### **Surges in Party Membership**

This book presents a comprehensive analysis of a remarkable and unexpected outcome of the 2014 referendum on Scottish independence.

 Despite defeat in the Scottish referendum, the two leading parties in the Yes campaign – the Scottish National Party and Scottish Green Party – experienced an extraordinary surge in membership. The book explains these events, examining the relationship between political parties and social movements, and it assesses the longterm consequences of the surge. Based on surveys of members and interviews with party and movement actors since the referendum, the book analyses the members' involvement in the 2014 referendum, their motives for joining a party, their backgrounds and political attitudes, and their behaviour as party members. A key component of the book is how the surge changed the parties – socio-demographically, ideologically and organisationally.

This book will appeal to scholars, students and observers of electoral politics, political participation, social and political movements, and political parties and their members, and more broadly to those interested in the debate on Scottish independence, British politics and comparative politics.

**Lynn Bennie** is Reader in Politics at the University of Aberdeen, UK.

**James Mitchell** is Professor of Public Policy at the University of Edinburgh, UK.

**Robert Johns** is Professor of Politics at the University of Southampton, UK.

### **Routledge Studies in British Politics**

Series editors: Patrick Diamond and Tim Bale *Queen Mary University, London, UK.* 

This series aims to promote research excellence in political science, political history and public-policy making, whilst addressing a wide array of political dynamics, contexts, histories and ideas. It will retain a particular focus on British government, British Politics and public policy, while locating those issues within a European and global context.

**Inequality in Britain**  *Alan Ware* 

**Why London is Labour**  A History of Metropolitan Politics, 1900–2020 *Michael Tichelar* 

**Neil Kinnock**  Saving the Labour Party? *Edited by Kevin Hickson* 

**The Brexit Effect** What Leaving the EU Means for British Politics *Gianfranco Baldini, Edoardo Bressanelli and Emanuele Massetti* 

**Institutionalised Dissent**  The Official Opposition in the UK since 1935 *Nigel Fletcher* 

#### **Surges in Party Membership**

The SNP and Scottish Greens after the Independence Referendum *Lynn Bennie, James Mitchell and Robert Johns* 

# **Surges in Party Membership**

The SNP and Scottish Greens after the Independence Referendum

## **Lynn Bennie, James Mitchell and Robert Johns**

First published 2024 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

and by Routledge 605 Third Avenue, New York, NY 10158

*Routledge is an imprint of the Taylor & Francis Group, an informa business*

© 2024 Lynn Bennie, James Mitchell and Robert Johns

The right of Lynn Bennie, James Mitchell and Robert Johns to be identified as authors of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

The Open Access version of this book, available at www.taylorfrancis.com, has been made available under a Creative Commons Attribution (CC-BY) 4.0 license.

This work was supported by the ESRC grant no. ES/N010590/1.

*Trademark notice*: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

*British Library Cataloguing-in-Publication Data* A catalogue record for this book is available from the British Library

*Library of Congress Cataloging-in-Publication Data*

Names: Bennie, Lynn G. author. | Mitchell, James, 1960–, author. | Johns, Robert, 1975– author.

Title: Surges in party membership : the SNP and Scottish Greens after the independence referendum / Lynn Bennie, James Mitchell and Robert Johns.

Other titles: Scottish National Party and Scottish Greens after the independence referendum

Description: New York : Routledge, 2024. | Series: Routledge studies in British politics | Includes bibliographical references and index.

Identifiers: LCCN 2023057660 (print) | LCCN 2023057661 (ebook) | ISBN 9781138332546 (hbk) | ISBN 9781032775395 (pbk) | ISBN 9780429446559 (ebk)

Subjects: LCSH: Scottish National Party. | Scottish Green Party. | Political parties—Scotland—History—21st century. | Referendum—Scotland— History—21st century. | Home rule—Scotland—History—21st century. | Self-determination, National—Scotland—History—21st century. | Scotland—Politics and government—21st century.

Classification: LCC JN1370 .B46 2024 (print) | LCC JN1370 (ebook) | DDC 320.941109/05—dc23/eng/20240319

LC record available at https://lccn.loc.gov/2023057660

LC ebook record available at https://lccn.loc.gov/2023057661

ISBN: 978-1-138-33254-6 (hbk) ISBN: 978-1-032-77539-5 (pbk) ISBN: 978-0-429-44655-9 (ebk)

DOI: 10.4324/9780429446559

Typeset in Times New Roman by Apex CoVantage, LLC

## **Contents**


### **Figures**


## **Tables**



viii *Tables* 

### **Acknowledgements**

We would like to thank three groups of people who supported our project and helped us understand an exceptional period in Scottish and UK politics. First, our interviewees who so generously gave up their time to talk about the membership surges and pre- and post-referendum politics. Second, party staff in the SNP and Scottish Green Party who offered invaluable advice and distributed our surveys. Third, the ordinary members of the SNP and Scottish Greens who completed those (lengthy) surveys and who perform such an important role in our democratic system. We are very grateful to you all.

### **1 Introduction**

### Party membership rejuvenated

Party membership has been the subject of much international research, mostly focused on membership decline. Across Europe and the democratic world, a peak in the membership of political parties occurred around the middle of the 20th century, and long-term decline followed (Katz and Mair 1992; Scarrow and Gezgor 2010; van Biezen et al. 2012). By the early 21st century, the act of joining a political party looked like a remnant of the past, an inevitable consequence of societal change (Mair 2013: 37–42; Dalton 2014: 46–48). Social, economic and technological development, the story went, meant a loosening of ties between citizens and parties and a reduction in the number of card-carrying party members. Parties, it was assumed, had adapted to this environment, appealing directly to voters using new campaign and communication techniques and attracting funds from sources other than members. From this perspective, there was reduced demand for party membership on the part of both citizens and parties.

In more recent years, however, decline in the membership of *some* parties appeared to have been stopped in its tracks and in several cases dramatically reversed. This book is about two such examples of party membership revival, exploring a remarkable and unexpected consequence of the 2014 referendum on Scottish independence. Despite defeat in the referendum, the two leading parties in the Yes campaign, the Scottish National Party (SNP) and Scottish Green Party (SGP), experienced an extraordinary surge in membership. In a few short months the SNP and Scottish Greens became parties composed mainly of new recruits, most of whom had joined online. A year on from the referendum, Scottish Green membership had multiplied by a factor of 6, from 1,500 to 9,000, what would prove a peak for this party. Over the same period, SNP membership jumped from 25,000 to 115,000, nearly a fivefold increase, and it would continue to grow to a peak of 125,000 in 2019. These surges were unprecedented in scale and pace, bucking wider trends in party membership. Other UK parties experienced member surges, notably the Labour Party, but these were less dramatic. Normally, party membership increases correspond with high-profile events like leadership contests or election campaigns, and often in the context of electoral success. These happened post-referendum and among parties on the *losing* side.

The striking initial upturn in SNP and Scottish Green memberships is depicted in Figure 1.1. This book aims to explain these events and to examine their

DOI: 10.4324/9780429446559-1

This chapter has been made available under a CC-BY 4.0 license.

*Figure 1.1* Membership of the SNP and Scottish Greens, 2003–2015

long-term consequences. The referendum acted as the catalyst, but what were the underlying political forces? What happened following the initial surge, and what was the long-term impact? And to what extent did the injection of members change the SNP and Scottish Greens, socio-demographically, ideologically and organisationally?

We consider how to interpret these developments in the context of party membership elsewhere. We argue that to explain the surges in memberships we must understand the role of a major event or trigger as well as the relationship between conventional parties and wider political and social movements. The 2014 referendum played an important role in raising the issue of Scottish independence, providing a lengthy campaign platform to project the policies and objectives of each party. And the referendum strengthened the relationship between the SNP and Scottish Greens, connecting them within a national movement for change. This involved the Scottish Greens moving beyond their core movement ideals, becoming a party more clearly supportive of Scottish independence.

Since 2014 there has been a UK-wide referendum on EU membership and an abundance of elections at state and sub-state levels. In the autumn of 2021, the SNP and Scottish Greens agreed to work together in government, publishing a 'partnership agreement' and a 'shared policy programme', building on an agreement in 2007 when the SNP formed a minority government with Greens' support. Together they committed to push for another referendum on Scottish independence. In early 2023 an SNP leadership contest revealed that the party's membership had fallen considerably. The Scottish Greens, meanwhile, had experienced fluctuation in their membership. Nevertheless, membership of each party remained at a much higher level than before 2014.

Before we outline the details of our study, we review important themes emerging from international research on party membership. We consider the benefits (and costs) of members to parties, and why declining numbers can be problematic. Then we consider party membership resurgence. We discuss spikes in party membership and what drives them, including connections that exist between parties and movements. These membership surges are not well-understood, but it is clear they present new opportunities and challenges for political parties.

#### **Party membership: decline and revival**

There are cultural and political differences in the way party membership is understood. Definitions vary by country and by party, creating difficulties for comparative research (Scarrow 1996; van Haute 2011; Ponce and Scarrow 2016). Membership can be conceived of as formal (fee-paying) or behavioural (such as when a voter attends a party event). For the SNP and Scottish Greens, membership means formal, fee-paying status, traditionally known as 'card-carrying membership'. Each party invites supporters to sign up but neither offers a separate low-intensity supporter option for those considering joining which offers some of the rights of full membership. The monetary cost of membership (the membership fee) is 'suggested' on the party websites and is highly discounted for different groups – students, lowincome groups, refugees and asylum seekers and so on. This financial payment brings rights and responsibilities. Entitlements include the right to vote in internal candidate and leadership selections and to contribute to policy development. Conditions of membership are not overly prescriptive. Members are expected to belong exclusively to that party and abide by the party's policies and rules.

#### *Membership decline*

Until recently, there was broad agreement that fewer people were joining parties. In the 1950s, roughly one in every ten European citizens belonged to a political party (Katz and Mair 1992). A body of work documented the long-term decline in aggregate membership across much of the Western world, observing a 'withering away' of parties as member-based organisations (van Biezen et al. 2012: 41). Mair (2013: 38) identified the 1990s as the critical period when the decline became 'unequivocal and seemingly unstoppable'. Van Biezen et al. (2012: 24) identified a 'low ebb' at the end of the first decade of the 21st century, whether membership was measured in absolute numbers or as a percentage of the electorate (the M/E ratio). It seemed that party members 'were doomed to disappear' (van Haute 2011: 14).

UK parties were especially badly affected by membership decay (Whiteley 2011; van Biezen and Poguntke 2014). Van Biezen et al. (2012: 33) report a loss of two-thirds of UK party members between 1980 and 2008, a decline of just over half a million (compared to an average decline of 50% in other European democracies). Similar conclusions were reached when studies used subjective measures of party membership (reported by respondents in population surveys) or objective measures (figures published by parties).

#### 4 *Introduction*

Until the development of centralised membership lists, however, parties lacked reliable data on their members and tended to overestimate numbers, meaning that apparent decline might, at least partially, have been due to inaccurate early reporting. And the pattern did not apply equally to all countries and parties. Decline mainly took place in mass parties (traditional parties strongly connected to social groups) in established democracies (Duverger 1954). Parties in new European democracies began with and maintained small memberships (Delwit 2011). In the United Kingdom, reliable figures on party memberships became available following the introduction of the *Political Parties, Elections and Referendum Act 2000*. This obliged political parties to deposit annual financial reports with the Electoral Commission, and most parties included membership figures.

A key debate is whether membership decline took place by design or necessity. Parties might see members and activists as more costly than beneficial, or they may be convinced of their value but find it difficult to recruit them. The dominant interpretation of membership decline was one of shrinking supply. Over time, societal, economic and cultural change weakened the relationship between citizens and parties (leading to partisan dealignment and declining trust), and this reduced the pool of individuals prepared to join a political party (Norris 2002; Morales 2009). Many saw change in the participatory preferences of modern citizens. Faucher (2015: 406) identified a personalised, ad hoc form of participation. In this context, pressure groups appeared more successful at recruiting supporters and members in a 'protest business' (Jordan and Maloney 2007). In turn this was connected to a new 'atomised' form of party membership (Seyd and Whiteley 2002: 214), with members less willing to engage in time-consuming activities such as delivering leaflets but more inclined to engage in individual acts like supporting a party on social media (Gauja 2015; Scarrow 2015).

Members, though, were viewed as beneficial and 'in demand' from the perspective of the political party. It was assumed that potential recruits were simply thin on the ground (Seyd and Whiteley 1992, 2004; Whiteley 2009). This implied that shrinking memberships had thoroughly negative consequences because rankand-file members performed important functions. Broadly, party members connect parties and society. If party membership is in decline, this inhibits parties' ability to perform traditional functions like democratic linkage, political representation and communication with voters (Lawson and Merkl 1988; Widfeldt 1999; Webb et al. 2022). Bale et al. (2020: 192) argue that parties in no way gave up on recruitment, because members were viewed unequivocally as 'an asset, not a liability'. Similarly, van Haute and Ribeiro (2022: 281) state that 'members still constitute a distinct and unique asset for parties'.

Parties accrue many organisational benefits from members. Perhaps most obviously, members bring income through the payment of membership subscriptions and additional donations. UK parties receive modest amounts of state funding compared to parties in other European countries.1 Some parties have alternative sources of funding, but many (like the Greens) rely heavily on income from members (Fisher 2018). The more active members help shape party policy – attending local meetings, policy forums and conferences. Members take part in the selection of election candidates and leaders, and they themselves form a pool of talent from which office-holders, candidates and leaders emerge. Indeed, the frequency of UK elections means parties require more candidates than ever before. And scholars identify the vague but important idea that members give parties organisational *legitimacy* (van Biezen et al. 2012: 26). The symbolic importance of a large membership is demonstrated by parties' tendency to over-report membership and to be reticent about decline.

 Members are also a valuable and direct electoral asset because they are loyal voters (Katz 1990). They campaign on behalf of parties, and they perform an ambassadorial role in local communities, networking with non-members and influencing opinion (Scarrow 1996; Seyd and Whiteley 2004; Fisher et al. 2014). These local linkages can mean that parties gain electorally from having members on the ground. (It is no coincidence that the SNP and to an extent the Scottish Greens have made significant advances in local government.)

European parties became increasingly concerned about the 'membership recruitment crisis' (Faucher 2015: 405). Supported by new technology, there was a shift from local to centralised recruitment. New recruitment mechanisms, especially electronic payment, allowed parties to experiment and develop discounted membership rates for different groups. Communication between leaders and members improved, and parties used data on members to target funding appeals. Bale et al. (2020: 18–19) identify three key developments: centralisation (the member's relationship is with the central party), digitalisation (the relationship is conducted using technology) and accessibility (the process is easy, quick and cheap). Some parties experimented with new 'affiliation options' (Ponce and Scarrow 2016: 679–80). These included registered supporters or 'friends' and 'virtual affiliation' designed to encourage those who wouldn't consider full membership to sign up for 'membership-lite', sometimes but not always for a fee (Hartleb 2013; Faucher 2015). These changes are summed up in Scarrow's (2015: 30–31) account of 'multi-speed membership parties', which identifies several forms of affiliation (including social media followers) and stresses the fluidity with which individuals might move from one to another.

By the early 2000s parties were implementing changes to internal decisionmaking procedures. Members gained new organisational rights, such as selecting party candidates and those who ran internal party committees (Cross and Blaise 2012; Cross and Katz 2013; Gauja 2017). Leadership selection was opened to party membership ballots (one member one vote), a process observed in Europe and beyond and in all party types (Kenig 2009). In some cases, non-member supporters were invited to participate in internal party decision-making. In candidate selection, some parties experimented with open primaries.

What this all meant for the location of power in parties, and specifically the role of the party member, was unclear. It could be argued that as membership rolls declined the individual member had never been more powerful, but these moves were commonly interpreted as enhancing the position of party leaders rather than a dispersal of power (Katz 2001). Faucher (2015: 416) described party memberships 'with vertical links to leadership (through regular two-way communication) but

#### 6 *Introduction*

little horizontal integration (without contacts with local activists)' who would be 'easily mobilised on specific issues or campaigns'.

#### *Surging party memberships*

Studies charting the decline of party membership underestimated underlying volatility – that is, short-term downturns and upturns in individual parties. Spikes in membership are not uncommon (Kölln 2016; van Haute et al. 2018). Researchers have connected these to electoral cycles and bandwagon effects, observing that electoral success brings new members (Whiteley and Seyd 1998; Fisher et al. 2006; Sierens et al. 2023). The impact of a party being in government is less clear and far from uniform. Entering government can boost recruitment (Sierens et al. 2023); but many governing parties lose members over time (Bartolini 1983; Widfeldt 1999). A Blair bounce in Labour membership gradually dissipated when the party was in government (Jones 2020; Seyd 2020).

Parties have seen waves of members joining to take part in leadership contests (such as the Canadian Liberal Party and Labour in the United Kingdom). And parties experience peaks in membership resulting from specific issues rising up the political agenda, an issue-attention effect (Downs 1972). Membership of the British Greens increased markedly at the end of the 1980s as environmental issues gained salience, only to fall away again (Rüdig et al. 1996). UK Independence Party (UKIP) membership surged in the mid-2000s but receded quickly when the issue of the United Kingdom's relationship with the European Union had been decided by referendum. There is a fluctuation in party membership that sometimes goes undetected in accounts of long-term decline.

In the wake of the international financial crisis and austerity policies imposed by European governments, some populist parties of the radical left successfully mobilised members (March and Keith 2016; Rooduijn and Akkerman 2017). Podemos in Spain attracted large numbers of recruits at the party's foundation in 2014 (Gomez and Ramiro 2019).<sup>2</sup> Parties of the populist radical right including Italy's Fratelli d'Italia and the National Rally in France (as well as UKIP) have seen spurts in recruitment. Whiteley et al. (2021: 645) suggest that these parties 'appear to have reversed a long-established trend of declining party membership'. However, numbers of populist party members in Europe remain small when compared to other parties (see Dassonneville and McAllister 2023: 19; van Haute and Ribeiro 2022: 286).3

The loss of members by older parties with historical traditions of mass membership largely explains long-term decline, a party lifecycle effect (Sierens et al. 2023). Yet the most striking developments, certainly in the United Kingdom, have been when new members have flocked to traditional parties, and sometimes in the context of electoral failure (Bennie 2016). Paradoxically, parties who have lost at the ballot box have occasionally gained many more members than those who have been on the winning side. This is a phenomenon of the past decade, with spikes in membership becoming more frequent and pronounced, suggesting a revival of party membership. In opposition, a Labour Party leadership election led to membership nearly doubling in 2015. The Liberal Democrats, electorally damaged by a governing coalition with the Conservatives, began to recover membership from 2015.

As in electoral politics, trends in party membership are becoming more difficult to forecast. Unpredictable events and triggers can mobilise members. Bale et al. (2020: 15) refer to these as 'catalytic moments'. Power and Dommett (2020: 508) identify events which act as a 'trigger' or 'moment of activation', persuading those who are supportive of a party to take the step to join. They highlight fluctuation in party membership rolls, suggesting that short-term events are as important as longterm trends, and argue that triggers can be highly context-specific, such as a party's exclusion from a national leadership debate (Ibid.: 507).

Referendums have the potential to act as 'catalytic moments', but prior to 2014 there was little sign of this in UK politics (as we will discuss in Chapter 2). Nor do comparable international experiences suggest such a link. Accounts of the referendums on sovereignty for Quebec in 1980 and 1995 contain no reports of unusual patterns of party membership, either during the campaigns or following the results (Keating 1996; Pammett and LeDuc 2001). Turnout in these referendums was high (86% and 94%), and the 1995 vote was extremely close (50.6% No), representing a very narrow defeat for the sovereigntists. Yet the campaign for change dissipated rather than creating a new momentum in party recruitment.

 The Catalan independence referendum in 2017 involved high-profile demonstrations and marches, but there was little evidence of citizens joining parties on a grand scale (Cetrà et al. 2018). A critical difference in this case is that that referendum was not recognised as legal by the Spanish government (McRoberts 2022: 190). However, there seems no prima facie reason to suppose that this fact should block any flow of members into the parties in the vanguard of this movement – indeed, if anything, we might expect it to trigger a stronger counter-mobilisation. However, there was no such spurt in party membership. 4

A feature of recent party membership surges is that they take place online. Continuing advances in digital technology have transformed the joining process, making it extremely quick and easy. This involves a form of instant gratification for the joiner. Add to this the speed with which members can communicate with others that they have joined, and it is easy to see how membership might have a selffueling, snowballing element resembling a 'flash flood' or 'flash activism' (Earl and Kenski 2022: 441). Parties can respond proactively following an initial surge, posting updates of numbers joining and inviting others to come on board, as did the SNP following the 2014 referendum.

Aspects of party membership surges require further research, especially the ways in which they impact on parties themselves. A membership surge has the potential to change the socio-demographic and ideological composition of a party. Three decades of research on who joined parties revealed the 'patterns of pervasive under-representation of groups' including young people and women (Young 2013: 78). Rüdig (2011) described party members as 'weirdos'. Research on dwindling party memberships across Europe suggested that members became less demographically representative of parties' voters (Scarrow and Gezgor 2010), but evidence was inconclusive (see van Biezen et al. 2012). The impact of the reverse process of surging memberships is similarly unclear (see Achury et al. 2020; Dassonneville and McAllister 2023).

As for why members join parties, this has been investigated at length (Seyd and Whiteley 1992; Whiteley et al. 1994, 2006) but much less so in the context of membership surges. Empirical studies suggest members can be influenced by social norms and participating alongside others, but the most common motivation is an expression of support for party ideals and policies (Cross and Young 2004; van Haute and Gauja 2015; Poletti et al. 2019). Of course, beliefs and views are shared by lots of people who don't join parties, leading Power and Dommett (2020: 508) to state that 'motivations are not enough' and must be accompanied by a prompt, or trigger, and processes which make it easy to join. The possibility remains that membership surges are explained by different reasons for joining. If motivations themselves are context-specific, they might vary in a post-referendum context. Among those who joined the SNP and Scottish Greens following the 2014 referendum, for example, we might expect a distinction between those who joined primarily as a means of achieving the policy end of independence and those who sought an outlet to maintain the participatory activities that flourished during the referendum. We explore this possibility later in the book.

A related question is how large numbers of new members impact on a party ideologically. New cohorts of members might strongly resemble established memberships, or they might realign the values and policies of a party. Some studies have addressed how new members of a party compare with established members ideologically and on party strategy (Mitchell et al. 2012; Gomez and Tonge 2016; Whiteley et al. 2019). While these suggest that an influx of new members can influence the profile and preferences of a party's membership, the wider picture is that new members strongly resemble the members they are joining.

Researchers are less certain about how a large intake of new members influences internal party democracy. Traditional studies considered the implications of membership *decline*. Mair (2013) observed a 'hollowing out' of internal party democracies, where power resided with leaders at the expense of grassroots members. A key question is whether internal party dynamics are reshaped by recruitment surges. Do new members demand more power? Parties born of social movements are likely to experience more of such demands from their members (Rüdig and Sajuria 2020). And modern technology offers the possibility of involving members more regularly in internal party ballots (Scarrow et al. 2022).5 However, these might actually enhance party centralisation (Chadwick and Stromer-Galley 2016; Gibson et al. 2017). The limited evidence available on surging memberships does not point to enhanced power for members. Seyd (2020: 19–20) examined the rhetoric of membership participation in Corbyn's Labour Party and concluded that little change occurred.

Many of the membership surges seen in the United Kingdom and elsewhere have been temporary, which raises the question of why members leave. Membership turnover is methodologically difficult to track, which explains the lack of research on the topic, but retaining members and persuading ex-members to rejoin are important to parties. Research on ex-members points to ideological incongruence and policy differences as the main reasons members leave, but perceptions of leaders and how parties are run matter too (Whiteley 2009; Bale et al. 2020: ch. 8; Barnfield and Bale 2022). Webb and Bale (2023) demonstrate that similar reasons motivate those members who switch between parties. 6

These are just some of the themes investigated by party membership specialists and some of the gaps in our understanding. Studies have been theoretically informed and rich in empirical detail, producing knowledge (and many detailed datasets) on key aspects of the party membership experience.7 There has, though, been a lack of research on the membership of sub-state parties, with some notable exceptions (see Mitchell et al. 2012; Gomez and Tonge 2016; Hennessey et al. 2019). Unsurprisingly, scholars take the firm position that members matter to political parties (and to democracy). Bale et al. (2020: 192) note that 'parties find it difficult to imagine how or even why they would carry on without members'.

#### **This study's aims and analytical approach**

To fully explain an exceptional example of political mobilisation, we argue that the relationship between conventional parties and wider political and social movements must be understood. Others have noted that referendums are sometimes linked to social movements, a way for campaigners to challenge the status quo (Qvortrup 2018: 1). della Porta and colleagues (2017: 1) refer to the Scottish independence referendum as a 'referendum from below' but don't address the reasons behind the party membership surge. Characteristics of social movements include a set of ideas or identities that bind participants, identifiable networks, organisational informality and spontaneous forms of activism/action repertoires (Tilly 1978; Diani 1992; Johnston 2014). Following the 2014 referendum, party membership appeared to be subsumed into the action repertoires of the Yes movement. The book will address key social movement dimensions identified by Tilly (1978) and others: the groups and organisations in the movement, the movement participants and the movement's ideas.

While the roots of some parties clearly lie in movement politics (see, e.g., Minkin 1978 on Labour), social movement approaches have not been widely utilised by party scholars in recent years. There has been analysis of contemporary movement politics in Europe and elsewhere – such as anti-austerity political movements – but relatively little work has been done on the connections between referendums, social movements and parties. In this book, social movement approaches will supplement the more traditional political science approaches on political participation, party membership and how parties organise, exploring the connections between conventional party politics and movement politics – an increasingly topical issue as voters turn to anti-establishment parties around Europe.

The book offers explanation of party membership in the SNP and Scottish Greens. Reporting on the results of an Economic and Social Research Council (ESRC)-funded study of the parties' members and an extensive programme of interviews with party elites and activists, the book examines the political and movement backgrounds of the parties' members, revealing ways in which they

#### 10 *Introduction*

 experienced the 2014 referendum, and how and why many were attracted to a political party. Motives for joining are examined, as well as demographics, views and attitudes of the parties' members. A key component of the book is an assessment of how – and how far – the membership surge *changed* the SNP and Scottish Greens, socio-demographically, ideologically and organisationally, addressing how the parties have been reshaped post-referendum. Finally, the book assesses the place of the parties in the modern independence movement. Specifically, the book aims to understand the following:


#### *Data and methods*

The book is the only comprehensive analysis of a dataset unrivalled in scale. Those data were collected via a multi-method ESRC-funded research study, 'Recruited by Referendum' (ES/N010590/1). Online surveys of the memberships of the two parties were conducted between September 2016 and March 2017, resulting in datasets of many thousands of members, allowing detailed analyses of member behaviour and attitudes. Questionnaires addressed the members' motivations behind joining a party, their other political and organisational affiliations, their opinions on a vast range of topics and their experience of being a party member. Some questions allowed written/free-form answers, to encourage the members to explain their views and motivations in their own words, rather than simply responding to the researchers' pre-constructed questions and answers.

These are supplemented by two other sets of data gathered via surveys in 2015 and 2018. The first set of data is derived from the campaign group Women for Independence, whose membership we surveyed in 2015. The second set of data comes from a survey of SNP members conducted in 2018. In June of that year, the SNP experienced a mini-surge in its membership, a 6% increase over 48 hours. This followed the SNP's parliamentary group walking out of the House of Commons in a row over post-Brexit reallocation of powers. We invited all these new members to complete a survey, with a focus on what prompted them to join. Full details of all these surveys with response rates (and discussions of non-response bias and weighting of data) are provided in the book's Appendix.

Throughout the book, we also draw on data from much earlier surveys of the SNP and Scottish Greens. A full membership survey of the Scottish Greens took place in 2002, and of the SNP membership in 2007–2008, and these data have been reported elsewhere (Bennie 2004; Mitchell et al. 2012). Comparison with the postreferendum survey findings allows us to assess the changing characteristics and attitudes among the parties' memberships.

We also conducted an extensive programme of face-to-face, semi-structured elite interviews with close to a hundred senior figures and activists in the two parties and across the national movement. Most of these interviews (75 of them) took place over three years (2016–2018) and focused on the perceptions of the referendum, why the surge took place, the impact of the surge on the parties and their organisational adaptation. These interviews were on the record and fully transcribed, allowing us to quote freely those directly involved in these events, but interview data are anonymised. A second batch of more informal interviews took place between 2021 and 2023, with a focus on understanding the long-term consequences of the surge on the parties and national movement. More details can be found in the Appendix.

Together these quantitative and qualitative components provide breadth, depth and richness to our portrayal of the two parties' members and their place in a national movement for change at a time of change. They allow us to make three central comparisons. First, the data enable comparison of the two parties in terms of members' characteristics, motivations and experiences. Second, the book compares long-standing (pre-referendum) and new (post-referendum) members. Third, the book compares the study's findings to those of the previous investigations of the SNP and Scottish Greens (Bennie 2004; Mitchell et al. 2012; Johns and Mitchell 2016). These three key comparisons provide an unparalleled opportunity to explore party change.

#### **Chapter outline**

Chapter 2 sets the scene by examining the history of referendums in the United Kingdom. We first explore the nature of referendum campaigns and any links with recruitment to political parties. Second, the chapter provides a detailed account of the 2014 referendum on Scottish independence, describing the national movement for change and the characteristics of the campaign. The chapter then provides a full account of the post-referendum surge, demonstrating its exceptional nature, and we document how membership developed in the years that followed (until 2023). This forms a foundation from which to explore the book's central research puzzles: Why did the surge take place and what has it changed?

Chapter 3 addresses the relationship between a conventional party and wider political and social movements, conceptually and empirically. It provides an account of the two parties prior to the referendum and the relationship between the two parties and others in the referendum. Tilly (1978) identified the 'action repertoires' of social movements and focused on their 'performance' aspects. We explore the ways in which the members experienced these alternative action repertoires during the referendum. Chapter 4 examines the characteristics and background of the members who joined the SNP and Scottish Greens. The chapter provides evidence of the socio-demographic characteristics across our key comparisons. An important question is whether the referendum attracted new, previously underrepresented groups – notably, women, younger people and the less affluent – into party membership.

The surge in membership saw new members far outnumber those who had previously been members. The central question of this study, explored in detail in Chapter 5, is what motivated the new members to join and how reasons for joining relate to the referendum campaign and outcome. We examine self-declared reasons for joining and how these relate to political attitudes and participatory motivations. Chapter 6 is a detailed exploration of members' values, political opinions and visions of independence, examining whether the surge changed the attitudinal profile of the parties. Did the new recruits see self-determination less as an end in itself and more as a means to anti-establishment and/or leftist politics, and how impatient were they for a second referendum on Scottish independence?

Chapter 7 investigates whether the new members were more *participatory* than traditional members, or active in different ways, perhaps carrying forward the unusual movement repertoires of the referendum campaign. The chapter provides direct evidence of the ways members connect with their parties, exploring the distinction between individual and collective forms of activism. The key objective of Chapter 8 is to understand the consequences of the surge for party organisations. The chapter includes the party elites' perspectives on intra-party benefits and challenges of expanding memberships, and it explores reforms to internal decisionmaking processes. We assess how the location of power in the two parties has been altered by the surge. Have party members been empowered or party leaders strengthened? The conclusion of the book (Chapter 9) considers the legacy of the independence referendum and the surges in party membership, and it includes an account of the 2023 SNP leadership contest. The relationships between the SNP and Scottish Greens and between pre- and post-surge memberships are set within a discussion of the pro-independence movement almost a decade on from the 2014 referendum.

#### **Notes**


#### **References**


#### 14 *Introduction*

Fisher, J. (2018) 'Party Finance', *Parliamentary Affairs*, 71 (1): 171–188.


### **2 From referendum to party membership**

#### **Introduction**

The 2014 Scottish independence referendum campaign was an extraordinary period in recent political history. The campaign was lengthy; it involved crossparty and party-group cooperation; and in its later stages it was a colourful 'festival of democracy' (McWhirter 2014: 14). Registration efforts enlarged the electorate, who became highly engaged, and the turnout (85%) was exceptional. The referendum also had potentially profound long-term effects on Scottish and UK politics. The constitutional debate reshaped electoral behaviour and party fortunes in ways few would have predicted (see Henderson et al. 2022).

Our key interest lies in the referendum's impact on party membership. In this chapter, we explore the relationship between referendums and party membership, arguing that until 2014 there was little sign that a constitutional referendum might spark party recruitment, either in the United Kingdom or elsewhere. Comparable sub-state level referendums – for example, those that took place in Quebec and, more recently, in Catalonia – did not have the same knock-on effects. The Scottish case of 'losers' being mobilised as party members following a referendum outcome has unique features, which we will outline in this chapter.

We begin by examining the history of referendums in the United Kingdom and then provide a detailed account of the 2014 referendum on Scottish independence, when a national campaign dominated by the SNP coexisted with an energetic grassroots movement. The chapter's next objective is to consider patterns of party membership over time. We document the early development of membership in the SNP and Scottish Greens, as far as is possible, because reliable data became available only in the 2000s. We examine the soaring of membership after the 2014 referendum, establishing exactly how and when this occurred, and we look beyond the surge to address whether the parties were able to retain their members. To provide wider perspective on the scale of the surge, we compare SNP and Scottish Green membership with other parties in the United Kingdom.

#### **Referendums in the United Kingdom**

In the United Kingdom referendums were traditionally viewed as an uncomfortable fit with parliamentary sovereignty (Butler and Ranney 1978, 1994). When

> DOI: 10.4324/9780429446559-2 This chapter has been made available under a CC-BY 4.0 license.

 manifesto commitments were superseded by a referendum, this was seen as incompatible with 'the doctrine of the mandate' (Bochel et al. 1981: 171). Referendums were perceived by governments as potentially risky, involving a loss of control. To date, only three UK-wide referendums have taken place: in 1975 (on the United Kingdom's continued membership of the European Community [EC]), in 2011 (on changing the United Kingdom's electoral system), and in 2016 (on whether to leave the EU). They are usually instigated by the executive, with an expectation of winning. The reasons for holding one vary, but they can be a process of seeking legitimation, a way of resolving divisions within parties or the result of pressure from below. Governments make the decision to hold a referendum with parliamentary consent.1

Referendums in the United Kingdom have mainly been used to decide on constitutional matters, and mostly at the sub-state level (Table 2.1). The first occurred in 1973 when the Northern Ireland border poll asked voters if they wanted to remain part of the United Kingdom. The referendum was controversial, boycotted by Irish nationalists, and the result maintained the status quo. In 1979, there were simultaneous votes in Scotland and Wales on devolution. The Labour government had proposed Assemblies for Scotland and Wales, but there was internal party resistance, and agreement on a referendum was reached. Against the wishes of the government, a Labour backbench amendment determined that a Repeal Motion would be moved if less than 40% of the eligible electorate supported the proposals (Bogdanor 1980). This represented a move away from the principle of decision by simple majority required in 1975, and it remains unique in the history of UK referendums.


*Table 2.1* Referendums in the United Kingdom (1973–2016)


*Source:* Compiled by authors.

\*32.9% of eligible voters, falling short of the 40% threshold

\*\*11.8% of eligible voters

Some reports of the 1979 campaign portray it as lively and engaging. Bochel et al. (1981: 170) note: 'An impressive number of meetings and debates was held in every corner of Scotland, which was awash with leaflets, pamphlets, stickers and posters'. However, the Yes side formed an uncomfortable alliance (Peran 1980). Those in favour of devolution had different constitutional objectives, no more so than SNP and Labour campaigners. For the SNP an Assembly represented a step in the right direction towards independence. Scottish Labour Secretary at the time, Helen Liddell, stated: 'We will not be soiling our hands by joining any umbrella Yes group' (Macartney 1981: 17). Labour campaigning was lacklustre, and their voters were exposed to mixed messages, which was crucial to the result and an early indication of the importance of party cues in referendums (see Lupia and

McCubbins 1998; Hooghe and Marks 2005). Turnout was 64%, similar to 1975 but lower (by more than 10 percentage points) than a general election at the time. While a majority of voters in Scotland supported devolution, the 40% rule thwarted the campaign. Welsh voters (59% of whom turned out) overwhelmingly rejected the idea of an Assembly.

 With the election of New Labour in 1997, referendums were back in vogue, specifically to deliver and legitimise constitutional change. The Scottish Parliament, the Welsh Assembly and the London Assembly were subject to referendum, as was the Northern Ireland Good Friday Agreement. The latter produced a turnout of over 80%, at the time the highest of any referendum in the United Kingdom. The 1997 referendum on Scottish devolution followed 18 years of Conservative government at Westminster, and there was cross-party support for the creation of a Scottish Parliament with tax-varying powers. Labour, Liberal Democrats, Greens and the SNP worked together under the umbrella organisation 'Scotland Forward', campaigning alongside civil society organisations, trade unions and some business figures and groups. They campaigned for a Yes-Yes or 'double yes' vote (this being a two-question referendum on setting up a Parliament and the Parliament having tax-varying powers). The pro-devolution camp was more united than in 1979. The Conservatives, having lost all their Scottish seats in the 1997 general election, were isolated in support of a No-No vote at the head of the 'Think Twice' campaign. Some campaigning was local, colourful and visible, but the duration of the formal campaign was short, certainly by the standards of 2014, and it was interrupted by the death of Diana, Princess of Wales (Denver et al. 2000). A 60% turnout was consistent with previous referendums.

A referendum on Welsh devolution was held a week later. The hope had been that a clear majority in Scotland would encourage support for a Welsh Assembly. Wales very narrowly voted in favour (50.3%) on a turnout of 50.2%. In 2004, a referendum in North-East England rejected an Assembly when almost 80% voted against on a 47% turnout. This had been intended to be the first in a series of referendums in English regions rolling out devolution but came unstuck with the result in the region anticipated to be most likely to be in favour.

The United Kingdom's vote on electoral reform in May 2011 was a damp squib of a referendum. There was limited support for the Alternative Vote (AV) even among Liberal Democrats who preferred the Single Transferable Vote (STV) but had negotiated a referendum on AV in their coalition deal with the Conservatives. Voters were uninspired – less than a third of the 42% who turned out supported the proposal. The Scottish Parliament election of 2011 took place on the same day as the AV referendum. The SNP achieved a majority in the Scottish Parliament, and a referendum on Scottish independence became a real prospect.

Other types of referendums have taken place in the United Kingdom. Local referendums have been quite common, from prohibition in the distant past to more recent votes on congestion charges and local building developments. In 2011, Aberdeen residents were consulted on a city centre garden project (Union Terrace Gardens): 52% took part, a majority voted in support, and councillors then rejected the idea. Since 2001 there have been more than 50 referendums on establishing local mayors in England (often rejecting the idea). Turnout was typically around 30% but on occasion surpassed 60%, usually when coinciding with an election, such as in Berwick-upon-Tweed in 2001.

Based on experience, then, there was little to indicate that a referendum campaign could provoke unusual grassroots or party membership activity, either during a campaign or following a result. Referendums were associated with constitutional change and were an occasional but conventional element of politics in the United Kingdom.

#### **The Scottish independence referendum 2014**

 The SNP manifesto in 2011 contained the commitment to hold a referendum on Scottish independence, and the Scottish Greens' manifesto supported a multioption referendum, but independence was not a significant issue in the election. The SNP had appealed to voters on the basis of competence and other policies, not independence (Johns and Mitchell 2016). The Scottish referendum was interesting constitutionally as its origins lay outside Westminster but was conceded by the UK government, believing it would easily win. David Butler noted that referendums in the United Kingdom 'are only going to be held when the Government of the day wants it or when it would be too embarrassing (because of past promises) to get out of it', describing this as 'a matter . . . of straight politics' (Butler 2010: para. 37). However, the government in London did not set the terms or timing of the debate and lost control of the agenda.

In January 2012, the SNP government produced a White Paper outlining their plans for a referendum. The decision to hold a referendum and the rules surrounding the process involved agreement between the Scottish and UK governments, and the Edinburgh Agreement was reached in October 2012. The date of the referendum would be set by the Scottish government, but it would take place no later than December 2014. There would be a single ballot paper, and voters would be given a choice between two constitutional outcomes (independence or staying in the United Kingdom). The UK government insisted on excluding an option of enhanced devolution supported by the SNP leadership, but agreement on a long campaign and extending the franchise to 16- and 17-year-olds was a victory for the Scottish government. The question which would be put to the Scottish people was agreed with the Electoral Commission: 'Should Scotland be an independent country? Yes/No'. A referendum date was set for September 2014 providing a long lead-in to the vote. At the time, polls suggested that independence would be easily defeated.

Under the terms of the Political Parties, Elections and Referendum Act 2000, the independent regulatory Electoral Commission designates lead organisations. Unlike in previous referendums, there was no public funding for the lead organisations. The lead campaigner on the 'Yes' side was 'Yes Scotland', and 'Better Together' was the equivalent on the 'No' side. Both were established in 2012 and later (in April 2014) became official designated lead campaign groups. There was no competition for status as lead organisation as happened in the pro-Brexit referendum campaign in 2016.

As the party of government and protagonist in the debate, it was inevitable that the SNP would dominate the campaign. While other organisations and parties had supported independence over the decades, by the time of the referendum the SNP had become synonymous with the independence movement. This benefited the SNP in elections, but the party was aware that it needed the support of others in the referendum, and it wanted a separate organisation to be established, with non-SNP figures, parties and groups working alongside the SNP. Yes Scotland performed this role. Dennis Canavan, former Labour MP and subsequently Independent MSP, was appointed chair, and its board was consciously cross- and non-party. But key links between the SNP and Yes Scotland were established with former special advisers to First Minister Alex Salmond moving over to Yes Scotland. The SNP encouraged its members and branches to get involved in local Yes groups.

The Scottish Greens had contained a range of opinions on Scottish independence. Robin Harper, the party's first MSP (and first Green parliamentarian anywhere in the United Kingdom), viewed independence as a distraction from the party's core green purpose (Harper 2011). Others in the party viewed the idea as inimical to the implementation of green policies, which require UK-wide and international approaches. Some were suspicious of the SNP which they viewed as un-green and materialist. Following the establishment of the Scottish Parliament in 1999, the party evolved to be supportive of independence, with prominent figures like Patrick Harvie leading the debate, although a section of the party remained unconvinced. A survey of the Scottish Green membership in 2004 revealed that half supported independence, 41% favoured independence in Europe, 9% favoured independence outside Europe and 37% preferred enhanced powers for Holyrood. The party was essentially tolerant of all views on Scottish independence within its ranks, but the body of opinion gradually moved in favour, members persuaded that this was consistent with the green values of sustainability and subsidiarity.

With the prospect of a referendum on the horizon, the Scottish Greens debated whether to campaign for Yes at their autumn conference in 2012. One attendee (Interview no. 62) described the atmosphere as 'heated' and estimated that 'about three quarters of the party wanted to campaign for yes, a quarter for no', but those in attendance at the conference voted 'fairly unanimously' to take part in the Yes campaign. The argument that won the day was that a referendum presented an opportunity for the party to showcase a radical green vision of an alternative Scotland. When co-leader Harvie addressed the conference, he encouraged members to get involved with Yes Scotland, stating that it was 'inclusive'. At the public launch of Yes Scotland in 2012, Harvie joined leaders from other independencesupporting parties, including the SNP and the Scottish Socialist Party.

The campaign was the most intense and lengthy in modern Scottish political history, and this is critical in understanding the events that followed. Formal campaign spending and funding rules applied to a 16-week period (30 May to 18 September), but the SNP's campaign amounted to a period of more than two years, generating extensive popular debate. Activists were being trained by Yes Scotland as early as 2012, and canvassing and leafleting took place from the summer of 2013. The legislation establishing the rules of the campaign – the Scottish Independence Referendum Act 2013 – was passed by the Scottish Parliament nine months in advance of the vote (compared to a period of three months for the AV referendum in 2011). In December 2013, the Electoral Commission began registering 'permitted participants' or 'registered campaigners', parties, groups and individuals planning to spend more than £10,000 during the statutory campaign period. Forty two groups were registered (Table 2.2). Supported by Yes Scotland, groups like the Radical Independence Campaign (RIC), National Collective and Women for Independence (WfI) would form relationships and networks with each other and with the pro-independence parties.

The long run-up to the referendum created opportunities to register voters. The Radical Independence Campaign was important in encouraging voter registration in deprived urban areas, those with traditionally low levels of electoral engagement, providing information, forms and envelopes before the


*Table 2.2* Referendum campaign groups registered with the Electoral Commission

*Source:* Electoral Commission (2014: 151)

registration deadline of 2 September. The electoral register contained nearly 4.3 million eligible voters, the largest ever for a Scotland-wide poll, also boosted by the extension of the franchise to include those aged 16 and 17 (Electoral Commission 2014: 60).

With its plentiful resources and experience, the SNP was the key player, nationally and locally (Table 2.3). As the driver of the referendum, the party set out the agenda that informed much of the long public debate. Yes campaigners had different visions of independence and different reasons for supporting a Yes vote (Geoghegan 2015; Mooney and Scott 2015), but the SNP's social democratic message dominated the campaign. The Greens advocated a radically transformed Scotland and a more equal and sustainable society.

The SNP dominated the traditional data gathering – the knocking on doors and canvassing of opinion – contributing to a database of voters called Yesmo. Most local Yes campaigners who made up the ground war were SNP supporters and members and were key to getting out the vote (GOTV) on the day of the referendum. Scottish Greens contributed some colourful campaigning, but activist numbers were low and patchy. Myriad other non-party organisations and individuals campaigned, new groups forming and evolving through the course of the campaign, but anti-independence groups on the ground were fewer in number.

As the campaign progressed, especially from the summer of 2014, organisation became more fluid and decentralised, with more community-based, grassroots activity. To a degree, there was a revival of traditional kinds of campaigning


*Table 2.3* Referendum spending by registered campaigners (over £10,000)

*Source:* Electoral Commission (2014: 48–49)

(Adamson and Lynch 2014). Lynch (2017: xiii) refers to 'very old school political campaigning based around conversations and shoe leather'. This pavement politics involved leafleting 'on an industrial scale' (Lynch 2017: 21).2 And public meetings drew particularly large crowds for modern times, attracting people who were unknown to the parties, and not only when these were addressed by high-profile Yes leaders as described here by one of our SNP interviewees (Interview no. 1):

There was a phenomenal turnout at public meetings – Millport [on the Island of Cumbrae] saw a quarter of the population turn out on an afternoon to a public meeting addressed by Nicola Sturgeon. She would speak for 15 minutes then take questions for an hour and a half. The people attending were not the usual suspects.

The grassroots activities on the ground – the action repertoires of campaigners – formed a colourful campaign landscape, especially on the Yes side. Innovative campaigning methods and political carnival were on display. Traditional town house–type meetings combined with campaign stalls, impromptu flash mobs, city marches and creative, cultural events (Adamson and Lynch 2014, 2019; Geoghegan 2015). There were choirs, concerts, flags and posters, fire engines and other Yes-mobiles. On the weekend before polling, WfI hosted a rally in Edinburgh, which was followed by a car cavalcade across the city. These campaign techniques, some of them 'old-style', were imaginative and fun.

The campaign saw extensive use of social media, with its good and bad features. It became an important organisational tool for campaigners, allowing them to spread details of national and local events among activists. Lynch (2017: 39) says Yes Scotland 'built a community through technology' with the sharing of material between Yes groups, although he argues that traditional campaigning was what really mattered. Social media was also important for crowdfunding; National Collective's 'Yestival' is an example of an event funded in this way.

Yes Scotland and local Yes groups produced campaign materials including badges, balloons, t-shirts, tote bags, car decals, posters and postcards, toys and giant foam hands. Parties also distributed a considerable amount of merchandise via their websites. This all helped create a lively and vivid campaign. As the referendum approached, it was difficult to be unaware in Scotland of the campaign. Visual displays were concentrated in some areas, but the campaign merchandise combined with home-made posters and signs to create colourful scenes in many parts of the country. For years following the referendum, Yes stickers could be seen on windows, cars and lampposts, remnants of an unusually high-profile campaign.

While the SNP played the key role, accounts of the campaign suggest that nonparty volunteers were also important, coming out in larger numbers as the referendum approached. Better Together and the pro-Union campaigners had more limited canvassing, placing more emphasis on the 'air war'(and on negative campaigning), while the Yes campaigners were more focused on the 'ground war' (Pike 2015). The unusual characteristics of this ground war – the colourful, innovative, old and new campaign methods and the networks of campaigners connected through social media – had a quality perceived by some as 'movement politics'. The Yes side appeared more organisationally fluid, creative and grassroots-based, prompting della Porta et al. (2017: 30) to refer to a referendum 'from below'. The campaign was perceived by many interviewees as having characteristics different from that of an election, 'not the same old same old' (Interview no. 66). Some described a movement at work, for example:

In the space of two years we managed to create a genuine movement for independence, not a coalition of organisations, not a combination of two political parties but a genuine social movement, which the Greens and the SNP were absolutely the core of as you'd expect. But it was a social movement and it was present in every single community. (Interview no. 74)

The established campaigners in the SNP tended to be more sceptical of the razzmatazz. A number of participants emphasised the importance of conventional data-gathering methods, one describing them as 'the steel in our campaign' and contrasting them with 'jumping and shouting'(Interview no. 13). Others suggested complementarity of old and new campaign methods. Where there appears to have been a significant and agreed difference with previous campaigns was that canvassing over this long period was more than the usual identifying support to GOTV but involved attempts to convert electors. One study participant (Interview no. 14) described it as 'more than a campaign':

People loved coming out and knocking on doors as this was about speaking to people, converting them and not just getting data. In 2013 it had been difficult to get people to engage but that changed in 2014. In 2014 people flooded the street stalls – people were very interested. People who were not activists were involved – we put out a leaflet and saw that a lot of people had put the leaflet in their windows on polling day.

As the referendum approached, the polls began to narrow. What had looked unachievable to independence campaigners suddenly looked possible (also to UK party leaders who rushed to commit to strengthening devolution). In the event, a clear majority (55%) voted to remain in the United Kingdom, but concentrations of support for Yes existed, notably in 4 of the 32 council areas – Dundee, West Dunbartonshire, Glasgow and North Lanarkshire. The turnout on 18 September was the highest recorded in a Scottish election or referendum since the establishment of universal suffrage (though not as high as the 94% turnout in the 1995 Quebec referendum).

As we have discussed, other referendums have generated interest, notably 1979 and 1997, but key indicators of political engagement suggest the 2014 referendum campaign was exceptional, and Yes voters were more participatory than No voters (Henderson et al. 2022: 32). There were, of course, societal conflicts, with different perceptions and experiences among Yes and No voters. For many observers and participants, though, this was a remarkable period, Tierney (2015: 226) describing

the independence referendum as 'an exercise in national public engagement' which enhanced the reputation of referendums more widely.

Those who had been active faced an abrupt cessation of their campaigning. In these circumstances, we might expect widespread disappointment and deflation. However, significant numbers gravitated towards independence-supporting parties. When the result was clear, Prime Minister David Cameron made a statement at the door of No. 10 Downing Street in which he outlined plans for English Votes for English Laws and the Smith Commission to work on strengthening devolution. This statement struck the wrong tone for many, contradicting the 'family of nations' narrative of the campaign, and it spurred on some of the Yes movement activists. Lynch (2017: 77–79) states: 'The combined effect of the referendum, the long mobilization of Yes and Cameron's support for English nationalism had a major effect on the SNP and the Greens'. That major effect was a dramatic and surprising upturn in party membership. This theme was reinforced by a number of our interviewees, for example:

David Cameron's speech – EVEL – reminded us why we'd been campaigning – he thought Scotland was back in its box. We remembered we'd been campaigning for fairness. And the way to get this was by joining a party. (Interview no. 13)

A factor that may have been relevant was the expectations of campaigners. At the start and for much of the campaign, support for independence languished well behind support for the Union. Dennis Canavan told the SNP conference in March 2013, 'I like climbing mountains'. This became a refrain of the campaign, acknowledging the challenge of the Yes movement. Matalin and Carville (1995: 142–143), US campaign strategists, argued that politics was an 'expectations game' in which success is 'not measured by actual results, but preconceived expectations'. While Yes activists were enthused and worked hard, defeat came as no surprise. Better Together activists were more relieved than joyous; they were deflated that their victory had not been as decisive as had appeared throughout much of the campaign. There was a sense that the Yes movement had come a long way and that the issue remained alive.

#### **Membership surges in the SNP and Scottish Greens**

For most of the SNP's existence, party membership estimates were unreliable. Traditionally, the party was decentralised with an emphasis on branch power, and one of the powers of local branches was autonomy over party membership. Members were recruited locally, and branches had control over the money raised by membership fees. Branches made an ill-defined contribution to the party centrally. Party headquarters had little knowledge of membership numbers because 'branches were very poor at sending records of members to HQ'(Wilson 2009: 4). It is remarkably difficult to trace the trajectory of the SNP's membership since its foundation in the interwar period.

#### 28 *From referendum to party membership*

Some accounts of the SNP in the 1960s suggest membership reached 120,000 following Winnie Ewing's 1967 by-election victory in Hamilton (Mansbach 1973: 185). Party sources now indicate that these estimates were inflated and that membership was more likely to have peaked at between 60,000 and 70,000 in the 1970s, declining to around 15,000 by the end of that decade (Mitchell et al. 2012: 23). Following the 1979 referendum and the SNP's general election defeat in the same year, Gordon Wilson, SNP leader in the 1980s, claims that membership rose from 14,087 in 1979 to 14,972 in 1980, a modest increase, and that membership dipped to 12,617 by 1981 (Wilson 2009: 206). Further decline occurred in the 1980s as the party argued internally over its ideological identity and approach to devolution. In the 1990s, support for the SNP began to improve as the party adopted a pragmatic approach to devolution, but membership wasn't significantly impacted in this period, with little sign that recruitment was boosted by the 1997 referendum. It seems that before 1999 the SNP experienced ebbs and flows in membership, which largely reflected the party's electoral standing, but details were derived mainly from personal accounts and are difficult to verify.

In the early 2000s the SNP claimed to have around 10,000 members (9,450 in 2003). The SNP was then a party of official opposition in the new Holyrood Parliament, transitioning into a party with the potential to govern. A critical development in the history of SNP membership occurred in 2004. Party reorganisation involved creation of a central party record of membership, a national register. From this point we can more reliably document the SNP membership. Figure 2.1 tracks membership rolls for both the SNP and Scottish Greens over the past two decades.<sup>3</sup> It shows that from 2004 the SNP gradually attracted more members, boosted by success in the 2007 and 2011 Scottish Parliament elections. Membership doubled to 20,000 between 2003 and 2011, and it continued to rise to 25,000 by the end of 2013. This was a long-term improving picture connected to the party's electoral and governing profile.

When the prospect of a referendum became clear following the 2011 election, the SNP had been in government for more than six years and had a reputation as a successful recruiter of members compared to other parties. In 2012, the SNP declared an aim of boosting membership to 36,000 by the time of the referendum and developed new membership recruitment materials, including a Member Handbook (SNP 2013: 30). With the referendum approaching, a membership of just over 25,000 was well short of the declared target but seen at the time as highly respectable.

In contrast, the Scottish Greens had a poor track record on membership. Green parties began to form across the world in the 1970s, and some were making an impact on political systems a decade or so later. Scottish Ecologists had formed in 1978, then part of the British Ecology Party, which faced almost insurmountable hurdles to political influence. In 1985, the party changed its name to the British Green Party and then benefited from a wave of interest in environmental issues at the end of the 1980s. In 1989, the party attracted 15% of the vote in a European election (at the time, the best Green electoral performance ever). While the voting system did not reward the Greens with seats, they gained from the publicity, and membership reached 20,000 in 1990 (Rüdig et al. 1996). Within a British-wide organisation, Greens in Scotland were part of this green wave, but the effect on membership was stronger in England.

The Scottish Greens amicably split from the UK Greens in 1990. There were approximately 1,200 Scottish Green members. Three years later, the party had lost 1,000 members (Bennie 2004: 25). The party existed as a tiny marginal force in Scottish politics until devolution threw it a lifeline. Robin Harper entered the Scottish Parliament in 1999, and by 2003 Green representation had reached the heights of seven MSPs. Membership numbers responded somewhat, rising from 350 in 1999 to 900 by 2005, but the scale of membership disappointed many in the party. One Green interviewee claimed that the party had always 'underperformed' in membership terms (Interview no. 61). At the end of 2013, there were 1,178 members, and there was no reason to believe that the referendum represented a recruitment opportunity. In August 2014, the Scottish Greens reported a membership of 1,500.

Neither party expected a dramatic upsurge in membership. Key to understanding events is the nature of the long referendum campaign during which unprecedented levels of political activism occurred but with a very modest increase in party membership. At no time *during* the campaign did it appear that the recruitment of new party members formed part of the action repertoires of the Yes movement. The referendum brought about a newer and broader movement towards independence, but this was not a movement that pursued its goal through party membership. Following the referendum result, however, membership surged dramatically, as can be seen in Figure 2.1. A key characteristic of the surge is that it occurred postreferendum and among parties on the losing side.

*Figure 2.1* Membership of the SNP and Scottish Greens, 2003–2023

#### 30 *From referendum to party membership*

Almost as soon as the outcome of the referendum was declared, new members began joining the parties in large numbers, and thereafter the figures began to snowball, with thousands of members joining per day, appearing to buck the trend of declining membership elsewhere. In the month following the referendum, membership of the SNP increased to 80,000. Peter Murrell, the party's chief executive, posted numbers on Twitter multiple times in a day, generating a sense of excitement and momentum. By the end of 2014, SNP membership had risen to 94,000 (multiplying, roughly, by four). By March 2015, the landmark figure of 100,000 was reached, and the 2015 general election boosted numbers further; 2,000 joined in the 24 hours following Sturgeon's participation in a televised UK leaders' debate. One year on from the referendum, membership stood at 115,000, a 460% increase on pre-referendum figures. At approximately 3.0% of the Scottish electorate (1 in 33), the SNP's membership arguably made it the kind of mass-membership party not seen in the United Kingdom for decades.

 In the four weeks that followed the referendum, membership of the Scottish Greens quadrupled (rising to more than 6,000). By the end of 2014 membership was nearly 8,000. By May 2015 the total had passed 9,000, representing an increase of over 600% on August 2014. A peak of 9,195 was reached in July 2015. Thus, the scale of the SNP surge was greater, and it attracted more attention, but proportionally the Greens' increase was the larger – starting from a much smaller base – and for a party with so few members before the referendum, the surge had more transformative potential.

In a few short months the SNP and Scottish Greens had become parties made up predominantly of new recruits. Exactly why these events occurred was unclear. Being on the losing side in elections typically supresses political efficacy and willingness to participate (Craig et al. 2006). In this case, losers seemed to be behaving like winners. In simple terms, the dramatic increase in party membership was provoked by an event (the referendum); it had the appearance of a bandwagon effect; and it was combined with conscious opportunism by the parties (especially SNP headquarters).

#### **The Brexit referendum and beyond**

Less than two years on from the Scottish independence vote, the United Kingdom experienced yet another referendum, on continued membership of the European Union. As with the referendum on Scottish independence, the Brexit referendum arose from extra-parliamentary pressure and a commitment to hold a referendum in the Conservatives' 2015 election manifesto. Cameron's gamble had paid off in 2014, but the result was much closer than he expected. Having prevailed in the electoral reform referendum in 2011 and the Scottish independence referendum in 2014, as well as in a general election in 2015, Cameron's luck ran out in 2016 (Glencross 2016; Sobolewska and Ford 2020).

The general election of 2015 had dramatically altered electoral allegiances in Scotland, the SNP winning 56 of Scotland's 59 seats with nearly 50% of the vote. In the 2016 referendum campaign, the nature of the debate in Scotland was different from that which played out in the rest of the United Kingdom. The SNP articulated the desire to remain in the EU (though its campaign was modest), as did the Scottish Greens and all the major parties in Scotland, and there was much less anti-immigration sentiment. The campaign itself was short, only six weeks in duration, with little of the public engagement and grassroots activity evident in 2014 (Henderson et al. 2022: 39–40). UK-wide turnout was 72%, and 67% in Scotland, significantly lower than in 2014.

Variation in voter behaviour across the United Kingdom was striking. Voters in Scotland (62%) supported Remain emphatically, with less clear majorities for Remain in London and Northern Ireland. However, support for Leave in England and Wales created an overall majority for change, feeding claims of a continuing democratic deficit in Scotland. Not all supporters of the SNP, Greens or independence were pro-EU. Academic research identified a significant group of Brexitsupporting SNP voters (Henderson and Mitchell 2018: 116).<sup>4</sup> But the apparent disregard for the wishes of Scottish voters provided further impetus for the independence cause. It was noted that a higher proportion of Scots voted for devolution in 1979 than UK voters supported Brexit.

 Our focus is the effect of these events on SNP and Scottish Green membership. As Figure 2.1 shows, the trajectories of the parties' memberships began to diverge two years on from the Scottish independence vote. Although the SNP had inevitably lost some of the 2014 surge recruits, membership continued to grow. Following the EU referendum (by August 2016) SNP membership reached 120,000. In 2018, a mini-surge occurred (reported by the party as 5,000 members in one day) following a walkout of SNP MPs from a Brexit parliamentary debate; by the end of that year, membership tipped over the 125,000 mark. For a period in 2018 the SNP membership overtook that of the UK Conservatives, making it the second largest party in the United Kingdom. At the end of 2019, SNP membership reached a peak of 125,691. This meant that 3.2% of the Scottish electorate were SNP members (1 in every 32 registered voters); in the United Kingdom, approximately 1.7% belonged to *any* party (Audickas et al. 2019: 5).

The SNP had been perceived as competent by voters, despite a mixed record in office, and membership formed part of a narrative of success. Recruitment had the appearance of unambiguous success. However, the accuracy of the SNP's reporting came into question. The party fell into a pattern of reporting waves of new members – 'X new members have joined in three days' and so on – without specifying total numbers, such as when it was claimed that over 5,000 members had joined, some rejoining, in the 24 hours following Nicola Sturgeon's appearance at the Holyrood Committee on the Handling of Harassment Complaints in March 2021. The desire to maintain positive headlines about the number of members speaks to the perceived importance conferred by a large membership, but SNP membership had begun to recede.

 In the SNP's financial accounts, the party indicated that membership had declined to 105,000 at the end of 2020, stating that financial hardship caused by the pandemic had led to some cancelled memberships, but in the same report it was claimed that numbers 'bounced back' to nearly 120,000 by May 2021 (SNP 2021: 13).

 However, the end of 2021 figure provided by the party was a puzzling 103,884 (SNP 2022: 11). The party's reporting was becoming less clear and less consistent. During the 2023 contest to replace Nicola Sturgeon as party leader, the party's headquarters was reluctant to release membership figures but admitted (on 16 March) that membership stood at only 72,186. In August 2023, the party's annual accounts indicated that membership had been 82,598 in December 2022 and was 73,936 in June 2023 (SNP 2023: 8). Taking these numbers at face value, SNP membership had declined by 30,000 over two years and by 50,000 since 2019. Even so, the party remained a much larger membership organisation than before the 2014 referendum.

Scottish Green Party membership had begun to decrease around 2016, despite some new members joining following the EU referendum. At the end of 2017, the party revealed in its financial reporting to the Electoral Commission that it had a little over 7,000 members. It referred to an 'expected fall out of part of the surge in new members following the Scottish independence referendum'(Scottish Green Party 2018: 4). By 2018 the party was no longer including membership numbers in these reports suggesting sensitivity about decline. Greens emphasise openness and transparency, and to suddenly omit these figures points to a perceived weakness internally. In the summer of 2019, the party's leadership contest revealed that there were fewer than 6,500 members, a fall of 30% from the post-referendum peak, with only 800 (12.5%) participating.

Nevertheless, the party was a bigger and more viable organisation than before the referendum. There had not been as dramatic a decline on the scale that some in the party had feared, in part because of a 'more professional' approach to managing the membership (Interview no. 67). By December 2020 numbers appeared to have stabilised at around 6,500, and events and electoral success always had the potential to boost membership. The election of a record number of Green MSPs in May 2021, the cooperation deal with the SNP and entering government, and the COP26 (United Nations Climate Change Conference) event staged in Scotland in December 2021 all had recruitment benefits, taking membership to 7,500 by the year's end. Membership then surpassed 8,500 in Spring 2022, its highest level since the post-independence referendum surge. This was interpreted by the Scottish Greens as a positive response to the party being in government. Membership, though, continued to fluctuate to a degree; it stood at 8,000 in October 2022, and 7,700 in April 2023.

#### **Membership of other parties**

The post-referendum surges in party membership were restricted to the independencesupporting parties. The small radical left Scottish Socialist Party (SSP) experienced a referendum-induced spike in membership akin to that of the SNP and Scottish Greens on a much smaller scale, with membership rising from approximately 1,500 before the referendum to 3,500 by November 2014, but other parties did not benefit from the referendum in the same way. Available data on parties opposed to independence suggest that only modest membership increases occurred around the time of the referendum.

Membership figures are not published by Scottish Labour, but media reports suggest that membership was below 13,000 at the end of 2013 and had risen to only 13,500 by November 2014, at most an increase of around 1,000 members (Hutcheon 2014). This is substantiated by the party's financial statements, which indicate that membership fee income increased only marginally in 2014, from £115,636 to £116,567 (Scottish Labour Party 2015). Membership appeared to then grow, to nearly 26,000 in January 2018, but then declined under Richard Leonard to 21,000 by early 2019 (Hutcheon 2019). In 2019, income from members fell to under £100,000, less than the corresponding income in 2014 (Scottish Labour Party 2019).

Labour peaks coincided with leader contests in Scotland and in the UK party, especially when Jeremy Corbyn was challenged for the leadership in 2016. Corbyn's leadership produced an impressive tally of over half a million Labour members across the United Kingdom – with membership numbers reaching 550,000 in 2017. The party was fond of describing itself as Europe's biggest socialist membership organisation. However, Scottish Labour benefited less from the Corbyn bounce. By February 2021, figures released on those eligible to take part in candidate selection for the Scottish Parliament election indicated that the party had fewer than 17,000 members. This may explain Scottish Labour's general reticence on its membership. In the past, full details emerged during leader contests, with the release of votes cast and voter turnout percentage. Now only the percentage of votes is reported, part of a broader trend among parties in Scotland. In the most recent Scottish Labour leadership election, Anas Sarwar won 57.6% of the membership ballot, but total membership numbers were not made public.

The Scottish Liberal Democrats usually publish membership figures in their annual accounts. Their membership remained static between the end of 2013 and 2014, appearing unresponsive to Scotland's referendum debate, but it increased by 25% to around 4,000 (from an extremely low base of under 3,000) by the end of 2015. This can be attributed to UK events. The Lib Dems (Liberal Democrats) had lost nearly a third of their members, many of whom opposed close cooperation with the Conservatives. Following the coalition, membership recovered, with many joining and rejoining around the time of the Brexit referendum and elections in 2017 and 2019. The party's 'Stop Brexit' campaign in 2019 proved better at attracting members than voters. However, the Scottish Liberal Democrats didn't benefit enormously from these surges. With fewer than 5,000 members at the end of 2020, they were a smaller party than the Scottish Greens.

In 2010, an internal Scottish Conservative report noted a long-term retrenchment in membership, from 40,000 in 1992 to 10,000 in 2010, and recommended a more proactive recruitment strategy (Scottish Conservative Party 2010). A year later, membership declined to below 9,000, which was revealed when Ruth Davidson became leader.<sup>5</sup> The Conservatives in Scotland attempted to use the referendum in 2014 to boost membership (the only party overtly to do so) asking 'Friends of the Union' to join, but to little effect. Along with other parties, the Scottish Conservative Party has been 'cagey' and 'elusive' in reporting membership statistics (Kennouche 2015). In February 2020, 6,498 Scottish Conservatives voted in a leader contest, but turnout was not published. 6

#### 34 *From referendum to party membership*

The secrecy of the parties and the resulting patchiness of the data make it impossible to compare with precision – or to extend Figure 2.1 to include other Scottish parties. Nevertheless, the broader picture is clear. The SNP is by far the largest party in Scotland based on membership and remains so despite the downturn. The surge into SNP membership also left it comparing favourably in size – at least proportional to the relevant electorate – with the UK parties. 7

As Figure 2.2 illustrates, other parties have enjoyed periods of successful recruitment at UK level.<sup>8</sup> Labour membership doubled in 2015; the Greens enjoyed a period of membership growth between 2013 and 2015; and membership of the Liberal Democrats doubled between 2015 and 2019.9 Even the Conservatives under Johnson claimed to have experienced a growth spurt. Plaid Cymru is not shown on the graph, partly because the numbers are much lower and partly because the data are too patchy to plot a trend, but it is claimed that membership increased by 25% following the election of party leader Adam Price in 2018, taking the party over the 10,000 mark. Price announced free membership to anyone joining that year.

The underlying causes of recruitment surges vary. Some cases look like a form of protest or reaction to electoral failure; some involve a more positive response to election success or new leadership; and some are accompanied by grassroots mobilisation, as with Labour's Momentum. But we should not overlook more prosaic factors related to the cost or ease of joining (a topic discussed in Chapter 5 on motivations for joining).

*Figure 2.2* Membership of UK political parties, 2011–2022

#### **Conclusion**

The membership surge among pro-independence parties following the 2014 referendum had some striking characteristics. The surge followed a long, active campaign; it was sparked by referendum defeat; it was enabled by technology; and it was exceptional in its pace and scale, especially by the standards of referendums which had taken place in British politics. The idea that a referendum might act as a trigger for party recruitment had not occurred to party specialists (or parties themselves), precisely because there was little evidence that such a relationship existed.

The closest that Figure 2.2 offers to the phenomenon examined in this book is the big surge into Labour in 2015. The Labour surge was followed by a slow unwind. Other upturns in Figure 2.2 were short-lived spikes. Party membership can surge and decline, often quite unpredictably. Until recently, the independence referendum influx looked different. Not only were the SNP and Scottish Green surges on an unprecedented scale, but they were also, especially in the case of the SNP, remarkably sustained. While headline figures say little about the numbers leaving, joining or rejoining (membership turnover), it is remarkable that, except for 2016–2017 when numbers were static, SNP membership grew every single year between 2004 and 2019. The SNP stands out as a particularly successful recruiter of members during this period.

More recently, of course, things have changed, with decline driven not only by political factors but also by external factors such as the pandemic and cost-of-living crisis. In their financial reporting, many parties, including those in our study, note the impact of these events as suppressing party membership. Nevertheless, while the SNP has lost members (and some credibility in terms of how it reports membership), it remains a successful membership organisation. The Scottish Green Party, meanwhile, has experienced greater membership fluctuation and is much smaller than the SNP but has succeeded in establishing a membership base which would have been unimaginable to the party before 2014. Such an expansion had potential to transform the parties demographically and organisationally, themes we return to later in the book. Next, though, we consider how these events are related to the interaction between political parties and social movements.

#### **Notes**


#### 36 *From referendum to party membership*


#### **References**


Glencross, A. (2016) *Why the UK Voted for Brexit*, London: Palgrave Macmillan.


Scottish National Party (2023) Statement of Accounts 2022. Available at www.electoralcommission.org.uk.

Sobolewska, M. and R. Ford (2020) *Brexit Land*, Cambridge: Cambridge University Press.

Tierney, S. (2015) 'Reclaiming Politics: Popular Democracy in Britain after the Scottish Referendum', *Political Quarterly*, 86 (2): 226–233.

Wilson, G. (2009) *SNP: The Turbulent Years, 1960–1990*, Stirling: Scots Independent.

### **3 Parties, movements and the 2014 referendum**

#### **Introduction**

Research on parties and social movements tends to exist in intellectual silos. Two leading social movement scholars described how those working in 'cognate areas of political contention' did so in 'cordial indifference to one another' and that 'transitions from one form of contention to another – the *dynamics* of contention – were left in the hollows between these oddly segregated scholarly specialities' (McAdam and Tarrow 2010: 529). They expressed most disappointment at the lack of interaction between movement and election studies, noting that social movement research focused on 'disruptive forms of contention' (Ibid.: 532).

In this chapter we focus on the relationships between parties and movements to explain developments in party membership in the SNP and Scottish Green Party. Analysing the independence referendum campaign and the subsequent rise in membership of the SNP and Scottish Greens requires conceptual tools and approaches drawn from complementary scholarship on social movements, elections and parties. The 'transition' from movement to party that followed the referendum offers insight into the motivations and values of those involved.

Before turning to our specific case, some definitions are required, noting conceptual similarities and contrasts between social movements and political parties. The chapter then provides an overview of the two parties prior to the referendum, placing each in the wider movement of which it was a part (SNP and the Scottish national movement, Scottish Greens and the environmental movement) as well as how the parties were related to other movements. Attention then turns to the 2014 campaign, and the relationship between the two parties and the movement dynamics which played a part in the referendum. We consider the referendum as a 'movement moment'. The chapter also introduces our survey data, revealing how those who would go on to join the SNP and Scottish Greens experienced the 2014 referendum campaign, as well as their wider engagement with movement politics.

#### **Movements and parties**

There are many definitions of social movements. In the broadest sense, a social movement is a form of collective action aimed at influencing societal change. When we refer to social movements, we are not referring to organisations, as with

> DOI: 10.4324/9780429446559-3 This chapter has been made available under a CC-BY 4.0 license.

pressure groups or political parties, but, in Heberle's (1951: 6) classic description, a 'commotion, a stirring among the people, an unrest, a collective attempt to reach a visualized goal'. Tilly (1978: 8–9) noted that the study of social movements had three related dimensions: the groups/organisations involved, events that make up the movement repertoire and ideas that provide coherence. Diani (1992: 7) highlighted the key aspects of social movements as networks of interaction, shared beliefs and solidarity and collective action on conflictual issues which take place outside the traditional institutional sphere. The emphasis on social phenomena – events taking place away from formal political institutions – helps explain why movement approaches have not been widely utilised by party scholars, although relations between political organisations (groups, movements and parties) have been examined by scholars (see della Porta et al. 2017; Fraussen and Halpin 2018; Muldoon and Rye 2020).

There is a tendency to view social movements as radical or even revolutionary in their aims. The very notion of 'contentious politics' that has animated much social movement research suggests disruptive and non-conventional forms of protest and objectives. Castells (1997: 3) refers to 'purposive collective actions whose outcome, in victory as in defeat, transforms the values and institutions of society'. But as with political parties, social movements are 'complicated aggregations of diverse groups and individuals'(Johnston 2014: 8). The goals of parties and movements overlap. There are radical parties and movements that either disdain or even eschew the electoral arena and many parties and movements that see electoral politics as only one focus of activity. Social movements may engage with elections, without contesting them but seeking to influence outcomes.

The ideas and values of social movements, as with political parties, have been studied in the context of mobilisation processes. Scholars emphasise collective ideas or values and feelings of belonging – a shared identity – and how these form network communities. della Porta and Diani (1999: 85) define identity as 'the process by which social actors recognize themselves – and are recognized by others – as part of broader groupings'. Different movements overlay each other, and individuals can experience a sense of belonging to several groups and movements. Gamson (2007: 243) argues that people belong to many communities or 'solidarity groups' and 'carry around with them various collective identities'.

Movement 'membership' differs from that of a party or group in being less clearly defined. Parties recruit formal members who pay membership fees and who in return have organisational rights and benefits. As della Porta (2007: 7) states, 'Whereas parties or pressure groups have somewhat well-defined organizational boundaries, with participation normally verified by a membership card, social movements are instead composed of loose, weakly linked networks of individuals who feel part of a collective effort'. Membership of a movement takes on a different meaning from that of a formal party but still suggests a connectedness or community. Anderson's (1991: 15) idea of an imagined community in his study of nationalism recognises people who could never know everyone else in the movement but 'in the minds of each lives the image of their community'.

A key observation is the importance of what Tarrow (1998: 18) referred to as the 'political opportunity structure', the opportunity external to any group that 'can be taken advantage of even by weak or disorganized challengers'. These opportunities open up access to power. In our case, the establishment of Scottish devolution and the outcome of one election started the process that triggered the remarkable increase in membership of the SNP and Scottish Greens. The SNP's overall majority in the 2011 Holyrood elections put an independence referendum on the agenda.

Events are also important. The referendum was an important event that brought attention to the party-movement relationship. It created challenges in terms of goals (beyond a simple majority for Yes in the referendum), organisational forms, campaign repertoires and relationships between hierarchies. It also became the springboard for the subsequent rise in party membership. Thus, referendums are of special interest and, as has been observed by della Porta et al. (2017: 31), the 'intersection' between referendums, social movements and parties requires investigation.

Elections demand organisation, and leaders seek stability in party organisation, an aspect of electoral professionalism (Panebianco 1988). Movements are associated with informal 'self-organisation', decentralisation and participation and the rejection of bureaucratisation. However, a strand of research, largely initiated in the United States, on movements' mobilisation of resources and changing organisational forms has relevance to this study. This work focused on how social movements marshal resources – people, organisation, money, expertise and support (McCarthy and Zald 1977). A distinction was made between professional social movement organisations and grassroots organisations. The former refers to organisations which become increasingly complex and may have full-time professional salaried staff. The latter refers to the more spontaneous, self-organised, organic groupings. The 'Yes' campaign combined these two organisational forms in an uneasy but ultimately successful balance, with social media providing a tool for communication and mobilisation, potentially surmounting inherent biases in conventional media and resources.

 When social movements become professionalised, this can come at a cost. A movement may lose its vitality and radical repertoire of activities and dilute its ideology, ethos or goals. Piven and Cloward's (1977) classic account of poor people's movements noted the way in which leaders seeking to preserve organisations will avoid disruptive activity. The distinction between organising and mobilising is seen as important (Block 2003). This echoes the 'organisation equals oligarchy' argument in the political parties literature. Michels' (1999 [1915]) argument that mass movements require organisational form to succeed has become the classic statement favouring political parties, but he also noted that larger parties experience difficulties in retaining participatory decision-making. Leaders may be reluctant to give up power and have interests they seek to protect. From this perspective, we might anticipate that the surge in membership of the two parties would lead to less participation and more power to the leaders. On the other hand, the new membership may have expected greater involvement having had the experience of referendum activity. These are questions we address in Chapters 7 and 8.

#### 42 *Parties, movements and the 2014 referendum*

 For many scholars, including those who have studied the SNP and Greens, movements seek to change society, not just manage it, while parties are more focused on winning elections and governing. Movements are less hierarchical and may seek to affect how people live their lives involving more profound change than most parties contemplate (Brand 1992: 81). We might expect these characteristics to be most evident during an election when party leaders are more visible and efficacy demands that the party operates less anarchically. For Brand (1992: 81), writing about the SNP 30 years ago, movements and parties should be seen as 'positions on a continuum, where there is no strict separation of characteristics, but there is a higher probability that certain of them will be at one end of the spectrum'.

#### **SNP and Greens before the referendum**

The SNP and Greens grew out of their respective movements. The SNP's origins lay in a perceived failure of the existing political parties and pressure groups, with criticism especially focused on Labour, to deliver a measure of Scottish home rule in the 1920s (Finlay 1994). Many of those who played a central role were figures in the 'Scottish renaissance' cultural movement bringing energy and colour but little political sense to the campaign for a Scottish Parliament. There had been and would continue to be a variety of organisations that sought to encourage the dissemination of Scottish literature and culture, part of a wider literary and cultural national movement. There was some overlap between these cultural figures and those aiming for self-government, but this did not mean that they had shared objectives. The dominant cultural movement was at best agnostic on self-government.

In his 1978 study of the Scottish national movement, Jack Brand (1978: 89) noted that nationalist movements are often influenced by poets or novelists. While key literary figures were important in the formation of the SNP in the 1930s, the 'paths of the poets and nationalists'subsequently diverged (Ibid.: 104). It was difficult to place the SNP within a broad cultural movement. In a further study of SNP members, Brand (1992: 87) found that only 17% of members mentioned any kind of cultural association and noted that the SNP had always been 'uneasy about the link with cultural nationalism'. He concluded that SNP members saw themselves in relation to other parties 'rather than as being a part of another sort of organisation'(Ibid.: 88). This distinguished the SNP from other similar parties in Wales and Catalonia, for example, which can be seen as more emphatically based in cultural and language movements.

For much of its early history, debate within the self-government movement focused on the most appropriate institutional form and strategy to be deployed (Mitchell 1996). A key part of debate was how the national movement ought to relate, if at all, to other movements, particularly the labour movement. This was in part related to goals. Self-government or even independence might be construed in a number of ways, but for most of its history the SNP was content to leave this relatively open. Supporters of independence worked alongside supporters of some form of legislative devolution. At various periods, other organisations would become more significant than the SNP in the campaign for a Scottish Parliament. The Scottish Covenant Association, Scottish Congress, Scottish Patriots, the Campaign for a Scottish Assembly/Parliament and the Scottish Constitutional Convention were all organisations within the broad national movement in the second half of the twentieth century. The key difference with the SNP was not so much about goals but means to achieving these goals. The SNP believed that only electoral pressure on the other parties would force them to deliver self-government, and latterly it hoped to win power itself.

We have previously argued that the SNP developed into an electoral-professional party having been an amateur-activist party after the establishment of devolution when 35 SNP Members of the Scottish Parliament were elected (Mitchell et al. 2012). The additional resources, not only MSPs but attendant staff and increased media interest, contributed to this transformation but also raised questions for the party. SNP members prided themselves in being a party characterised by decentralisation, participatory democracy and collective leadership. In essence, it had characteristics of a social movement. Over the course of devolution this changed, with the party becoming centralist with a strong, even dominant, leader. Electoral success from 2007 when the SNP became the largest party in Holyrood smoothed the development of electoral-professionalism.

The SNP came to dominate the wider movement for independence due to its electoral success, but since devolution two other parties have achieved representation in Holyrood that could claim to be part of the wider national movement. The Scottish Socialist Party and Scottish Greens each support independence. The high point of Holyrood representation for non-SNP pro-independence parties was 2003 when the Greens won seven and the SSP won six seats in the 129-member Parliament. While both the SNP and Scottish Greens supported Scottish independence, there were nonetheless significant differences between the two parties ideologically and also in their organisational forms and ethos.

In the early years of devolution, the Scottish Greens included a significant number of members who opposed independence. Among them was Robin Harper, the first Scottish Green MSP. Harper joined the party in 1985 becoming convenor and secretary at a time when it had 35 members, and 5 members attended its annual general meeting (Harper 2011: 10). The absence of electoral success limited the resources available to the party. The Greens had no corporate financial support and relied on small individual donations largely from members, which ensured that its social movement type approach was as much forced upon the party as it was ideologically preferred.

A study of Scottish Green membership in 2004 found multiple and overlapping membership of environmental and other organisations, placing the Scottish Greens firmly within the broader green movement (Bennie 2004: 148–150). Half of the Scottish Green membership had been members of Greenpeace and 40% Friends of the Earth. This finding challenged assumptions that the party competed with other environmental organisations for members. The Greens incorporate a plurality of views on goals and strategies, with one senior member interviewed for this study maintaining that the party 'tolerates dissent and encourages people to speak' and had always done so (Interview no. 58), though this has been disputed by at least two of its former MSPs. If we accept Brand's (1992) conception of parties and movements, then the Greens have always been further towards the movement end of the continuum, the SNP having moved towards the party end post-devolution.

By 2014 the Scottish Greens remained a small party with only two Members of the Scottish Parliament elected in 2011 and 14 councillors (out of 1,223) elected in 2012. This limited the resources available to the party, and it relied on voluntary activists to fulfil many functions that would be performed by professional staff in other more successful parties. Like the SNP of old, the Scottish Greens took pride in the party's participatory, decentralised ethos. Greens championed community politics, decentralisation and participation. It might be expected that the relationship between how a party governs itself should be consonant with the party's philosophy, but such a philosophical underpinning is challenged the nearer a party gets to power. Independence was not as central to the Scottish Greens' concerns as it was to the SNP's concerns. The Greens originated as a party with roots in the environmental movement. Their vision of independence was consequently one that placed environmental considerations at the forefront.

Other than those elected as SNP MSPs in 2011, only the two Scottish Green MSPs supported independence. These two parties were the largest organisations, though one was much bigger than the other, committed to campaigning for independence in advance of and during the 2014 referendum. Before the referendum campaign was launched there were only a few other fringe organisations campaigning for independence. Many non-aligned individuals and new groups emerged that coalesced in 'complicated aggregations of diverse groups and individuals', to borrow Johnston's (2014: 8) definition of a social movement. Many of those who were unaffiliated joined existing or ad hoc campaign organisations allowing for overlapping membership of the multitude of bodies. The referendum revived the notion of a national movement beyond the SNP. What was unclear was whether these organisations and mobilisation would continue to exist and if so in what form after the referendum.

A number of key changes created the conditions for the dramatic rise of the independence movement. There had long been a party, groups and a movement supporting Scottish self-government and independence. The creation of the Scottish Parliament in 1999 created the possibility that the SNP might become not only the largest party but also, though this was deemed highly unlikely at the time, have an overall majority. The SNP's electoral success in 2011 triggered the independence referendum, although the party's success in elections had been explained less by its commitment to independence than perceptions that it had been competent as a minority government since first coming to office in 2007 (Johns et al. 2010).

But there was a further and immediate contextual factor. The SNP first assumed office in 2007 just as the Great Recession hit the world economy. In 2010, David Cameron was Prime Minister, and his government in London embarked on a policy of austerity, which had serious implications for the devolved bodies relying on funding from the UK Treasury. The interplay of austerity and recession with constitutional politics would be significant in the independence referendum. This backdrop alone would not explain the surge in membership but, as ever, the 'concatenation of events, policies, and trends' (Rokkan 1999: 171) came together in the referendum and its aftermath. A background of economic dislocation had been important in earlier phases including the SNP's victory in the Hamilton byelection in 1967 (Mitchell 2017: 23–56).

The idea that recent movements have roots in anti-austerity politics has been explored in the social movement literature. Research on anti-austerity protests across Europe found that some mobilisations were rooted in traditional 'old' labour movements while others were protests by radical groups using what appeared to be new, less hierarchical and more spontaneous forms of mobilisation (Peterson et al. 2015).

#### **Referendum as movement 'moment'**

The literature on constitutional change refers to 'constitutional moments' (Ackerman 1984, 1993). These are not brief transitory affairs but may last for years or decades, periods during which substantial change occurs, not necessarily in the formal constitution but in its practice. Such constitutional moments are transformative and distinct from what Ackerman (1984: 1022–1023) referred to as 'normal politics', the former being 'intermittent and irregular politics of public virtue associated with moments of constitutional creation' and the latter periods when 'factions try to manipulate the constitutional forms of political life to pursue their own narrow interests'. The extent to which such constitutional moments can be distinguished from 'normal politics' is debatable (Mitchell 2012). Change precipitated by active social and political movements from below can be a feature of 'normal politics' (Anderson 2013; Benvindo 2015).

Nevertheless, a referendum can create a 'constitutional moment' and an opportunity for social movements. Dufour and Trainsel (2014: 256) describe the movement for sovereignty in Quebec, referring to the 'alliance of convenience' between the left and nationalists that 'crystallized through the creation of a sovereignty movement that was both nationalist and social democratic'. This movement included organisations and individuals with shared 'common frames of reference (practices, values, beliefs and a common identity) united around the advocacy of a particular option – Quebec sovereignty' (Ibid.). Following a turnout of over 90% in the October 1995 referendum and a slim defeat for the sovereigntists, tensions between the conservative, social democratic and radical elements of the 'Partners for Sovereignty' became more evident.

In social movement research, as in constitutional politics, 'moments' tend to be more long-term affairs than might be understood in common parlance. There have been global moments of protest. The 'moment of 1968', for example, is seen as pivotal for social movements in many countries (Klimke and Scharloth 2008). Even '1968' was a longer affair than a single year. Social movement and constitutional moments are often linked by virtue of involving significant change, the former focused on causes, the latter on consequences. 'Moment' is a vague term used in a variety of ways, referring to different time frames. The debate on Scotland's constitutional status can be seen as a long, drawn-out moment though independence was never much in focus until 2011, apart from brief periods in the late 1960s and mid-1970s. The 2014 referendum created a unique moment when Scotland's constitutional status combined with wider debates on the nature of Scottish society and economy to dominate the political agenda. It was unique not least in the levels and nature of political engagement.

Some key considerations informed the nature of the referendum campaign and how supporters of independence mobilised support. The trigger for the referendum had been the SNP's triumph at the 2011 Holyrood elections. As the governing party, the SNP had civil service resources to draw up the lengthy white paper on independence, but it needed to organise and mobilise opinion across Scotland. The SNP leadership knew that it needed to reach beyond support for the party. It was anticipated that the press would be overwhelmingly hostile to independence, and therefore alternative 'politics from below' would be required. While the SNP had long experience of a hostile press, it had received considerable press support in the 2011 election though this was for the SNP as a party of government, not for its constitutional preference. In the event, only one (Sunday) newspaper came out in support of independence during the referendum.

A key decision taken – or forced upon the 'Yes' campaign given the circumstances – was an emphasis on the promotion of a grassroots campaign to complement that led by the SNP nationally. This would require local campaigning, often in parallel with and geographically contiguous with local SNP branches. In 2012, the SNP reported that it had 259 'accounting units' – branches and affiliated organisations – across Scotland. This network of branches would provide the base for local activity, but the SNP leadership realised that this would be insufficient. The Greens were a much smaller party with members concentrated in a few locations, and their network was much more limited. Party activists combined with those from organisations supporting 'Yes', including some that were geographically focused and also those that were functionally based including Farmers for Yes (see Chapter 2). The campaign materials used by the various registered groups were fairly similar though the issue emphasis varied.

The duration of the campaign worked to the advantage of Yes Scotland, permitting it to develop a large base of activists. While SNP activists initiated most local 'Yes' campaigns, and most activists, especially initially, were from the SNP, there was a conscious effort at local as well as national level to avoid it being seen as an SNP campaign. In many instances, the local chair was chosen for having no public affiliation to the SNP. More activists emerged in the closing months of the campaign, and our interviews suggest that these were often people unknown to the parties. This influx of campaigners brought energy although many new activists were inexperienced.

Johnston (2014: 7) noted the 'common error'of mistaking the 'organisations of a movement for the movement itself'. This became evident during the long independence referendum. The campaign was wider than the range of groups involved and brought many people into the campaign who had previously not been involved in any organisations. There were people who played little or no part in local Yes Scotland or other activities but who may have campaigned individually and then sought an institutional home after the referendum. Such people may have seen themselves as part of the broad movement and been active on social media, speaking to family and friends about the issues with limited or no links to any organisation.

#### **Party members' political participation**

Our survey of SNP and Scottish Green members conducted in 2016–2017 allows us to draw a picture of the members' backgrounds of involvement in organisations. Table 3.1 presents data on the survey respondents' membership of non-party voluntary organisations, comparing those who had joined the parties pre- and postreferendum. As previous research has suggested, members of political parties are often members of other organisations, especially those within the wider movement. Consistent with this, Table 3.1. indicates that the Scottish Greens had a significantly larger proportion of members who were involved in 'cause' campaign groups, even accounting for class and education differences, although the proportion among surge joiners was smaller, suggesting that the party post-referendum was reaching out beyond its traditional and relatively active base. There were few similar outlets for SNP members who might have wanted to be active in the national movement, especially noting the earlier point about the relative lack of overlap between cultural and political movements. There was less difference between the parties when it comes to involvement in charities, professional bodies and trade unions.


*Table 3.1* Involvement in non-party group activity by party and cohort

#### 48 *Parties, movements and the 2014 referendum*


Table 3.2 moves from group-based to other forms of political activity. Reported levels of activism are high, perhaps implausibly so (although this may be as much about response bias – the inactive being less likely to respond to our survey in the first place – as about social desirability bias leading members to exaggerate their levels of activity). The key point is that reported levels of activity are similar among surge joiners, who appear 'more of the same' and not a cohort tempted into political activity for the first time by the referendum. But there remains an appreciable gap between the two parties, with Scottish Greens more likely to have engaged in the activities in the table – including previous party membership.

 The table shows that one in six SNP and as many as one in three Scottish Green respondents had previously been members of another party. A higher proportion of Green members (8%) had been members of the SNP than vice versa (2%). Most commonly in both cases they had moved over from the Labour Party. Approximately half of all former party members had belonged to Labour, 10% and 14% of all SNP and Scottish Green respondents, respectively.<sup>1</sup> It is striking, however, that the referendum didn't especially persuade former members to rejoin. Of those who joined following the referendum, a tiny proportion (2%) report being 'returners'. Significantly more had been members of a different political party in the past.2

We now turn to the members' participation in the referendum campaign itself. Table 3.3 reports on a vast range of potential activities, reflecting the intense and varied nature of campaigning and activism. Elsewhere, we have reported on these activities clustered into a few categories (Bennie et al. 2021). Here we present the full range of the data. Once again, the levels of activity reported are very high, but the fact that some very high-effort or unusual activities – like delivering speeches or writing on wish trees – record far lower percentages reassures that respondents were not simply ticking every box.


*Table 3.3* Activities during the referendum campaign by party and cohort

Here the partisan gap is reversed: SNP members were likelier to have done most of the activities in the table, reflecting the close relationship between the SNP and Yes campaign, although the gap is often small. Cohort comparisons reveal that those who would go on to join the parties were just as likely as existing members to have discussed the referendum (online or offline), but they were less likely to have been involved in the more active types of campaigning such as attending meetings and canvassing. The surge members were connected to the campaign through informal activities, while existing members participated more. We would expect this to be the case because during the campaign the surge members weren't yet connected to a party infrastructure. However, those who joined the parties postreferendum were more involved in the campaign than Yes voters; for example, they were much more likely than Yes voters to have displayed a window poster during the campaign (Bennie et al. 2021: 1190). This suggests significant engagement among surgers and a feeling of activity that needed an outlet post-referendum.

Table 3.3 includes data on the members' involvement with some of the more prominent campaigning groups who were registered with Yes Scotland. Involvement was very broadly and loosely defined to include donating or simply following the organisations online as well as active participation. Unsurprisingly, the official campaign group was widely followed and supported, especially among SNP members. Involvement in the other movement-style groups was less common, but connecting with these groups was clearly a feature of the referendum experience for many, even more so for the Greens and just as much so for the post-referendum joiners in each party.

The scale and breadth of the activities shown in Table 3.3 reinforce the impression, formed by even seasoned campaigners and observers, that the referendum campaign was different from previous campaigns. The extraordinary turnout, including with higher registration, was the ultimate demonstration of these levels of public engagement. Moreover, compared to an election in which campaigning is monopolised by parties, the Yes campaign was somewhat different, and there are some grounds for referring to it as a movement given its disparate, decentralised and participatory nature.

This operated alongside a more traditional SNP campaign, run and fronted by the party's leadership. We should not overstate the movement-like flavour of the Yes campaign or the referendum as a whole. The print and especially broadcast media focused on the SNP leadership. Televised debates between First Minister Alex Salmond and Better Together chair Alistair Darling attracted large audiences. Over 900,000 people tuned in to the Scottish Television debate on 5 August 2014 and over 840,000 to a BBC debate on 25 August. This was four times the number of viewers for the Scottish party leaders' debate during the 2010 general election and twice as many as those who watched the UK party leaders' debate in that election (Mitchell 2016: 95). It would be wrong to see the 'Yes' campaign as entirely or even dominated by grassroots activity. The SNP leadership was much more visible than any other group in the campaign.

What emerged was a grassroots campaign combined with the policy of independence largely defined and framed by the SNP. As one senior non-SNP figure stated, the SNP had provided the mandate for a referendum and would initially form the government of an independent Scotland, so it was therefore appropriate that the SNP should determine the currency to be adopted though that could change with the election of a new party to government (Interview no. 59). This did not mean that the 'Yes' campaign accepted all SNP policies but acknowledged the leading role of the SNP. The Scottish Greens placed greater emphasis on environmental matters and supported a separate Scottish currency. It was simply not possible to have the kind of message discipline that has marked SNP campaigns in recent years. If we consider the campaign along a continuum, it was much more of a movement-style campaign – less hierarchical and more participatory – than a normal election campaign, but it was still dominated by the SNP.

The existence of grassroots-style activity and SNP leadership during the campaign was widely acknowledged by our survey respondents, as can be seen in Table 3.4. The campaign experience was clearly a positive one for large majorities in both parties. There was broad and strong agreement that the campaign felt bottom-up, and 'more like a movement than a party'. At the same time, the important role played by the SNP is recognised. And the way the membership, including new members joining after the referendum, experienced the campaign suggests some small differences between the two parties. The SNP membership were more inclined to think that the campaign gave women the opportunity to be involved. Scottish Green members were somewhat more likely to see the campaign positively, as more exciting and less disappointing. This may reflect Greens' greater appreciation of the process over the outcome, with the party's ideas given a platform during the referendum campaign.

As an organisation, Yes Scotland folded immediately after the referendum was over. Some of the groups registered with the Electoral Commission continued to exist as did some local groups, and indeed others were resurrected with the hope


*Table 3.4* Perceptions of the referendum campaign by party and cohort

of a second referendum. But in the immediate aftermath of the 2014 referendum, a body of activists found themselves without an institutional home, and some sought refuge in either the SNP or Scottish Greens.

#### **Members and movements**

Our surveys in 2016–2017 probed those who had joined the parties about their movement identities and participation (Table 3.5). When asked which movements they belonged to, the vast majority of SNP members saw themselves as part of the independence movement though 'movement' in this case may simply be seen as synonymous with being in the SNP. Less than half of the Scottish Green members identified as part of the independence movement, and notably only 54% identified as part of the environmental movement. There was little difference between the two parties in identifying with the anti-nuclear movement, reflecting the SNP's decades-long opposition to nuclear power and weapons. While a quarter of Greens identified as part of the feminist movement, only 13% of the SNP members did so. In both cases, the overwhelming majority of these were female – even among Greens, only 9% of the male respondents identified as part of the feminist *movement* (although 15% identified as a feminist when asked about their ideological identities – see Chapter 6). It may be that some of the members identified with other movements and causes not listed in our question, such as campaigns for social justice. Some of these attitudes are explored in Chapter 6.


*Table 3.5* Movement identities and the efficacy of activities by party and cohort


When it comes to the efficacy of campaigning, we see that the SNP members appear markedly more convinced by conventional politics such as party membership and voting, and Scottish Greens tend to be more positive about the impact of campaigns to change personal behaviour, including lifestyles and making ethical choices, more movement-like activities. These data suggest that few of the Green respondents viewed party membership as a positive force for change, whether the members had joined pre- or post-referendum. The new cohort of SNP members were more sceptical than established SNP members about party-related activities like membership, donating to and being active within a party, which might be explained by a lack of integration into party life.

#### **Conclusion**

Given the referendum campaign was quite different from an election in terms of activities, communication and organisation, with elements borrowed from the social movement tool kit, we require concepts and ideas drawn from a wider literature than familiar political parties approaches to understand this unusual phenomenon. This does not mean dispensing with party politics, given the dominant role played by the SNP, but recognising the 'transitions from one form of contention to another – the *dynamics* of contention' (McAdam and Tarrow 2010: 529).

Kriesi (2015) outlines three main ways of analysing party-movement relations. The first sees parties as part of the political context in which movements operate, with some parties more open to movement demands and more likely to form alliances with movements. Parties are viewed as part of a political opportunity structure. The second approach understands parties as emerging from and being located within movements, seeing them as social movement organisations. This is the idea that social movements *encompass* individuals, groups and organisations. Political parties, from this perspective, can be viewed as constituent parts of a movement. Third, movements can be studied in terms of their potential to shape or reconfigure party systems or to transform individual parties.

Each of these is relevant in our case. The SNP's 2011 election victory provided the context for an independence referendum, which in turn created opportunities for a wider Yes movement. Traditionally, both the SNP and Scottish Greens have been viewed as integral parts of wider movements, and together and with others they formed the pro-independence movement. Finally, these events had the potential to transform the parties involved. There is no doubt that the referendum proved a key moment for the membership of the two parties. That the membership surge occurred after the referendum and after defeat is notable. The referendum was the catalyst for an increase in membership despite or because independence was defeated following the most intense and longest political campaign in modern Scottish politics.

The referendum exhibited evidence of the kinds of activities associated with social movements, but this needs to be put into context. The SNP and the usual party activities were still dominant. Leadership in terms of strategy, ideas and messaging came from the SNP. While different emphases were evident among Green politicians and activists, as indeed from some within the SNP, there was remarkable willingness to allow the SNP leadership to direct the campaign. But what was happening in communities was less under the control of the SNP leadership. The campaign is best understood as two complementary campaigns that managed to operate relatively harmoniously, but the SNP had the lead role.

As the campaign progressed, its grassroots nature became part of the narrative. This makes it difficult to distinguish between the myth of a social movement campaign and its reality. It is easy to find evidence of social movement campaigning but, whether this was most effective or not, it tended to be given a high profile, portraying the campaign as broad, highly democratic and rooted in Scotland's diverse communities. It was in the interest of the SNP leadership to emphasise – even exaggerate – the breadth of support for Yes to avoid the campaign being seen as one dominated by the party. The kind of professional campaigning that the SNP had exhibited in elections, and which had proved very effective, was less visible during the referendum campaign but remained important.

#### **Notes**


#### **References**

Ackerman, B. (1984) 'The Storrs Lectures: Discovering the Constitution', *Yale Law Journal*, 93: 1013–1072.


### **4 Who joined the SNP and Scottish Greens?**

#### **Introduction**

Politics is beset by a lack of diversity in the backgrounds of those who participate. The politically engaged and active are often the well-educated and well-resourced. From community-level volunteering to protest and membership of interest groups, studies have consistently revealed a resource bias (Barnes and Kaase 1979; Parry et al. 1992; Verba et al. 1995). This pattern is pronounced in the United States but observed in other democratic countries too. And the rise of digital forms of engagement has done little to narrow the 'participation gap' between the most and least affluent; if anything, the well-resourced are more able to take advantage of new opportunities to participate, resulting in 'unequal voice' (Dalton 2017: 17). Research on British politics points to worsening participatory inequality, with a decline in working-class electoral participation over two decades (Evans and Tilley 2017).

The problem is no less acute when it comes to membership of political parties, highlighted by a wealth of empirical evidence (Widfeldt 1999; Gallagher and Marsh 2002; Heidar and Wauters 2019). There may be differences between ideological types of party – in Britain, the Labour Party attracts more working-class members than the Conservative Party does – but, as a collective, party members are likely to be well-educated, middle-class, middle-aged, white and male, characteristics which are associated with high levels of political efficacy. Van Haute and Gauja's (2015: 194) volume on party membership in Europe, which included survey evidence on 57 political parties, concluded:

Party members are remarkably similar in their social characteristics. A typical party member, irrespective of the country in which they reside, or the party to which they belong, is an older male who is more highly educated and better off financially than the general population.

Our previous research showed that Green members were younger and more middle class than SNP members, but neither party challenged the general picture. An important question for this chapter is whether a sudden and unexpected rise in membership changed the parties' profiles. We begin with a brief theoretical

> DOI: 10.4324/9780429446559-4 This chapter has been made available under a CC-BY 4.0 license.

account of political representation, before outlining the perspectives of our interviewees on how the surge changed the parties. Then we turn to our survey data, addressing whether, according to this evidence, the referendum encouraged underrepresented groups to join the SNP and Scottish Greens. The three key comparisons outlined earlier in the book will guide the analysis: a comparison of the two parties, a comparison of pre- and post-referendum memberships and a comparison of the parties' members over time. The chapter will provide evidence of the key socio-demographic characteristics of the members, including gender, age and class, and it will examine the levels of congruence between the parties' members and voters.

#### **Representation and political parties**

Aregularly cited theoretical account of representation is that of Hanna Pitkin (1967). This conceptualised four forms of representation: (1) formalistic (the rules and mechanisms guiding elections and the selection of representatives), (2) descriptive (the extent to which the social characteristics and experiences of representatives reflect those they serve), (3) substantive (representatives acting in the interests of those being represented) and (4) symbolic (when the represented perceive that they are being well represented). These interact, such as when descriptive representation enhances substantive representation (Phillips 1995; Dovi 2002; Celis et al. 2008; Childs 2008). In the context of women's representation, for example, it is argued that when representatives have similar experiences and characteristics to those they represent, they are more likely to prioritise and act on these interests when shaping policy, although the relationship between descriptive and substantive representation involves many mediating factors (Mansbridge 2003; Childs and Lovenduski 2013).

These questions are relevant to party membership because parties are representative institutions, and those who join influence the way parties make decisions and the policy they devise, sometimes in government. Party members have a say in the selection of party personnel, including candidates and leaders, and they help determine the direction of party policy. For these reasons the social make-up of members has consequences for representation and for democratic legitimacy (Kittilson 2013; Heidar and Wauters 2019).

While we would not expect party members to reflect the wider population precisely, we might expect members to perform a basic inclusive function of representation. We might assume party members would have something in common with their party's voters in policy terms (a substantive form of representation). In other words, a degree of member-voter congruence is expected. If the gap or disconnect between parties and voters is too large, it has the potential to drive down confidence in parties and democracy (Allen 2018; Weinberg 2020). This requires awareness of who become members of a party, not just who become the party's elected representatives. The less representative a party's membership of its electorate, the more difficult for a party to provide political linkage, to connect with large, diverse parts of the population (Widfeldt 1999).

Empirical studies consistently indicate that political parties suffer from a lack of descriptive representation. What is less clear is how this is influenced by fluctuation in the size of a party's membership. The expectation might be that declining memberships would make parties even less representative of voters (Young 2013: 77: Heidar and Wauters 2019: 2). However, the evidence is mixed. Widfeldt's (1999: 222) study of representative linkage in Sweden in 1960–1994 found a familiar pattern of decline in the number and activity of members, but there was 'little evidence of a decline in social representativeness' with the exception that parties were appealing less to young people. Scarrow and Gezgor (2010) confirmed the ageing of European parties but found that members in the 2000s were *more* like voters in other key respects (income, gender, education, religiosity and trade union membership) than members a decade earlier. Scarrow and Gezgor (2010: 839) concluded:

Party memberships may be shrinking, but at least so far this has not meant that parties' grassroots are becoming some kind of odd subculture, no longer able to provide legitimacy because they are too different from the rest of society.

Our focus is on how a sudden and unexpected *increase* in members of a party influences the composition of its membership. Bale et al. (2020) found little to connect rising memberships with greater member-voter congruence, either descriptively (social backgrounds) or substantively (policy preferences). Nevertheless, the abrupt and massive expansion of the SNP and Scottish Greens was so large that it had the potential to reshape the memberships. The pro-independence parties which emerged following the 2014 referendum contained new recruits who far outnumbered the existing members (by a ratio of at least 5 to 1). This suggested to observers at the time that change in the memberships was possible, even likely, with a larger presence of previously under-represented groups.

Those interviewed in the SNP and Scottish Greens describe an initial uncertainty about the backgrounds of the people joining. Some had expected that new members would be more socially diverse and representative of the population at large and that they would 'look more like Scotland', as one SNP figure put it (Interview no. 7). Another referred to the presence of 'an SNP member in every street in Scotland' (Interview no. 5). Some were concerned that the new members might be different and transform the parties in unpredictable ways. A Scottish Green interviewee recounted: 'What if thousands of people who have joined our party aren't actually what we understand Greens to be, because they're now 80% of the party?' (Interview no. 74).

Many interviewees spoke of broad change in the memberships, suggesting increased diversity. A Scottish Green interviewee claimed the party attracted 'a much broader section of people' (Interview no. 72). The SNP interviewees tended to mention the recruitment of more women and young people. Interviewee no. 31 stated: 'There's definitely more women than when I joined. Hardly any women involved back then. A lot of younger people came through from the referendum – the young people were the energy'. The Greens, too, highlighted the recruitment of young people, but many claimed that the party was successful at attracting women before the 2014 referendum. The Greens also pointed to a rise in LGBT+ members. Some suggested the party attracted more working-class people post-referendum and that there was a more even geographical spread of members from across Scotland. One discussed an increased presence and voice of women and (some) minority groups in these terms:

We have much more representation of people with protected characteristics – in the Equalities Act – transgender, LGBT, particularly people with disabilities. They have much greater membership and interest and voice in the party. I think we've probably increased the number of women – the Women's Network is strong – much bigger than it was; more active than it was. Where we haven't scored is black and Asian ethnic minorities. (Interview no. 60)

 While many subscribed to the idea that the surge enhanced diversity, there was also recognition of persistent 'challenges of diversity'. Many of the Greens interviewed were forthright about limited progress on ethnic and class diversity, one referring to 'huge challenges for the Greens' (Interview no. 69). Interviewee no. 66 stated: 'We're still pretty white, pretty middle class, and very well-educated', and they referred to the organisational dominance of the central belt Greens: 'At the national level, the SGP is very dominated by Edinburgh and Glasgow . . . and the rural membership and the rural branches feel forgotten about'. The branch structure of the Greens expanded on the back of the surge, allowing the party to claim it was a 'national party' (see Chapter 8), but some of those we spoke to perceived a lack of members beyond the central belt as an unresolved problem.

The SNP interviewees were much less likely to highlight problems of diversity among their members. However, as the larger party with more members and activists throughout Scotland, many noted the difficulty of assessing change in the wider membership, suggesting that change might have occurred among activists who are visible, with more women and young people coming forward in some branches. Interviewee no. 55 commented: 'That's difficult to say. The thing is even if you only have a handful more women, a handful more young people active at the local level, that feels like a qualitative difference'. Thus, perceptions of change were influenced by what interviewees were observing in branches and at party events, which might not reflect the wider memberships. A Scottish Green interviewee noted: 'Because of the sheer numbers, there are more women, more LGBT people. To be honest, I think it is just numbers and not proportion' (Interview no. 68).

Some participants robustly questioned the idea that the fundamental character of the memberships had been altered. An SNP interviewee described the new members as 'a cross-section, not drastically different from existing members' (Interview no. 22). Some described a process of moving from not knowing who the new members were to observing meetings and conferences and seeing that the surge members were much like the old ones: 'They looked like everyone else. They sounded like everyone else other than that they'd never been to a party conference' (Interview no. 49). A Green interviewee describes a realisation that this was 'just

a bigger version of the same party' (Interview no. 74). Another spoke of a sense of 'familiarity': 'A lot of them looked and felt and sounded really, really familiar in terms of the kinds of people they were, the kinds of things they said. Even jobs that they were doing' (Interview no. 75).

Our interview data revealed varying perspectives on the backgrounds of the new members and whether the memberships were reshaped by the surge, although the dominant view was that the parties had attracted more but not necessarily different types of members. To profile the memberships, we now turn to our survey data. As explained in the book's Appendix, we use data weighted by time of joining to obtain more accurate profiles of the full party memberships.

#### **The demographics of party membership**

Traditionally, women have been significantly under-represented among the rank and file of political parties. International (mainly European) research has suggested that women make up around a third of those who join parties (van Haute and Gauja 2015). Of the different party families, left of centre, social democratic and green parties appear better at persuading women to join (Kittilson 2011). These parties are more likely to adopt feminised party structures or internal party mechanisms to ensure equal representation. This sometimes translates into success in attracting women as members. Gauja and Jackson (2016: 372) found that exactly 50% of Australian Greens were females. Regionalist and nationalist parties, on the other hand, have been less good at attracting women as members (and voters) and less open to making internal organisational choices that ensure equal representation of women internally (Mazzolini and Mueller 2017). Chapter 8 considers candidate selection in the SNP and Scottish Greens, but in the early years of devolution neither party had strong measures to achieve gender equality among MSPs, unlike Scottish Labour.

Very few parties achieve actual gender equality in their memberships. The first major study of the Conservatives in the early 1990s identified gender parity among members – well ahead of Labour at the time (Whiteley et al. 1994: 50; Seyd and Whiteley 1992: 39). Decades later Bale et al. (2020: 33) found that 39% of all party members in Britain were women. The Labour Party with Corbyn as leader had improved its ability to recruit women, as had the Green Party (of England and Wales); in 2017, women made up 48% of these parties' memberships. However, according to the same study, 38% of Liberal Democrats, less than a third of Conservative members and only a quarter of UK Independence Party members were female. This suggests no secular improvement in women's representation within British political parties.

Turning to the parties of interest here, our previous SNP survey in 2007–2008 estimated the proportion of women at only around a third (32%). There may have been specific reasons for this: men were more likely than women to support Scottish independence and were more sympathetic to then leader Alex Salmond (Johns et al. 2012). As for the Scottish Greens, previous estimates of the proportion of women in this party range between 45% in 1990 and 37% in 2002 (Bennie 2004). Thus, the Scottish Green membership was closer to gender equality but not all that close. Comparing these two parties leads to the same conclusions as from a broader comparison: there are structural barriers which keep the overall average of women's party membership some way below 50%, and then there are partyspecific political factors which generate variation around the average.

These earlier survey statistics are provided by way of comparison in Table 4.1, but the main purpose is to present the results of our 2016–2017 survey and to see whether the 2014 referendum modified the gender profiles of the parties. The surge had the potential to increase the number of women and to produce a gender rebalancing, a closing of existing gender gaps, among the parties' memberships. This would imply that the parties were more successful at recruiting women relative to men, which would be consistent with the narrative of quite a few of our interviewees. One SNP elected representative talked of the 'feminisation' of the party membership (Interview no. 36). A Green interviewee referred to the party attracting 'particularly young women'(Interview no. 61). While this may have been their experience, the survey evidence allows us to consider whether this applied across the parties as a whole.

Table 4.1 shows that, in both parties, the proportion of women among the postreferendum joiners was 9 percentage points larger than among those already in the party. The scale of these influxes means that the overall gender profile of each party changed considerably. According to our survey, 38% of SNP members were women, up appreciably on the 2007–2008 survey although still well short of parity. In their survey a year or so later, based on a smaller sample but a higher response rate, Bale and colleagues (2020: 3) estimated that 43% of SNP members were women, and the party itself reported a figure of 44% in its annual accounts (SNP 2018: 3). Our own survey of the 2018 'mini-surge'recorded a figure of 39%, similar to the main surge. Overall, the figures vary a little but paint a similar picture: improvement but a continued under-representation of women in the party's base.

By contrast, the Scottish Greens look to have achieved parity, with more or less half (49%) of their members women and a small female majority among those who joined in the surge. As with the SNP until recently the Scottish Greens had not recorded the gender identification of members, and so we are unable to compare our results with any official party data. On this survey evidence, however, they


*Table 4.1* Gender of members by party and cohort

can be added to the small list of parties achieving gender equality. When it comes to gender, then, there is at least some basis for interviewees' impressions that the surge left the parties' memberships looking more representative of Scotland as a whole. The differences across the parties persist, but there is evidence of a massrecruiting referendum reaching beyond the 'usual suspects'. In neither case were the differences transformational. And they say little about women as activists and leaders within the parties or how they affect policy agendas (see Childs 2013: 82).

Respondents were asked about their sexuality: 6.3% of the SNP members and 11.7% of Scottish Greens described themselves as LGBT+. When comparing preand post-referendum members there is little difference. New SNP recruits are a little more likely to be LGBT+ (7%) than existing members (6%), whereas the pattern is reversed among the Scottish Greens, with 13% of pre-referendum and 11% of surge joiners identifying as LGBT+. This suggests that the surge made the parties a little more similar, but the difference between them remains noticeable. According to official statistics, the proportion of UK adults identifying as LGBT+ in 2017 was much lower, at 2.1% (Office for National Statistics 2023). 1

The difference between the two parties is likely to derive partly from the Scottish Greens being a more middle class, degree-educated party, but it probably also derives from the party having a high-profile community of LGBT+ members, activists and even leaders. And, since parties recruit few members in the youngest age groups in which minority sexual orientations are more common, it is possible that membership disproportionately attracts LGBT+ citizens. This is not specific to party membership, however. We asked a slightly differently worded question of Women for Independence members in 2015 and found that 9.4% described themselves as 'gay, lesbian, bisexual or other'.

We turn now to the subject of age. Political parties find it difficult to overcome the idea that they are irrelevant to young people (Pickard 2019). Studies show that those who join and become active are much more likely to be middle-aged or older than in the first flush of youth. The average age of a party member in Europe has been estimated at 53, with roughly a third of members over the age of 65 (van Haute and Gauja 2015: 295). Research points to a decades-long process of ageing among European party members (Scarrow and Gezgor 2010). And green parties around the world are subject to the same trends; Gauja and Jackson (2016: 372) found that the average age of the Australian Greens was 53.

Bale et al. (2020: 35) suggest that a very small proportion (only 6%) of UK party members is below the age of 25. Corbyn's Labour built a reputation for attracting young members, supported by images of events well attended by young people, but the reality is that the age composition of Labour's membership didn't change much. Many more people joined Labour, including young people, but the age distribution remained skewed towards older age groups, with a mean age of 53 (a recurring figure!). Of all party members in that study, only the Greens had an average age below 50 (Bale et al. 2020: 34). This is all evidence that political parties experience problems recruiting younger people. The referendum, though, might have acted as a catalyst for change in the SNP and/or Scottish Greens, a view expressed by many of our interviewees. Commenting on the change in voting age, one noted 'a huge difference in the numbers of young people feeling that they could get involved in adult politics' (Interview no. 51).

Table 4.2 compares the SNP and Scottish Greens, and as expected our survey showed the Greens to be the younger party. The mean age of SNP members was 54; for the Scottish Greens it was 47. Those aged under 35 made up less than 10% of the SNP membership but closer to a quarter of Green members. More than half of the SNP members were 55 or over, compared to a third of Green respondents. The Greens thus buck the usual tendency of party members to be disproportionately older people since, according to the 2011 census in Scotland, 38% of the adult population was over 55.

Next, we examine more closely how the parties might be changing, if at all. There are again two angles on this: comparing the overall sample in our survey with the results from a previous data collection and comparing pre- and post-referendum members within each party. First, comparing the SNP membership in 2016–2017 with the 2007–2008 membership, we find a mixture of similarities and differences. There is little difference in the first three age categories (below 45), belying any notion of post-referendum transformation. The larger proportion of over-65s in 2007–2008 should be interpreted with a degree of caution for methodological reasons. The earlier survey was fielded by mail, whereas the later survey was fielded online, the former method likely more congenial to the oldest respondents. However, the vast majority of people in all age groups are online – and so it is probably fair to say that the 4.5-year difference in mean age between the 2007–2008 and 2016–2017 surveys at least partly reflects a real difference. There is also evidence that the surge brought in at least some people from the elusive 18–34 age bracket. This was even truer of the further influx of members joining the SNP following the Brexit walkout of its MPs, 17% of whom were under 35 and only 15% of whom were 65 or over.

As with the SNP, the post-referendum Scottish Green joiners look younger than established members. Again, though, the change is far from dramatic and is driven by an expansion of the middle categories – notably, the 35–44 age range – rather


*Table 4.2* Age of members by party and cohort

than a surge of young adults. The mean age of Green members has hardly altered, and those aged 55+ actually constitute a slightly larger group than those back in 2002. Nor is there evidence that the Green surge recruited proportionally more young *women*. Among 18- to 24-year-old Greens, women are significantly *under*represented: 70% are male compared to 50% across the party.

It looks like the parties recruited a marginally higher proportion of young people following the referendum. The new recruits showed the same skew towards middle and older age groups, and hence the surge did little to change the overall age profile of the parties. It is worth noting that, among all party membership studies in the United Kingdom, only those of the Greens have ever recorded a mean age below 50, but the Greens are like other parties in that their age profile is far from representative of the wider population. Of course, this is not just about parties; youth under-representation applies to campaign groups and voluntary organisations too. Our 2015 survey of Women for Independence revealed that only 2% were 25 or younger and more than two-thirds (67%) were over 50, and their mean age was 54.

The age profile of the party memberships explains some of their other characteristics. Relatively few had the responsibility of caring for children below the age of 16 (19% of SNP and 26% of Scottish Green members). In the SNP, members were more likely to be caring for sick, disabled or elderly adults than for pre-school children. More than one in five people living in Scotland are disabled (Equality and Human Rights Commission 2016). In our surveys, a similar proportion – 24% and 20% of SNP and Green members, respectively – considered themselves to have a disability or long-term health problem. In this respect, the party members appear representative of the general population.

Studies since the 1990s have pointed to a stubbornly low proportion of black and ethnic minority members in all major parties. Bale et al. (2020: 41) suggest that these make up no more than 5% in any party, despite the growth of ethnic diversity in the population of the United Kingdom. Attracting members from more diverse ethnic backgrounds was a challenge identified by our interviewees, and our survey data confirm this under-representation: 97% of SNP and 98% of Green respondents described themselves as white. If this pattern looks even more pronounced than among UK party members, it is partly because Scotland is markedly less diverse: minority ethnic groups made up around 4% of the population according to the 2011 census, a third of the figure for England. There are, at least, signs of progress in the SNP given that members from minority ethnic groups accounted for just 1% of our 2007–2008 survey sample.

We also asked about the religious identities of the members. The large majority – over 70% of SNP members and over 80% of Greens – did not identify with any religion or denomination. As Table 4.3 shows, these proportions are well up on those recorded in the previous surveys of these parties, especially in the case of the SNP where a survey less than ten years earlier indicated that fewer than half of the members disclaimed any religious affiliation. Given the speed of this change, it seems unlikely to be due only to the broader secularisation of Scottish society and the generational replacement of religious with non-religious members (McCrone

#### 66 *Who joined the SNP and Scottish Greens?*


*Table 4.3* Religious affiliations by party and cohort

2017) although those underlying processes will play a part. The cohort comparison in Table 4.3 shows that post-referendum joiners were slightly more secular than existing members, but existing members were already markedly less religious than their counterparts of 2008 or 2002. As we show in Chapter 6, there was a shift towards a social liberal or progressive politics among these memberships which predated the surge, and that probably attracted disproportionately those who identify as secular.

Another noticeable change over time is that Church of Scotland identifiers, who outnumbered Catholics by almost 4:1 (38% to 10%) in the earlier SNP membership survey, are now barely more numerous (10% to 7%), and the gap is even narrower among post-referendum recruits (8% to 7%). However, this is clearly much more about a collapse in the Protestant percentage than about the SNP making the kind of inroads into Catholic Scotland that it has achieved electorally (see Johns and Mitchell 2016: chs. 2, 8). In the case of the Scottish Greens, the story beyond secularisation is largely one of continuity and diversity. Rather few identify with the major denominations, and indeed Scottish Green members are almost as likely to identify as Buddhists or Quakers – traditions linked to principles of pacifism and environmentalism – as to identify as Church of Scotland or Catholic.

#### **Socio-economic status**

Research over three decades has shown that those who join political parties are disproportionately well-educated. Back in 2007–2008, around a third of SNP members were graduates, a similar proportion to that found in Labour and the Lib Dems (Liberal Democrats) at the time but not as large as the striking three quarters of Scottish Greens recorded in 2002 as being educated to at least degree level. Environmentalist parties have a reputation for attracting *the* most educated members (van Haute and Gauja 2015). According to our most recent surveys (Table 4.4), the


*Table 4.4* Educational background by party and cohort

picture in the Scottish Greens remains the same: almost four in five members were university-educated – more than 40% had a postgraduate degree! – and those who left school at 16 were in a very small minority.

The SNP's membership is more like other political parties. Bale et al. (2020: 39) estimate that around half of UK party members are graduates, and the proportion in the latest SNP survey was just short of that at 46%. This marks an increase of 11 percentage points from the previous survey in 2007–2008. As with religion, there are several potential sources of this change: shifts in wider society, a long-term (pre-surge) change in the kinds of people to whom the party and its policies appeal and then the surge itself. Table 4.4 shows that the surge recruits were especially likely to be graduates. However, the difference is not large, and again a comparison between pre-surge members in the 2016–2017 survey and those in the 2007–2008 survey points to a membership that was already changing before the referendum. Also as with religion, this change can plausibly be linked to the SNP's growing commitment to socially liberal and internationalist positions that appeal disproportionately to graduates.

Education is of course strongly linked with social class. One expression of that is in the final row of data in Table 4.4, which shows that one in ten SNP members and more than one in five Scottish Green members attended a fee-paying school – compared with an estimated 4% of the population in Scotland (Leask 2016). These proportions are no lower among surge recruits. To some extent, this reflects the long-standing middle-class skew in party membership. This bias is a routine feature even of parties professing to represent the interests of the most disadvantaged in society (Seyd and Whiteley 2004; Bale et al. 2020).

Table 4.5 presents objective indicators of social class via the occupational profile of the members. As is conventional, if respondents were not in work at the time of the survey (whether due to unemployment or, more likely, retirement), they are classified according to their previous occupation. The survey did not include the very full battery of questions needed for a detailed classification, but the broad picture is clear from the three-category version of the classic Goldthorpe schema used in the top half of Table 4.5. The first point is simply to confirm that both


*Table 4.5* Occupational class and sector by party and cohort

memberships – like those of most parties in European politics – are dominated by those with professional (or what market researchers would call AB) occupations. This is only slightly less true of the SNP, in which working-class members make up a slightly larger minority (21% rather than 14% in the Scottish Greens) but are still far outnumbered by those in professional occupations.

When it comes to the surge, the influx into both parties was clearly middleclass. It brought basically the same kinds of people into the parties as were already there, although there is a hint of broadening in the Green recruitment of those from working-class occupations. And while the Scottish Greens are discernibly more middle-class in occupations, they are not more affluent. In both parties, the median annual household income (before tax) was in the £30,000 to £39,000 band, and 11% of SNP but only 7% of Green members were from households with incomes over £80,000. This tallies with the lower panel of Table 4.5, which shows that while Scottish Green members may have been in professional occupations, these were often in the public or the charitable or voluntary sectors, whereas SNP members were more likely relative to the Greens to have private-sector occupations. Green respondents were nearly twice as likely (17.6%) as their SNP counterparts (9.5%) to be in temporary or contracted work.

When we shift from 'objective' class, based on our classification of members' occupations, to their own subjective class identities, the SNP is more working-class than middle-class. The results in the upper panel of Table 4.6 are striking. There are many SNP members in professional occupations who nonetheless identify as working-class. One reason is that those who have experienced upward mobility may continue to identify with their family upbringing. However, class identities are also political, and the contrasts between the two parties in Table 4.6 – set against the relative similarity of their objective class profile – suggests that SNP members have more political interest or motivation in identifying as working-class. The same is implied by the fact that working-class identity was actually *more* common in the 2016–2017 survey than in the 2007–2008 survey. (In our follow-up survey of 2018


*Table 4.6* Subjective class identities by party and cohort

recruits to the SNP, the gap was even more pronounced, with 56% identifying as working-class compared to just 26% as middle-class.) This trend is hard to explain through sociology alone, not least because the growth in working-class identity has occurred across all three of the *objective* class categories in Table 4.5. Nor is it the result of the surge, as the cohort comparison here makes clear: working-class was also the most common identity among those who had joined before the referendum. In the case of the Scottish Greens, the surge involved a shift away from middleclass identity, but as Table 4.5 indicated, this may be because the surge brought some more objectively working-class members into the party.

Another feature of class identities (as opposed to occupational categories) is that they may not be very strongly held or may be outright rejected. A test of the strength of class identities is to pit them against national identities, asking respondents whether they felt they had more in common with an English person of the same class or a Scottish person of the 'other' class.2 Across the sample, 20% chose the former and 54% the latter, the remainder claiming that they could not choose. In both parties, the gap was narrower among working-class identifiers, suggesting that they feel that class identity more strongly than those identifying as middle-class, but national identity predominated in both cases.

#### **Origins and geography**

In the 2007–2008 survey, we found that more than one in ten (11%) of SNP members were born outside Scotland (Mitchell et al. 2012: 65). The more recent survey indicated that this had risen, including a growth in the proportion of members born in England – by far the largest group among members with a birthplace outside Scotland (Table 4.7). According to the 2011 census, 83% of Scottish residents were born in Scotland and 9% in England, so on this characteristic the SNP membership reflected the wider population. The other change is the increase (albeit only to 3%) in the share of members born elsewhere in the EU.

The Scottish Greens often recruit members who have moved to Scotland, and around three in ten (29%) of our survey respondents reported being born in England. There are also larger contingents from elsewhere within and beyond the United Kingdom. Generally, the profile indicates diversity and high levels of mobility.

#### 70 *Who joined the SNP and Scottish Greens?*


*Table 4.7* Birthplace of members by party and cohort

*Table 4.8* Experience of living outside Scotland by party and cohort


However, the surge expanded the Scottish-born share of the membership. Some of the members born outside Scotland had family connections with the country. In the case of the SNP, our survey indicated that 40% of members born elsewhere had at least one parent born in Scotland, but this was true of only 17% of Greens born outside Scotland.

Another feature of both parties' memberships is the proportion with experience of *living* outside Scotland. The survey defined this as being outside Scotland for a period of six months or more and, as Table 4.8 shows, this was the experience of around half of the SNP's membership and two-thirds of the Greens' membership. The reason for the gap between the parties is that Scottish Green members were more likely to have been born outside Scotland. If we confine the analysis only to those born in Scotland, then the probability of having lived elsewhere is about the same – at 43% for the SNP and 44% for the Greens. England was the most common answer to the follow-up question asking respondents where outside of Scotland they had lived, but large numbers had lived outside the United Kingdom. The cohort comparison for the SNP reveals that surge recruits were very like existing members in this respect, and indeed the party's overall profile is little different from that in our earlier survey. It is within the Scottish Greens that the surge changed the picture slightly, with an appeal based more on independence attracting proportionally more members who had always lived in Scotland.

Finally, we examine where respondents lived within Scotland at the time of the survey. Only 4% of SNP and not even 1% of Scottish Green members were living outside Scotland, mainly in England. For Table 4.9, we exclude these and report the distribution of the remaining members among the eight Scottish Parliament electoral regions. To put these results in context, we also show the regional breakdown of the parties' list votes in the 2016 Scottish Parliament election,<sup>3</sup> as well as the Scottish electorate as a whole to account for the fact that the regions are not quite equal in population size.

The SNP's vote is remarkably evenly distributed across Scotland, but the membership is more concentrated in the urban centres: Glasgow, Lothian (dominated by Edinburgh) and Dundee. Despite a respectable vote share in the South of Scotland, the party attracts very few members in that region. Compared to the previous survey of SNP members, there is a similar shift as in the party's electorate away from the north and east and towards the central belt, especially Glasgow (Johns and Mitchell 2016). This shift was accelerated by the surge. The post-referendum recruits stood out somewhat in being disproportionately likely to come from Glasgow. However, this shift was not pronounced because there was already a skew, one we expect to find in party membership, towards urban centres where education levels are higher and participation is more practical.

This pattern is even more pronounced in the Scottish Green Party with more than half its members living in either Lothian – Edinburgh long having been the Greens' most fertile soil – or Glasgow. The SNP is more successful than the Greens


*Table 4.9* Regional distribution of parties' membership and electorates

in recruiting members in small towns or rural areas. We also asked respondents to describe their place of residence in urban-rural terms, and the most common answer among SNP members – chosen by 40% – was 'small city/town'. The most common response among Green members was 'big city', with 36%, and a further 13% living in the suburbs of such a city.

#### **Conclusion**

This chapter has been a story of continuity, not change. The surge altered rather than transformed the membership profiles of the SNP and Scottish Greens. The Greens did appear to achieve gender parity, with nearly half of the members women. While the proportion of women in the SNP increased, the interviews suggested more change. The Greens remained the younger party, but neither party was attracting the very young. The class composition of the two parties was as might be expected based on previous studies. Both parties were well-educated and objectively middle-class, especially the Greens, but SNP members displayed a stronger working-class identity, an identity that was more pronounced than that in 2007– 2008. Large numbers of members (a majority in both party cases) had lived outside of Scotland for six months or more. This finding, along with where the members lived – the urban-rural classification – was again consistent with past research. Half of the Greens lived in a big city or in a city's suburbs/outskirts – precisely the same proportion as in 2002. Combined with the results outlined in chapter 3, which revealed a typical profile of well-networked organisational membership among our respondents, there is an overwhelming familiarity in these findings.

Our study is consistent with the general profile of party membership found in many other studies. This suggests significant barriers to widening democratic engagement with ingrained 'patterns of exclusion and under-representation' (Young 2013: 78). Apart from the gender equality achieved by the Greens, there is little sense that the surge in membership had a major impact on the parties' sociodemographic make-up. The new members resembled the existing members, suggesting that a lack of diversity in the backgrounds of party members is reproduced even in the context of surging membership.

From the parties' point of view, while inroads into under-represented groups would no doubt have been celebrated, the 'business as usual' picture painted by the data in this chapter is not necessarily troubling. Indeed, there are advantages to parties in recruiting those with more time, experience and money – as is the case with older and more middle-class members. It is from a normative or democratic standpoint that under-representation of key groups – women, minority ethnic communities, low-income groups and young people – is more problematic. While research on 'non-institutionalised' types of participation like demonstrating and signing petitions suggests that these may engage women and young people to a greater extent, the overall story of political participation is one of under-representation of large sections of society (Whiteley 2011; Dalton 2017; Pickard 2019). The surge analysed here does not challenge this understanding.

#### **Notes**


#### **References**


### **5 Motivations for membership**

#### **Introduction**

The answer to the question of why people join political parties might seem obvious – they support the party and its objectives. It might seem even more obvious in a context like this one, when surging membership was prompted by a long and intense debate about Scottish independence. The idea that people joined the SNP or Scottish Greens after the referendum as an expression of their support for independence appears uncontroversial and borne out by the results in this chapter. It is also consistent with previous research. The finding that ideological motivations are the single biggest driver of joining is close to universal in party membership studies (Cross and Young 2004; Gallagher and Marsh 2002; van Haute and Gauja 2015; Poletti et al. 2019; Whiteley et al. 2019). Our own 2007–2008 study of the SNP confirmed that, for the members themselves, a belief in independence overshadowed all other reasons for joining (Mitchell et al. 2012: 73).

Things are more complicated than this, however. Even after those spectacular expansions, membership remains the pursuit of a small minority. More than 1.6 million people voted 'Yes' in 2014, and everything about the campaign and subsequent polling evidence indicates that most of them felt strongly about the issue (Henderson et al. 2022). Yet only a small fraction joined the SNP or the Scottish Greens, either before or after the referendum. So ideological commitment may be a necessary but not sufficient condition for joining. Other aims – such as direct influence over policy, opportunities for participation, fitting in with peers – may persuade supporters to become members.

This echoes the rational choice framework, which has been prominent in research on group and party membership (Olson 1965; Whiteley 1995; Wilson 1995). Its key insight was that ideological goals alone could not explain an individual investing time and effort in joining a party, because that individual's membership would have a vanishingly small impact on the likelihood that the party achieved those goals. However frustrated a supporter of independence in the hours and days following the referendum, they were very unlikely to make any material difference to its prospects by joining the SNP or Greens. The goal of independence might have been closer than ever or set back by a generation or more but, either way, it was out of the individual's hands. The rational choice is to free ride off the

> DOI: 10.4324/9780429446559-5 This chapter has been made available under a CC-BY 4.0 license.

efforts of existing members and other joiners – that is, unless membership offers incentives available to joiners regardless of whether ideological goals are achieved. Those *selective* incentives – so-called because they are only available to members – are, on the rational choice reading, what persuades the ideologically sympathetic to take the further step into membership.

In this chapter we assess what past theorising and research can – and cannot – tell us about the motivations for joining during a surge like this. We describe a combination of three factors: the ideological conviction and emotion generated by the referendum campaign, the relative ease of joining in an online world and the fact that the surge became a collective movement in itself. Consistent with previous studies, we find the first of these to dominate decision-making. With the SNP as with the Greens, these ideological motivations go beyond support for independence to take in a broader world view that independence is believed to serve. Another echo of previous research is that we find rather little evidence of members joining to pursue ambitions for political participation. This may seem surprising given the participatory flavour of the referendum campaign, but it is important to bear in mind the timing of the surge: it reflects reactions to the referendum *result*.

#### **Surge motivations?**

In an influential early study of organisational joining, Clark and Wilson (1961) distinguished between three types of incentives: purposive, material and solidary. The first type, purposive incentives, relating in this context to the party's purpose or goals, are exactly those described earlier. That they are important is not in dispute; that they are sufficient for membership (notwithstanding the rational choice critique) remains a possibility for two reasons. First, potential joiners do not necessarily consider or accept that their individual membership does not make a difference, just as many citizens believe that their vote makes a difference, however implausible this is on a strict calculus of costs and expected benefits (Dowding 2005). Many of those joining the SNP or Scottish Greens – whether before or after the referendum – may well have believed that this would contribute to achieving independence. Second, the sense of pursuing such a cause is itself a selective incentive available only to joiners. There is satisfaction in contributing rather than simply free riding, even if that contribution is more about expressing support than achieving change.

Of the remaining two types of incentives, material and solidary, the former – more tangible rewards such as financial or career benefits, including the possibility of electoral office – has been proven empirically to be less important than the latter (Whiteley et al. 1994; Bale et al. 2020). Solidary or more psychological benefits come in various forms. Membership can offer social connectedness, conformity to norms within an individual's social network, the opportunity to associate with like-minded individuals or a sense of altruism from contributing to the democratic process or the perceived national interest (Bale et al. 2020: 79).

Three features of this research need noting here. The first two are methodological. In a field dominated by surveys of members, there are reasons to suppose the evidence to be skewed away from personal incentives and towards more expressive or ideological motivations. One of these is social desirability bias. Citing collective goals tends to present members in a more favourable light. But this is not only or even mainly about respondent dishonesty. One of the most pervasive tendencies in human psychology is to assume that our behaviour is driven by our deeper goals and attitudes rather than 'shallower' features of a situation (Nisbett and Wilson 1977; Lodge and Taber 2013). 'I joined that party because I share its goals' is a natural inference for members to make.

Second, very few studies have had access to non-members, yet this group is key to confident causal inference about why people really join parties. As noted already, hundreds of thousands of strong supporters of Scottish independence did *not* join a party sharing that ideological goal. If those who *did* join report strong support for independence as the reason for doing so, then this report seems incomplete. One possibility is that joiners are simply those whose support was the strongest. One of the few studies that was able to compare joiners with non-joiners found that the most powerful predictor of making the step into membership of British parties was the strength of identification with that party (Poletti et al. 2019). Next most powerful was ideological extremity (defined as distance from the midpoint on a left-right scale). This supports a kind of threshold model whereby a certain strength of expressive or ideological motivations prompts joining. In that same study, however, a wide range of other selective incentives also predicted joining, suggesting that strength of agreement with a party's goals is the central narrative but not the whole story.

The third point is that most party membership studies were snapshots of membership, taken in what might be called 'normal times'. As such, they were not well placed to capture anything distinctive about surging membership. We are interested not only in why SNP and Scottish Green members joined those parties but also why so many of them joined at a particular time. On the one hand, we would not expect underlying motivations to be so different: the referendum-driven nature of the surge links it clearly to the ideological or purposive goal of independence. On the other hand, the surge meant that the act of joining became something of a collective as well as an individual act, and part of the motivation may have been to join that collective.

It is difficult to understand 'surge motivations'. Membership surveys have tended to ask about why people join, rather than why they joined at a particular time. And surge motivations are probably in the same psychological category as material motivations. Survey respondents are likely to downplay them, to researchers and themselves, in favour of more purposive, ideological motivations. Examining the closest thing British politics offers to a comparable case, the Corbyn-driven surge in Labour membership, Whiteley et al. (2019) highlight a range of ideological and political motivations but do not test for or speculate about surge-specific motivations.

In their wider study, however, Bale et al. (2020: 88) note the role of leadership contests in prompting influxes into all three major UK parties and pinpoint each as an event 'which, if seen as a high stakes occasion by enough potential members, can, almost in and of itself, encourage an influx of members'. They take this as 'a useful reminder that, while [ideological] incentives matter when it comes to joining, so, too, do triggers' (Ibid.: 89). As we noted in Chapter 1, Power and Dommett (2020) also highlight the importance of context-specific triggers, observing that, while parties are often surprised by the events that trigger an influx of members, they then quickly seek to capitalise on them to maximise the scale of any surge.

The relevance of this to our study is plain. What is slightly less clear is what was the trigger or event in this case. Given the timing of the surge, it was not the referendum campaign (or event within it). By the eve of the referendum, SNP membership had increased only marginally compared to the figure reported for 2013. Scottish Green membership had grown rather more in percentage terms, from 1,178 in December 2013 to 1,500 in August 2014, but this 27% increase was of the same order that parties might normally expect during election campaigns. Interviewees from both parties reported consistently that it was the referendum result that prompted a huge and unexpected surge.

But these surges did not take place overnight. It took five days for the SNP membership to double in size, two weeks for it to treble in size, and numbers continued to climb over the following months. This raises the possibility that the membership surges themselves became a motivating force or trigger. Of course, there is prima facie circularity in arguing that 'people joined because people were joining'. However, once the referendum result is recognised as the initial causal trigger, it is reasonable to suggest that the surge became partly self-fuelling. As Yes supporters saw more and more of their number reacting to the frustration of defeat by joining a pro-independence party, that option was both highlighted and legitimated. Moreover, by following suit, they were not only joining a party but joining a surge – a mass movement in a literal sense.

This process has two likely key drivers. The first is publicity. The surges into the SNP and Scottish Green Party seem likely to have been fuelled by their extensive coverage in the traditional media and on social media. We noted in Chapter 2 that SNP chief executive Peter Murrell took to regularly tweeting the latest updates to the party's membership figures. In interviews, many senior figures cited this as both a deliberate and a successful attempt to maximise the size of a surge that the party had not seen coming. In addition to such 'top-down' surge recruitment, social media also facilitates the second driver: a more horizontal process whereby joiners can report the fact and encourage others in their network – either directly or just by example – to join and swell the surge. This kind of social network recruitment has long been recognised as a catalyst for collective action (Verba et al. 1995) and, whereas mobilising into more intense forms of political participation often requires networks of quite strong ties, even the rather looser and weaker networks on social media are enough to inform and persuade a wide audience about party membership. In short, party membership went viral in some networks.

There is at least indirect evidence of this from a survey question asking whether, when members signed up, they knew other people who joined at the same time. Among those who reported joining in the week after the referendum, 58% of SNP and 60% of Green recruits said that they did. The corresponding figures from more 'normal'times – defined here as outside the referendum year – are 28% for the SNP

and 18% for the Greens. This provides some support for the notion that knowing other joiners was part of the driver of mobilisation.

#### **The cost of joining**

Online registration has drastically reduced the time and effort needed to join a party (Scarrow 2015: 130–132). The online forms at party websites are barely more arduous than those completed to access Wi-Fi at a café. This is not only about ease but also immediacy: speedy online sign-up from the sofa can capture those seized by a moment of zeal (or even a moment of intoxication – one Scottish Green member reported in the survey that he joined the party because he was 'drunk and annoyed') that would not have survived a more demanding procedure. It may even capture those for whom joining is a less than momentous decision. In interviews, staff involved in processing Scottish Green Party membership applications reported some cases of duplicate applications from people who signed up more than once during the immediate surge, second time round seemingly forgetting that they had already done so. Of course, many new members will have given the decision plenty of thought and would have had patience for a more onerous registration process. Our interviewees reported the parties being overwhelmed by the volume of requests to join in the immediate aftermath of the referendum, with online systems crashing and new recruits then calling the parties' offices, meaning that some persistent members joined by telephone. The point is that online sign-up brings both impulsive and committed types into a party.

The financial costs of joining were also limited. At the time of the surge, there was no set fee for SNP membership: the party recommended payment of £1 per month, but this was optional. Even among those who did pay, a fee of £12 per year was well below the average for parties in Europe. The Scottish Green annual fee of £36 was a good deal higher, but discounted rates meant the average member paid £22, and for a party traditionally made up of the relatively affluent middle classes, it was unlikely to prove prohibitive (and was payable by monthly instalments). Both parties offered big discounts for the unwaged. In short, fees were small enough to be consistent with an argument about low-cost joining and certainly too small to have much impeded the surge.

Why is this discussion of the cost of membership relevant for a chapter on motivations? After all, we would not expect survey respondents to report that they joined a party because it was cheap, and ideological reasons are likely to be cited instead. First, if the costs of joining are nearly zero in terms of money, physical effort and time, then the calculus supposed to scupper rational choice models – that the costs exceed the expected benefits of participation – becomes less clear-cut. Second, members need not necessarily expect that much from membership for the benefits to exceed the costs. Third, just as the psychological benefits of membership outweigh the material gains, it may be that the bigger barriers to membership are also psychological. Those many strong supporters of independence who did not join the SNP or Scottish Greens were probably daunted less by the time and effort of joining and more by a feeling that party membership was in some sense not for them.

#### 80 *Motivations for membership*

The second point has important implications for motivations. The general argument made at the outset was that ideological motivations are not sufficient to induce individuals to pay the costs of joining a party; they need to get something more out of membership, such as the enjoyment of campaigning or the feeling of connectedness with like-minded people. The lower the costs, the less would-be joiners should demand from membership in terms of added selective incentives. Applying this argument to our case, suppose that many of those brought into the SNP or Scottish Greens were Yes supporters for whom surge joining was primarily an immediate and low-cost way of expressing continued strong support for independence and frustration at referendum defeat. We would not then necessarily expect them to be as driven by material, solidary and general participatory motivations as those who had joined their party at an earlier time and had not been carried into it on a surge.

This would not make surge joiners unusual on a broader perspective. Rather, it suggests that they fit the mould of a newer type of party member, described by Faucher (2015: 407) as 'individualised'(see Chapter 1). More effortful and collective engagement – canvassing, delivering leaflets, regularly attending party meetings and so on – was always a minority pursuit in most parties (Scarrow 2015). But this seems (insofar as data allow comparison over time) to have become still more so, not least since social media and smartphones opened up new channels for individual activism (Gibson et al. 2017). The growth areas, such as following and supporting parties on social media, or offline activities such as displaying posters, are, as Webb et al. (2017: 69) sum up, 'all things that can be done without actually leaving one's home'.

It may seem strange to suggest that a referendum campaign noted for intense and collective political participation triggered an influx of people seeking a lowintensity and individualised form of party membership. For some joiners a highly participatory referendum campaign will indeed have been a gateway into highly participatory membership of a party. However, it is important not to confuse the vocal and visible minority with a quieter majority. Results in Chapter 3 confirmed that, while large majorities of surge recruits engaged in discussion of the referendum and majorities put up posters and wore badges, the more high-intensity activities were the preserve of a minority. Much referendum involvement was of an expressive form that did not require people to leave their house. This could well be what surge joiners were seeking in party membership.

#### **Self-reported motivations**

Surveys offer a number of different methods of assessing members'motivations for joining. Three of these are used in this chapter. The first and most direct is to ask respondents why they joined. The second is to ask more general questions about the perceived role and appeal of membership and to infer from strongly positive assessments that these drew members into their party. A third approach is to ask respondents what they aspired to do within their party at the time of joining.

We begin with the first approach. This section reports on an open-ended question asking members to explain in their own words up to three reasons why they joined, and a closed-ended question presenting respondents with a list of possible reasons for joining and asking respondents to rate the importance of each of those reasons in their decision. Table 5.1 presents the results from the open-ended question for each party.1 Almost all respondents gave at least one reason and around half gave all three. The table presents two sets of coding, one based on the first reason given and the other based on all reasons. The individual codes are clustered into broader types.


#### *Table 5.1* Reported reasons for joining by party

#### 82 *Motivations for membership*

The first result is the most predictable: that independence is by far the most commonly cited reason among SNP members. By coincidence, the 44% of first mentions is the same percentage as in our 2007–2008 survey. But that gives a misleading impression of stability because the post-referendum survey included a lot of responses which are clearly grounded in independence but are clustered here under 'referendum triggers'. Independence is now overall a more central motivation for SNP members than before. Indeed, reporting the earlier survey, we were struck by the fact that more than half of respondents gave a primary reason other than independence for joining. Some of those, such as criticism of Westminster and wishing to further Scottish interests, are connected to independence. Others, such as a general liking for the SNP and dislike of other parties, belong more to everyday party politics. Those reasons were to the fore at a time when the party had just taken office at Holyrood for the first time – and had done so partly by downplaying the independence issue in its offer to voters. All these other reasons just mentioned are visible in the 2016–2017 data but in smaller proportions than a decade earlier and are more likely this time to be cited as a secondary reason for joining, reflecting the changed context.

 For SNP members, referendum triggers accounted for 16% of first reasons given and of all reasons combined. Greens were less likely to mention these triggers, but a sizeable one in ten mentioned them as a primary reason for joining. A notable feature of those triggers is how varied they are in terms of the range of feelings reported: anger, frustration and regret but also some more positive and purposeful responses. Negative and positive reactions were often combined in the same answers which, although it made coding a difficult judgement call about which predominated, did highlight the mixture of emotions felt in the immediate aftermath of the result. Here are three illustrative examples, all ultimately coded as 'Urge to do something positive': 'Joining was the only thing that made me feel better about the referendum result. It cheered me up' (SNP member); 'There was a void left by the referendum vote and I wanted to keep up the momentum for positive change in Scotland' (SGP member); 'I just felt that I had to *do something* after the referendum' (SNP member). As might be expected, very few members spontaneously expressed a surge motivation for joining, though a handful in both parties did cite boosting party membership. An SNP member said, 'I had always voted SNP but suddenly wanted to be part of the movement I felt was sweeping over Scotland'.

As in the previous SNP survey, but this time even more strikingly so, the participatory incentives discussed at the outset of this chapter were not prominent in the responses – even as secondary motivations. Where they were mentioned, the participation in question was of the less intensive variety, either contributing financially or – the most common response among SNP members – simply 'to support' the party. In relative terms, active involvement was much more likely to be mentioned by Green members; in absolute terms, though, it was still rarely mentioned even by them. This is consistent with the point made several times in this book that the average party member is not especially active (see Chapter 7).

The biggest difference between the SNP and Scottish Green responses is the relative importance of nationalist and other ideological motivations, notably environmentalism. The gap is predictable but its scale is noteworthy. The large majority of this sample of Green members had joined after the referendum – most of them in the immediate surge – yet only one in ten reported independence as their primary reason for joining the party (including the small number who explicitly said they had been seeking a different vision of independence from that offered by the SNP). As we show in the next chapter, this is not because of widespread scepticism about independence among Scottish Green members; it is instead an illustration of their different expressions of priorities compared to SNP members. Those priorities range well beyond the environment, too. Plenty of respondents endorsed the Greens' social justice agenda and more cited a general match between their own views and what the party stands for; for example, 'It most closely matches my personal aspirations and principles' or, more laconically, 'I like the policies'.

Even though the surge was proportionally just as big (in fact larger) in the Scottish Green Party as the SNP, it is hard to say why joiners of the former were less likely to mention referendum triggers. It may have to do with the Greens being a less familiar party until coming to greater prominence during the referendum campaign. Those inspired to join after the vote, even if driven by the same commitments and frustrations as those surging into the SNP, may have been thinking more about the ideological or broader party appeal of the Greens. It is noticeable that there are more mentions of leadership – notably praise for Patrick Harvie – by Green than by SNP joiners. This is noteworthy given the traditional scepticism in environmental parties about hierarchy and leadership and suggests – as Chapter 8 discusses – that the Green recruits might be accepting of the need for leadership and their own subordinate role as members.

Table 5.2 presents a cohort comparison which goes beyond the pre- and postreferendum split, distinguishing those who joined pre-referendum, those who joined in the main surge and those who joined sometime after the referendum, from January 2015 onwards. Perhaps the most striking result is just how few of the surge joiners cited referendum triggers. This highlights the distinction between motivations and triggers (see Power and Dommett 2020). It seems highly likely that those


*Table 5.2* Reported reasons for joining (all reasons, categorised) by party and cohort

joiners were indeed triggered by the referendum but, from their perspective, the motivation for joining was usually a purposive commitment – to independence in the case of the SNP and to environmental and social justice concerns in the case of the Scottish Greens. Another interesting question concerns those who joined after the immediate surge (from 2015). Was this a delayed reaction to the referendum or do they more closely resemble those who had joined in more 'normal' times? The answer looks more like the latter. Relatively few (especially among Greens) reported referendum triggers, and there was more mention of general party ideology, policy and leadership. In both cases, it is notable how little this group reports independence- or referendum-related motivations. As shown in the next chapter, this is partly because they are less strongly supportive of independence.

In the closed-ended question, respondents were asked to rate (on a scale from 0 to 10) the relevance of several motivations. Figure 5.1 shows the results for SNP members, broken down into the same three cohorts as in Table 5.2. Each bar represents the mean importance attributed to that reason in members' decisions to join. The long bars at the top of the graph confirm what the open-ended question reported; that ideological purpose – primarily independence but also social justice and environmental motivations – dominates members' understanding of their decision-making. If we are seeking other factors that convert ideological sympathy into membership, then leadership should not be overlooked – it was typically given a rating well over the midpoint. What is less clear is the role of the referendum as a prompt. On the one hand, a sense of 'maintaining an exciting movement' was deemed important by many respondents – 26% of SNP members gave it the maximum point on the 0–10 scale. However, the cohort differences on this

*Figure 5.1* Mean importance ascribed to motivations for joining the SNP by cohort

motivation – indeed, throughout the graph – are meagre, which might cast doubt on whether it can specifically explain surge joining. However, pre-surge members were also caught up in the experience of the referendum and could reinterpret their initial decision to join through that lens. All in all, the sense of a movement or collective does look like a significant prompt for many surge recruits. As we have noted elsewhere, this sense of belonging to a movement doesn't require direct interaction with others (see Bennie et al. 2021). The comparatively low ratings for 'mixing with other like-minded people' and 'maintaining referendum relationships' are notable. It seems that for many such relationships did not exist or at least played little role in the decision to join. Finally, and less surprisingly, very few respondents endorsed the surge hypothesis that they joined in part because so many others were doing so.

Figure 5.2 reports parallel results for Scottish Green members. The surge and decline in the importance of independence seen in Table 5.2 is visible here, too, as is the importance of other ideological motivations. Strikingly, equality and social justice weighed at least as heavily as environmental considerations in Green joining decisions. This is a point we return to in the next chapter. The other point to note is that the more movement or associative motivations that appear lower down the graph were generally deemed less important by Scottish Greens than by SNP members. This may seem counter-intuitive given the association between green parties and participatory democracy. Motivations for joining the Scottish Greens were more about expressing support for the party's ideology, policy and leadership than a desire for active participation.

*Figure 5.2* Mean importance ascribed to motivations for joining the Scottish Greens by cohort

#### 86 *Motivations for membership*

Some indirect evidence for this argument comes from correlating the importance given to the 'maintain exciting movement' motivation with surge joiners' reports of political activity during the referendum. Those correlations are all rather weak, indicating that active participation was not a precondition for movement belonging. The correlations (across both parties) were actually a little stronger with the *lower*-intensity participation such as online activity (*r* = 0.17) and visual display of posters and car stickers (*r* = 0.20) than with the more active forms like canvassing (*r* = 0.11). Echoing Chapter 3, this confirms that many felt part of an inspiring movement despite having had rather little activist involvement in the referendum campaign.

#### **Understanding surge joiners**

The purpose of this section is to look more closely at the kind of members that the surge brought into the parties. It begins with a comparison of three groups of people: those already in the SNP by referendum day, those who joined in the surge, and 2014 Yes voters. The Yes voters are accessed via the Scottish Referendum Study and give us a means of comparing actual and potential joiners and thus examining conjectures about why the surgers signed up. The aim is to test the threshold model mentioned earlier: that is, the notion that it is a certain strength of ideological conviction or fervour that prompts membership. We confine this comparison to the SNP because Scottish Green supporters (hence potential Scottish Green members) make up a small proportion of Yes voters as a whole and because the ideological convictions involved there are more diverse.

The surveys offer two parallel measures of strength of opinion. One is the strength of emotional reactions to the referendum outcome; the other is what might be called cognitive polarisation, measured by disagreement with the statement 'Although I was on one side of the debate, I must admit that the other side had some strong points'. The results are shown in Figure 5.3. Anger stands out as the strongest reaction – especially among surge joiners. For all that the surge is sometimes characterised as a positive and purposeful response to defeat, it looks fuelled less by excitement and hope and more by negative emotions. The other notable point is that surge joiners look more like existing members than Yes voters as a whole. The same is true when we look at the 'other side had some strong points' question. The proportions disagreeing with the statement were 64% among prereferendum joiners, 61% among surge joiners and 44% among Yes voters. This supports the idea that the heat of the referendum and dejection of defeat tipped many more people over the threshold at which support for independence becomes strong enough to motivate SNP membership.

Conclusions about the similarity of surge joiners and existing members should not be pushed too far. While they may have reacted similarly to the result in September 2014, there remains the key difference that the latter had already found a motivation to join the SNP – in many cases long before the referendum was even mooted. This may simply mean that they reached the 'threshold' sooner, but it may mean that their decision was driven not just by strength of support for independence

*Figure 5.3* Emotional reactions to referendum outcome by SNP cohort and among all Yes voters

but also by other motivations, as opposed to the more purely ideological groundswell that brought in surge joiners. It might be that those joining the party in normal times were seekers of conventional party activity, while those joining in the surge were looking for an outlet for their anger and a way of expressing their support for independence.

We look at this in more detail in Chapter 7, comparing levels of *actual* party activity across cohorts. In this chapter on motivations, we are interested in the kinds and levels of activity that members were *seeking* when joining. One relatively direct route to understanding participatory motivations is a question asking members how they would like to be involved in the future. Given the timing of the survey, the responses will reflect a combination of ambitions when joining and experience since doing so, but they give us a sense of whether surge joiners are different in terms of their self-declared appetites for activity. Going by Table 5.3, the answer is that they are. Across both parties, each of the first four rows shows the same pattern: surge joiners have less appetite for activity than either pre-referendum joiners or those who joined after the immediate surge. The rebound among 2015 joiners looks like something of a return to normal. If so, then surge joiners were indeed unusual in their limited interest in party activity. The one exception is the bottom row, supporting online campaigns, where surge joiners are the keenest cohort. This again suggests that these were people attracted to online rather than offline activism.

#### 88 *Motivations for membership*


*Table 5.3* Percentage wanting to be involved in various activities by party and cohort

#### **Conclusion**

The data presented in this chapter is consistent with the central message from party membership research that people are driven to join a political party by a powerful commitment to its ideological goals. Had we generated a word cloud from the open-ended responses in Table 5.1, 'Independence' would have dwarfed anything else in the graphic. Whether this should be classified as a purposive or expressive motivation is hard to say because the theoretical distinction blurs in practice. Few members will think their contribution pivotal to securing independence, but few will think it completely irrelevant, so the two motivations work in combination. Both the SNP and the Scottish Greens offered joiners the opportunity to express their strong support for independence *and* to continue the fight for it after disappointment at the ballot box.

What this case adds to existing scholarship is suggestive but also interesting. A first point involves triggers. While the importance of these is clear from the timing and suddenness of surges like those examined here, this chapter provides direct evidence both of members acknowledging that they were 'triggered' and of the role of anger in this process. It also seems highly likely that, while triggered by the referendum, the surges then became self-fuelling. The scale of the influx, the view of party elites that they were contributing to this momentum, the importance of horizontal recruitment networks, not to mention what we know about the psychology of group behaviour – all of this makes it likely that the surge was itself a part of what persuaded people to join the two parties. However, this is very hard to substantiate using membership survey data, and the evidence here is limited and indirect.

Another point concerns the rather more modest participatory ambitions of the surge joiners. The differences are not large, but we report clear evidence that the surge is not simply a spike in demand for membership but an influx of at least some people with a different conception of membership. That conception is broadly the one described as 'individualised' by Faucher (2015). The very visible grassroots activity of the referendum campaign risks obscuring the fact that most Yes supporters were engaged in less intense (though still politically important) activities such as trying to persuade peers and displaying posters. Those are the lower-key activities that many surge joiners were seeking in their new-found party.

If this seems to offer little by way of selective incentives, consider two further arguments. The first concerns the emotional impact of the referendum. 'Peacetime' joiners of the two parties may have needed a combination of ideological conviction and the material or solidary incentives that come with membership, even if past research shows the first motivation to predominate. The referendum campaign intensified the ideological motivation to the point where little else was needed but the trigger of the result. The second returns to the point about the cost of joining a party in just a few minutes via an online form. The further those costs fall, the less substantial the tangible benefits need to be to make joining rational.

There is a risk of exaggerating the difference between surge joiners and existing members. Throughout this book, we report more similarities than differences, and much of the data in this chapter conforms to that pattern. If we emphasise the lack of activist ambition among surge joiners, it is in part to counter the myth that the grassroots activists of the Yes campaign simply moved en bloc into these parties. In fact, the average new joiner is like the average party member from any cohort, with limited appetite for high-intensity participation. Moreover, even if the new recruits are a little less participatory in relative terms, they are so many in absolute terms that the surge hugely boosted the total activist capacity of both parties. Questions remain about whether, in the event of a second independence referendum, surge joiners would participate more and in different ways as party members.

#### **Note**

 1 A coding scheme was developed using a randomly selected 500 responses from each party; the full scheme was then applied to a further 1,000 responses (taking us close to the limit of the Green sample size), and those results are reported in the table. If a reason does not appear in Table 5.1, it is because it did not appear in any of the 1,000 responses on which the coding scheme was based. If a reason appears with a row of zeros, this is a matter of rounding down.

#### **References**


### **6 Identities, ideologies and independence**

#### **Introduction**

The previous chapter confirmed that a commitment to independence was the main motivation for those who joined the SNP and Scottish Green Party in the postreferendum surge. This raises questions for the current chapter which profiles the political identities, values and opinions of the parties' members, new and old. What explains their support for independence? Does independence mean different things to different joiners? This second question is not primarily about constitutional detail. While there are many forms that independence could take (Mitchell 2020), we might not expect rank-and-file party members, even if rather more politically engaged than the average citizen, to have very distinct preferences. We might expect their support for the broader goal to be stronger than their opinions about its precise form. These questions about why members support independence focus on what they think independence offers to Scotland.

There is a distinction between independence as an end in itself and independence as a means to some further political end. In the case of the SNP, there has long been a co-existence of end-in-itself motivations with a variety of ends to which independence was seen as the means. The latter take in single issues, such as the retention of oil revenues or the removal of Trident, and broader ideological goals such as the rejection of Thatcherism or the promotion of social justice. The distinction is an over-simplification but is nonetheless potentially useful. It is especially pertinent when leaders eschew the label 'nationalist' and are explicit about independence as a means of achieving other than constitutional goals. We might expect Scottish Green Party members to see independence from the perspective of achieving environmental goals, for example. On the other hand, the parallels in the scale and timing of the SNP and Scottish Green surges suggests that the two sets of joiners, each exposed to the same campaign, might have plenty in common when it comes to what they want out of independence.

Our attitudinal profile of the two parties' memberships begins with identities. While national identity is the most obviously relevant in this context, the increasing entanglement of Scottish with left-wing identity makes the latter another important focus. Those ties were drawn more tightly during the referendum campaign and, via a focus on austerity, were given a more prominent expression in policy

> DOI: 10.4324/9780429446559-6 This chapter has been made available under a CC-BY 4.0 license.

terms. That makes it worth examining whether these party members are markedly to the left in terms of policy attitudes as well as identities, unlike the wider Scottish public which tends to see itself as more left-wing than its specific opinions would suggest (Henderson 2014).

The SNP has espoused 'independence in Europe' since the late 1980s, and the Yes campaign in the referendum was predominantly pro-European Union. However, the European Union gave Yes voters little encouragement during that campaign, and many supporters of independence went on to vote Leave in 2016 (Johns 2018). Moreover, in our 2007–2008 survey of SNP members, almost one in four chose independence outside the EU over independence within it (Mitchell et al. 2012: 105). That raises two questions. Does there remain a seam of hostility to EU membership among SNP members? And were the new joiners more pro-EU? They were brought into the party at a time when support for EU membership was linked to Scottish independence, but in the years before the referendum the issue was less prominent in SNP discourse than it had been in the early 1990s.

A broader question is raised by the fact that the Remain/Leave battle was not only – not even mainly – about the EU. Instead, it crystallised a deeper divide in values and outlooks that had been opened up by immigration and related changes in British society (Sobolewska and Ford 2020). Whether this division is characterised as social conservatives versus liberals, 'somewheres versus anywheres' or isolationists versus internationalists, what is clear is that the SNP and Scottish Greens both positioned themselves squarely on the liberal Remainer side of the argument. The chapter addresses whether the parties brought their members with them and whether the new recruits were particularly likely to share these values.

 The final section of the chapter looks in more detail at members' attitudes to independence. We also look at how strongly and for how long members have supported independence, and how far they prefer it to 'devo max' and other devolution-short-of-independence options. Finally, we examine members' views on the road ahead and the strategic questions – especially concerning a second independence referendum – that their parties face.

Throughout the analysis, as in previous chapters, we compare the two parties but also pre- and post-referendum joiners within the parties. The latter comparison is useful for gauging whether the big influx of members shifted the ideological centres of gravity, or the values and priorities of the parties. In the case of the SNP, referendum and election survey data enable a third comparison: between the party's members and its electorate. This provides a useful benchmark for judging the attitudinal profile of the membership as a whole. It also allows us to examine whether the surge brought in members well to the left and impatient for independence or instead a set of members who were if anything more ideologically moderate and more like 'ordinary' voters.

#### **Identities**

Party membership is, among other things, an expression of identity. While constitutional preferences are not a direct function of national identity, they are interrelated. Most Scots feel at least some sense of British identity but, since a feeling of Scottishness is much more widespread in the electorate as a whole (even among many No voters), this link between national identity and support for independence has potential for the pro-independence side. Not surprisingly, national identity was a powerful predictor of voting in the 2014 referendum (Henderson et al. 2023).

Rather like party identities, national identities were assumed to take root in childhood and then to shape subsequent political thinking and attitudes (Hierro 2015). Recent research has considered the possibility that national identities – or, at least, survey measures of those identities – are less stable than this suggests (e.g., Charnysh et al. 2015; Bochsler et al. 2021). This is more likely where co-existing identities are perceived to be in tension, such as in Catalonia in the past decade or so or Scotland since the calling of the 2014 referendum. Pressure is placed on those with dual identities – such as those identifying as supporters of both independence and Labour or as an independence supporter who feels British as well as Scottish – to choose between them, either weakening or perhaps even rejecting the secondary identity altogether (Hierro and Gallego 2018). If supporters of independence felt persuaded or pressured to disclaim any sense of Britishness, then we would expect that to be especially true of SNP and Scottish Green members – and perhaps particularly those who were converted to independence during the referendum campaign.

Table 6.1 shows how the two parties' memberships responded to the classic Linz–Moreno question, a measure frequently used in research in this field (Guinjoan and Rodon 2016). The table compares these responses with the previous SNP membership survey and then with the electorate as a whole courtesy of the Scottish Election Survey of 2016.

The SNP shows the profile we might expect from a party with the goal of Scottish independence. Most members reject Britishness altogether and, of the rest, almost all place it second to Scottishness. This is clearly not the consequence of the independence referendum. The SNP's membership has slightly *more* of a British identity compared with that in the 2007–2008 survey. The 'Scottish, not British' percentages among pre- and post-referendum recruits were 74% and 68%, respectively. These large majorities contrast with the SNP voting base in 2016, 43% of whom opted for 'Scottish, not British'. While overwhelmingly Scottish in terms of primary identity, the SNP electorate contained more who asserted some British identity than who rejected it.


*Table 6.1* Linz–Moreno national identity responses by party

#### 94 *Identities, ideologies and independence*

The Scottish Greens have a very different national identity profile. Part of the reason why was highlighted in Chapter 4 – a much larger proportion of their members were born in England. However, this cannot account for all the difference in 'Scottish, not British' identities between the two parties. Whether born in England or Scotland, Scottish Greens are markedly more likely than SNP members to have some British identity. In this respect, Green members closely resemble both Green voters and the Scottish electorate as a whole. The most distinctive feature of Green members was their tendency to choose 'Other' and, while 'English' accounts for a few of those responses, many were explicit rejections of national identity or labels. Several claimed to be 'citizens of the world'. Overall, the results in Table 6.1 indicate that many Scottish Greens evade association with national identity while supporting Scotland as a distinct polity that should be independent.

The obvious criticism of the Linz–Moreno question is that it is a measure of relative rather than absolute identity and it is ambiguous about the strength of the identities in question. Someone 'more Scottish than British' might feel both identities quite strongly, feel neither very strongly or feel far more Scottish than British. We therefore turn to separate seven-point scales (from 0 to 6) measuring how strongly respondents reported feeling Scottish and British. These provide a subtler gauge of cohort differences. For a further comparison, providing useful background for the later discussion of support for EU membership, Figure 6.1 also reports the strength of European identity.

*Figure 6.1* Strength of Scottish, British and European identities by party and cohort

The main contrast between the parties – SNP members being much more polarised on these two identities than Scottish Greens – was already clear from Table 6.1. If we compare pre- and post-referendum joiners of each party, Scottish Green members recruited by the independence referendum felt a little more Scottish on average than those already in the party. Given that the difference between the cohorts is small, this suggests pre-existing in-group identities driving party joining rather than the reverse process whereby supporters of pro-independence parties 'learned' to reject an out-group identity (see, e.g., Sczepanski 2023). The data also reveal an embrace of Europeanness among those in the pro-independence parties, with SNP members barely separated from the decidedly internationalist Greens on this yardstick. We return later to the question of whether these 2016– 2017 survey data reflect conversion of SNP members to Europeanness.

The evidence that British identity is felt only weakly or not at all by many party members, especially in the SNP, raises the question of why Britishness has so little appeal. It could be that the national symbols or political institutions of the United Kingdom are meaningless to respondents or that they have little sense of affinity with other parts and people in the United Kingdom. The SNP was often accused of being anti-English in the past by critics, though far less so in recent times, with little evidence of anti-English sentiment among party elites. It was virtually nonexistent in respondents' open-ended explanations of why they joined these parties. On the other hand, perceived divergence in electoral behaviour between England and Scotland, long visible in election results and then underlined by the EU referendum, is part of a wider narrative of difference that extends beyond constitutional arrangements to an argument that the people of the two nations want different things politically (e.g., McIntosh et al. 2004).

When asked which nationalities respondents had most and least in common with (Table 6.2), it was found that around one in five SNP members stated that they had least in common with English people. By contrast, very few Greens said the same, and that gap between the parties persists even accounting for Scottish Green members being proportionally more likely to be English themselves. Members of both parties, however, were more willing to see commonality with Welsh people and strikingly more so with Irish people. This is likely due to closer family and historic ties with Ireland in many instances.

A quarter of respondents, especially Scottish Green members, were reluctant to rank or compare nationalities. This finds an echo in Table 6.3, which reports responses to a question asking whether members subscribe to various ideological identities. A large difference between the parties is that very few Greens, including the recent joiners, consider themselves nationalist. What is harder to assess is whether the roughly three in five SNP members who identify as nationalist is low or high. This reflects a long-standing issue within the SNP on the nature of the party and its relationship with the national movement. The results in Table 6.3 suggest Greens have negative associations with nationalism, attitudes which SNP members are unlikely to be oblivious to it. The SNP's leaders emphasise that the 'N' stands for 'National' and are irritated by opponents referring to the party as the 'Scottish Nationalist Party'. One SNP MP interviewed explicitly rejected the

#### 96 *Identities, ideologies and independence*


*Table 6.2* Peoples with whom members feel most/least in common by party

*Table 6.3* Ideological identities by party and cohort


label 'nationalist' (Interview no. 36), and Nicola Sturgeon followed her predecessor in expressing discomfort with the term (MacDonnell 2017). We might expect that those who accept the 'nationalist' label may also accept traditional nationalist stances. For instance, those members (of each party) who described themselves as 'nationalist' were twice as likely to report feeling least in common with English people.

There was an appreciable *negative* association between identifying as 'nationalist' and identifying as 'internationalist'. Indeed, 'nationalist' was negatively correlated with every other identity in Table 6.3. This suggests that it denotes commitment to independence as an end in itself rather than a means to another end. Those joining the parties for instrumental reasons, such as getting rid of nuclear weapons, were less comfortable with calling themselves nationalists. There is also some evidence of this from the cohort comparison in the table. Those who joined the SNP after the referendum were appreciably less likely to choose 'nationalist' but more likely to identify with a range of other identities. Among Scottish Greens, the parallel is with environmentalism and ecologism: both were less popular among new joiners, who were likelier than existing members to identify as 'socialist' or 'feminist'. None of these cohort differences is especially large but the pattern seems clear. The referendum campaign associated independence with a progressive outlook – a kind of constellation of left, environmental and internationalist positions (Hepburn and Rosie 2014) – and brought those with that outlook into both parties. Each party looks slightly broader, rather less sharply focused on its core (nationalist or environmental) identity. Nonetheless, the core identity remains 'nationalist' for SNP members and 'environmentalist' for Scottish Green members. The influx made two different parties a little more similar.

 Left-right offers an unrefined but useful summary of this apparent shift in a progressive direction. Respondents were asked to place themselves on a scale from 0 (left) to 10 (right). The results are outlined in Table 6.4, again broken down by cohort but this time also reporting the overall results, comparing the parties as a whole with each other and with their voters using 2016 Scottish Election Study (SES) data. That first comparison confirms that Green members' selfidentities lie appreciably to the left of the SNP's. Two-thirds of Greens were in the leftmost three categories, compared to just one-third of SNP members, and virtually no Green members chose the midpoint let alone a position to the right of it. This is not because the SNP membership is particularly centrist – their mean position of 3.1 is well to the left of the midpoint and more than a point to the left of the average SNP voter. Half of SNP voters in 2016 placed themselves either at or to the right of the midpoint. With both parties, as is typical, the members believe themselves to be more radical – and thus less diverse ideologically – than

voters. Table 6.4 gives a sense of whether the post-referendum surge shifted these parties to the left, according to self-placements. There was no shift among Scottish


*Table 6.4* Left-right self-placements by party and cohort

Greens; the membership was already far to the left, and much more so than the SNP. With the SNP, there is some sign that the new members see themselves as more left-wing, but the difference is limited – the cohorts are just 0.3 points apart on the 0–10 scale. The new members were left of centre, as was the existing membership. However, the SNP identity profile in Table 6.4 is much more left-wing than in our 2007–2008 survey, when around 50% of members placed themselves to the left, 25% in the centre and 25% to the right. Of the processes that could have driven this change, the surge into the party appears to have made only a minor contribution. Since the earlier survey, the SNP appears to have disproportionately recruited left-wing identifiers and lost right-wing identifiers, or existing members' identifications may have changed.

#### **Attitudes and policy preferences**

A recurring feature of the referendum campaign was the association between Scottish identity and left-wing or 'progressive' politics. However, as noted earlier in the chapter, this often applies more to general left-wing identity than to specific left-wing policy opinions. Both matter as a party is constrained by its members' ideological self-image and by their policy preferences. The latter are the focus of this section. Some issues loom large enough in politics that it is worth examining members' positions on those specific questions. Others may be less salient in everyday politics but useful in assessing members' broader ideological or value orientations. As usual, we are interested in comparing both within and across the parties, paying particular attention to whether the referendum surge shifted the ideological base of the memberships and to how much common ground there is between the two parties – a pertinent question given their record of working together in Scotland's political institutions.

We begin with three policy attitude batteries intended to assess broader values or ideological positions (Figure 6.2).1 The overall scores are measured on a scale from 1 to 5, coded so that the lower score is the more progressive – left-wing, environmental and liberal – position. The three key results are familiar from the earlier analysis of ideological identities. First, Scottish Green members are more progressive than SNP members on all these measures. The gap is predictably largest on the environmental scale but appreciable on all three. Second, this is not because the SNP members are especially centrist. The SNP mean scores are invariably well below the midpoint score of 3. Third, the post-referendum influx of a large group of broadly progressive independence supporters meant that the SNP membership shifted slightly in a left-liberal direction, while the Greens, already quite far in that direction, shifted slightly back the other way. Again, then, the referendum brought the two parties a little nearer to each other while leaving the differences still clearly visible.

The fact that the parties worked together in the independence referendum and would later agree the SNP–Green deal that led to two Green MSPs becoming junior ministers makes these differences particularly interesting. Although this is a question that matters only as far as the party elites feel constrained by the opinions

*Figure 6.2* Mean scores on three value scales by party and cohort

of their members. On a range of statements, about redistributing wealth or endorsing immigration or the threat posed by climate change, there was widespread consensus. The difference was just that SNP members were more likely to say 'agree' while Greens said 'strongly agree'.

Points of disagreement include 82% of SNP but only 32% of Scottish Green members agreeing that 'the Scottish government's priority should be to boost economic growth'. The statement that 'taxes on business should be cut to strengthen Scotland's economy' was supported overall by SNP members but was decisively rejected by Greens. There were also hints of an SNP more divided on liberalauthoritarian issues than either the party's elite or the Green membership: 25% of SNP but just 5% of Green members disagreed that 'the death penalty is never justified'. SNP members supported Scotland's membership of NATO, while Green members on balance opposed it. Although this is in line with their parties' respective policies, it is worth noting that SNP members had supported NATO membership in the 2007–2008 survey, at a time when their party was still committed to withdrawing. Finally, a clear majority of Green members claim to support the transfer of powers from the Scottish government to local authorities, while members of the SNP were split down the middle on this issue.

One area of consensus both within and across the parties is Brexit. Figure 6.3 presents the percentages reporting a Remain vote in the June 2016 referendum, broken down by party and cohort. Those who did not vote – typically because

*Figure 6.3* Remain voting by party and cohort and compared with 2016 voters

they were ineligible – are excluded. We compare the parties' members with their voters as surveyed for the 2016 Scottish Election Study. This means working with EU referendum vote intentions because the SES post-election survey was conducted around a month before polling day in June. (This is not ideal, but Remain vote intentions in this May survey were the same 62% as in the eventual vote in Scotland.)

All categories of members reported voting overwhelmingly to Remain. Virtually no Green respondent voted Leave, while around one in eight SNP members did so. As the voter comparison makes clear, the SNP's electoral base contained a much larger contingent of Leavers: indeed, rather more proportionally than in Scotland as a whole, perhaps reflecting the party's pre-referendum electoral strength in the more Eurosceptic north-east. The SNP's existing members were already more pro-EU than the party's voters; the surge widened the gap further. And the minisurge in the SNP membership in 2018 brought in another tranche of members who were overwhelmingly (89%) Remain and were even more likely than respondents in the 2016–2017 survey (64% compared to 56%) to agree that 'we should look to increase the number of immigrants coming into Scotland'.

While we would expect members of these two parties to remain committed to independence, there is a subtler gauge of the relative importance of the two referendum issues. This comes via a question asking people to rank various options for governing Scotland. The results for both parties – first preferences and second


*Table 6.5* First and second constitutional preferences by party

preferences by first preferences – are shown in Table 6.5. When it comes to the SNP, the strength of support for independence within the EU is hardly surprising. It might be thought more conspicuous that one in seven SNP members would prefer 'devo max' to independence, and further devolution was more popular than independence outside the EU. It is with the Scottish Greens that the overwhelming first preference for independence is more striking, given the party's different history. Whether this reflects independence sceptics being won round or simply leaving the party is hard to say, but the constitutional consensus is striking.

Among the SNP members favouring the party's official line, second preferences split 56:41 in favour of independence outside the EU. Among those whose first choice is independence outside the EU, second preferences go almost 2:1 in favour of independence in the EU. In both cases, then, there is appreciably more willingness to compromise on the Brexit issue than on independence. The reverse is narrowly the case among Scottish Greens. Of the large majority favouring independence in the EU, a majority – 53% – choose further devolution over independence outside the EU. This suggests limited appeal of independence on the wrong terms.

A significant change in first preferences of SNP members is evident since a parallel question in the 2007–2008 membership survey – independence outside the EU then enjoyed 22% support. This may suggest that Eurosceptics are likely to be over-represented among those leaving the party or that some members have changed their view on the EU since that earlier survey.

#### **Independence**

Figure 6.4 summarises responses to a question measuring strength of support for independence on a 0–10 scale. A slightly more refined cohort comparison allows us to pick out those who joined in the immediate post-referendum surge (and distinguish them from those who joined in 2015 or later). The first point to make is that, while independence may have been the first preference for just as many Scottish Green members as SNP members (see Table 6.5), it is not as deep a commitment for the Greens. The differences are slightly exaggerated by the compression of the vertical axis to focus only on the top half of the scale, and we should not lose sight of the fact that support for independence is widespread in both parties. Nonetheless, the gap is clear. Equally clear is that the referendum surge brought in a higher proportion of independence-driven members to the Scottish Greens – but the Greens remain less strongly committed to independence than those who joined the SNP. Surge joiners of the SNP are indistinguishable on this criterion from the existing members, who were either strongly driven by independence all along or became so during the referendum. Finally, there are signs that those who joined after January 2015 have a slightly weaker commitment to independence, although the downturn is very slight in the case of the SNP. In the Scottish Greens, those who joined at the height of the surge are more clearly the most pro-independence cohort.

*Figure 6.4* Strength of support for independence by party and cohort


*Table 6.6* History of support for independence by party and cohort

 The surge in membership after the referendum followed a similarly unexpected surge in support for independence during a campaign which for a long time had looked like an easy win for No. This raises the question of whether the membership surge came partly among those who had been won over to independence or whether the referendum was instead a trigger for party joining among those who had been long committed to independence. Table 6.6 shows, by party and cohort, the proportions of members who had either always supported independence or been won over (the analysis excludes the small minority placing themselves below the midpoint on the support-for-independence scale).

Scottish Greens are much more likely to be converts to independence, having previously been either ambivalent or opposed. And there is very little by way of cohort differences in that party: long-standing Green members were not all long-standing supporters of independence. In the SNP, meanwhile, surge joiners look more different from existing members and, while the timing of their conversion is not explicit, these data suggest that the referendum campaign won some of them over to independence and then to party membership. Of those who joined after the immediate surge, a significant minority (17%) had previously opposed independence. This suggests that the campaign first attracted those who had been tempted but unsure, and then the surge – and post-referendum politics more generally – was able to attract not just doubters but also some erstwhile opponents.

Why do members support independence? At several points in this chapter, we have contrasted independence as an end in itself and as a means to some other end. The latter assumes support for independence which is based on instrumental reasons. These could be about specific issues and policies or about Scotland's broader ideological direction. There are a number of potentially instrumental reasons based on an assumption that an independent Scotland would be more likely to deliver certain policies.

A range of instrumental possibilities were captured in an open-ended question in which respondents were invited to give up to three reasons for supporting independence. The question was only asked of those who had indicated a strength of

#### 104 *Identities, ideologies and independence*


*Table 6.7* Reasons for supporting independence (open-ended responses coded) by party

support for independence at 6 or above on the 0–10 scale used for Figure 6.4. The responses were coded to allow for a rich variety of responses, and this resulted in 27 categories, which are grouped under broad headings. Table 6.7 presents these findings, based on the first reason provided by respondents.<sup>2</sup>

 Among SNP members, by far the most cited reason was a variation of Scottish control of Scottish affairs, classified as 'national motivations' in Table 6.7. 'Make our own decisions' was its most common expression. There were many references to 'self-determination'. This was also the most common reason among Green members, but they cited a variant of local decision-making almost as often as a variant of self-determination. Some used the term 'subsidiarity'. One made the vivid point that 'a small fire is easier to manage than a large fire'. Greens were less likely than SNP members to refer to Scotland's history, potential or nationhood. Several SNP entries in that category simply stated that 'Scotland is a nation', with the implication that nations should be states sometimes added but often left implicit. Green members were the more likely to point to broad ideological reasons for supporting independence, such as a belief in social justice.

Alongside these arguments for independence, there were also references to what Scotland should separate *from*. Among these, there were more references to the Westminster system and to the Union than to Conservative governments, though several responses implicitly conveyed anti-Conservative sentiment. The rejection of Westminster and the Union was at least as common among Green members.

Basing Table 6.7 on the first reason given probably underestimates the ideological and instrumental motivations in the second half of the table. Respondents were more likely to cite core nationalist principles first and then to proceed to the benefits of independence. However, even in the additional reasons given by most respondents, there was the same clear preponderance of what were clustered in the table under the heading of 'national motivations'. These data suggest independence is viewed more as an end in itself than a means to an end. Instrumental or issue-specific motivations for independence were less often articulated by the parties' members. The removal of Trident was rarely cited, and the most common entry was a general assertion that independence would make Scotland wealthier or in some way make for a 'better future'. When an ideological reason was primary, it was likely to be social justice or equality or a reference to Scotland as being more left-wing or simply having 'different values'. Barely 1% of Green members specifically cited environmental factors as a reason to support independence.

In both parties, there were some differences between existing members and those who joined after the referendum, whether in the initial surge or from 2015 onwards. Those who joined post-referendum were less likely to report 'traditional' nationalist reasons for supporting independence and likelier to report an ideological purpose. Three quarters (75%) of existing SNP members offered 'national' justifications for their support of independence compared with 59% of the surge members. In the Greens, 19% of existing members cited broader ideological reasons for supporting independence, but 29% of the surge members did so. This makes sense given that the referendum campaign had involved unusually detailed ideological discussion of arguments for and against independence.

In terms of intra-party politics, members' reasons for supporting independence perhaps matter less than the urgency with which they want to achieve this goal. The question of when to hold – from one perspective, when to risk – a second referendum has been debated since 2014. For an SNP whose post-referendum leadership was clearly cautious on these questions, a membership impatient for independence might create significant tensions. We asked respondents about the timing of a

*Figure 6.5* Preferred timing of holding second independence referendum by party

second referendum on a scale from 0 (as soon as possible) to 10 (only when confident of victory). The results are shown in Figure 6.5. The mean score for each party was over 7 suggesting a cautious position. SNP members are on average no more impatient than the Greens but they are more likely to be found at either end on the scale, at both 0 and 10. Those at the left-hand end of the scale represent that part of the SNP membership that is often referred to as pressuring the party leadership into pushing harder for a second referendum. It is a subset that exists but is hugely outnumbered in a broadly pragmatic membership.

The surge joiners were not demanding an immediate referendum. Indeed, if we calculate the mean scale score by cohort, there is no statistically significant difference in either party between those who were members before the referendum and those who joined in the surge. The exception to this is that the members in the newest cohort, those who joined in 2015 onwards, were slightly keener on an earlier referendum. Their mean scores were 6.73 in the SNP and 6.99 in the Scottish Greens. This may reflect the different contexts. The later joiners were more likely to cite Brexit as their reason for supporting independence, and those who cited Brexit were on average the most in a hurry (as measured on the scale presented in Figure 6.5).

Table 6.8 shows the almost unanimous rejection of the notion that the 2014 referendum settled matters for a generation. The SNP members feel more strongly, but Scottish Greens are also almost all agreed. For a clear majority in both parties, opinion poll support would be a sufficient condition for calling another referendum. These responses come from a survey after the Brexit referendum, which changed the circumstances and may have affected views.


*Table 6.8* Opinions on basis for a second independence referendum by party and cohort

#### **Conclusion**

Scholars of nationalism around the world typically refer to it as a 'thin-centred' ideology, meaning that it has a core but lacks any specified view on other matters such as redistribution of wealth or environmentalism (Freeden 1998: 750). In this chapter we explored the ideologies of the two parties with particular emphasis on the place of other ideologies in the SNP and the place of nationalism in the Scottish Greens. A key problem in analysing survey research on this topic is that many respondents fail to distinguish between core beliefs and themes that relate to campaign messages, thus ends and means become entangled. Those already persuaded of independence, and especially those campaigning for it, are given reasons for supporting independence beyond the core objective. These will include instrumental reasons to support independence. This comes across very clearly in our data.

When we compare the SNP membership's attitudes in our previous survey with the SNP after the surge, we find that the party remains much as before. It would be extraordinary if it had not changed at all, but the change does not conform with any suggestion that the SNP was transformed by the surge. Changes were small and tended to reflect the messages that dominated the Yes referendum campaign. It is unsurprising that this should have happened given the intensity and duration of the campaign which had presented the SNP with the most sustained coverage of its views (and leading members) in the party's history. It would be more surprising had the SNP attracted a massive surge in members who included people who supported alternative policies and ideological positions. The notable aspect of the attitudes of SNP members was the willingness to provide views across a range of matters rather than stick with an old formulation that these were matters to be determined by the people of Scotland. Scottish nationalism as articulated by the SNP may be a thin ideology, but it is one to which other ideological positions have been firmly attached.

The Scottish Green Party did not have as high a profile as the SNP in the referendum. Nonetheless, in such a long campaign and given the high levels of engagement by senior members of the Scottish Greens it too had an exceptional opportunity to project the party's views. Patrick Harvie, Scottish Green co-leader, was a very active and articulate campaigner. The extent to which the public were previously aware of the party's view on Scotland's constitutional status is unclear, but the referendum gave it an opportunity to sell its message and relate its environmentalism to the constitutional debate. Given its members could be expected to have joined the Scottish Greens because of its environmentalism, we find that the Greens are also a party with a much broader ideological base. In each case, it is clear that the parties have a core, but these cannot be detached from other attitudes and ideological positions.

Surge recruits might, in ideological terms, be described as 'the same, only slightly more so'. The new recruits to both parties share the same constellation of left-wing attitudes, economically left-wing and socially liberal policy preferences and an internationalist outlook including support for EU membership. But this barely shifted the left-leaning consensus in each party. In the SNP, on issues like gay marriage, there is a small constituency of support for the conservative views defended by Kate Forbes in the 2023 leadership election, but these are minority positions and are typically held much less strongly than the desire for independence that brought members to the party. The one significant change was the proportion of Scottish Green Party members who now supported independence. Previously, there were many members of the Scottish Greens who opposed independence. The surge in membership post-referendum brought not only large numbers of new members but new members who were unambiguously in favour of independence.

#### **Notes**


#### **References**


Evans, G., A. Heath and M. Lalljee (1996) 'Measuring Left-Right and Libertarian-Authoritarian Values in the British Electorate', *British Journal of Sociology*, 47 (1): 93–112.

Freeden, M. (1998) 'Is Nationalism a Distinct Ideology?', *Political Studies*, 46 (4): 748–765.


### **7 Activism and action repertoires**

#### **Introduction**

This chapter explores party activism in the SNP and Scottish Greens – what the members do within and for their parties. Previous studies have revealed that few members of parties are intensely active, and passive or minimal engagement is the norm (Scarrow 1996; Seyd and Whiteley 2004; van Haute and Gauja 2015). Heidar (1994: 72) noted that 'party members rarely participate in party activities'. This is true even when there are considerable opportunities for members to become involved and parties encourage participation internally. Activism, though, is multi-dimensional – it takes many forms – and what is meant by the term is sometimes contested. Activism can vary by party and according to election cycles, with members more likely to be active during an election campaign (Cross 2015: 62). Some party activists stay highly engaged for a political lifetime; others are active intermittently; many 'burn out' and are replaced by new cohorts. The basic distinction between members and activists – what Duverger (1954: 90–91) termed 'militants' – appears unrefined, unable to capture the variety of possible contributions to party life.

This chapter offers a detailed picture of activism in the SNP and Scottish Greens, that is, the extent to which, and the different ways in which, the members get involved and connect with their parties. We examine internal party activities like attendance at party meetings and helping with party fundraising events. Our focus is on how the 2014 Scottish independence referendum and membership surge impacted on the nature of activism within the two parties, once again comparing the two parties' members and the pre- and post-referendum membership cohorts. It is important to remember that the new members joined the parties following an unusual campaign – with its novel forms of grassroots activism – and we aim to understand how these events shaped the patterns of party activism. Were the new members more (or less) participatory than traditional members? Were they active in ways which suggest a carrying forward of the movement repertoires of the referendum period?

Chapter 5 on motivations for joining suggested that the surge members might be more comfortable with individualised forms of behaviour like digital participation and less keen on traditional face-to-face activism, and this expectation is broadly

#### DOI: 10.4324/9780429446559-7

This chapter has been made available under a CC-BY 4.0 license.

substantiated by the analysis in this chapter. This all meant the parties experienced and gained from an increase in the numbers of activists, but active members represented a smaller proportion of the memberships. The chapter includes an account of the characteristics of activists, profiling who becomes active and why, including analysis of their identities and opinions. We find little evidence that the parties' activists are motivated by radical and uncompromising views which are unrepresentative of the wider memberships or challenging to the party leaderships.

#### **Members and activists**

Active members perform important internal- and external-facing roles within their parties. Activists run local parties, raise funds, attend party conferences, stand for party positions and as candidates, and they influence the direction of party policies and strategies, although the *extent* of this influence is debatable (a subject we address in Chapter 8). Activists represent a public face of political parties by interacting with voters in their local communities. At times, party activists have been viewed as a 'troublesome layer' (Mair 1994: 16). This is due to a perceived tendency, noted by May (1973), that activists are uncompromising and more radical on some issues than leaders or passive members, although empirical studies have not always found evidence of this claim.

Some parties have outside sources of funding and campaign support, suggesting that activists may be less needed than in the past. Research has shown, for example, that parties sometimes rely on non-member supporters to perform campaigning roles, such as delivering election materials (Fisher et al. 2014; Scarrow 2015; Webb et al. 2017). Nevertheless, decades of scholarship suggests that activists remain indispensable to parties (Whiteley and Seyd 1998, 2002; Webb et al. 2020). Without activists, who would organise party fundraising events, or canvass electoral support, or come forward as party officials and leaders?

Activism has various dimensions. In their Labour study, Seyd and Whiteley (1992: 87–94) identified contact activism (contact with other party activists and attendance at meetings), campaigning (involvement in internal and external campaign groups) and representation ('elite party activists' including office-holders and elected representatives). In our own earlier study of the SNP, we proposed that activism took three forms: grassroots activism like attending party meetings and electioneering (traditional), holding office in the party (office-holding) and contributing money to the party (financial) (Mitchell et al. 2012: 91). A further distinction has been made between internal activities, which are about maintaining the party organisation, described by Gallagher and Marsh (2002: 81) as a party 'talking to itself', and external activities, the more public-facing roles, mainly election campaigning – delivering leaflets, displaying posters – but also wider community involvement and some movement types of activity such as attending a rally or march. Of course, many of these activities will be interrelated, such as when local meetings involve planning for elections.

Traditional classifications require revision because the modern party member is likely to be active in different ways from before. Most obviously, there has been an increase in digital involvement (Margetts et al. 2016; Gibson et al. 2017; Mellon and Prosser 2017; Dommett et al. 2021). There are many activities that now take place online, some of which connect members with local parties, others with the national party; for instance, members take part in online leadership selections and make financial donations to parties, a form of participation that might be described as 'armchair activism'. Recent research has highlighted that activism varies in intensity and commitment and has taken on these new forms (Bale et al. 2019; Webb et al. 2020). Bale et al. (2020: 99–100) describe three types of activism: a low-intensity type of activism which can be done from home, including using social media or displaying a poster; medium-intensity activism such as attending meetings and delivering leaflets; and high-intensity activism, which includes organising, canvassing and standing as a candidate.

It is possible that activism resulting from the referendum experience had a different character or added component, namely a movement dimension. Activism is closely connected to what social movement scholars call 'action repertoires', the tools and actions available to movement organisations and participants (McCarthy and Zald 1977; Tarrow 1998). In parallel with party activism, the action repertoires of social movement campaigners have expanded to include the digital sphere (Bennett and Segerberg 2012). Activities can be extremely diverse and vary according to movements and campaigns. As we established in Chapter 3, many of our survey respondents had engaged in movement politics in the past, and in 2014 they observed and/or experienced a referendum campaign brimming with movementstyle events and activities. However, few had participated in face-to-face activities during the campaign. Most engaged more loosely with the campaign, and it was existing SNP members who dominated the core campaign activities like canvassing and delivering leaflets. As we saw in Chapter 5, reasons for joining were driven more by the aim of policy change (independence) and ideology than a desire for participation. Nevertheless, those who would go on to join once the referendum was over did feel part of a movement for change.

It has been argued that activism responds to participatory opportunities offered by more democratically organised parties (Heidar 1994; Diamond and Gunther 2001; van Haute and Gauja 2015). This relationship is not clear and made more complicated by internal democracy that takes place online. Work on parties with a largely digital infrastructure reveals low to moderate levels of participation (Hartleb 2013; Gomez and Ramiro 2019).1 We might expect there to be higher levels of activism in successful and governing parties, with enhanced feelings of efficacy motivating activism. Again, the evidence is lacking: long-term governing parties often experience a decline in their activist base. Some research suggests that societal change leads activism to fluctuate over time. In the 1990s a 'spiral of demobilization' was identified by Whiteley and Seyd (1998), but two decades later Bale et al. (2020: 99) reported a rise in the proportion of party members thinking of themselves as 'more active than five years ago', a perception that was most pronounced in Labour and in the SNP.2

All considered, we would expect both the SNP and Scottish Greens to be parties where members perceive opportunities to get involved and actively

 contribute. It is worth noting that very few empirical studies support the idea that members of green parties are significantly more active than other parties' members (Bennie 2015; Bale et al. 2020; Rüdig and Sajuria 2020). However, studies identify the SNP as a party with a participatory ethos and an active membership compared to other parties (Brand 1992; Mitchell et al. 2012; Bennie 2015; Bale et al. 2020). Our interest lies in the new generation of members in the SNP and Scottish Greens.

#### **Traditional activism in the SNP and Scottish Greens**

A rise in the absolute number of activists does not necessarily represent an increase in the proportion of members who are active. This critical distinction was recognised by our interviewees, as exemplified by one who noted: 'There's no sign that increased membership has been matched with increased levels of activity. The vast majority of members who have joined don't come to meetings' (Interview no. 10). There was a widespread perception of increased activism locally but a decrease in the activity rate of the average member. Another of our interviewees observed that new members 'were not necessarily keen to become active' and identified a group they called 'sleepers', those 'not active in elections for the most part but are waiting for the next referendum' (Interview no. 1).

This is consistent with previous research on new recruits to parties. It takes time to be socialised into the way of activism, and that socialisation cannot happen without engaging with other active members. Our previous research showed that the SNP enjoyed a vibrant activist base and that the party could rely on large numbers of members to support its election campaign activities. Nevertheless, as with other parties, only a minority of members would ever be 'very active', and new members were especially inactive (Mitchell et al. 2012: 95). This does not preclude the possibility that the post-referendum joiners were inclined to be active in *different* ways because they joined on the back of an unusual referendum experience.

Table 7.1 presents our more recent data on attendance at a local party meeting, a common measure of party activism. Following the surge in membership in both the SNP and Scottish Greens, large numbers of new members attended party meetings. Some queued to gain entry in the very early days of the surge. The parties indicated that this was short-lived and attendance at meetings stabilised at a more


*Table 7.1* Attendance at local meetings in last year by party and cohort

manageable level soon after, one SNP interviewee reporting that 'huge numbers joined and lots came to early meetings but fell away' (Interview no. 15).

Three points are notable in Table 7.1. First, the majority of respondents report not having attended a local meeting in the previous year, a finding that resembles those of previous studies. Question wording was different in the 2007–2008 SNP study, but we found then that 53% of SNP members never attended these meetings or attended less than once a year. In 2002, 53% of Scottish Green Party members reported that they hadn't attended a local meeting in the previous year. Second, the Green members appear marginally less participative, but the overall pattern of engagement is very similar in the two parties. Third, the newly recruited members (in both parties) are markedly less likely to have attended a meeting than those who were members prior to the referendum.

Table 7.2 reports time spent on party activities (on average each month). Most members spend no time at all on them. There is a remarkable parallel between the parties, with the SNP members appearing slightly more active than the Greens. And in each party the new members spend less time on party work than established members. This table includes findings from previous studies, and these closely match the more recent study, although there is a suggestion that the SNP in 2016–2017 had a *slightly* more active membership with fewer respondents saying they spent no time at all on party activities. The overall pattern of activism is very similar at the different time points, and among the 2016–2017 respondents the pre-referendum members' more intense commitment to party activities stands out.

We next examine how the members perceive their own level of party involvement – whether they describe themselves as very active, not at all active and so on (Table 7.3). The established pattern is repeated: the average party member does not think of themselves as active; the two parties are alike, but the SNP members see themselves as a little more active; and the pre-referendum members are more likely to describe themselves as active than newly recruited members (the much bigger group).


*Table 7.2* Time spent on party activities by party and cohort


*Table 7.3* Self-perceived activism by party and cohort

This table on self-perceived activism signals that the proportion of activists in the SNP has fallen since 2008, but the more objective measure of activity in Table 7.2 suggested little difference between 2007–2008 and 2016–2017. One way of reconciling the two is to suggest that a highly participatory referendum changed members' benchmark for what it means to be very active. It is possible that the same amount of involvement is now less likely to be described as 'very active'.

#### **Other ways of being active**

The findings so far closely parallel previous studies, pointing to a highly predictable pattern of activism, rather than a transformation brought about by an unusual referendum campaign. This is hardly a picture of a radically energised activist base. However, there remains the possibility that the new members might have been attracted to other types of activities. A reluctance to attend local meetings might be perfectly understandable if a member's motivation for joining was a desire to maintain the excitement of the referendum. We can examine a wider range of campaigning activities, which include traditional, on-the-ground and in-person activities, but also those which can be undertaken from home with no face-to-face interaction with others – in the case of online discussion and donations, from an armchair or sofa. We asked whether the members had taken part in any campaigning activities since the 2014 referendum (either in an election or in the EU referendum). The results are displayed in Table 7.4.

On some activities, SNP respondents score more highly than Greens – such as displaying a poster and canvassing – but on others, such as delivering leaflets or donating to party funds, a higher proportion of Greens report having taken part. Overall, the most regularly undertaken activities of the members (both SNP and Green), and in descending order, are displaying posters, discussing elections via social media or online forums, attending meetings, making financial donations and delivering leaflets.

Poster displays are by far the most regularly reported of the behaviours – even more than cyber-activities. These are associated with traditional party campaigning, at its peak in Britain in the 1950s and 1960s, when labour-intensive activities were common and party colours were displayed prominently in local communities. Over the decades, there was a decline in these activities. Voters became less

#### 116 *Activism and action repertoires*


*Table 7.4* Campaign activities since 2014 by party and cohort

attached to parties and more averse to publicly declaring commitment to them. As a result, posters and visual displays have become a less common sight in elections, with much of the colour fading from campaigns. <sup>3</sup> The prominence of posters in windows in the 2014 referendum campaign suggested a kind of renaissance of a traditional form of electioneering. The Scottish Referendum Study revealed that 29% of Yes voters (but just 8% of No voters) displayed a poster during the campaign. As seen in Table 7.4, this compares with over 50% of those who would go on to join the SNP and Scottish Greens and two-thirds of existing SNP members.

Poster displays have been widely reported by other studies, but they are generally regarded as a relatively unimportant, low-intensity form of behaviour. In the early 1990s, 65% of Labour members had frequently displayed a poster (Seyd and Whiteley 1992: 95); but only 19% of Conservative members (Whiteley et al. 1994: 258). More than two decades later, Bale et al. (2020: 100) found, again, that Conservative members were least likely to have displayed a poster in the 2015 and 2017 general elections, while over 50% of Labour and SNP members reported doing so. The most prolific were SNP members in 2015, with 68% publicly declaring their political opinion in a window display (Ibid.). The high incidence of this activity among SNP members points to a legacy of the referendum campaign when posters and other visual displays like stickers and badges were widely observed. It suggests that allegiance to the SNP among its membership is entirely compatible with public displays of support.

We are particularly interested in whether the post-referendum joiners show signs of being active in different ways from established members. Table 7.4 suggests that the pre-referendum members are generally more active. There is a pronounced gap between the SNP's pre- and post-referendum members in displaying posters. Studies have shown time and again that new recruits are less active, and the members in our study appear to fit this expectation. The exception to this is online participation. The new members, particularly in the Greens, are more likely to engage in online activity, reflecting the rise of social media evident during the referendum (see Shephard and Quinlan 2015).

In Chapter 5 on joining, we reported on how members *wanted* to be involved in their parties. Online support was far and away the most attractive to the members – around seven in every ten said they wanted to support online campaigns/petitions. The surge joiners were the keenest of all to take part in these activities, and they were considerably less attracted to conventional activities. These preferences are reflected in the actual levels of activity outlined here. All members were attracted to the kinds of participation seen during the referendum campaign like 'supporting online campaigns'. The new members, though, were less involved in conventional party activities. We also asked about the perceived 'effectiveness' of these different activities and found parallel results. The newest members were the more sceptical about the impact of some conventional activities like election campaigning and financial contributions. SNP members were considerably more likely than the Greens to view online participation as efficacious.

We attempted to gauge the social media engagement of the members more closely by asking respondents whether they talked about politics on a social media platform (Table 7.5). Post-referendum Scottish Green members look the most politically engaged via social media. A caveat about our data gathering is worth reiterating here. Our data were generated via an internet survey, and it is likely that we recorded the responses of some of the more digitally active members. On the other hand, we know that a large majority of members are connected digitally to their party.

We can also examine the proportions of members who take part in internal party votes, a key 'power' of party members which is now conducted almost fully online (Table 7.6). Most of our respondents said they had not participated, and again we see the discrepancy between pre- and post-referendum members, with the established members appearing more involved. In the case of internal elections, new members are bound to be less knowledgeable about party candidates and internal party procedures. However, the table points to relatively high levels of member engagement: 46% of SNP members and 38% of Scottish Greens said they had taken part in an internal vote since joining. These figures are higher than those generally reported by the parties. We return to this theme in Chapter 8 and in the conclusion of the book.


*Table 7.5* Use of social media e.g. Twitter, Facebook, Instagram by party and cohort

#### 118 *Activism and action repertoires*


#### *Table 7.6* Member participation in internal party votes by party and cohort

#### **Dimensions of activism**

It appears that the proportion of the parties' memberships who were active declined following the surge, certainly if our focus is on conventional forms of activity, and this is consistent with the opinions expressed by many of those we interviewed. Nevertheless, activism is multi-faceted, and the question to be addressed in this section is whether intra-party participation became even more multi-dimensional with the growth of membership. Members can be active in very different ways, and the contribution of an individual member can change over time. We are interested in whether activism was in any way reshaped by the referendum and the large intake of new members.

We can examine this more closely through factor analysis of the correlations between the various forms of activity. This provides a better understanding of the patterns of engagement and the relationship between traditional party activism and other less institutionalised forms like participating in online discussion. As noted earlier, our previous study of the SNP identified three core dimensions of activism – traditional grassroots activities, holding party office and making financial contributions. Analysis of data from the more recent study, displayed in Table 7.7, shows an even more simplified picture. With both parties the range of activities could be accounted for by two underlying factors. The first of these might be labelled 'traditional' or 'conventional' party activism and includes attending meetings and functions, canvassing voters, holding office within the party and voting in internal elections. The second factor involves displaying posters and discussing party politics online. The distinction seems to be less about traditional versus non-traditional and more about collective versus individualised or atomistic forms of activism. This is even more clearly the case with the SNP where financial donations also loaded on the second factor. With the Greens, donating money correlated more strongly with the standard activities. This suggests that Green donations are more likely to come from stalwart activists keeping the party afloat, while SNP donations come from across the party including ordinary members.

On the basis of this analysis we calculated simple indices of collective and individual activism for each party. Figure 7.1 shows how levels of activity compare across the two dimensions across the parties and between pre- and post-referendum


*Table 7.7* Summary of factor analyses identifying dimensions of party activity

*Figure 7.1* Levels of collective and individual activity by party and cohort

joiners. There is clearly more activity of the individual kind. This is not surprising because these are low-intensity, low-cost activities. On average, around half of respondents reported engaging in these activities, compared to only around one in five respondents across the more traditional and more intensive activities such as attending meetings or canvassing voters. Amid the general tendency already observed for the new members to be less active, it seems that this is particularly the case with those high-intensity collective activities. This may be because these require longer socialisation within the party. It could also reflect newer members' preference for less traditional repertoires of activity. The differences are small, in any case, echoing our recurring theme about the non-transformational nature of the surges. The joiners resemble typical party members in terms of their activity – and their inactivity.

#### **Who are the activists?**

 Who are the barely a fifth of members who do the heavy lifting in parties? These members express views at meetings and conferences and contribute to shaping the image of a party. We consider the socio-demographic, attitudinal and past activity profiles of these activists in terms of collective and individual activism.<sup>4</sup>Table 7.8 reports the results for the socio-demographic variables, empty cells indicating no significant relationship. There are many blanks in the table, and relationships are typically weak even where there were significant differences. These variables can together explain little of the variation across members in activity levels. This is particularly true of the SNP, within which activists are strikingly representative of the party's membership. All of this echoes the findings from the Bale et al. (2020: 106) study of UK party membership, which found few demographic or socio-economic predictors beyond the tendency for middle-aged members to be more active. There is some sign of that when it comes to collective activity, but again the differences


*Table 7.8* Results of regressions predicting party activity: socio-demographic variables

are small and, in the case of the Scottish Greens, not statistically significant. We do not see the same age pattern when it comes to individual activity. Among Greens, younger members were more engaged in individual activities, perhaps due to the digital component. Another sign of individual activity reaching a different part of the Green membership comes from the gender findings, with women disproportionately likely to be involved. Collective activity in the SNP shows the more usual gender gap, although again it is not wide.

Education is the most consistent pattern, associated, although not strongly, with more of both types of activity. It might have been thought that the civic skills fostered by education are a more important resource for the collective and generally more demanding form of party activity (Verba et al. 1995). If so, then education may be playing a different role – perhaps digital literacy or a more engaged online network – when it comes to individual activity. But it is also possible that greater literacy in the conventional sense and confidence foster the kind of online discussions that are important for individual activity. Meanwhile, there is little sign that income or social class play much of a role. Finally, it is perhaps surprising that having children under 15 or other caring responsibilities does not appear to inhibit activism of either kind in each party. There is a tendency for disabled SNP members to participate less in collective activity, but there are signs that disabled members in both parties seek to compensate with more accessible forms of individual activity.

Many of the attitudinal variables explored in Chapter 5 are added to the model to identify how far activists are representative of the broader membership in terms of identities, values and policy preferences (Table 7.9). Again, empty cells mean


*Table 7.9* Results of regressions predicting party activity: identities and attitudes

*(Continued)* 

#### 122 *Activism and action repertoires*



no significant relationship between that attitude and levels of activity, and this is the case for the most part with national identity. Feeling British does not inhibit activity. There are also few strong effects of class identity apart from the suggestion that those who feel middle-class are particularly likely to get involved in conventional activity in the Scottish Greens. Feeling part of various movements encourages activity in both parties, as do some ideological identities, although here there are evident contrasts. Identifying as a 'nationalist' is not only far more common in the SNP, it also predicts activity. And identifying as social democratic rather than socialist predicts activity in the SNP. The unimportance of environmentalist and ecologist identities may be unsurprising among SNP members but might have been expected among Scottish Green members. These members may be more concerned about specific policies than identity: further down the table, the greenness of member's policy attitudes proves to have the expected positive effect on activity levels within that party. Interestingly, identifying as 'radical' predicts collective participation in both parties, suggesting that the intensity of views propels activism. Internationalism predicts activity across the board.

The results for left-right self-placement are striking given that, as we saw in the previous chapter, the new members in particular see themselves as to the left. This does not mean that left-wing members are *less* active; it simply means that there is no tendency for those to the left (or right) to be disproportionately active. The liberal-authoritarian, or social liberal-social conservative, dimension appears to do more to predict the levels of activity in the SNP. Those towards the more conservative end are less likely to be active. As we might expect, in the Scottish Greens support for environmental policies appears to spur both collective and individual activities, but these policy preferences are also positively associated with both types of activism in the SNP.

Activists are not more radical in ways that would undermine the party leaderships, as previous research on the SNP showed (Brand 1992; Mitchell et al. 2012). While strength of support for independence has a mild positive association with activity, there is no sign that those who want a second referendum more urgently are driving collective activity within the party though they are a little more likely to engage in individual activity.

A recurring theme in studies of political participation, especially party activism, is that socio-economic status and ideological motivations explain much but not everything. Activity also reflects opportunities, experiences and habits (Webb et al. 2020). Our data, presented in Table 7.10, suggest that family and other social networks do not play much role in determining activity. Prior membership of another party was the only significant predictor of activity in this category, and its main effect was to boost individual participation more, at least among that large minority of Scottish Greens who had previously been a member of a different party. By contrast, there is consistent evidence that those members more accustomed to political activity elsewhere, especially within voluntary organisations, were more active within both parties. The stronger predictors were involvement in local community or charity groups rather than the more ideologically driven organisations like environmental or other cause groups – which is echoed by the weaker impact of protest participation on party activity, though this does have a positive influence and especially for the collectively active Greens.

By far the strongest effects in this table – and in all the regressions – are those of independence referendum activities. There is a clear pattern of members replicating their referendum activities within the parties. Levels of individual activity during the referendum – putting up posters, discussing the referendum online, and so


*Table 7.10* Results of regressions predicting party activity: past activity

### 124 *Activism and action repertoires*

on – do nothing to predict collective activity levels in the parties at the time of the 2016–2017 survey, and there was little spill-over in the other direction from collective referendum participation to individual party activity. The type and quantity of participation in the referendum has been replicated by the party members. Having belonged to more groups within the wider independence campaign translates into less collective but more individual activity within the SNP.

The final three variables in Table 7.10 measure various motivations for party membership or activism. There is a contrast between the first two. Reinforcing the point about networks, those who joined partly to mix with like-minded people are now markedly more likely to engage in collective activity within both parties. However, among those for whom maintaining the exciting movement of the referendum campaign was a central motivation, activity levels are appreciably lower, confirming our previous findings that experiencing a sense of belonging to a movement does not necessarily translate into active involvement (Bennie et al. 2021). Finally, those who believe that party membership and activity are effective at 'influencing decisions in society' are more likely to engage in that activity, especially collective activity.

#### **Satisfaction gained from activism**

A final task for this chapter is to examine levels of satisfaction gained from participation in party activities. We can examine the correlation between a given activity and reported member satisfaction. There are three key variables in this final analysis, each of them a composite of multiple questions from the survey:


The first two are used to predict the third in a series of multiple regressions by party and cohort (each including controls for age, sex, class, income, education and urban/rural residence). The standardised coefficients from those regressions, akin to correlations, provide a broadly comparable indication of how far high- and low-intensity activity satisfies the different cohorts of joiners. They are plotted in Figure 7.2.

There are instructive comparisons across parties and cohorts. Within the SNP, there is a clear break with the referendum. Among existing members, highintensity, collective activity leads to satisfaction. Among surge and indeed all

*Figure 7.2* Effects of activity indexes on satisfaction with membership by party and cohort

post-referendum joiners, the more low-intensity activity they had undertaken, the more satisfied they were. What tend to keep those members happier are the loweroctane activities like displaying posters and discussing online. They resemble the more 'individualised' members. The pattern among Scottish Green members differs in a couple of important respects. First, their pre-referendum members are even more clearly activism-inclined than their SNP counterparts, and they are if anything dissatisfied the more individualised their membership experience. Second, none of the Green cohorts show the negative correlation between activism and satisfaction that we see in the SNP surge joiners. But there is the same pattern that low-intensity activity becomes increasingly able to satisfy the Green joiners. The influx into the Scottish Greens may have tempered the participatory ethos of that party's base.

#### **Conclusion**

The scale of the surges meant the parties had more activists on the ground in the years following the referendum, and the importance of an increased number of activists should not be underestimated, but the new recruits were less inclined to be active than those who were already members when it comes to conventional ways of participating such as spending time on party work and attending party meetings. Online discussion was the only type of activism where the new recruits, especially in the Scottish Greens, scored more highly than the established members. This is a sign that new members prefer online engagement and participation 'from a distance'. These are forms of engagement traditionally classed as low-cost and involving lower levels of commitment. The distinction between two underlying dimensions of activism – collective (conventional) and individual (atomistic) activism – is instructive. The former relates to in-person forms of activity which require time, effort and working with others, the latter involves activities which can be conducted in isolation and are much less intensive and time-consuming. Individualised activities are by far the most popular among the two parties' memberships, and even more so among the new members.

The data suggest that the individualised forms of participation are more accessible to those who may find it challenging to engage in traditional activism. Our socio-demographic profile of the activists in the SNP and Scottish Greens contained few surprises, but there are signs that individualised activities are disproportionately attractive to young and disabled members of each party and women in the Greens. This type of activity is important, not least because of its inclusive and accessible qualities.

We might have expected that those who actively involve themselves in policymaking and campaigning (the high-intensity, collective element) would be the most committed in terms of values and policy preferences. Yet we found little evidence that activists in these parties are motivated by radical and uncompromising views which are unrepresentative of the memberships or challenge the leaderships. For instance, activists do not appear particularly left (or right) wing in their policy preferences. Nor do we observe an urgent demand for a second independence referendum. Movement identities generally encourage party activism. In the SNP, identifying as a nationalist and being socially liberal enhances the chances of being active. For Greens, being 'radical' and strongly supportive of green policies and being middle-class are the key predictors of collective activism. The most striking effects on activism were to be found in independence referendum activities. To a significant extent, members replicated their referendum involvement within the parties they joined. Being collectively or individually active in the referendum campaign encouraged similar patterns of involvement in the parties. The surge didn't generate particularly participatory motivations or active involvement.

The overall picture presented in this chapter is that the average party member is not very active, and the two parties are very alike in their patterns of activism. The most common activities – in both parties – were displaying posters, the only activity reported by a majority of respondents, followed by online discussion. The evidence suggests that the more recent joiners resembled their predecessors in terms of their patterns of activity and inactivity, with only around a fifth of members inclined towards collective activities. The post-referendum surge did not change the essential character of activism and member involvement. These findings have implications for the parties in that they should expect few party members to be intensely active, at least in a conventional sense, but they can assume that many more are prepared to engage individually.

#### **Notes**


#### **References**


### **8 Organisational challenges and opportunities**

#### **Introduction**

The extent to which political organisations can overcome oligarchic tendencies has been debated at length. Michels (1999 [1911]) argued that mass movements require organisational form to succeed. He also claimed that parties ineluctably transform grassroots participatory forms into top-down oligarchies, spawning numerous studies testing this 'iron law of oligarchy' (Michels 1999 [1911]: 365). Notably, one of the conditions that Michels suggests leads an organisation to have an oligarchic structure is success in recruiting new members. As parties grow, 'The provisional must then give place to the permanent, and dilettantism must yield to professionalism' (Michels 1999 [1911]: 107). Military analogies are frequently adopted in these discussions. Michels (Ibid.: 79) refers to a 'degree of caesarism' to 'ensure the rapid transmission and precise execution of orders'. Such claims can be tested in parties with deep grassroots – parties like the SNP and Scottish Greens. Members of each party would challenge the 'iron law of oligarchy' and indeed the characterisation of grassroots democracy as dilettantism.

A five or sixfold increase in membership is likely to require a radical overhaul of any organisation. The opportunities of having a bigger base of active members and the financial benefits of a much larger membership are obvious. However, political parties are, as Ignazi (2017: 278) observed, 'loci of competition and conflict for power', so a substantially larger membership has the potential to disrupt existing internal power structures. The extent to which this occurred in the SNP and Scottish Greens is the subject of this chapter. We might anticipate that the surge in membership would lead to less participation and more power to the leaders. On the other hand, the new membership may have expected greater involvement having had the experience of the referendum.

Katz and Mair (1993) distinguish between three distinct faces of parties: the party on the ground, the party in central office and the party in public office. Kleidman (1994) focused on the relationships between volunteer activists and professionals in social movement organisations and identified different patterns of professionalisation: the inhibition or erosion of voluntary activism, substitution of volunteers with professionals and professionals facilitating volunteers. Panebianco (1988) observed an increased role for professionals in political parties, identifying

> DOI: 10.4324/9780429446559-8 This chapter has been made available under a CC-BY 4.0 license.

the electoral-professional party. A key question is whether and how the professional and voluntary elements interact, and this will be addressed when considering the SNP and Scottish Greens.

We start by considering the organisation of the two parties before 2014 to provide a base for comparison. Each party could credibly claim to have been internally democratic compared with their parliamentary opponents. The immediate impact of the surge is then discussed, followed by the extent to which change occurred in internal structures. This involves examining the position of the leadership and senior office-holders, governance and policy-making and the selection of candidates. The initial burst of enthusiasm settled, and our research allowed us to follow the process of change that ensued from excitement and enthusiasm to facing up to challenges, a return to something approximating what had gone before, and then greater centralisation. We examine the extent to which this represented a hollowing out of internal democratic institutions, a process identified by Mair (2013) when he cautioned against the dominance of party leaders at the expense of grassroots democracy.

#### **Before the surge**

Pedersen (1982, 1991) referred to the 'lifespan'of parties and the crossing of thresholds as parties gain influence, with consequences for party organisation. Gunther and Diamond (2003: 188) refer to movement parties with 'fluid organizational characteristics', including newly created parties 'prior to their institutionalization'. The implication is that parties with a tradition of collective decision-making face pressure to abandon principle in pursuit of votes, with the demands of electoral competition taking precedence over internal party participation. There can be a shift from amateur-activist to electoral-professional (Panebianco 1988).

 This was true of the pre-devolution SNP, which invested much in its participatory self-governing internal organisation. The party's transformation can be dated from devolution, and becoming a governing party was a further step in the direction of professionalisation at the expense of grassroots democracy. After forming a government for the first time in 2007, the party in public office became central in the SNP. Over a short period of time, the importance of this face transformed intra-party democracy and power, disrupting the relationship between the party on the ground and the party in central office. Our previous research concluded that the SNP had become an electoral-professional party (Mitchell et al. 2012). After many years in government, this process would develop further as power moved increasingly to the leadership.

Until 2004, the SNP did not have a 'leader' though the party 'convener' had come to be seen as leader. The convener's role was relatively open to interpretation, influenced by the skill of the incumbent, but from Alex Salmond's first period as convener came to be seen as little different from the leader in other parties. John Swinney, who succeeded Salmond in 2000, was undermined by the party's structures when he was challenged by a little-known activist. One of Swinney's significant legacies was to overhaul the party constitution and formally create the post of party leader, also making it more difficult to challenge an incumbent.

Change had initially come about informally, that is, the re-interpretation of the role under Salmond, then formally with the change to the party's constitution in 2004. Alex Salmond returned to lead the party that same year. In announcing his decision to stand for the leadership again, Salmond declared that he was 'not just launching a campaign to be SNP leader. Today I am launching my candidacy to be the First Minister of Scotland' (BBC 2004), signalling a significant shift in interpretation from his first period as party leader. In 2007, Salmond became First Minister and the SNP entered government for the first time, considerably enhancing the status of leader.

Formally, the SNP followed the 'archetypal model' of policy development, which starts at local branch level and working groups before reaching national level in party conference and/or alternative national fora (Gauja 2013: 119). But the extent to which branches participated in this process varied. Campaigning was their key function. SNP conferences had been lively affairs but became duller, more controlled by the leadership, and served as rallies in which leading front-bench spokespersons were given a platform. Policy remained the prerogative of the party's annual conference in theory, but the leadership in public office assumed the key role in policy-making with conferences acting more to legitimate leadership decisions.

The Scottish Greens had held out as an amateur-activist, social movement–type party but, given the experience of sister parties elsewhere, this may have been a function of operating on the fringe of parliamentary politics in much the same way the SNP had been amateur-activist before devolution. Pre-devolution, the Greens existed as a tiny political actor with little expectation of surmounting any 'threshold of influence'(Pedersen 1991). The party lacked formal organisational structures and was run by a small number of volunteers. Devolution in 1999 enhanced the public face of the party when a Green Member of the Scottish Parliament (MSP) was elected. Although the party's representation was precarious from the outset – one MSP in 1999, seven in 2003, two in 2007, six in 2016 and eight in 2021 – it has had continuous parliamentary presence.

By the 2000s changes to the party infrastructure were being introduced. In 2004, the Scottish Greens moved from a single party convenor to co-convenors (the constitution stipulating one female and one male). A conscious decision was taken to reject the term 'leader', reflecting scepticism about concentration of power, but Robin Harper and then Patrick Harvie consistently had much higher public profiles than their co-convenors. Having MSPs in the Scottish parliament meant the Greens gained staff support, but the party in central office remained small.

In 2007, the Scottish Greens reached a 'cooperation agreement' with the SNP that identified areas of policy agreement, including independence, and included a commitment that the SNP would consult the Greens on the 'broad shape of each year's legislative and policy programme' and budget. The Greens would support the SNP minority government on key votes and ministerial appointments in return for the SNP nominating a Green MSP to chair a parliamentary committee (SNP/ Scottish Greens 2007). It would be a precursor to cooperation during the independence referendum. But this agreement did not bring the Scottish Greens into the same relationship with government power that had transformed their sister

parties. It allowed the Scottish Greens to retain their innocence and internal party democracy.

Organisational debates within the Scottish Greens were at this time muted affairs compared to those taking place in other green parties, a reflection of the party's distance from governmental power. On the face of it, party organisation was rooted in a participatory ethos. Conference was 'the supreme decision-making body' of the party, open to all members (not delegates). The Greens displayed the characteristics of grassroots participatory intra-party democracy more than any other party that has been represented in the Scottish Parliament since its establishment. However, low levels of membership represented a small pool from which activists and candidates emerged, and most members would never attend a party meeting or conference. Conference prior to the surge was the domain of a small number of activists, a gathering where 'everyone knew each other' (Interview no. 73).

#### **Immediate impact of the surge on party organisation**

SNP and Scottish Green membership started to rise within hours of the referendum declaration. There had been some expectation of increased membership, as commonly happened *during* elections but not *after*, nor on this scale. Applications started to come in after the result became known and then massively increased after Alex Salmond held a press conference the following morning. Salmond's acceptance of the result while promising that the 'dream will never die' was seen by SNP headquarters staff and others as contributing to the surge. According to one well-placed figure, the SNP might have expected up to 50 new members per day (Interview no. 45). Some Scottish Greens thought that they might double their membership (Interview no. 60). Others feared a loss of members after the referendum. The Greens had considered encouraging new members with a special rate of £5 for one year (at that time the average member paid £22). This was shelved when it became clear that enquiries were pouring into the party (Interview nos. 62, 64).

 Officials in both parties reported surprise at the scale of the surge (Interview nos. 45, 60). One referred to thinking there was something wrong with the technology (Interview no. 64). The first challenge for the parties was simply dealing with the new mass memberships. The SNP hired call centres to handle phone enquiries (Interview no. 45). Volunteer activists were recruited into the party's headquarters to help. One senior SNP official referred to a system struggling to cope, describing how 'the phones were ringing off the hook, the system . . . collapsed'(Interview no. 45). The Greens were even less prepared and did not have the SNP's spare capacity or resources. There was organisational chaos, with some potential members logging onto the England and Wales Green Party website by mistake (Interview no. 62). Some new members with IT skills were working on facilitating membership processing 24 hours a day for a period (Interview no. 60). A senior party figure (Interview no. 64) described the events:

Here's what happened. We sat here the day after the referendum. 1500–1700 members joined. When somebody joined, we'd hear a beep. And we heard, beep, beep, beep, beep. They were joining at the rate of 3 or 4 per minute. And it just went on and on. At that point we had to add the members manually. There were about 14 people over the weekend – mainly volunteers – in the office, sitting in corridors, all typing madly. It was exhausting.

SNP membership was recorded at headquarters with information about new members (and a share of membership fee) passed on to local branches, and a similar process took place in the Greens. The impact locally was evident at branch meetings following the referendum. New premises had to be found for meetings. One branch of the SNP saw attendance rise from 'on a good night 12 people' to 230 at the first meeting following the referendum (Interview no. 21). An Edinburgh Green member (Interview no. 63) described an increase in branch attendance from 50 before the referendum to 350 the following week.

The surge had a dramatic impact on attendance at party conferences. The increased SNP membership gave local branches increased delegate entitlement for national conference and other national fora. In many cases, this meant five times more members could attend conferences. The predominant view was that new members should not be excluded from conference, not least as SNP conferences had become rallies and larger audiences appealed to the leadership. Few branches were able to send their full complement of delegates, and rarely did delegates meet in advance to debate issues and decide on how they should vote. Green conferences before 2014 were small-scale affairs. The party's 2012 conference had only 100 attendees. The October 2015 conference in Glasgow had 700 (800 registered), described by the Greens as their 'biggest conference ever'. New members were welcomed – all were entitled to attend – and offered sessions on 'how conference works', an attempt to absorb them into traditional decision-making processes.

The surge provided the parties with a significant financial boost. The SNP had long relied on members' donations though it had received substantial sums from wealthy backers during Alex Salmond's leadership, including donations to the Yes campaign. Post-surge, as one senior activist stated, SNP headquarters were 'sitting on a lot of money' and had not 'had a chance to think about what to do with this' (Interview no. 10). For the Greens, 'it basically changed everything', the party in central office having been 'run on a shoestring' between 2007 and 2014, with any staff on part-time and temporary contracts (Interview no. 64). Almost overnight the party had more money than ever before and was able to take on a team of staff to replace volunteers, transforming core party functions like maintaining a membership database, communication with members, website support and the organisation of conference, 'so staff now do what volunteers did before – more professionally' (Interview no. 64). The Greens updated their IT infrastructure and encouraged new members to pay by direct debit, although this took time.

The SNP had experienced a spike in income in 2011, the year it won an overall majority in Holyrood, and significant sums of money were channelled into the Yes campaign during the referendum. As a result of the post-referendum surge, the SNP's income from membership fees grew from £586k in 2013 to over £1.3m in 2014. In 2015, the first full year after the surge, this income was over £2.7m. 1 As for the Scottish Greens, their 2013 income from membership subscriptions was only £24k, but this climbed post-surge to £183k in 2014 and 2015.<sup>2</sup> It is clear the surge brought large sums into the parties from membership subscriptions and members contributed financially in other ways, as documented in the parties' annual accounts, including via fundraising and legacies.

The Green surge coincided with an increase in the number of local branches 'where there's always been a big gap in the map' (Interview no. 61). Pre-2014, some of the party's activists had become involved nationally because of a lack of local branches (Interview no. 74). Following the referendum new branches were created in Ayrshire, Central Scotland and the Highlands, but growth benefited existing areas of strength. One party official observed: 'This has boosted areas of Edinburgh and Glasgow more than anywhere else. We've always struggled in other areas and still do – Aberdeen, North-East, Dundee to an extent. This partly reflects the concentration of the population, but we've always had problems beyond the central belt' (Interview no. 64).

One of the challenges faced by local parties was making branch meetings interesting. Some effort was made to engage with the new members including leaving time aside for guest speakers and debating policy, but these were concentrated in a few branches. As one senior SNP figure noted, local parties struggled to retain the interest of new members, describing: 'the tedium of political meetings – it is hard to sustain interest at meetings which discussed organisation. We try to get speakers to keep up interest but that's difficult – having to compete with EastEnders or Coronation Street' (Interview no. 15). Only a small proportion of new members attended branch meetings, and the initial interest tailed off within a year of the surge, although attendance 'did not decline to previous levels' (Interview no. 1). Most branches were able to return to previous accommodation. In some places the local branch was refreshed with new office-bearers and activists. Elsewhere, presurge local office-bearers remained in post.

 There remained strong SNP self-discipline in the immediate aftermath of the referendum with activists wary of criticising the leadership: 'People are sensitised to the game. The unwritten responsibility' (Interview no. 36). The overwhelming feeling was that the new activists and members refreshed the party and were easily integrated. Only a few examples of tensions were mentioned in the early period after the surge. One interviewee observed: 'The only significant change was that we were bigger' (Interview no. 15). The absence of any ideological challenges assisted with organisational stability. According to one senior figure who had joined the SNP in the surge, 'people joined the SNP to enhance it, to keep it the same. I mean I didn't join to disrupt it or even to change it. It was already a vessel into which I could pour myself and I think others felt the same way', and this was felt to contrast with those who joined Labour as part of Momentum (Interview no. 36). A typical comment was that the surge did not involve 'entryism', it was 'not like Militant or Momentum' (Interview no. 10).

 Among Scottish Greens there was some initial distrust of new members. In the early weeks and months of the surge, there were reports that the party's old guard felt threatened and behaved like 'a private members' club' (Interview no. 72). One interviewee refers to a branch convenor who described the process as 'traumatic' because of worries over how this might change the party (Interview no. 69). Very soon this perception changed. A senior member refers to a 'politically coherent' surge, noting: 'It's really striking that the kind of people who've come in were Greens already I think in the wider sense' (Interview no. 75).

Many pre-surge Green activists were completely burnt out. The party had relied on a small core group 'who kept the party going, over a period of twenty years, and there might not be a Scottish Green Party if they hadn't really banged the drum and given up a high proportion of their time', and these people were exhausted (Interview no. 73). People who had joined in the surge could advance quickly because in a small party 'it's quite easy to get known'(Interview no. 73). Some new members 'got heavily involved in branches and committees', described as 'a whole new broom' in some national committees (Interview no. 62). However, there was now an obvious need to reform internal party organisation. The party had been organised to accommodate a few hundred members when volunteers knew each other and had not had to seriously consider questions of accountability and internal democracy. Party structures 'were already creaking at the seams' (Interview no. 75).

 It was thought that the new members would bring different experiences and understandings of political campaigns and organisation into the parties, but there was an expectation that many would not engage beyond paying their membership fee. One SNP MSP maintained that sustaining membership is 'easy' but 'getting them involved is the challenge' (Interview no. 20). An SNP staff member stated: 'I didn't get any sense these people were wanting to get active . . . my sense is they were not willing to canvass or be on a street stall but might be willing to share an SNP press release or even just tell their kids or speak to their neighbour, speak to people on the bus' (Interview no. 45). There was speculation among those interviewed that new members were 'sleepers' who might be activated in the event of another independence referendum (Interview no. 1).

#### **Leadership asserted: the hollowing out of party democracy**

The transition from Alex Salmond to Nicola Sturgeon was a classic case of successful succession planning. There was no need to hold an election when Salmond resigned as Sturgeon was the only candidate nominated. This was the first time that a new leader had been chosen unopposed since 1960. In lieu of a leadership contest, the SNP headquarters organised a series of rallies across Scotland addressed by Sturgeon between late October and early December 2014, which attracted large crowds. This included addressing around 12,000 supporters in the SSE Hydro in Glasgow alongside singers and entertainers (BBC 2014) with echoes of the referendum campaign. But while the effort to establish her leadership used repertoires from the referendum, Sturgeon's leadership involved a return to 'normal politics' and put the SNP back on the electoral-professional trajectory.

A contest was held for depute leader when Sturgeon became leader. It was won by Stewart Hosie MP, with 55.7% of the party's membership taking part. Hosie resigned two years later by which time the SNP was coming to terms with the surge in its membership. The candidates in 2016 offered different views on organisational matters. Angus Robertson, leader of the SNP MPs in the Commons, won the contest, but only a third of the members voted. Tommy Sheppard, who had been elected an MP in 2015 after joining the SNP in the aftermath of the referendum (and had previously been a senior Labour Party official), also stood. The consensus among those interviewed was that Sheppard had been the most impressive candidate at hustings. He had played a significant part in the 2014 referendum campaign both locally in Edinburgh and nationally. Sheppard hoped to garner and mobilise the grassroots vitality in future campaigns. Robertson was more top-down.

Robertson was on the party's right, and Sheppard was a socialist, but the main issue that had divided them during the contest was reform of party organisation. Ironically, Sheppard rather than Robertson embraced the traditional ethos of the SNP by insisting that membership was its 'biggest asset' and that there was a need for new structures to allow members to become more involved including increased roles in policy-making. As the party's 'basic unit', he argued, the branches should engage in more political discussion and activity. Sheppard also favoured building alliances beyond the SNP (Sheppard 2016). As a relatively new member he gave voice to the party's past social movement ethos. Sheppard's view of the relationship between voluntary activists and professionals was that professionalisation should facilitate rather than erode voluntary activism.

Robertson resigned as depute leader after losing his Parliamentary seat in 2017, which lead to the third election for the post in three years. Three candidates stood: Keith Brown MSP, Councillor Christopher McEleny and Julie Hepburn. Brown was elected leader with 55% of votes after McEleny's second preferences were allocated. He had decided to stand against Sturgeon's wishes and was removed from the Cabinet in a reshuffle, a decision explained publicly as allowing him more time to devote to his party responsibilities including preparing for a second independence referendum. On this occasion, the SNP did not release figures on the number of members who had voted, but newspaper reports suggested that only about a quarter of the party's members had participated (*Sunday Herald* 17 June 2018). The party continued to insist that it was retaining members, but its new base was slowly eroding.

The SNP experienced 'de facto presidentialisation' under Salmond and then Sturgeon, with the leader gaining increased resources and authority, and there was more emphasis on 'leadership-centred' electoral politics (Poguntke and Webb 2005: 5). This has been a common feature in political parties across liberal democracies. While being in government assists this process, presidentialisation can occur in parties in opposition. Government, party and electoral politics each contributed to the SNP leader's power, distinct from the party's formal constitutional arrangements. The status and resources available as First Minister combined with authority based on winning elections added significantly to the party leader's reputational power.

Greens throughout the United Kingdom and Europe have long accepted the importance of 'leadership'. The demands of modern politics and campaigning combine with a media ill-suited to parties with collective, rotating leaderships, applying pressure to be conventional. Green parties have adapted to these demands, with a 'realistic', pragmatic approach to leadership, moving away from rotation of leading offices. Rüdig (2008: 219) described 'the loss of the taboo associated with the concept of a single party leader'. The Scottish Greens provide a good example of such pragmatism. At the time of the 2014 surge, the party's co-convenors were Maggie Chapman and Patrick Harvie. Harvie had been male convenor since 2008, and he attracted most media attention despite the party's efforts to offer alternative spokespeople in response to media queries. This created some frustration in the party, but it was recognised that Harvie was a good media performer.

The Scottish Greens' 2015 Autumn conference agreed to an internal review of party structures. This began in earnest following the 2016 Scottish Parliament election (when six Greens were elected), and it examined the party's leadership and committee structure. In 2017, the party recommitted to the principle of joint or shared leadership. For each shared office 'at least one of the members serving must be a woman' (Scottish Green Party 2020: 28), a move aimed at increasing female representation. When the party adopted a new constitution in 2019, co-convenors become co-leaders subject to a membership election every two years, and there would be a time-limit on the leadership.<sup>3</sup> These rules came into practice at a time when Harvie had already served as co-convenor for a decade and during the 2014 referendum campaign had been treated by the media as de facto party leader. The first leadership election took place in August 2019 when Patrick Harvie and Lorna Slater were elected, the latter having joined the Greens as part of the surge. While all members were entitled to vote, 800 did so, a participation rate of 12%. The party did not make public at the time the total membership figures, which suggested some sensitivity about falling membership.

On coming to power in 2007, the SNP leadership had gained new authority having led the SNP to become Scotland's largest party but also in having the resources of government at its disposal. The resources, access and visibility afforded by winning the election imbalanced the relationship between the three faces of the party though this was a cost most members were happy to pay for victory. The combination of winning an overall majority in 2011 against the odds and the prospect of an independence referendum contributed further to the reputational power of the party leadership vis-à-vis the party on the ground.

The SNP's central bureaucracy was formally the servant of the National Executive Committee (NEC), but the NEC in recent years simply rubber-stamped decisions made by the leaders. Its central role was managing the party's membership and branch structure, organising conferences and implementing strategy devised by the leadership. There were strong links with a few key figures in the SNP's central bureaucracy who advised the leadership, drafted manifestos and helped develop strategy and policy while the SNP was in opposition. On winning the 2007 election, a small core moved from headquarters into roles in government as special advisers. This contributed to the hollowing out of the party's central bureaucracy. There is nothing new in a small core of individuals playing key roles in the SNP, even when it operated on the fringe of Scottish politics in the 1960s (Crawford 1982). What changed was that the party in public office assumed far greater significance.

In October 2016, the SNP began a process of reviewing its constitution and rules leading to a revised constitution agreed at its conference two years later. Party conference formally remained the 'supreme governing and policy-making body'. Well before these post-surge changes, the annual conference had increasingly adopted the style of a rally rather than a decision-making forum. Occasional set piece debates would be held such as on whether an independent Scotland should be a member of NATO in 2012. By 426 votes to 332, the SNP agreed that an SNP government would maintain NATO membership but not host nuclear weapons. While this was evidence of conference delegates determining a key policy change, it was a change the leadership wanted, to offer a more attractive policy in the forthcoming independence referendum; policy change was legitimised through a conference decision.

In 2018, the SNP abolished National Council, which had been the key decisionmaking body between annual conferences, and replaced it with regional steering committees in the eight Holyrood List regions. Each region would elect two members to the SNP's NEC. The reconstituted NEC would be much larger with 14 national office-bearers, 16 regional representatives, 2 parliamentarians elected by SNP conference, representatives of the two parliamentary groups, representatives of councillors and affiliated groups (Black, Asian and Minority Ethnic [BAME] members network, Disabled Members group, Federation of Student Nationalists, Out for Independence, Scots Asians for Independence, Trade Union Group and Young Scots for Independence). This unwieldy body would be easily controlled by the SNP leadership and headquarters. National Assemblies would meet in each region annually providing a forum for debate but not as decision-making bodies. National Councils were supposed to have been reinstated in 2021 but had not met by the time Murrell and Sturgeon resigned in 2023.

Challenges to the leadership were mounted in 2020 when Douglas Chapman MP successfully replaced incumbent Treasurer Colin Beattie MSP, and a number of other places on the NEC were won by critics of the leadership. Chapman later resigned along with three members of the SNP's Audit Committee. Joanna Cherry MP also stood down from the NEC criticising the 'secretive and evasive' and a 'menacing atmosphere' preventing her from fulfilling her mandate to 'improve transparency and scrutiny'.

 A further governance review was established at the behest of depute leader Keith Brown in September 2020. A final report was completed in August 2021. It maintained that the SNP had operated on the basis of 'goodwill, strong networks, peer support, shared knowledge and self-discipline', but these 'old ways of doing things had become obsolete in a mass membership organisation' (SNP 2021: 3). It recommended more transparency including 'outcome of business' reports from meetings. The Finance section noted the need to restore confidence in the party's financial governance and that it needed to 'think seriously about transparency and accountability in addition to financial probity' (Ibid.: 6). Reporting after Chapman's resignation, it recommended that the Treasurer should have access to 'sufficient information to execute' duties to provide 'financial oversight, appropriate scrutiny, and strategic guidance' (Ibid.). SNP headquarters staff complained that the review left them feeling demoralised and unappreciated, signalling tension with the party in central office. The report was quietly sidelined.

The Scottish Green Party's internal organisation was examined in some detail by the structural review, which produced a new constitution in 2019. The party's main decision-making body had been the National Council, which consisted of up to 50 people, including parliamentarians and party representative group members, meeting four times a year. Interviewees highlighted a party structure that was 'clunky', unresponsive and in need of reform, with the large unwieldy National Council but no formal party executive (Interview no. 75). A Working Group produced a set of proposals which was ratified at an Emergency General Meeting in Spring 2019. An NEC would be created, to be overseen by Council. This was described as the Council setting strategy and the NEC implementing strategy. The NEC became responsible for delivering the party's strategic plan. New committees designed to support the membership were created, as well as the new positions of co-leaders. In principle, this all meant the party would be more 'nimble', able to respond more quickly to emergencies and unplanned elections. In practice it meant that the party adopted a structure akin to a conventional party, with a stronger leadership. This was widely described within the party at the time as 'streamlining' (Interview no. 73). It coincided with the party's increased capacity to employ staff including regional campaign coordinators.

Some in the party complained that decision-making had become too centralised, insufficiently focused on the party on the ground, one criticism being that the national party had taken more control of membership fees, when these used to be more generously shared with local branches. It was argued that 'the party has concentrated on parliamentary and national work at the expense of building vibrant local branches' (Bright Green 2020). Conferences (annual general meetings), too, were criticised for being more about the leadership presenting policies than the membership deciding on policies.

The Greens remain different from other parties, still more amateur-activist than electoral-professional. Local branches actively shape policy by introducing motions to conference/general meetings, motions which are not necessarily supported by the leadership. Conference remains open to individual members who can also attend 'policy weeks' and pre-conference workshops. And the Greens remain a party where activists can quickly rise through the ranks. Interviewees suggested that those who joined following the referendum are now running the party, although the 'old guard' are over-represented among the parliamentarians (Interview no. 62).

A not inconsiderable challenge for the Greens as for any party is that they find it difficult to engage members. The average member does not reply to party emails and requests, and few members attend local meetings, conferences or even online sessions. Online events are largely presentations by the party's leading figures, and there is a feeling in the party that these events have enhanced leadership power. Gauja and Kosiara-Pedersen (2021) argue that digital decision-making can facilitate and encourage member engagement – participatory capacity is in theory enhanced – but if few members embrace these opportunities, leader power is strengthened. Activists in the Greens sometime express disappointment at the lack of engagement of ordinary party members. This certainly presents problems for local branches, some of which are deflated with poor attendance of office-bearers in many cases. As in other parties, this is convenient for the party leadership who have the support of passive members.

#### **Candidate selection**

Sartori maintained that the selection of candidates for public office distinguished parties from other campaign organisations (Sartori 1976: 64). Selecting who is most capable of advancing the party's objectives while maintaining inclusivity and representativeness can be a fine balance. In the SNP and Scottish Greens, members choose candidates subject to rules set by the parties at the centre.

In March 2015, the SNP conference agreed that when an incumbent constituency MSP stood down, the NEC could require an all-women shortlist and at least one woman should be on the shortlist for the successor. It gave the NEC power to add candidates to a shortlist. Similar requirements would apply to regional lists rankings. This ensured that 43% of SNP candidates were women at the 2016 elections though only 35% of SNP MSPs who were returned. At its 2019 conference, the SNP required all-women shortlists in constituencies where an incumbent SNP MSP was standing down. There would be the possibility of adding BAME, women or disabled potential candidates to a constituency and the possibility of applying 'zipping' – alternating women and men – to regional lists. This reflected the emphasis on equality that had already been achieved in the SNP Cabinet, and the proportion of women SNP MSPs would rise to over 50% in 2021. The 2020-2021 pandemic added new challenges as engagement could only take place online. Campaigning was restricted to online hustings, there was limited contact with members through an authorised email system, and candidates relied on personal contacts and the use of a website, personal Facebook page and personal twitter account.

 Controversies arose in the selection process before the 2021 Holyrood elections. The most public controversy was over additional rules governing incumbent SNP MPs seeking to contest a seat in Holyrood and how much potential candidates could spend. These controversies focused on one contest. Edinburgh Central had been won in 2011 by the SNP, but Ruth Davidson, then Scottish Tory leader, took the seat in 2016. It was high on the list of SNP target seats. Marco Biagi, the previous SNP incumbent, was under pressure to put himself forward again not least as it looked set to be a heated contest between Joanna Cherry KC, MP for Edinburgh South West, which includes part of the Central Edinburgh Holyrood constituency (Commons and Holyrood constituencies are not coterminous), and Angus Robertson who lost his Moray seat in the north of Scotland at the 2017 Commons election. The NEC ruled that any incumbent MP – at that time Cherry was the only SNP MP publicly considering this move – would have to stand down from the Commons in time to allow a by-election on the same day as the Holyrood elections. This was dubbed 'Cherrymandering' by critics of the leadership, and Cherry withdrew from the contest. Biagi and others seeking to become candidates in other constituencies complained that those with access to money were able to fund expensive online campaigns. The party's formal rules suggested support for under-represented groups, but these details meant that Robertson, a middle-class, middle-aged, white, heterosexual male had an advantage over two gay candidates, winning the candidacy and the seat. As is often the case, the devil is in the detail, and purported principles can be undermined.

Another controversy, though one that only subsequently came to light, surrounded the 'equalities mechanism' that applied to regional list selection. This empowered the NEC to reallocate positions on the list. It was used to remove Joan McAlpine, an incumbent MSP who came top of the list in South of Scotland in 2016. McAlpine had been a long-time feminist and an independent-minded backbencher who occasionally spoke out against SNP government policy. McAlpine was subject to additional vetting before the 2021 selection and was replaced in the top position by another more loyal woman who was given preference given her disability as someone who suffered from diabetes. The SNP NEC had taken legal advice and had been warned against adopting the procedure, but the NEC split in favour of ignoring the advice with the casting vote of Kirsten Oswald MP, SNP business manager, who had been appointed by Sturgeon. McAlpine was refused access to the number of votes cast (McAlpine 2023).

The Scottish Greens had attempted proactive gender-balancing in candidate selection, adopting formal equality mechanisms, but these were not always successful. Party rules had determined that 50% of candidates in winnable seats must be women and that a zipping mechanism would be used in regional lists. These rules were widely accepted by the Greens, which is not to say that the party did not experience controversy in candidate selection. Before 2021 (when five of eight of the party's MSPs were women) these mechanisms had largely failed to deliver gender equality in those elected to the Scottish Parliament. In the 2016 election, the party stood in all eight regions, women headed four, but only one of six Green MSPs elected was a woman. This led to some recriminations and internal tensions. And there was also some unhappiness over the placing of national party figures at the head of regional lists at the expense of local activists less well known to ordinary members when voting to select candidates.

Some viewed the outcome of the 2016 election as 'bad luck', but it led to some changes to candidate selection procedures. The zipping process was modified so that women could not be 'zipped down', that is, if member votes determine that three women finish top of a list, they remain in place and are not alternated with men. The party remains committed to 50% of candidates in winnable seats being women and 40% as a minimum overall, and has removed a previous stipulation that men should make up at least 40% of candidates. The party describes this as having 'no maximum ceiling on the number of women candidates who can be selected'. As in other parties, gender-balancing requirements combine with practical support for members and local branches. The Women's Network has been a particularly high-profile and active internal party group, encouraging women to participate internally and supporting them as candidates. There has been an increase in the number of women standing as Scottish Green candidates.

#### 142 *Organisational challenges and opportunities*

While candidate selection is a power exercised by ordinary members, the process involves a relationship between local and central party. Sometimes local Green party branches struggle to abide by central party rules but not because of any ideological objection. There can be problems selecting women in 50% of target wards/constituencies due to a lack of interest from members in some areas, resulting in difficulty attracting potential candidates. In these circumstances, branches seek special permission from Council to divert from the policy.

In 2020, a dispute erupted in the Scottish Greens though not directly over candidate selection. Andy Wightman – high-profile campaigner for land reform who joined the Greens in 2009 and became an MSP in 2016 – had been selected as candidate for the 2021 Holyrood election. He resigned from the Greens in December 2020 claiming that 'the mood changed' at a time of debate over sex, gender and public policy and that the party leadership and senior officials had become intolerant of diverse opinion (Wightman 2021).

#### **Perceptions of party organisation and membership**

As we established in Chapter 2, the Greens initially appeared to have more difficulty retaining their new members than the SNP, but membership recovered somewhat by 2022. Over time the SNP appeared to experience a significant decline in membership though the precise timing of this is unclear. In 2023, the party was forced to concede that it had lost members, and it was assumed that a significant loss had taken place in the previous year. Our data are based on an earlier period. Levels of membership satisfaction, including perceptions of party organisation and party leadership should be read in this context. They provide an opportunity to test the idea that surge joiners may have been frustrated by their experience of party membership, and that this was a particular issue for the Scottish Greens, as suggested by more than one of our interviewees, for example:

There certainly have been issues with people who joined the party and just did not have anything like an accurate idea about what a political party is for. A lot of people joined the party kind of expecting it to function like a movement organisation. We're a political party, our goal is to elect people. It is electoral politics we engage with. We do movement politics, we campaign on non-electoral issues like rail nationalisation and stuff, but ultimately it's about electing Greens who'll do these things. (Interview no. 74)

There was little sign of discontent at the time the survey was conducted. Table 8.1 reports a series of specific and more general evaluations of party membership and points to widespread positivity. In both parties, very large majorities expressed confidence that leaders represented the interests of members, and only small minorities lamented a lack of member influence. One explanation for the widespread satisfaction might be the third row in the table, in which majorities of respondents, especially in the SNP, felt that inactive members are still making a contribution. This point was as widely accepted among surge recruits as among the long-standing


*Table 8.1* Evaluations of membership by party and cohort

members. There was also widespread satisfaction with the party leaderships. Nicola Sturgeon was popular with all cohorts within the SNP, and Patrick Harvie was also widely popular and particularly so among referendum recruits who, as we saw in Chapter 5, were the more likely to cite leadership as a motivation for joining. This is consistent with the claim that many surge recruits would be content with followership rather than activism. Certainly, none of these data point to an influx of highly participatory recruits being frustrated by a lack of opportunities for activity within these parties.

However, there is a noticeable difference between the SNP and the Greens, which shows up most clearly in the evaluations at the bottom of the table. Scottish Greens, particularly those who joined in 2014, were markedly less likely to report that membership had fully lived up to their expectations. They were also more likely to report at least some uncertainty about still being a member in the future. While we do not know how well individual respondents predicted their own behaviour, these data do tally with the membership statistics of the post-referendum period, which suggest that the SNP was better than the Scottish Greens at retaining surge recruits (even if it would then suffer heavier losses in later years).

#### **Conclusion**

The surge had an immediate impact on the two parties, but it would take time before the full impact became clear. While this differed in extent, a tendency towards more control by the leadership emerged. When we consider Katz and Mair's (1993) three faces of parties – the party on the ground, the party in central office and the party in public office – we see that the relationship between the three faces of each party was altered over time. In each case, power shifted towards the professional staff and full-time elected politicians. The party in public office became more important. The party on the ground remained important not least in providing foot soldiers at elections but its policy-making role diminished, more clearly in the case of the SNP.

In the SNP, the party in public office and party in central office worked as one and did so at the expense of the party on the ground. There had been a debate on this during the second SNP depute leader contest when Tommy Sheppard had advocated reforms that would have involved the professionals facilitating volunteer activism and decision-making. But instead, the SNP carried on further along the route from amateur-activism to electoral-professionalism. The SNP hollowed out its internal democratic structures. The concentration of power in a small group around the leader undermined not only democratic decision-making but also accountability mechanisms. In the SNP's case, there was a diminution of the membership's power. But the new members had joined not to change the SNP but because they supported it and that meant the message and policies of the leadership. The SNP had taken Scotland closer to independence than most people imagined likely in 2011, and there was a sense that this goal was within sight. Members put their trust in the leadership at least so long as independence looked likely. Selfdiscipline created a culture of quiescence and trust but that started to fray as time went by. By the time Sturgeon stood down as leader, more voices were being raised questioning strategy and party management.

Greens in many countries have undergone change from amateur-activist to electoral-professional in much the same way as the SNP has done. Like its sister parties, the Scottish Green Party has moved away from the movement-party model with its strong commitment to direct participation and 'fluid organizational characteristics' (Gunther and Diamond 2003: 188). Concentration of power in the hands of party leaders is evident, though this process is not as strong as in the SNP. While Scottish Greens insist that they will avoid the kind of changes seen in other parties, the surge in membership tests the capacity to retain a participatory democratic basis and entering government does so to an even greater extent. We might surmise therefore that a surge in membership does not in itself alter a party's internal organisation fundamentally. Attaining or having the prospect of attaining governmental power appears more likely to involve reaching a threshold affecting a party's internal power structures.

The assumption that parties ought to be democratic raises familiar questions as to what is meant by democracy. Tensions exist between representative and participatory democracy. The role of deliberation, extent of participation in key decisions and ultimately empowerment of members involve debates and disputation. Over half a century ago, Sidney Verba (1961: 220–221) warned about 'pseudo-participation', which creates an impression of participation but is limited to endorsement of decisions by leaders. Scarrow et al. (2022: 204) refer to 'pseudo-democratic theatre', a similar process whereby membership participation ultimately boosts the position of party leaders, but they also note that this outcome is not inevitable.

#### **Notes**


#### **References**


#### 146 *Organisational challenges and opportunities*


### **9 Conclusion**

### The legacy of the independence referendum

The surge in Scottish National Party and Scottish Green Party memberships after the 2014 independence referendum warrants attention for a number of reasons. The sheer scale of the surge alone merits our interest because it clearly bucked the general trend in liberal democracies of parties losing members. Also governing parties tend to lose members over time, and the SNP had been in office for seven years. We might also have expected that following defeat in the referendum, the parties that campaigned for the losing side would have become despondent and be more likely to lose than gain members. Few organisations – political or otherwise – are likely to be unaffected by such a massive increase in membership. How would these parties cope? Who were these new members, and why did they join? What were their views, and what would they contribute to the parties? We know that the independence referendum was a key factor as the surge came immediately after that event, but the precise role required to be explored.

The nature of the referendum campaign was important in understanding the events that followed. The SNP leadership had understood the need for a broadbased referendum campaign. It could not be fought like an election. While party cues are important in mobilising support in a referendum, the combined SNP and Green vote in 2011 would not be enough to win. The Labour vote was the key target, and the creation, or resurrection, of a broad national movement was required to make it easier to attract such voters. This was facilitated by the rules governing the conduct of referendums requiring the creation of an umbrella campaign organisation. Yes Scotland was consciously created to project an image of being more than the sum of its party political parts. A feature of this period was the language used to describe the campaign for independence. It was consciously described as a 'movement' – not the 'national movement', as in the past, but the 'Yes movement'. As our research found, there were some people who joined in the surge (and indeed some who became elected representatives of the parties) who eschewed any association with 'national' or 'nationalism', preferring to describe themselves as supporters of independence but not nationalists.

We have found that many supporters who did not participate in the campaign later joined one of the parties. The referendum became an unplanned retrospective recruitment exercise. The referendum was the catalyst for the surge, but there was no expectation at the time that the referendum would increase the memberships

DOI: 10.4324/9780429446559-9

This chapter has been made available under a CC-BY 4.0 license.

of the two parties so dramatically. The energy and momentum generated during the campaign might have dissipated after the declaration, but our interviewees and survey respondents reported a sense of disappointment, in some cases anger, mixed with belief that significant progress had been achieved and that independence was in sight. The result was interpreted less as a defeat than a step towards victory by many campaigners. The role of the parties, especially the SNP, was to provide agency in channelling activists and sympathisers. SNP headquarters provided frequent updates on membership, which were widely reported on social media. The surge became a news story in itself, and this helped sustain the surge. Yes campaign groups across Scotland provided a network which facilitated a snowball effect.

The movement that developed during the referendum campaign was channelled into the two parties, leaving little trace of what had been witnessed between 2012 and 2014 other than a massively increased membership of both parties. This is not to say that the Scottish independence movement beyond the two parties completely ceased to exist (see below). Some of the campaign groups formed before the 2014 referendum disbanded, but others continued, and new groups would later be established. However, the grassroots energy created by the referendum campaign dissipated, and party membership became a more obvious route for ardent independence supporters to express those strong feelings.

As Chapter 5 revealed, the post-referendum recruits were motivated by ideological or expressive reasons involving a belief in independence or what could be achieved with independence, and significant numbers acknowledged that the referendum in some way 'triggered' their decision to join. Few of the new SNP or Scottish Green members had been active in the campaign, and few had ambitions to be active in their party. They largely joined for broader ideological reasons and were prompted to do so by the referendum outcome.

Developments occurred as our research proceeded, leading us to conduct a further survey when the SNP experienced another surge in membership in 2018. This was modest only by comparison with what had happened four years earlier but would otherwise have been worthy of study in its own right. This wave of new recruits, whose commitment to independence was if anything even more intense, came into the party in response to a different event but confirmed the importance of understanding the role played by triggers in the decision to join a political organisation. Those who responded to these triggers – the types of people who joined the SNP and Scottish Greens – were much like the members of old. There was no dramatic transformation in the membership profile of either party.

As well as interest in studying and comparing each party's experience, we were interested in the relationship between the parties. Among the parties' memberships, we found differences of views – Greens had a different perspective on economic growth, with more emphasis on the environment, and they placed themselves more to the left – but there was commonality of attitudes on many issues. Chapter 6 showed that members see themselves as having a constellation of economically left-wing, socially liberal and pro-EU values. This has enabled a working relationship between the parties in government.

Back in 2007, the SNP formed a minority government with 47 of Holyrood's 129 MSP and reached an agreement with the two Green MSPs. That agreement gave the Greens convenorship of one of Holyrood's committees that the SNP was entitled to and a commitment to be consulted. It was a tentative step and one that made cooperation easier during the independence referendum. It was agreed after the 2021 Holyrood elections that the Scottish Greens' co-leaders would become junior ministers in the SNP government. The deal fell short of a formal coalition but extended the cooperation that existed between 2007 and 2011. Green MSPs would not be bound by collective ministerial responsibility across all aspects of Scottish government policy.

The scene appeared to be set for further cooperation in the event of a second independence referendum. The prospect of another referendum was the dominant issue for the Scottish government under Nicola Sturgeon. In early 2023, Sturgeon resigned as SNP leader and Scottish First Minister after nine years. This was followed by a series of revelations and controversies regarding the internal workings of the party, not least concerning membership figures and organisational changes that occurred after the surge. We have been able to track the changes and assess the impact of the surge over the period up to and just after Sturgeon's resignation.

#### **Parties and movements**

Tilly (1978: 8–9) outlined the key analytical dimensions of social movements: groups and organisations involved in collective action, events that make up the movement's action repertoire and the ideas unifying the movement. These offer a useful framework for considering the extent to which a movement emerged during the referendum and persisted thereafter or whether the movement dissipated or was swallowed up by the parties. But first, we should recall the pre-referendum period.

The SNP and Greens grew out of wider political movements. Older studies of the SNP tended to locate the party within the broader national movement (Hanham 1969; Webb 1977; Brand 1978). A striking feature of the SNP has been how it assumed primacy, almost monopolising national movement forms from the late 1960s onwards. While minor parties and organisations supporting independence existed over the late twentieth century, none came close to assuming the importance of the SNP. The party/movement distinction created tensions *within* the SNP. Should it become a political party with a clear position across a range of socioeconomic matters as well as campaign for self-government, or should it focus exclusively on the latter?

On gaining parliamentary representation, and especially with the prospect of forming a government after devolution, that tension was resolved by necessity. SNP elected representatives and a (prospective) SNP government could not stand aloof from everyday policy choices, but the tension was only partially resolved. Over time, and especially with the advent of devolution, the SNP's movement– type decentralised participatory structures were replaced by a more hierarchical electoral-professional model. The SNP followed many other parties in changing its internal power structures in pursuit of public office. The activists were perceived to be more radical, holding back progress and limiting the autonomy of leadership. By 2011 there were few other national movement organisations.

The Scottish Green Party has origins in the wider environmental movement and in many respects resembles the SNP of old. For much of its history, it relied on unpaid volunteers and had very limited resources. As with the SNP, a decentralist participatory ethos was embraced. Its electoral prospects were severely limited, though the electoral system adopted for the Scottish Parliament gave it an opportunity to gain a toehold in parliamentary politics. The Greens contested only regional list seats and gained their first member of the Scottish Parliament in the first elections to the parliament in 1999. Parliamentary representation has been continuous but erratic, in much the same way as the SNP's had been in the Commons from 1967 to 1997. Parliamentary representation gave the Greens resources and a profile.

Both parties had preferred collective styles of leadership. Each had refused to have a 'leader' preferring a 'chairman' then 'convener' until 2004 in the SNP and 'convenor'(s) in the Greens though de facto leaders emerged in each case post-devolution. Each had a decentralised and participatory ethos with members jealously guarding against leadership efforts to by-pass the membership. Conferences were lively events, and policies were debated thoroughly. There were even ideological similarities. The SNP had long opposed nuclear weapons and power and had a strong environmentalist strand. Both leaned to the left, at least in rhetoric. Scottish Green Party support for independence was the key issue that facilitated cooperation after 2007 and became stronger during the independence referendum and subsequently.

Jack Brand (1992: 81), writing about the SNP over three decades ago, suggested that parties and movements should be seen as positions on a continuum. This study confirms that parties and movements are far from being institutionally discrete but also that the parties are not just an organisational manifestation of a movement or even one distinct part of a movement. There is considerable fluidity in the relationship. These processes of change have not involved a straightforward transition from one type of political actor into another but a political shape shifter capable of adapting to changing contexts and needs.

#### **Social movement groups and organisations**

The referendum broadened out the groups and organisations involved in campaigning for independence. Other smaller political parties that favoured independence, though they played a limited role in campaigning for it until the referendum, worked alongside the dominant SNP during the independence referendum. Various other groups were established during the referendum often with the support of the umbrella campaign organisation Yes Scotland though some of these had limited members. This suited campaigners who sought to project an image of a broadly based campaign, but membership often overlapped with the pro-independence political parties.

The Yes movement's new organisational forms barely survived the referendum campaign other than in the massively increased membership of the SNP and to a much lesser degree the Scottish Greens. This involved a return to something like the status quo ante but with the SNP partly sharing organisational form of the independence movement with the Greens as most of the Yes groups during the referendum were little more than paper organisations or had the sole purpose of campaigning in the referendum. Yet many of our interviewees maintained that the independence movement continued beyond the referendum and that it was organisationally separate from but overlapping with the SNP and Scottish Greens. Our survey respondents also perceived and felt part of this broader movement for change.

Over time, though, those movement organisations which lasted beyond the referendum wound down. Members of the Radical Independence Campaign (RIC), which played a significant part in registering voters in communities with low levels of electoral registration, were part of an electoral alliance with the Scottish Socialist Party (SSP) and others that contested the 2016 Holyrood elections as RISE (standing for Respect, Independence, Socialism and Environmentalism) (Freeman 2016). But RISE won only 0.5% of the vote across Scotland's regions though there were 800 people at its launch in August 2015.

Women for Independence (WfI) was set up in 2012, but its campaign activities focused on the period from International Women's Day in 2014 (8 March) through to the referendum. It was a loose group and carried on and gained membership after the referendum. It did not become formally constituted or charge a subscription until after the referendum. By 2016 it had 900 members. One of its founders described how it had been a 'pro-women' organisation before the referendum but became a feminist organisation later. It had links with political parties including 13 SNP MPs who were elected in 2015. Our survey in 2015 found that over 90% of members of WfI had voted SNP in 2011 and just under 60% were SNP members with about a quarter of those joining the SNP in the previous year. Around 10% of WfI members were members of the Greens, half of those joining after the referendum.

WfI had a broader agenda than independence and campaigned against proposals for a new women's prison in Greenock and aimed to reduce the number of women in prison. It prides itself on being a flat, participatory organisation. Members were divided on whether it should have a national leader, with around equal numbers agreeing and disagreeing (although almost 40% were uncertain). There was strong support for the importance of local groups and local campaigning and a strong feeling that members should have a chance to have a say in WfI meetings. But while WfI looked like a classic decentralised, participatory organisation run by volunteers, it had a crisis when its former treasurer, an SNP MP, was jailed for embezzling WfI and local SNP funds. WfI is one of the few groups that remain in existence which owes much to its broader focus and activities. Even this organisation, with a broader focus than independence, has struggled to maintain the engagement of its activists and members. Post-pandemic, the organisational vitality of the group has become diminished, with many activists burnt out and de-energised.

Nearly a decade on from the 2014 referendum, the extent to which a modern independence movement exists other than through the SNP and Scottish Greens

### 152 *Conclusion*

might be debated. On the one hand, new groups have emerged post-2014, some from older versions of Yes groups (Petrie 2023). All Under One Banner (AUOB) was formed in the aftermath of the referendum and organises marches and rallies across Scotland. Believe in Scotland (BIS) began in 2019 but had its roots in the referendum's Business for Scotland. Individual campaigners and online bloggers continue to press for independence and/or another referendum. Each of these includes members of the pro-independence parties. Alex Salmond's Alba Party was formed in 2021. The National, established in the months following the 2014 referendum, plays a role in the debate as Scotland's only pro-independence newspaper. These could all be viewed as movement organisations. On the other hand, none begin to match the organisational scale and creativity of the independence movement in the run-up to the 2014 referendum, and there is overlapping membership or engagement in each case. The movement organisation qualities of the 2014 campaign were discussed in Chapter 3. The bottom-up and spontaneous elements coexisted with top-down party style organisation, and this created a highly unusual type of campaign. What we have observed since that time is much more in the category of 'politics as usual'.

#### **Events and repertoires**

Many novel activities, or re-activated repertoires from much earlier SNP campaigns, were highlighted both in the media and the Yes campaign during the independence referendum. For the SNP, the referendum was a back-to-the-future moment with the 'razzmatazz' and fun of old-style SNP campaigns. Greens brought the kind of imaginative campaigning that had long been a part of the environmental movement. A group of 'creatives' – artists, writers, singers and other performers – operated under the Yes Scotland banner bringing a range of social movement repertoires to the campaign. The colour and novelty of 'wish trees', poetry, songs and 'flash mobs' were reported across social media, copied and adapted, and all contributed to a lively and movement-style campaign. As Hank Johnston (2014: 74) has noted, 'Social movement researchers have long recognized that cultural artifacts play important roles in mobilization processes'. The medium became the message.

These events and repertoires were not directed from any central organisation but were spontaneous grassroots activities. In this sense, they represented the selfgoverning, decentralised ethos that characterises some social movements. They were also often joyful and may have been important in maintaining the interest and momentum of campaigning for many involved as much as generating publicity and support. The Yes campaign was keen to convey a spirit of optimism (it made much of Yes Scotland's address on Hope Street in Glasgow) in contrast to accusations that its opponents engaged negatively – 'Project Fear' as it was described by one of the Better Together anti-independence campaign group (Pike 2015).

The independence movement's practices became one of its defining features. As a number of social movement scholars have noted, these practices and performances create a common experience that becomes embedded, part of a movement's myths and belief systems (Johnston 2014: 92). It has been argued that 'movements produce culture, and cultural changes are an important product of collective action' (Meyer and Whittier 1994: 281). Social action as theatre instils a sense of identity, community and belonging. This performative aspect to social movements may have been more myth than real, but its mythical status matters.

While our survey found that few members actually took part in much of this performative theatre, it was recalled by nearly all of our interviewees. It is easy to overstate the extent to which these repertoires existed and understate the extent to which traditional party style campaigning, notably canvassing and Get-Out-The-Vote (GOTV) activities, was involved. The myth of novel repertoires – regardless of actual extent – played into the movement's sense of itself. But these repertoires were not adopted by the parties after the surge. Those who joined the parties would not see a continuation of the festival-type activities that were viewed as a distinguishing feature of the Yes campaign. The members who joined after the referendum did not greatly influence the election campaign styles of either party.

#### **Unifying ideas**

But a movement is not only made up of organisations and repertoires. It consists of ideas that have broad appeal. The ideas that unified the Yes campaign and carried through into the parties were already evident in each party before the referendum. Independence was the key unifying idea in the SNP as we saw in our earlier study of that party. In the past, the main motivation for joining the Scottish Greens had been concern for environmental issues, and Scotland's constitutional status divided opinion in this party. The surge in Scottish Green membership on the back of the referendum has meant that independence has more appeal. As we reported in Chapter 5, other ideological motivations appeared more important for joining the Greens. That only one in ten of those who joined the Scottish Greens after the referendum referred to independence as their primary motivation might suggest that independence is not the unifying idea that might be expected if the party is to be seen as part of a broad independence movement. But this is not the whole story. Chapter 6 demonstrated that support for independence was widespread in both parties. The surge brought in independence-driven members to the Scottish Greens, but they were less strongly committed to the idea of independence than those who joined the SNP. The Greens had different ideological *priorities*, but they viewed independence as a way of delivering those aims. This broad support for independence among Scottish Greens made them a different party from before the referendum. Independence was articulated as a means to Green ends. Independence became a more unifying idea, both within the Scottish Greens and between the two parties, but not one that changed the ideological character of the parties fundamentally, meaning they remained different parties.

In their discussion of the Quebec sovereignty movement, Dufour and Traisnel (2014: 257) noted that it consisted of 'two types of protest (social protest and nationalist protest)', which had been distinct but 'quickly found common ground and common cause: taking control of the Quebec state and sharing a vision – an independent Quebec'. It would be wrong to suggest that the independence referendum fused two similar protests in Scotland. Evidence from previous studies shows that the SNP and Scottish Greens did not so much alter as use the referendum campaign to promote their ideologies. But this was a campaign and this meant that messages were expressed in broad-brush terms. While the movement leaned to the left, some within it leaned much further left than others, and it managed to accommodate others who leaned rightwards. The one unifying idea was independence, but there were tensions that were largely hidden for the purposes of campaigning on some aspects of what was meant by independence, for example, the currency and external relations. There was sufficient overlap in ideas across the two parties in left-right and conservative-liberal spectrums to allow for cooperation after the referendum, and this was clearly seen in the parties' memberships, as we discussed in Chapter 6. It is conceivable that some members of each party could comfortably move into the other, but that did not make the parties the same ideologically.

#### **SNP–Green deal**

There had been no need for an SNP/Green deal in 2011 as the SNP had an overall majority in Holyrood. However, the SNP found itself in a precarious position again after it lost its overall majority in 2016 and again failed to win an overall majority in 2021. This meant there was a prospect of losing confidence votes in Holyrood, which led to talks in Summer 2021 between the SNP and Scottish Greens on a new cooperation agreement. The process was carefully steered by the Scottish Green leadership. The party's new constitution outlined that if a coalition or cooperation agreement was proposed, this would trigger a process of consultation with members, but the precise nature of that consultation was open to interpretation, ensuring the leadership had some latitude. Online information events/Q&A sessions for members (national and regional) took place. The party's co-leaders presented the proposal as an important opportunity for party influence and development such as that enjoyed by other green parties and building on an existing relationship with the SNP (having backed four SNP budgets in the Scottish Parliament as well as the earlier cooperation agreement). During these sessions, activists expressed a desire to work on the links between the parliamentary and voluntary party, improving internal democracy and communication with members.

 Scottish Green members backed the deal (the Bute House Agreement) in an online extraordinary general meeting in September 2021 (Table 9.1). It required the agreement of the membership with a simple majority vote plus two-thirds majority of the party Council. Both elements must approve a deal. Approximately 20% of members took part. Harvie and Slater became junior ministers.

There was little evidence of dissatisfaction with the Green leadership among the wider membership, but some discontent has existed since 2014 among activists concerned about a concentration of power. An internal group called 'Greenroots' had campaigned for more grassroots democracy. When the party was forced to conduct meetings online due to the pandemic, concerns were raised that these events were more like presentations by the leadership than exchanges between members


*Table 9.1* Result of 2021 Scottish Green membership vote on political cooperation agreement

and leaders. These concerns culminated in a challenge to the leadership in 2022 with a proposal to separate the party's leadership from ministerial office. This provoked a wide-ranging discussion about the importance of dispersing power in the Greens. Activists called for a more radical model of internal democracy, and some suggested there existed a conflict of interest between responsibility to party and government. Discontent was defeated with a two-thirds majority against the motion, but a clear message of dissent had been sent to the leadership from a third of members.

The SNP also allowed its members to vote on the deal with the Greens. In August 2021, 95% of those who participated voted in favour of the deal. But only a tiny proportion of its members voted: 715 backed the deal, 84 opposed and 9 abstained in a membership that the SNP claimed was over 100,000. This meant that approximately 0.7% of SNP members had voted for the deal. Opposition to the deal grew within the SNP and became an issue in the 2023 SNP leadership contest with two of the three candidates critical of the arrangement.

#### **Sturgeon resigns**

Sturgeon's announcement in March 2023 that she was resigning as SNP leader and the SNP leadership contest that followed revealed tensions that had been simmering in the party. It became clear that SNP membership figure claims were false. The party was forced to admit that its claim to having 100,000 was inaccurate. The SNP had 72,186 members who were eligible to vote in the contest. In August 2023, the party published its annual accounts and revealed that membership had stood at 82,598 in December 2022 and as of June 2023 was 73,936 (SNP 2023: 8). Based on the party's own figures, membership had declined by 30,000 over a period of around two years. The SNP remained a much larger membership organisation than before the 2014 referendum, but an attempt to control the narrative at the cost of truth and accuracy created much negative publicity for the party and led to the resignation of Peter Murrell, the SNP's chief executive (and Sturgeon's husband).

Three SNP Members of the Scottish Parliament gained the necessary 100 nominations drawn from at least 20 SNP branches to allow them to stand for the leadership: Kate Forbes, Ash Regan and Humza Yousaf. Yousaf had the support of the outgoing leader and all other members of the SNP cabinet except Forbes and depute leader Keith Brown, and the endorsements of 35 of the SNP's MSPs and 21 of its MPs. Forbes was endorsed by 5 MPs and 16 MSPs. Regan had the support of one MP. Yousaf was keen to portray himself as the 'continuity candidate'.

Yousaf sought to frame the contest in terms of conservative versus progressive ideas aligning himself with the latter and presenting Forbes, his main opponent, as conservative given her views on abortion, gay rights and trans rights. Forbes sought to frame it as a contest in terms of governing competence and argued that Yousaf had shown little competence as a minister and she was critical of Sturgeon's record in office. There were differences on the agreement with the Scottish Greens, with the deal only likely to survive in the event of Yousaf's victory. Regan frequently referred to the arrangement as 'the tail wagging the dog'. Each candidate emphasised their support for independence, and each claimed to be able to deliver independence within five years. The key tension on the constitutional question was how this was to be achieved with Regan proposing a revival of the broadbased referendum campaign and the immediate establishment of an Independence Convention if elected leader. She said she approached a range of independencesupporting organisations but received 'no response' from the Scottish Greens. Forbes was critical of the lack of preparation for a second independence referendum under Sturgeon.

In the SNP leadership election, 70% of members voted, and 96% of these cast their vote online. Without survey data, we cannot know for certain the reasons for Humza Yousaf's narrow victory (Table 9.2). Given the support from across the SNP leadership and advantage in having early warning that a contest would take place, the narrow victory was far from a ringing endorsement. Our research suggests that this cannot be taken to mean that around half the SNP's membership is socially conservative and critical of the agreement with the Greens, although subsets of these opinions no doubt exist. The most plausible explanation for the outcome relates to who was deemed most likely to deliver independence regardless of views on other matters. The SNP has always seen itself as a movement campaigning for independence, whether part of a wider movement or the dominant organisational form of the movement, and a political party seeking public office. The tension in this relationship became clear over the period after the referendum as the SNP in public office struggled with the challenges of governing.


*Table 9.2* Result of 2023 SNP leadership election

This was evident in internal tensions on how the party was managed (see Chapter 8), which came to a head shortly after Yousaf's election when Murrell and the SNP treasurer were arrested, though not charged, by the police investigating accusations into the SNP's finances and then later when Sturgeon herself was arrested and again not charged in the ongoing investigation. Critics who had questioned the lack of transparency and accountability under Sturgeon had been marginalised by the SNP leadership but appeared to have been vindicated when a 'Governance and Transparency' review took place that resulted in reforms agreed by the SNP conference in October 2023. But the review proposed minimal changes that did little to reverse the SNP's top-down, electoral-professional structures.

#### **Final conclusions**

This book has been about an exceptional example of political mobilisation. Party members remain a small minority, as one of our interviewees (Interview no. 70) self-effacingly highlighted with this statement: 'We're a *tiny* minority of the population – both members and then those who are involved. We are the weird people. We are the strange ones'. Members and activists might be unusual in political behaviour terms, but they are the beating heart of political parties and are generally motivated by idealism. As we have seen, a surge can take place unpredictably and suddenly in a digital and social media age. The combination of ease of access, speed in 'spreading the word' and demand triggered by events can produce interesting results. There is an analogy here with the volatile nature of vote shares in dealigned electorates. Just as an increase in a party's appeal can suddenly expand its vote share, the same could prove true of its membership now that the path to joining is so much smoother. Based on this, we might expect more surges – probably not as dramatic but possibly the kind of spikes that might disrupt any broader downward trend in membership – at least when there is a significant trigger such as a referendum.

In the case of the SNP and Scottish Greens, the surge followed an unusual campaign and disappointment for those who advocated change in an independence referendum. After being on the losing side in the independence referendum, the SNP and Scottish Greens were compensated with a surge in membership. But this surge did not alter either party ideologically or organisationally as dramatically as the numbers might have suggested. Many of those who joined had not been active during the referendum and would not play an active role in party membership, and they were broadly in agreement with the positions of the party they joined. Most of these new members resembled party identifiers in their support without any intention of participating in party activities. If we conceive of support for a party as a series of concentric circles with super activists in the inner core, working out to inactive members then outside to party identifiers and then to people who voted for the party, then what appears to have happened is that many people we would normally classify as party identifiers, but not members, crossed the threshold to join the party they supported. Some intended to be active and were for a period, but as time went by such activism receded.

#### 158 *Conclusion*

These new members brought much needed money into the two parties, but the parties were left vulnerable to the kinds of fluctuations that the SNP had experienced in previous decades when membership had risen only to fall back after initial enthusiasm subsided. This weakness was evident when the SNP sought to defend a seat when a by-election was held in October 2023. Even at low points in the past, the SNP usually managed to mobilise activists for a by-election campaign. The expectation of those interviewed in the year or so after the surge had been that the new members would provide a formidable active base, but the SNP had to pay a company to deliver leaflets in the by-election and at a time when it was struggling financially. Both parties, though, have significantly more members than at the time of the independence referendum.

When we compare the two parties along the amateur-activist to electoralprofessional spectrum, the Scottish Greens may have moved in the direction of electoral-professionalism, with more focus on the national leadership, partly due to the media profile of the two ministers, but the party sits much closer to the amateur-activist end of the spectrum than the SNP does. The Greens appear to have handled the surge more successfully, in the sense that membership is close to post-referendum levels, but membership has been subject to fluctuation. Both parties have lost members who joined in the early surge. What is unclear is how many of the surge members remain or have lapsed and rejoined. Future research should follow the lead of Bale and colleagues (Bale et al. 2020, ch. 8; Barnfield and Bale 2022), examining decisions to leave a party and, in turn, giving us an idea of the 'decay rate' of spikes and surges like this one.

Jane McAlevey, activist and academic, referred in her memoirs to 'movement moments'. These are when:

 large numbers of people are willing to drop what they are doing, forget that the utility bill won't be paid on time or that they will miss their favorite TV shows or their daughter's soccer games or their gym session or whatever, forget about how many hours of sleep they think they need every night, and go do some stuff they would never have imagined they could. . . . People get in this unusual state either because they are truly pissed off and there is no other option, or because for some reason the horizon of what they think they are capable of achieving suddenly expands – or, most likely, a combination of both (McAlevey 2014: 11).

 But she cautions that 'movement moments don't last forever, and it is much easier to snuff them out than to keep them lit. Everything depends on optimism: the optimism organizers call "raised expectations"'(Ibid.). The surge occurred when many supporters of independence felt both disappointed and hopeful and wanted to channel these feelings generated by the referendum campaign. The SNP under Nicola Sturgeon sought to maintain the momentum by frequent injections of expectation that another referendum was imminent while maintaining strict control of her party. But the movement moment passed.

#### **References**


Johnston, H. (2014) *What Is a Social Movement?* Cambridge: Polity Press.


## **Methodological appendix**

We gratefully acknowledge the financial support of the Economic and Social Research Council (award ref. ES/N010590/1). The grant funded our project 'Recruited by Referendum'. The empirical analyses in this book are based on two main sources derived from this project:


This appendix provides more details on each in turn. Further methodological details on the main survey, and the data themselves, can be accessed via the UK Data Service.

#### **1. Membership surveys**

#### *Questionnaire content*

Many of the questions were taken from existing sources, including our own previous studies of the SNP and Scottish Green memberships, previous party membership surveys in the United Kingdom and further afield, and public opinion surveys, notably the Scottish Election Study and Scottish Social Attitudes series, which allowed for comparisons between the parties' memberships and their voting bases. We also developed new questions, especially about respondents' experience of the 2014 referendum, designed to shed more light on the drivers of the surge. We gratefully acknowledge advice and comments on the questionnaire design from both academic colleagues and staff from the two parties.

One key feature of the questionnaires is that we adopted a core-plus-modules design whereby respondents all answered a large common core of questions – the approximately two-thirds of the questionnaire deemed most central to our research interests – but were randomly assigned to either Version A or Version B of the


*Table A1* Summary of questionnaire content

survey, each of which contained smaller modules of questions which were of interest but were thought less central to the project aims. This approach enables us to expand the coverage of the survey while keeping the task for individual respondents reasonably manageable. This explains why the sample size reported in some of the tables is roughly half of what it is for other analyses.

Table A1 summarises the content of the questionnaires and the order in which the topics were asked about. We present just one table because, aside from one or two inevitably party-specific questions (such as those about voting behaviour in the SNP's depute leadership election), the questionnaires were parallel across the two parties.

#### *Fieldwork and samples*

The surveys were scripted online in Qualtrics by the researchers and fielded via that platform. Individual links were generated and sent to the parties, whose administrators then distributed these links to their own emailing lists. The online mode barely constituted a choice when compared to face-to-face or telephone surveying: the former would have been hugely expensive even for much smaller samples, while the latter would have meant a much lower response rate and much tighter constraints on the length of the questionnaire. In practice, the only choice was between an online survey and the postal mode that we used for the 2002 Scottish Greens and 2007–2008 SNP membership surveys. In the previous SNP study, the postal mode delivered a response rate of 54% (51% for the 2002 Green study). However, this was an expensive process involving three reminders (with an obvious environmental impact), and systematic analyses of mail response rates suggest that they have fallen by an average of 20 percentage points since that period (Stedman et al. 2019). Without the response rate advantage for postal surveys, the huge efficiency advantages of online surveys – including much greater flexibility in questionnaire design – was deemed more important.

Of course, both methods would have involved non-response biases, with mail survey samples skewed towards older respondents and web survey samples skewed the opposite way. With the digital divide narrowing all the time, the former bias seems likely to have been stronger, which is another reason why we opted for the online mode. What this does mean, however, is that comparisons with those previous postal surveys need to be undertaken with caution. In the main text of this book, we state the relevant caveats at several points. They are particularly applicable when comparing on a variable like age. Comparison is likely to be less distorted on variables like attitudes or activities since there is no particular reason to suppose online and postal respondents to be especially different on a variable like left-right position or likelihood of canvassing for the party.

The main SNP survey went into the field on 9 November 2016, a reminder was issued on the 20 December 2016 and the survey was closed on 2 February 2017. The Scottish Greens survey went into the field on 20 September 2016, a reminder was sent on 17 February 2017 and the survey was closed on 14 March 2017. In each case, the large majority of responses were within a few days of the initial email out, and the large majority of those remaining came within a few days of the reminder. At least by recent standards, this was a time of relative calm in Scottish and wider British politics – it followed the EU referendum but preceded the calling of the 2017 general election. So responses are unlikely to have been disrupted by events while the surveys were in the field and, crucially, comparison should not be distorted by the slightly different fieldwork periods across the two parties.

Links were sent to all party members on those emailing lists. This amounted to 77,778 contacts for the SNP and 8,110 for the Greens. There was therefore no sampling involved: we sought to survey the entire memberships of the parties. Sampling, a necessary evil in most survey contexts, was unnecessary because of the zero marginal cost per respondent in these online surveys. This does not mean that there was no sampling bias, however, because in both cases – especially that of the SNP – these mailing lists contain fewer contacts than the parties were reporting at the time as their total memberships. The gaps are partly due to some members not being accessible online, although both parties reported that this applied to a very small fraction (well below 10%) of their memberships. The SNP gap was also attributed by the party to some upheaval during a change of database, which meant that some members'details were not accessible at the time of data collection. (In the light of the controversy at the time of the party's 2023 leadership election, in which the electorate was revealed to be more than 30,000 short of the party's most recent membership report to the Electoral Commission in December 2021, there might be suspicions that our sampling frame of around 80,000 members was also rather closer to the actual membership numbers than the figures reported at the time. We have no particular basis for such suspicions, however, and instead find it more plausible that a plunge in SNP membership took place between 2021 and 2023.)

The total numbers and response rates were 16,101 and 21% for the SNP and 1,775 and 22% for the Scottish Greens. These response rates are low in absolute


*Table A2* Charity donations voted for by survey respondents

terms but not relative to those obtained from other online membership surveys, with which they are broadly in line.

While there was no payment to respondents for participating in the survey, the final question offered them the opportunity to vote for a charity to receive its share of a £2,000 fund – £1,000 per party – that we would donate on respondents' behalf. The donations made were as per Table A2.

#### *Non-response bias and weighting*

Response rates of around 20% create the potential for significant non-response bias: that is, for those responding to the survey to be unrepresentative of the membership in some key respects. The most obvious and probably the strongest biases will be on the cluster of variables measuring engagement and participation. For one thing, only those engaged enough to read the party's emails will have found out about the survey. More broadly, responding to surveys is itself a variant of political participation and is driven by many of the same factors – day-to-day interest in politics, strength of opinions and availability of free time – that drive political activity. For this reason, as highlighted at various points throughout the book, our estimates of the extent of members' activity – whether in the party, during the referendum campaign, or more generally in politics – are likely to be overestimated.

Much of the time in this book, however, we are interested less in the level or average of a single variable like party activism and more in how that level or average varies across the two parties or across cohorts within each party. And non-response bias is much less problematic for such comparisons. First, since the response rates were similar for the SNP and Scottish Greens, cross-party comparisons are valid to the extent – probably quite a large extent – that the factors driving non-response are similar across the parties. Similarly, provided that it was the most engaged among both the pre- and post-referendum joiners that responded, a comparison of those groups is also unlikely to be severely disrupted.

Nonetheless, it is worth considering methods to mitigate the impact of nonresponse bias. The standard method of correcting for such bias is weighting. This is straightforward in the case of opinion polls or voter surveys because of the

availability (via the census or official statistics such as election results) of accurate information about the target population. It is much harder here because there is very little information available about the target populations in this context, that is, the overall SNP and Scottish Green Party memberships. Even age and sex data were not always available or reliable, and so it is not possible to adjust for any tendency for, say, middle-aged or female respondents being likelier to respond. These tendencies are unlikely to be strong, though, in any case. Basic demographics are less likely sources of bias than the engagement variables mentioned earlier. However, since we have no reliable data on, say, the average number of hours that members invest in party activity or the percentage of members that canvassed for Yes in the 2014 referendum, we have no means of adjusting for the fact that the more active members and the referendum canvassers will be over-represented.

A more promising weighting variable is the timing of members' joining. This is useful because we might have expected – and Table A1 confirms – that longstanding members in each party were more likely to respond to the survey. Those who joined during the post-referendum surge were less likely to respond and therefore constitute a smaller proportion of the survey samples than of the parties' memberships (especially in the case of the Scottish Greens). By up-weighting the surge-joining cohort and down-weighting the existing members, we would achieve a more representative profile of the total membership in each party.

The difficulty is in estimating the scale of up- and down-weighting required. Precise calculation is not possible because the parties did not supply for each individual the timing of their joining. This timing can only be inferred from the aggregate membership numbers reported by the party, and this inference involves some simplifying assumptions. Our own simple assumption was that there was no leaving of the parties during the post-referendum periods under study. For example, if overall numbers increased by 25% between two time points, we assume that one-fifth of the resulting total membership is new. In normal times or between two distant time points, this assumption would be highly implausible: any net increase would be the result of a larger influx partly offset by members who leave. Even during the post-referendum surge, the assumption will not be 100% accurate: there will have been some surge joiners who quickly drifted away. Given both the buoyant mood in the parties and the short time period under study, however, it is plausible to assume that the very large majority of any increase in membership consists of new recruits. The key point is that the assumption need not be 100% correct. So long as it is broadly correct, weighting based on it will mean a more representative sample than if we used unweighted data. We therefore calculated a cohort weight, based on the data in Table A3, meaning that the survey results match the estimated cohort breakdown of the *membership* rather than the cohort breakdown of the survey *samples*.

There are several types of analysis in the book, and it is worth clarifying when that cohort weight is used.

1. Univariate analyses, profiling each membership or comparing the memberships on a single variable – a referendum behaviour, an attitude, a perception


*Table A3* Profile of survey sample (including cohort comparison with membership rolls)

of membership and so on. For these, we use the cohort weighting variable described before and calculated based on the proportions in Table A3.

2. Cohort comparisons, in which we compare members by the period of their joining the party. These rarely use the full breakdown in Table A3: much more often, there is either a simple pre-post referendum comparison or a three-category breakdown in which the surge joiners are separated from those joining in 2015 onwards. These are unweighted analyses because the weighting variable would have no effect on the results. (If we want to compare, say, the percentage of Remain voters among the various cohorts, the results would not be affected by a weighting variable, which simply changed the relative *size* of the cohorts.)

Multivariate analyses, typically regressions, predicting a behaviour, attitude or perspective were based on a range of independent variables. In line with long-standing methodological advice (and taking into account the fact that the cohort weighting variable is rarely very strongly correlated with the dependent variables estimated), we use unweighted data for these analyses (see Winship and Radbill 1994).

#### *Survey of 'mini-surge' joiners, 2018*

When the SNP membership surged again – by 5,000 in less than 24 hours – after its MPs' walkout at Westminster over the repatriation of powers from the EU following Brexit, the party agreed to field a greatly slimmed-down version of the original survey among the new recruits. This 'mini-surge' survey was in the field between 21 and 28 June 2018, and 4,115 responses were received.

The response rate to the survey is hard to calculate because the distribution was this time handled by the party, and we were not provided with a sampling frame and hence a denominator for response rate calculations. We do know that the SNP was reporting 118,200 members as of April 2018 and 125,500 as of August 2018, an increase of 7,300. What we do not know is: (i) how many of those joined *after*  the mini-surge survey was distributed (in which case the response rate denominator should be smaller because these people could never have responded) and (ii) whether this net increase of 7,300 reflects a bigger surge into the party partly offset by departures (in which case the denominator should be larger because there were more new joiners than the net figure suggests).

To calculate an indicative response rate, we need assumptions about these unknowns and, as before, we use the simplest assumptions: (i) that all of the 7,300 increase had happened by the time of the survey and (ii) that there were no departures and so the net increase equals the total influx. Again, neither is likely to be entirely accurate but both are largely plausible: mid-June was clearly the time of the big surge into the party, and it was generally a time of political gain for the party in a way that makes significant departures unlikely. (Moreover, if the assumptions are both slightly out, they will be so in cancelling directions as far as the response rate calculation is concerned.)

Our response rate estimate for the mini-surge survey is thus 56% (i.e., 4,115/7,300). This is appreciably higher than for the main survey, and there are at least two possible explanations. First, that main survey was at least two years after the referendum and the resulting membership surge, whereas this follow-up was landing in the inboxes of members who had joined just days before. (We had applied to our funders, the Economic and Social Research Council, for an emergency grant to get to respondents much sooner after the referendum surge but were turned down.) Second, the fact that this survey was distributed more directly by the party may have made respondents perceive it as a higher priority than the more academic enterprise that the main survey probably appeared to be.

#### *Survey of Women for Independence*

We also surveyed members of the campaign group Women for Independence (WfI). The survey was conducted in 2015, between 15 August and 30 September. The academic researchers constructed the questionnaire in consultation with WfI organisers, and the online survey was sent to members and supporters (those with email addresses) by the group. Information on the total number of individuals invited to participate in the survey was not made available, but we know that there were approximately 850 WfI members at the time. Overall, 536 responses to the survey were received, producing what looks like a very healthy response rate of 63%. However, around a third of the respondents to the survey described themselves as supporters rather than members. Supporters were defined in the questionnaire as those receiving newsletters and emails, and members as those with a subscription to the organisation. Therefore, the response rate among *members* is likely to have been more like 40%. The distinction between members and supporters, though, was not necessarily all that clear to the survey participants, so our estimates of response rate cannot be precise.

The survey explored the backgrounds and referendum experiences of the respondents, their political opinions and behaviours, their involvement in WfI and their views on political parties. WfI's activities during and after the referendum set it apart from the political parties, but the organisation included many party members. The survey results were informed by interviews at the time with senior members as well as discussion following a presentation of the findings at the WfI National Council in January 2016. In the book, the survey data are used mainly as a point of comparison in our account of the socio-demographic characteristics of the SNP and Scottish Green Party members reported in Chapter 4, but further interviews with WfI organisers and activists were undertaken between 2016 and 2018 (see the next section), and these inform our analyses more widely.

#### **2. Elite interviews**

We conducted an extensive programme of face-to-face, semi-structured elite interviews with senior figures and activists in the two parties and across the national movement, close to 100 in total. Interviewees included elected representatives (councillors and MSPs), party staff, local party and community activists, and special advisers (or SPADs) and MPs in the case of the SNP. Participants were selected according to their closeness to the events being investigated, many having roles connected to party membership and organisation or direct experiences of the membership surge, and an attempt was made to reflect the experiences of the parties and campaigners across Scotland. Interviews took place at party conferences, at party offices, on University premises, in some public spaces (coffee shops and restaurants), and occasionally in the homes of interviewees. Interviews varied in duration, ranging from approximately 40 minutes to more than two hours. Some interviews came about following the recommendations of those interviewed early in the study, a 'snowballing' technique (see Handcock and Gile 2011). Interviews were conducted in two phases:


#### **References**


### **Index**

Note: Page numbers in *italics* indicate a figure and page numbers in **bold** indicate a table on the corresponding page.

accessibility 5, 41, 77, 126, 137, 141, 162 Ackerman, Bruce 45 activists: active engagement 50, 80, 110, 114, 118, 151 (*see also* party membership); armchair activism 112, 114, 116; flash mobs 7, 25, 152; grassroots 8, 17, 21, 24 – 26, 34, 50, 89, 110 – 111; interviews 9, 11; radical 122, 126, 129; socio-economic status 120, 123; training 23, 25 – 26 (*see also* Yes groups); voluntary 44, 129, 136 affiliation option 5, 10, 46; *see also* religion Alba Party 152 Alternative Vote (AV) 20, 23 Anderson, Benedict 40 appendix 160 austerity/anti-austerity policies 6, 9, 44 – 45, 91 AV *see* Alternative Vote Bale, Tim (et al 2020) 4 – 5, 7, 9, 59, 61 – 63, 65, 67, 77, 116, 158 Better Together 21, **23 – 24**, 25 – 27, 50, 152 Bochel, John (et al 1981) 19 Brand, Jack 42, 44, 150 Brexit: re-allocation of power 10; referendum 22, 30, 33, 106; Remain/Leave campaigns **19**, 31, 92, 99 – 100 British: identity 92 – 95, **121**, 122; politics 35, 57, 61, 77 British Ecology Party/British Green Party 6, 28 Brown, Keith 136, 138, 156 Butler, David 21 Cameron, David 27, 30, 44 Canavan, Dennis 22, 27

Carville, James 27 Castells, Manuel 40 Catalan independence referendum 7 Catalonia, Spain 17, 42, 93 centralisation 5, 8, 130 Chapman, Douglas 138 Chapman, Maggie 137 Cherry, Joanna 138 Clark, Peter B. 76 class identity 68 – 69, **69**, 72, **121**, 122 climate change 32, 99 Cloward, Richard 41 coalition 7, 20, 26, 33, 149, 154 cohort comparison 66, 69, 71, 83, 97, 102 commitment: ideological 67, 75, 88; to independence 21, 44, 84, 91, 96, 102, 148; manifesto 18, 30 community: activism 25 – 26, 44, 57, 111, 167; connection 40, 63, 123, 153 Conservatives 7, 20, 30 – 31, 33 – 34, 61, **104** conviction 75, 86, 89 Corbyn, Jeremy 8, 33, 61, 63, 77 Cross, William P. 5 Darling, Alistair 50 data collection 13n7; comparison 11, 49, 58, 62, 64, 67; gathering methods 10 – 11, 24, 26, 117; member 4 – 5, 32, 47, 50, 164 (*see also* cohort comparison); unreliable 4, 24 Davidson, Ruth 33, 140 debate: constitutional 17, 21 – 22, 30 – 31, 108, 138; labour movement 42, 45 – 46; leadership 7, 30, 132; membership decline 4; televised 50 decentralisation 41, 43 – 44

decision making process 5, 41, 76, 84, 104, 130, 132 – 133, 138 – 139, 144 della Porta, Donatella (et al 2017) 9, 26, 40 – 41 Diamond, Larry 130 Diani, Mario 40 digital technology 5, 7, 57, 110, 112, 117, 121, 137, 157 diversity 57, 59 – 60, 65 – 66, 69, 72 Dommett, Katharine 7 – 8, 78 Dufour, Pascale 45, 153 Duverger, Maurice 110 Edinburgh Agreement 21 education 47, **67**, 71, 121, 124 election: campaigns 1, 51, 78, **88**, 110 – 111, 113, 153, 158; general 20, 28, 30, 50, 116; Holyrood 41, 46, 140, 142, 149, 151; party leadership 6, 33, 106, 137, 156; results 95; security 144 electoral: behaviour 17, 95; pressure 43; professionalism 41, 43, 130, 135, 139, 144, 149, 157 – 158; reform 20, 30 Electoral Commission 4, 21 – 23, **23**, 24, 32, 51 environmentalism 66, 83, 97, 107 – 108 EU (European Union): membership 2, 92, 94, 108; referendum 31 – 32, 95, 100, 115 European Community (EC) 18 events: cultural 25, 42, 47; online 139 – 140, 154, **155**, 156, 161; that trigger action 7, 139, 149, 152 Ewing, Winnie 28 expectations game 27, 124, **143**, 158 Facebook **117**, 140 Faucher, Florence 4 – 5, 80, 89 financial: costs 3, 79, 129; crisis 6; donations **88**, 112, 115, 118; hardship 31; report 4, 32 – 33, 35 fluidity 5, 150 Forbes, Kate 108, 155 – 156 Gamson, William A. 40 Gauja, Anika 57, 61, 63, 139 gender: demographics 58 – 59, **62**, 121; equality 61 – 62, 72, 141; transgender 60 getting out the vote (GOTV) 24, 26, 153 Gezgor, Burcu 59 Good Friday Agreement 20 GOTV *see* getting out the vote Gunther, Richard 130

Harper, Robin 22, 29, 43, 131 Harvie, Patrick 22, 83, 108, 131, 137, 143, **143**, 154 Heberle, Rudolph 40 Heidar, Knut 110 Holyrood: committee 31, 149; elections 41, 46, 140, 142, 149, 151 identity 40; British 92 – 95; commonalities 58, 69, 91, 95, **96**; dual 93; internationalist 67, 92, 95, 97, 108; left-wing 91 – 92, 98, 108; national/nationalist 69, **81**, 91 – 93, **93**, 94, **96**, 96 – 97, **121**, 122, 126; self-placement 94, **97**, **121 – 122**, 122, 154 ideology 41, 75, **81**, **83**, 84 – 85, 107, 112; *see also* motivation incentive 76 – 77, 80, 82, 89 inclusive 22, 58, 126 independence movement 10, 12, 22, 44, 52, 54, 151 – 152 intersection 41 Jackson, Stewart 61, 63 Johnston, Hank 34, 44, 46, 152 Katz, Richard S. 5, 129, 144 Kriesi, Hanspeter 53 Labour Party: government policies 18; member surges 1, 6 – 8; social characteristics 57 – 58; spending **24** leadership: assertion 135 – 136; challenge 126, 129; contests 5 – 6, 12, 33, 77; party/SNP 6, 21, 46, 50, 54, 106, 122, 137 – 138, 140, 142 – 143, 155 – 156; role 77, 84 LGBT+ **52**, 60, 63 Liberal Democrats (Lib Dems) 7, 20, **23 – 24**, 33 – 34, 61, 66 Lynch, Peter 25, 27 Mair, Peter 3, 8, 129 – 130, 144 mass movements 41, 129 Matalin, Mary 27 May, John D. 111 McAlpine, Joan 141 Michels, Robert 41, 129 motivation: ideological 75 – 77, 80, 82, 85, 123, 153; reason for joining **83**, *84 – 85*, **88**; as self-reported 80, **81**; surge motivations 77

movement: anti-nuclear 52; identities

**52 – 53**, **121**, 126; independence 10, 12, 22, 44, 52, 54, 151 – 152; poor peoples

41; social (*see* social movements)


research: academic 31, 166; British politics 57; methodology 76 – 77 (*see also* appendix); party membership 3, 7 – 9, 72, 75, 113; social movements 39, 45


Tarrow, Sidney 41 Think Twice campaign 20 Tierney, Stephen 26 Tilly, Charles 9, 11, 40, 149 Trainsel, Christophe 45, 150 Trident, removal of 91, **104**, 105 triggers: referendum triggers **81**, 82 – 84; understanding of 2, 7 – 8, 78 Twitter 30, **117**, 140

UK Independence Party (UKIP) 6, *34*, 61

United Kingdom (UK): electoral failure 6, 34; electoral reform 20, 30; ethnic diversity 60, 65, 69, 71; political institutions 40, 95, 98; referendums 17, **18 – 19**, 21, 30, 65; voter behaviour 31


Wales: Green Party of England and Wales (GPEW) **34**, 132; Plaid Cymru (nationalist political party) 34; Welch Assembly **19**, 20


Yes groups: activities **49**, 151 – 152; campaign merchandise 22, 25, 46

