seek to drive forward adoption. Our analysis has shown that local actors’ response to elite desires is unlikely to be uniform, and may result in behaviours not only contrary to, but also potentially damaging to elite party strategies. Although it is easy to conceive of party members and activists as passive, docile agents for the goals of elites, our analysis has shown that the process of exercising control is by no means straightforward and that diverse power distributions can exist in with parties. Rather than presuming simple controlled-interactivity in line with elite intentions, it is therefore necessary to delve deeper to understand the precise extent of central party control.

Footnotes

Acknowledgements

We would like to thank Professor Justin Fisher for providing access to Election Agent Survey data. We would also like to thank the participants of the workshop ‘Cyber parties: New parties, ICTs and organization in the digital age’ in Valencia for providing feedback on an earlier version of this draft

We would also like to acknowledge support from the Economic and Social Research Council, grant number: ES/N01667X/1

Conflict of interest

The authors have no conflict of interest to report.

REFERENCES