0
ECPR Joint Sessions of Workshops
University of Antwerp, Belgium
10 - 15 April 2012
PRIMARIES AND POLITICAL PARTIES IN EUROPE
A PROPOSAL FOR A TAILORED ANALYTICAL FRAMEWORK
Giulia Sandri (University of Oxford)
Antonella Seddone (University of Cagliari)
Work in progress (there is no point in quoting this version)
1
1. Parties’ rationales for externalizing candidate and leader selection procedures
Most recent literature that analyzes parties from an organizational perspective focuses often on the
concepts of intra-party democracy and party organizational democratization (Scarrow, 1999; Scarrow and
Kittilson, 2003; LeDuc, Niemi and Norris, 2002; Bosco and Morlino, 2007). Le Duc (2001) and Rahat and
Hazan (2007) underline that the most used instrument for implementing this ‘democratization’ process is
the enhancement of the inclusiveness of the methods for candidate and party leadership selection. The
actors endowed with candidate and leader selection powers are the central actors in the functioning of the
party according to many authors (Gallagher and Marsh 1988, Marsh 1993; Massari, 2004; Hazan and Rahat,
2010). At the moment, the most inclusive method identified by the literature for selecting candidates for
elections or the party leader is represented by party primaries, i.e. internal direct elections by party
members and (sometimes) supporters and voters (Cross and Blais, 2008 and 2009; Kenig, 2009). Although
the literature on primaries is quite extensive, especially concerning the development of this instrument
within the US political system (see, for example, Ranney, 1972; Norrander, 1989; Palmer, 1997; Morton and
Gerber, 1998; Hopkin, 2001; Ware, 2002; Cohen et al., 2008) the analysis of the implementation of primary
elections outside the US and in particular within the European context is not equally developed (Heidar and
Saglie, 2003; Valbruzzi, 2005; Lisi, 2009; Wauters, 2009; Pasquino and Venturino, 2009 and 2010; De Luca
and Venturino, 2010, Seddone and Venturino, 2011). Nevertheless, this instrument has been adopted by
parties active in several European countries such as, for example, Finland, Denmark, France, Spain, Greece
and Italy (Laurent and Dolez, 2007; Lisi, 2009; Kenig, 2009; Mavrogordatos, 2005).
In particular, with regard to the European (or non-US) political context, the main dimension of primary
politics to be taken into account according to tailored analytical frameworks is the one dealing with party
organizational structures and particularly party membership role. Literature on party politics generally
argues that primaries represent a further step in the organizational evolution of political parties. Following
the analysis proposed by Katz and Mair (1993, 1994; 1995; 2002; 2009) we can easily see how parties have
progressively and strategically reduced the size of the “party on the ground”. The party in public office has
taken over the organizational role of mass membership. Political parties seem to find new legitimacy in the
participation in government rather than in social integration and encapsulation of voters and activists: the
result is a shift in the mobilizing dynamic of intra-party politics. In this perspective, party organizational
changes such as internal democratization processes and the adoption of primary elections are often
analyzed in relation to the evolution of party model and in particular of the model of mass party theorized
by Neumann (1956) and Duverger (1961). Trying to attract the median voter (Downs, 1956), political parties
are argued now to target their political message for all the electorate, adopting thus a catch-all approach
(Kirchheimer, 1966). The old organizational structures, rooted in the grass-roots membership and
ideologically distinctive, have been replaced by this new logic in the mobilization of party supporters. These
evolutions are often though to interact with (an in some cases explain) parties’ organizational shifts
towards (at least apparent) greater internal democracy. At any rate, when studying the explanatory factors
and the potential consequences of the adoption of primary elections by European parties, it appears rather
relevant to take into account previous literature on party models and party organizational transformations.
Moreover, other political dynamics, which to some extent are also specific to the European context, might
play a role in the analysis of the main dimensions primary politics. The processes of party personalization
and professionalization in communication strategies have been long described by party politics literature
and are argued to have replaced party ideological strength in its role of main instrument of interparty
2
competition (Panebianco, 1982; Poguntke and Webb, 2005). These new tools for mobilizing voters may
allow to attract new quotas of the electoral market, but do not guarantee a loyal and stable electoral
support (Dalton and Wattenberg, 2000), thus affecting negatively the transformation of voters into activists
(Raniolo, 2004; 2006). If we take into account the growing evidence on the generalized decline of party
membership and of election turnout in Western Europe (Scarrow, 2000; Scarrow, 1996), it seems that this
new approach to electoral and party mobilization might not be as effective as it was intended to be in
terms of outcomes of the new mobilizing strategies it entails (Mair and van Biezen, 2001; Mair, van Biezen
and Poguntke, 2011). Furthermore, the increasing spread of anti-party and anti-politics feelings among
citizens and voters, as well as the decline in the levels of political trust witnessed recently in many
advanced democracies, seem to strengthen the idea of the emergence and gradual deepening of the gap
between parties and their supporters (Bardi, 1996; Poguntke, 1996; Poguntke and Scarrow, 1996; Scarrow,
1996; Dalton, 2008).
Furthermore, primary elections are a recurrent theme in the debate about internal democracy, parties and
their organizational changes (Mair, 1994; Katz and Mair, 1995; Scarrow, 1999; Seyd, 1999; Katz, 2001;
Ware, 2002; Bolleyer, 2011). Especially within North American literature, there seem to be on this issue a
deeply rooted prejudice (Ranney, 1972; May, 1973). The idea is that primaries could lead parties to an
organizational and possibly electoral decline because they are thought to entail a gradual weakening of the
control exerted by the party leadership on the recruitment procedures and on the organizational
boundaries of parties, as well as to strengthen internal divisions and the autonomy of candidates from
party central bodies (Hopkin, 2001). Indeed, if candidates are selected by a wider electorate, they will be
responsive to this larger selectorate and not to the party (Ware, 2002). But then, in the European (or
rather, non-US) context, the literature recently noticed an increasingly extensive and strategic use of this
instrument for selecting candidates or party leaders. On the basis of this empirical fact, increasing scholarly
attention has been paid to the phenomenon and several explorative analyses of the political economy of
candidate and leadership selection rules have been recently developed (Hazan, 1997; Le Duc, 2001; Hopkin,
2001; Pennings and Hazan, 2001; Rahat and Hazan, 2006; Kenig, 2009). In the last decades European
parties have adopted a wide range of different types of internal elections and party politics scholars are
currently attempting to understand and conceptualize both the causes and the consequences of such
organizational changes.
In fact, primary elections for choosing party leaders and candidates are becoming usual events for
European parties. Although apparently foreign to European political culture and party systems, open (to all
voters) primaries for selecting party leaders and candidates to elected offices have been increasingly used
by European parties in the last decade (Kenig, 2009; Cross and Blais, 2012a). Moreover, closed primaries
(also called the “One Member One Vote” system) have been adopted by parties in many Western and
Eastern European countries since the 1970s. Straying from the original model of US primaries, where the
direct primaries have been used for more than two centuries (Ranney, 1972; Palmer, 1997; Cohen et al.,
2008), European parties have re-adapted primaries to their needs. This adaptation process is mainly due to
the different role played by parties in the national party systems within the European context and it is also
related to the different challenges that parties have to face in the contemporary European societies. The
organizational nature of parties in Europe is very far from the one existing in the US. The very concept of
party membership, so crucial in the study of primary elections, has a completely different meaning in the
US and the European political context (Stone, Rapoport and Schneider, 2004; Heidar, 2006). The linkage
between voters, parties and representatives is developed on the basis of significantly different dynamics in
the two contexts.
3
Moreover, the effects of the adoption of primary elections on parties’ electoral dynamics are also highly
contested within the US literature on the subject. In fact, the question that arises here concerns the
electoral gain in promoting primaries. Literature on primaries is controversial on the issue of the electoral
appeal of candidates selected through primary elections. If the literature is quite varied on the trade-off
between the openness and inclusiveness of the candidate selection process and the electability of these
candidates, the question of the negative impact in electoral terms of these inclusive internal elections is still
under discussion. The differences in the ideological positions of the general electorate and the selectorate
mobilized by primaries (Kaufmann et al., 2005; Norrander, 1989) are generally thought to explain the
argued negative effects in electoral terms. In this perspective, the results of primaries in terms of
participation could be difficult to be valued by the general electorate (Adams and Merrill, 2008). Other
studies focus the attention on the negative stance and the aggressive discourse of primary campaigns. The
mutual de-legitimization between primary candidates could disclose and emphasize internal conflicts and
unsolved rivalries (Djupe and Peterson, 2002; Haines and Rhine, 1998; Peterson and Djupe 2005). Finally,
other stances on this issue refer to the concept of divisiveness (Atkeson, 1998; Bernstein, 1977; Born, 1981;
Hacker, 1965; Hogan, 2003; Johnson et al., 2010; Kenney, 1988; Kenney and Rice, 1984; Makse and Sokhey,
2010; Piereson and Smith, 1975; Ware, 1979; Wichowsky and Niebler, 2010). The idea is that divisiveness in
primary elections could demotivate party members and supporters. In fact the high level of
competitiveness could affect the electoral choices of loser candidate’s supporters leading to their electoral
defection.
In terms of party image, primary elections can be considered as a tool used by parties in order to
compensate the loss of legitimacy towards the electorate, to regain political credibility and to attract new
supporters. Thus, parties are thought to provide more internal decision-making power to their grass-roots
members and supporters as an incentive to their own membership to mobilize internally and to present a
public image of being open and ‘democratic’ (Mair, 1994, Seyd, 1999; Scarrow, 1999; Scarrow, Webb and
Farrell, 2000). Primaries represent a new pattern in the relationship between parties and their supporters.
On the one hand, the adoption of internal direct elections contributes to incentive the internal mobilization
of members already enrolled in the party, proposing new activist proceedings that in some way could
represent a re-edition of the traditional mobilizing strategies of mass-based parties. On the other hand,
primaries provide new opportunities for participation to those citizens less inclined to intra-party,
traditional activism. In fact, the open and inclusive character of this instrument incentive new typologies of
political participation, which do not require any formal affiliation to the party, but instead develop an
intermittent participatory behaviour that concerning in particular voters interested by cognitive
mobilization (Dalton 2008). In this perspective of political economy of leadership and candidate selection
methods, primaries are considered mainly as a tool used and promoted by parties with the specific goal of
building a new relationship with supporters that is subsequent to their own catch-all electoral strategies.
These two main dimensions of primary politics outlined here and concerning the internal and external
consequences of such instruments, still need to be systematically and theoretically addressed within the
European contexts. Hazan and Rahat (2007 and 2010) describe four dimensions for evaluating the
functioning of leader and candidate selection methods: participation, representation, competition and
responsiveness. Even though the literature on the political consequence of leadership and candidate
selection processes outside the US is quite varied (Obler, 1974; Rapoport, Abramowitz and McGlennon,
1986; Hazan, 1996; Rahat and Sher-Hadar, 1999; Meirowitz, 2005; Cermel, 2007; Maravall, 2008; Hazan,
2002; Barnea and Rahat, 2007; Rahat et al., 2008, Rahat, 2009; Kenig, 2009), it is also highly fragmented,
while cross-national and comparative studies have been developed only very recently. Moreover, the
4
analyses of candidate and leader selection methods and other intra-party elections have been generally
limited to a description of the current and previous situations (Faucher-King and Treille, 2003; Heidar and
Saglie, 2003; Seyd and Whiteley, 2004) or to an evaluation of the influence that the party on the grounds
can exert through these processes (Scarrow, 1999; Wauters, 2003).
The aim of this paper is thus to launch a scholarly debate on the most pertinent way for approaching the
analysis of primary elections within the European political context. We attempt at providing an exploratory
analytical framework for documenting and evaluating the impact of primary elections on political parties.
Given the new relevance of this organizational and political instrument within European party politics, it
appears rather crucial to find appropriate analytical tools for understanding its functioning and its
consequences. The classic literature and theoretical approaches to the study of primaries has been
elaborated on the basis of the American experience. Given the significance of the difference between the
structures of European countries’ political systems and those present in the US, we need to elaborate new
analytical schemes that stemming from the classical US theoretical and empirical models could be adapted
to the European electoral, party and political context in the most effective way. In this paper, we will thus
try to propose a few operationalization reflections, hoping to stimulate a scholarly debate on the subject.
Our aim is to understand which are the main changes triggered by the adoption of these inclusive
procedures for selecting leaders and candidates, both in terms of external (electoral) and internal
(organizational) dynamics.
In fact, the literature is often conflicting with regard to the organizational and electoral changes entailed by
the adoption of primary elections, especially open ones. For instance, on the one hand some scholars have
underlined the political shortcomings with regards to internal cohesion that may be associated to the
introduction of this instrument for selecting leaders and candidates: primary voters are usually considered
to be more ideologically extreme than the general electorate and this might entail significant consequences
in terms of candidates electability in general elections (Key 1956; Lengle 1981; Polsby 1983; Colomer 2002).
On the other hand, other studies pointed out the fact that (open) primaries could negatively affect the
candidate loyalty to the party, because his/her nomination is legitimated outside the party, directly by
primary voters (Hopkins 2001). Then again, other scholars argue that primaries allow party central
leadership bypassing the control of middle-level elites and local organizers and thus tend to increase their
autonomy and their power (Katz and Mair, 1993 and 1995). Conversely, some studies argue that the
differences between the selectorate and the electorate in terms of ideological positions are not that
relevant (Geer 1988; Kaufmann et al. 2003) and that candidates selected through primary elections actually
do not tend adopt extremist platforms, because they aim at preserving their ideological loyalty to the party
(King 1999; Hansolabehere et al. 2001).
5
2. Primary elections in the European context
Before presenting an overview of the potential consequences of primary elections on European parties, it
might be useful to briefly provide a general definition of the concept. Party leader and candidate selection
through direct election is often labeled as open or closed primary election, although the adoption of the
terms of “primaries” or “primary elections” for designing those specific internal elections, either open to all
members or the wider electorate, is somewhat controversial in the European literature (Fabbrini, 2002;
Valbruzzi, 2005 and 2008; Lisi, 2009; Castaldo, 2009; Gangemi, 2009; Pasquino and Venturino, 2010;
Lefebvre, 2011). In fact, Hazan and Rahat (2006) define as non-party or open primaries all the methods for
leader and candidate selection that are open to all voters. They define, on the other hand, as party or
closed primaries (or also OMOV systems, ‘one member one vote’) all the methods for leader and candidate
selection that are open only to party members (Valbruzzi, 2005; Rahat and Hazan, 2007). In the American
literature, on the contrary, primary elections are defined as those elections in which party members or
voters select candidates for a subsequent election (Ware, 2002). As Ware pointed pout in his work on the
American direct primary, nowadays in the US in almost every state the “selection of candidates for public
office takes the form of an election that is organized by a state agency, rather than parties themselves, and
is subject to state law” (2002:2). From this basic definition of primary elections within American political
settings, we can easily see the three main points of variation between the American direct primaries and
the direct election of party president within European parties:
1. The instrument has been elaborated mainly in the framework of the nominating procedures for electoral
mandates rather than for party internal mandates;
2. The primary election is thought to be regulated by (state) law and organized by external bodies and not
directly by the party;
3. There is a direct and essential link between the nominating procedure and the subsequent (general)
election. This representation link is reflected in the requirement of an overall correspondence between the
primary selectorate and the general electorate. According to some authors, then, we can define as primary
elections those electoral competitions open not only to party members but also to registered voters and
sympathizers (Fusaro, 2006: 44).
We are aware that using the same concept for identifying two objectively different phenomena such as
primary election in the US and internal elections for selecting the party leader within Western European
parties might represent a clear case of “concept misformation” (Sartori, 1970: 1038). Nevertheless, we
underline here the absence of specific research tools for studying such new organizational features of
European parties and also the undeniable similarity of several organizational dimensions and of various
symbolic aspects between the internal direct elections for selecting the leader and candidates in Europe
(both OMOV systems and those open to the electorate at large) and the primary elections for selecting
presidential and gubernatorial candidates in the US (Rahat and Kenig, 2011). Although European primaries
for selecting the leader and/or the candidates for office are mainly organized and financed by parties
themselves, in many cases the regulation of such instruments is provided by state law, as for instance is the
case in Germany or Finland (Ranney, 1981: 81). In many other cases, also, the selection body is open to the
entire general electorate and not only to party members. Moreover, beside such practical and
organizational similarities as those outlined above, the main argument in favor of using the concept of
6
primaries within the European context, in particular for studying the elections for selecting party leaders, is
that the linkage with subsequent elections might be latent but is generally present. Although the party
leader is not technically and automatically a candidate for general or presidential elections at the moment
of the selection procedure, in most parties the leader is automatically designed as candidate for Prime
Minister (or President) in subsequent elections (Valbruzzi, 2005; Kenig, 2009b and 2010). In conclusion, in
this paper we propose few suggestions on the potential development of an analytical framework for
studying in a comprehensive way the various dimensions of primary politics, taking into account at the
same time all those party internal ballots used either for choosing the leader or the party candidates for
office and open either to registered members or to party voters.
3. Exploratory suggestions for evaluating primaries’ impact on political parties
In this section, we elaborate a preliminary, exploratory proposal of an analytical framework for explaining
in a comparative perspective the impact of primaries on political parties both internally (on other
organizational features and settings) and externally (on their electoral performance). Tables 1 and 2
summarize our suggestions for identifying the most pertinent indicators for assessing the impact of
primaries on political parties. In fact, within North American literature, there seems to be a lack of
consensus on the actual consequences of primary elections on other party organizational dimensions
(Ranney, 1972). According to some authors, primaries inevitably lead parties to organizational decline,
because they entail the explosion of party organizational boundaries, while exacerbating as well internal
divisions and increasing the degree of candidates’ autonomy from party central bodies (Pomper, 1977;
Born, 1981; Wattenberg, 1998). Moreover, the effects of the adoption of primary elections on party
electoral dynamics are also highly contested within the studies on the American experience. In fact, the
question that arises here concerns the electoral gain in promoting primaries. We have seen in the first
section that he theoretical models are quite varied on the trade-off between the inclusiveness of the
candidate selection process and the electability of these candidates. These two main dimensions of primary
politics, concerning the internal and external consequences of such instruments, still need to be
systematically and theoretically addressed within the European contexts.
Given these gaps in the existing literature on the European cases and the contradicting evaluations of the
US literature, the fact that the increasing use of primary elections in Europe has triggered relevant
organizational and strategic changes within parties (Hopkin, 2001) and the lively political and scholarly
debate that ensues, we develop a proposal of analytical framework. The framework looks at two distinct
aspects of the consequences of primaries: organizational aspects and electoral ones. Thus, fiver different
dimensions will be analysed: membership role and internal factionalism, leadership type and autonomy
(internal aspects), electoral performance of the party, candidate responsiveness and party image (external
aspects). For each dimension of analysis we suggest a set of empirical indicators. The measurement of the
variables proposed here might be implemented on the basis of original data on party organizational
structures, on electoral data (votes and turnout at subsequent general elections, votes, competitiveness
and turnout of the primary elections) and on secondary data for measuring internal factionalism, leadership
style and autonomy.
7
Table 1. Operationalization of the main consequences of the adoption of primaries: internal aspects.
Impact of primaries on political parties
Dimensions
of primary
politics
Research questions Indicators Empirical methodology
INTERNALLEVEL
MEMBERSHIP
RQ. To what extent
primaries enhance
participation by
attracting new
members?
1. Evolution of membership figures
2. Degree of membership mobilization
(members/primary voters)
Aggregate data and
meso (party) level data:
membership figures;
primary turnout
figures, aggregate
electoral data
Individual level data:
qualitative interviews,
quantitative surveys
(candidates and party
members)
RQ. To what extent
primaries enhance
ordinary members’
mobilization within the
party?
1. Absolute membership turnout in
primary elections
2. Degree of involvement in primary
campaigns
3. Evolution of the degree of
participation in intra-party activities
before and after primaries
RQ. To what extent
primaries affect
members’ attitudes
towards the party?
1. Evolution of the degree of party
identification
2. Sense of political (external) efficacy
3. Members’ attitudes towards primary
elections’ rules and functioning (degree
of satisfaction with regard to the
nature of the selectorate, the
nomination rules and primary
campaign)
RQ. To what extent
primaries affect internal
party loyalty?
1. Attitudes towards the selected
nominee of losing candidates’
supporters
2. Perspective candidates’ voting
intentions in general elections
3. Perspective candidates’ attitudes
towards party reaction to primary
results
4. Perspective candidates’ attitudes
towards the degree of competitiveness
of primary elections
LEADERSHIP
RQ. To what extent
primaries strengthen the
party leadership?
1. Leaders’ longevity in office
2. Share of preference votes obtained
in general elections by the
leader/candidate selected through
primaries
3. Degree of cohesion of the dominant
coalition (measured by endorsements)
Party statutes and
constitutions, official
communication and
statements, media
briefs, documents
regarding party
conventions
Electoral data at
aggregate level
Qualitative interviews
Endorsement data
RQ. To what extent
primaries weaken party
cohesion and stimulate
internal conflict?
1. Evolution of the number of party
congressional proposals and motions
2. Evolution of the degree of
divisiveness in party communication
8
Table 2. Operationalization of the main consequences of the adoption of primaries: external aspects.
Impact of primaries on political parties
Dimensions
of primary
politics
Research questions Indicators Empirical methodology
EXTERNALLEVEL
CANDIDATE
RESPONSIVENESS
RQ. To what extent
primaries shift the focus
of candidate/leaders
responsiveness and
accountability towards
the external political
arena?
1. Evolution of the parliamentary
behaviour of candidates:
1.1. Degree of consistency of
candidates’ legislative initiatives with
party manifestos and policy agenda
1.2. Candidate voting behaviour in
parliament (degree of deviation from
party indications)
Legislative proposals,
roll call voting data
Individual level data on
candidate behaviours
(quantitative surveys)
ELECTORAL
PERFORMANCE
RQ. To what extent
primaries foster party
electoral performance?
1. Index of participation (primary
elections turnout/share of party votes
at previous general elections)
2. Index of mobilization (primary
elections turnout/ share of party votes
at following general elections
3. Share of preference votes obtained
in general elections by the leader or
candidates selected through primaries
Electoral results
(primary and general
elections)
PARTYIMAGEANDPUBLIC
OPINION
RQ. To what extent
primaries convey a more
democratic image of
political parties?
1. Evolution of the degree of political
trust and of trust in parties
2. Voters’ attitudes towards the degree
of competitiveness of primary elections
3. Evolution of the degree of
competitiveness of primary elections
4. Degree of “negativity” in campaign
communication
5. Party central elites’ renewal rate
Qualitative interviews
and secondary data
Data on party
organization (offices in
central bodies)
Data on candidate
profiles
Individual level data:
quantitative surveys
(public opinion, voters,
candidates and party
members)
9
4. Internal level
Primary elections could affect parties at organizational level even more strongly than general elections. The
potential consequences of primary politics on party internal structure are due to the nature of the
instrument itself. The selection of leaders and candidates through an inclusive selectorate constitutes a
new competitive arena for party internal factions. Especially in the case of open primaries, the exogenous
legitimation of the selected candidates or leader and the participation of voters and supporters could
supply an opportunity for minority factions and for individual ambitions to defy the party leadership and its
dominant coalition. This means that within primary competition there is more than the mere question of
nominations. Thus, in order to understand the functioning of main dimensions of primary politics, we need
to focus on party organization and particularly on the evolution of the internal relationship between
members and the party, both at organizational level (by looking at the changes in party rules, boundaries
and membership size) and at individual level (by looking at members' attitude towards their role within the
party and towards this participatory procedure). Also, a relevant dimension to be taken into account is
represented by the leadership's ability to deal with the political issues raised by primary elections and to
face the risk of a potential shift in the internal distribution of decision-making power.
A. Primary elections’ potential impact on membership
Primary elections, both open and closed ones, might represent a modern transposition of (or a viable
alternative to) the old participatory procedures of mass parties, where members and activists played a
crucial role within party organization. The (at least apparent) devolution of decision-making powers to the
wider membership and party supporters (in the case of open primaries) entailed by this kind of internal
elections could strengthen activists and ordinary members' propensity to get involved in intra-party
activities. Instruments of intra-party democracy such as primaries could also affect party membership
structures by attracting new members. With regard to these two points, nevertheless, it is important to
distinguish between the potential effects of open and closed primaries. If the impact on internal
participation of the devolution of decision-making power to ordinary members through the adoption of
OMOV systems for selecting candidates and leaders might be considered to be more direct (more
instruments for participation stimulate higher involvement) and easily measurable, the impact of open
primaries to the attitudes and behaviours of ordinary members can be considered more controversial.
Members might negatively evaluate the inclusion of party voters and simple supporters within internal
decision-making, especially with regard to such crucial and symbolic decisions as the selection of
candidates for office and leaders. These negative attitudes could then discourage members’ involvement in
intra-party activities.
Furthermore, in the case of both open and closed primaries, the high degree of inclusiveness of the
selection procedure might be considered as a threat to the traditional internal functioning of the party. This
could entail a differentiation in the attitudes towards the party according to membership duration, and
thus between old and recently affiliated members, and also according to the degree of previous
involvement in party activities, and thus between active and passive member (Abramowitz, McGlennon and
Rapaport, 1986; Russell, 2005; Van Holsteyn and Koole, 2009). Particularly in the case of closed primaries,
one of the most relevant political opportunities triggered by the adoption of primaries is the development
10
of a new kind of relationship between party and its members. Primaries, by evocating a more direct,
participatory mode of political participation, could indeed incentive intra-party mobilization of affiliated
members and stimulate new forms of involvement of supporters and voters (in the case of open primaries).
However, if the evolution of overall turnout in open primaries over time might represent a pertinent and
straightforward indicator of the impact of this instrument on party's mobilizing appeal, in the long term the
consequences on membership internal mobilization and role within party structures appear to be more
complex to assess.
a) To what extent primaries enhance participation by attracting new members?
Given that one of the main reasons for parties to adopt intra-party democracy instruments such as primary
elections is to open their organizational boundaries and to stimulate enrolment, the first indicator we
suggest to explore is obviously the evolution of overall membership figures in order to assess to what extent
the adoption of primaries attract new members and whether these new cohorts of affiliates remain active
within the party or end up being “instant members” (Rahat and Hazan, 2007). The second indicator is the
index of membership mobilization, which is calculated, for the case of open primaries, by dividing the
number of affiliated members by the overall figures of primary voters and which grasps the participatory
reaction to primaries through the incidence of activists on the overall primary turnout. In the case of closed
primaries, conversely, the index is measured by the share of members that participated to the internal
election over the total membership figures.
Beside the quantitative evaluation of the participatory dimension of primary politics, it might be useful also
to address the evolutions in the quality of intra-party involvement that are argued to be linked to the use of
primary elections. The adoption of primaries for selecting leaders and candidates could trigger significant
evolutions not only in the quantitative aspects of intra-party mobilization but also in the kind of activities
carried out by activists and ordinary members. If we argue that primaries could represent a way to provide
an effective replacement to the traditional channels of intra-party participation no longer available (or
viable) within contemporary parties, we can consequently assume that the recent evolutions in the nature
and type of membership involvement within parties that use primaries can be explained mainly by the
adoption of these inclusive procedures of decision-making.
In terms of empirical methodology, this can be measured only at individual level, by collecting data on the
intensity, quality and type of intra-party mobilization through tailored quantitative surveys of party
members, and possibly panel surveys repeated over time. Such data would contribute in evaluating
whether primaries and the wave of political mobilization they usually entail could have consequences on
the long term, especially open ones and in terms both of campaign activities and turnout. These data could
also contribute in assessing whether, conversely, primary elections represent a tailored participatory event
exclusively related to the issue of candidate or leader selection, without any impact on members'
propensity to mobilize within the party beyond participating in internal elections. Thus, we need to collect
data on the relationship between members and the party, focusing particularly on the evolution of
membership internal mobilization. This will lead us to the next research question.
b) To what extent primaries enhance ordinary members’ mobilization within the party?
Besides the assessment of the evolutions in the overall degree of membership mobilization, also the
evaluation of the qualitative dimension of membership involvement appears to be relevant here. This
means trying to understand to what extent primary elections could contribute in reshaping the nature and
11
type of intra-party participation. Besides the rather straightforward evaluation of the absolute membership
turnout in primary elections, which focuses on the the mobilization appeal of primaries (both open and
closed) on the members, one can look also at other aspects of their behavior within the party. For instance,
we propose to look at the degree of involvement in primary campaigns (particularly in the case of open
primaries). Tailored quantitative surveys of party grass-roots members could shed a light on the intensity of
involvement and on the participatory reaction to this intra-party democracy instrument of ordinary
affiliates. A third possible indicator is represented by the evolution of the degree of participation in intraparty
activities before and after primaries, which could contribute in assessing whether participating in
primary elections represents for ordinary members just a one-time political activity within a pattern of
intermittent mobilization, or this could trigger a pattern of more stable and intense participation,
enhancing thus intra-party activism in the long run.
A further step in the analysis of the consequences of the use of primaries on party membership is to assess
not only the evolutions in membership behaviors, but also in their attitudes towards the instrument itself
and other attitudes towards the party. It would be pertinent to investigate members’ perception of
different dimensions of primary politics and their level of satisfaction with these particular internal
elections. We know from literature that primaries represent a new way for parties to distribute collective
and selective incentives to members in order to foster their participation, so it might be useful to
empirically assess to what extent this new balance of incentives could change (or even damage) the
relationship between grass-roots members and party leadership. In this case too, quantitative surveys
could represent the most effective empirical tool, because they allow analyzing in detail members'
attitudes and perceptions on a wide range of aspects of intra-party organization, such as for instance the
role played by party elites within the so called “invisible primaries”, namely the process of definition of
rules on passive and active selectorates, or the type of primary campaign conducted by perspective
candidates or leaders. We are thus interested in the point of view of members on the consequences of
internal democratization. Very few studies have explored intra-party democracy, and particularly primaries,
and their consequences from the point of view of the actors mainly concerned by these reforms, namely
rank-and-file ordinary members (Scarrow, 2005; Lyons, 2009; van Holsteyn and Koole, 2009).
c) To what extent primaries affect members’ attitudes towards the party?
The third dimension that we have identified as relevant for understanding the impact of primaries on the
relationship between parties and their members is the one related to membership perception of intra-party
democracy. The set of indicators we suggest here are thus aimed at investigating not only members' degree
of satisfaction with the instrument of primaries per se, but also other attitudes towards the party such as
the degree of ideological identification and the sense of belonging to the organization. Quantitative surveys
of grass-roots members remain the most useful empirical tool for collecting this kind of data. We suggest to
develop three indicators with regard to this dimension: the first is represented by the level of party
identification, which evaluates the degree of ideological proximity between members and their party on
the classical left-right axis or concerning its policy agenda and thus with regard to the degree of party
ideological representativeness and responsiveness to membership demands.
The second indicator deals with members' sense of (external) political efficacy and specific support (trust)
for the party, both in general terms (with regard to members' role within the party) and, more specifically,
concerning party rules on nominations, candidacy and voting systems. The sense of external political
efficacy concerns the evaluation of the responsiveness of party organizational structures to members’
demands and to the individual’s reported belief that their participation matters. In particular, it deals with
12
members’ perception of the relevance of their vote in primary elections. The latter variable, more
specifically, links the way the individual perceive his/her role within the organization he/her belongs to and
the way he/her exploits the opportunities for mobilization provided by the organization’s structural
features. The sense of (incumbent-based) political trust and of specific support for the party deals with
members’ satisfaction with party performance, in terms of representation and internal democracy
(Gamson, 1968; Craig et al., 1990; Niemi et al., 1991).
A third indicator contributes in analysing members' evaluation of the structural aspects of primaries. This
indicator, namely members’ attitudes towards primary elections’ rules and functioning, deals with the
degree of satisfaction of grass-roots members with regard to the nature of the selectorate, the nomination
rules and the type of primary campaign. These organizational changes are giving members more say over
outcomes: but are members satisfied with primaries? The analysis of the process of rules definition and
adaptation to the reactions of the party base might allow grasping the ability of party elites to address
primary results.
d) To what extent primaries affect internal party loyalty?
The fourth research question that we suggest to address, with regard to primaries' impact on membershipparty
relations, concerns the electoral strategies implemented by members in reaction to primaries. Also
this research question is developed at individual level. If we consider, as explained above, that the
dynamics of primary competition tend to spur factions and party internal conflict, we can also suppose that
this enhanced internal conflict could affect members’ electoral choices in general elections. Thus, it appears
rather relevant to identify which are the main voting intentions of primary voters (either members, in the
case of OMOV systems, or primary voters, in the case of open primaries) in case of defeat of their chosen
candidate (either for party leadership or perspective candidate for general elections). In fact, in the case of
primary elections for selecting the party leader, the latter is often automatically designated as the party
candidate for prime ministerial office at subsequent general elections. Thus, members’ voting intentions in
general elections could also be affected by the outcome of primary elections for selecting the party leader.
This is a field of analysis highly debated within the literature on primary elections politics. Some scholarly
analyses blame the divisiveness in primary competition for the potential negative performance of the party
at subsequent general elections, underlining the possible risk of defection (either abstention or dissenting
vote) in general elections of those primary voters that supported the losing primary elections candidates
(Piereson and Smith, 1975; Kenney, 1987; Johnson et al., 2010). Other interpretations focus rather on the
degree of “negativity” (i.e. the strategy of trying to gain electoral advantages by referring to negative
aspects of other perspective candidates) of primary campaigns for explaining the potential increased gap
between the party and its members (and primary voters) with the result of potential defection in general
elections (Ansolabhere, 1997; Djupe, 2002; Peterson and Djupe, 2005). With regards to this point, it might
be helpful to look at the voting intentions of losing candidates' supporters towards the winner, both in the
case of primaries for selecting the leader and for selecting candidates for general elections. A quantitative
survey of the voting intentions at general elections of primary voters (open primaries) or party members
(closed primaries) remains the most appropriate empirical strategy concerning this indicator.
Nevertheless, given that the research question dealing with the electoral strategies of party members in
reaction to primaries involves also the party and its internal organization, we suggest taking into account
also the reaction of party elites to primary elections' results. In particular, it would be useful to analyse the
communication strategies and reactions of party leadership and central bodies concerning primary results,
13
assessing whether the selected candidate or leader obtains the official support of the whole party or,
especially in the case of highly divisive primaries, the nomination process triggers a crisis in the internal
balance of power. We suggest looking at three different indicators, measurable mainly through qualitative
semi-structured interviews: the (losing) candidates’ voting intentions in general elections; the candidates’
attitudes towards party reactions to primary results; the candidates’ attitudes towards the degree of
competitiveness in primary elections.
B. Primary elections’ potential impact on leadership and internal power distribution
Also concerning the internal dimension of primary politics, it appears relevant to take into account the
potential consequences of primary elections on the other side of the membership-party relation, namely
the one of leadership. Especially in the case of parties using open primaries for selecting the party leader,
the role of the latter is nowadays legitimized through a procedure which is highly inclusive yet external to
the party organizational boundaries. This fact might affect the dynamics of leadership's loyalty towards the
party. Thus, it might be useful to explore more in depth the extent of the evolutions in the leadership-party
relations, in order to assess the scope of change in the role played by leaders and in the features of their
legitimacy towards party members, voters and supporters. With respect to this point, it is obviously
necessary to distinguish between the effects on leadership’s role of primaries used as candidate selection
procedures and of primaries used as leadership selection procedure. It is necessary also to distinguish the
effects of open and closed primaries. For the moment and for simplicity reasons, we will focus on the
effects of open primaries on leadership-party relations.
We consider that the dynamics and consequences on party leadership's accountability and reponsiveness
of closed parimaries might be more complex to apprehend. This is because the leadership accountability
dimension of primary politics could be affected more strongly, within this type of primaries, by the
organizational specificities of the selected case. We will develop further the reflection on the identification
of possible indicators for analyzing the consequences of closed primaries on leadership accountability in
the next version of this paper. As an example, we can look at the case of the UK Labour and at the various
studies that tested consequences of the adoption of the Electoral College method for selecting the party
leader: several authors underlined that these organizational reforms resulted in a decreased accountability
of the party elites towards members and activists and in an increased centralization of decision-making
powers in the leadership’s hands (Shaw, 1996; Quinn, 2005; Russell, 2005; Faucher-King, 2008).
Several scholars underlined the potential (negative) effects on the accountability and responsiveness
dimension of this inclusive method for selecting candidates and leaders (Lisi, 2009). Indeed, if candidates or
leaders are selected by a wider selectorate as it happens in open primaries, there are high chances that
they will become more responsive to the ideological positions and demands of this larger group of selectors
rather than their party. They might also end up feeling more accountable to the primary voters that elected
them than to the party central bodies and middle-level elites. Thus, this process can ultimately affect party
internal cohesion, leadership loyalties and the capacity of the leader to exert its control over the
organization. A potential outcome of such dynamics might be that the leadership would be legitimized by
political actors external to party structures, namely primary voters that may or may not be party members,
but less so internally, with consequences on its support by members and middle-level elites. This could
affect the profile, quality and duration in office of the leadership. Hence, it might be useful to clarify these
potential implications on the basis of both quantitative (surveys, electoral data at aggregate level) and
14
qualitative research tools (semi-structured interviews and endorsement data).These techniques could help
in investigating the evolution of the degree of approval and support for the leadership over time and across
different groups. We identified two main macro-dimensions of analysis on the issue of party leadership, the
first one is related to the leadership’s effectiveness in managing exogenous participatory dynamics that
could generate shifts in the internal balance of power, the second one concerns its ability to manage
potential internal conflicts triggered by the results of primary elections.
a) To what extent primaries strengthen the party leadership?
One of the most debated hypotheses about primary elections concerns the idea that the inclusiveness of
this selection procedure could entail a consolidation of leadership control over internal decision making
processes. The inclusiveness assured by primaries could generate a direct link between the leadership and
either voters (in the case of open primaries) or members (in the case of closed primaries), hence bypassing
the role of party intermediate level elites and activists in controlling elites’ decisions. In other words,
primaries allow elites to rearticulate their power within the party, by diminishing the role of traditional
party organizational structures and by providing an external legitimation (in the case of open primaries).
Internal democratization reforms that broaden the party organizational boundaries to include, in such a
crucial decision such as the selection of the leader or the candidates, all individual party members, also
those that are less socialized within the party or are less active, or in some cases even party voters and
supporters, could correspond to a strategy for “emasculating” activists and increasing in the end the elites’
control over party decision-making. Several authors, from Mair (1994) to Katz (2001) to Webb (1994)
argued that the adoption of primary elections for selecting the leader or the candidates for office would
increase the leadership autonomy and boost the concentration of decision-making powers in its hands.
In order to understand the impact of primaries on leadership autonomy and power, we suggest focusing on
few indicators. The first one concerns the leader’s longevity in office and is obviously related only to the
case of primary elections for selecting the leader. This indicator might be useful particularly for analysing
highly divisive primaries, because in this context the internal cohesion could be jeopardized by primary
dynamics and outcomes and consequently the leadership, and particularly its capacity to run internal
decision-making processes, could be weakened. The second indicator regards the electoral results gained
by the nominees, either the selected leader or candidates for elections (namely the share of preference
votes obtained in general elections by the leader/candidate selected through primaries). This indicator is
aimed at measuring the electoral appeal of the leader or candidates, comparing his/her electoral
performance considering his/her modality of selection. This indicator might contribute in understanding the
consequences of the outcomes of a given primary election because it grasps the degree of
“personalization” of party politics triggered by the use of primaries. Personalization of politics is one of the
most useful concepts in understanding current political processes, and primaries - even more than general
elections - bring out the relevance of the role of personal charisma, individual profile and personality of
perspective candidates and leaders. Hence, it might be useful to analyse the personal contribution of the
leader or candidate selected through primaries to the party electoral results.
The third indicator deals with the extent to which party elites can steer primary voters’ preferences and is
measured by analysing elites’ endorsement towards the primary candidates (Cohen et al. 2008; Mayer and
Bernstein, 2011). This indicator could help in measuring the degree of cohesion of the dominant coalition
within the party because, as previously suggested by several studies applied to the American case (Cohen,
Karol, Noel and Zaller, 2002 and 2003), endorsements to primary candidates seems to represent a useful
strategy for mapping party internal factions. Collecting data about elite endorsements allows to identify
15
internal networks and patterns of power distribution and thus to analyse the role of party organization and
its dominant coalition within the primary election process. The map of party internal factions that could be
elaborated on the basis of such empirical explorations contributes in measuring the overall level of elite
cohesion.
b) To what extent primaries weaken party cohesion and stimulate internal conflict?
Building on the components highlighted in the previous sub-section, this dimension looks at the long term
impacts of primaries and takes into account not only the power and functioning of the dominant coalition
within the party but more generally it looks at the overall degree of intra-party cohesion. The aim is to
investigate the internal consequences of primaries in terms of the relationship between the leadership and
its opposite factions. Taking into account the potential divisiveness triggered by primary competition and
its effects on party organization, we identified three pertinent indicators aimed at measuring the level of
internal cohesion. We suggest adapting and exploring some ‘traditional’ indicators, such as for instance the
evolution of the number of party congressional proposals and motions, which could contribute in clarifying
the internal level of party cohesion in ideological and policy position terms. Also the analysis of party
political communication could be helpful with regard to this point. The qualitative and quantitative content
analysis of party elites’ political speeches during and after primary campaigns could provide useful
information on the level of negativity of the latter. The study of the evolution of the degree of divisiveness
in party communication could thus contribute in explaining the impact of primaries on the leadership-party
relations, helping in interpreting the dynamics of primary competition.
5. External level
As widely assessed in empirical terms by the literature on the American experience, the effects of the use
of primary elections for selecting leaders or candidates go well beyond the organizational dimension and
concern primarily the external side of party politics. Leaders and candidate selection procedures entailing a
highly inclusive, participatory electoral contest often characterized by high political personalization might
affect in a crucial way the overall political representation process and this from several points of view. The
first component is the one related to responsiveness and concerns the changes in the focus of leaders and
candidate accountability. Especially in the case of open primaries, the legitimacy of the nominee comes
from an external body and this could generate a shift in the dynamics of internal accountability by
marginalizing the role of control played by party central bodies. Accountability concerns the obligation of
leaders and candidates selected through primaries to answer for their political decisions and the control
that their principals can exert upon their actions. The adoption of open primaries can transform the
features of their principals, which change from party middle-level elites, central bodies or members
(according to the previous procedure of selection) to primary voters. A direct link with the selectorate is
created and this is also true in the case of closed primary elections, which shift the focus of accountability
from party elites (usually the parliamentary party group) or central bodies to the wider membership, either
active or not. The second component of this dimension is represented by the impact of the use of primaries
on the electoral performance of the party. The third and last component is represented by the reaction of
public opinion to the adoption of primaries and concerns the need of political parties to address such issues
as the spread of anti-party and anti-political feelings, which could undermine their legitimacy as
instruments of political representation, by redefining their public image.
16
A. Potential impact of primary elections on candidates’ responsiveness and accountability
With regard to this first component of the external dimension of primary politics, we suggest that unlike
nominations by party elites, which incentive candidates and leaders to take into account elites’ indications,
policy positions and claims in their legislative behaviour, primary elections due to their highly inclusive
nature tend to transform the dynamics of both responsiveness and internal accountability. Hence, we try to
suggest here a few exploratory indicators that can contribute in understanding to what extent the
behaviour of nominees (candidates for elections and/or leaders) selected by primaries remains loyal to
party elites’ indications and, conversely, to what extent the emergence of external legitimation processes
generate nominees’ political attitudes distant from party ideological positions and policy agenda. With
regard to this component of the external dimension of primary politics, thus, we have identified one main
research question:
a) To what extent primaries shift the focus of candidate/leaders responsiveness and accountability towards
the external political arena?
Concerning the potential changes in the party base of leaders and candidates’ legitimacy, we identified
found two main indicators that could contribute in understanding the implications of the adoption of
primaries on this variable, namely the parliamentary behaviour of the nominees selected through primary
elections. Thus, it is worth noting that this first research question is more pertinent in the case of primary
elections for selecting candidates for office rather than party leaders, given that it makes little sense to
explore the degree of loyalty of leaders towards their own party in terms of parliamentary behaviour. The
first indicator deals with the degree of consistency of candidates’ legislative initiatives with party
manifestos and policy agenda and measures their propensity to follow the party line as opposed to their
tendency to promote other policy priorities. In empirical terms, this indicator can be measured by the
exploration of the type and number of individual legislative proposals and initiatives and also by collecting
individual level data on self-reported behaviour in parliament though quantitative surveys. These data
would be aimed at assessing the degree of individual behaviours consistence with party manifestos and
ideological positions, on the one hand, and with party policy agenda on the other hand. The second
indicator deals with the candidate voting behaviour in parliament and thus grasps the degree of deviation
from party indications within the legislative process. In empirical terms, this indicator can be measured by
the analysis of patterns of roll call voting.
B. Potential electoral impact of primaries
The main aim of primary elections is to mobilize party members, in the case of closed primaries, or
sympathizers and voters, in the case of open primaries. In the first case, the inclusion of the whole
membership within such a crucial decision-making process such as the selection of candidates or the leader
could trigger an increase in the degree of electoral mobilization of party activists and ordinary members,
particularly with regard to the campaign. This could consequently affect the electoral competition both at
primary and general elections level. In the case of open primaries, the inclusion of sympathizers not
formally affiliated to the party within the internal decision-making processes (and in particular in the
leadership and candidate recruitment processes) could expand the support for the party both at electoral
17
and organizational level. Therefore, the question that arises here concerns the electoral gain in promoting
primaries. It is thus relevant to investigate the electoral appeal of candidates and leaders selected through
primary elections. This is the dimension of primary politics that has been most widely explored within the
literature on the American experience and thus we dispose of a large array of potential relevant indicators
and variables identified by previous studies which can be easily adapted to the European context. We will
thus just provide here a brief overview of the most pertinent ones.
Nevertheless, the literature on US primaries is controversial on this issue, especially with regard to the
consequences of primary elections aimed at selecting candidates for office. If there are several studies that
theoretically and empirically explored the trade-off between inclusiveness of candidate selection
procedures and the electability of these candidates, the issue of the negative impact in electoral terms of
the most inclusive internal elections is still strongly debated. The differences in the ideological positions of
the general electorate and the selectorate mobilized by primaries are generally thought to explain the
negative effects in electoral terms of the adoption of primary elections (Kaufmann et al., 2005; Norrander,
1989). In this perspective, the fact that the usually high turnout in primary elections could trigger an
increase in the electoral consensus for the party in general elections could be difficult to grasp by the
general electorate (Adams and Merrill, 2008). Other studies focus on the contrary on the negative stance
and the aggressive discourse of primary campaigns. The mutual de-legitimization between primary
candidates could disclose and emphasize internal conflicts and unsolved rivalries (Djupe and Peterson,
2002; Haines and Rhine, 1998; Peterson and Djupe 2005). Finally, other scholarly opinions on this issue
generally refer to the concept of internal divisiveness (Atkeson, 1998; Bernstein, 1977; Born, 1981; Hacker,
1965; Hogan, 2003; Johnson et al., 2010; Kenney, 1988; Kenney and Rice, 1984; Makse and Sokhey, 2010;
Piereson and Smith, 1975; Ware, 1979; Wichowsky and Niebler, 2010). The idea is that the high level of
divisiveness which often characterizes primary elections could demotivate party members and supporters
in the following general elections. The collection and analysis of electoral data concerning both primary and
general elections could contribute in clarifying these issues. With regard to this component of the external
dimension of primary politics, thus, we have identified one main research question:
a) To what extent primaries foster party electoral performance?
Investigating primaries’ impact on electoral performance requires to develop the analysis on two different
levels: the party level and the individual candidate level. In fact, we need to understand whether it is the
party or the individual candidates that gain the greater electoral advantage from the use of primaries. Thus,
the first indicator we suggest here concerns the potential impact of the degree of participation in primary
elections on party results in general elections. Although primary elections turnout could be equally relevant
for explaining the party performance in general elections both in the case of closed and open primaries, it
appears rather clear that this indicator is more pertinent in the case of primary elections open to the whole
party electorate. We suggest thus to focus here on the case of open primaries and to develop an index that
could take into account party electoral results and primary elections participation, namely an “index of
participation”, which is calculated by dividing the turnout in primary elections by the share of party votes at
previous general elections. The index measures thus the proportion of party voters that bothered to show
up and vote in primary elections, giving information about the primaries’ ability to mobilize party voters.
Another way to look at this issue is to measure, on the contrary, the proportion of primary voters that
decided to support the party also in the following in general elections. In this case, we suggest the use of an
“index of mobilization”, which is calculated by dividing the primary elections turnout by the share of party
votes at the following general elections, giving information on primaries ability to attract new voters.
18
The second level of analysis concerns the electoral appeal of candidates or leaders nominated by primaries.
The question that arises here deals with the extent to which primaries foster the individual electoral
results. In order to explore empirically this issue, we suggest to develop a “personalization index” which
measures the share of preference votes obtained in general elections by the leader or candidates selected
through primaries. Although this indicator could be also used with regard to the internal dimension of
primary politics, and in particular for assessing the degree of autonomy of leaders or candidates selected
though primaries (see Table 1), it is at the same time a pertinent instrument for measuring the electoral
impact of these intra-party democracy procedures.
C. Primary elections’ potential impact on party image
It is generally argued in the literature that anti-party and anti-political feelings are strongly rooted in postmodern
society. Post-materialism, new forms of political participations, higher levels of education and
greater availability of cognitive resources have led to a sort of self-sufficiency in the exercise of
representation by citizens and voters. At the same time, the literature on party linkage argues that there is
nowadays an increasing distance between parties and society which generates a relevant gap in the
political representation process. The latter is also supposed to trigger the emergence of a generalized
feeling of discontent, distrust and rejection towards parties (Gamson, 1968; Dalton and Wattenberg, 2000;
Dalton, 2002; Norris, 2002). The issue at stake here is whether the adoption of such intra-party democracy
instruments as primaries, thanks to their transparent and inclusive procedures, could compensate this lack
of legitimacy of parties towards the electorate. In this view, participatory instruments such as primaries are
adopted in order to improve the public image of political parties (Dalton, 2004). The question that arises
here is thus the following: to what extent primary elections, which are supposed to strengthen internal
democracy, can compensate the growing lack of legitimacy of parties with regard to the electorate at large?
a) To what extent primaries convey a more democratic image of political parties?
The empirical assessment of the impact of primaries on the public image of political parties is rather
complex and might require a multidimensional approach. We propose to develop the analysis concerning
three very different aspects: the political attitudes of citizens towards political parties and primary
elections; the features of primary processes that incentive the competitiveness of the internal electoral
contest (divisiveness, negativity); the potential organizational changes at the level of party elites. These
three components concern thus very different dimensions: individual attitudes, political communication,
organizational features of the selection procedures and party organizational aspects. To start with the first
component, we propose to focus on indicators measuring the individual attitudes towards the party and to
test their correlation with indicators measuring voters’ opinions on the use of inclusive candidate and
leadership selection methods. Thus, on the basis of quantitative instruments such as public opinion and
voter surveys, it would be helpful to assess the evolution of the degree of political trust and of trust in
parties and to assess the link of the latter with the citizens’ satisfaction with primaries and with individual
perceptions of the democratic nature of primary elections.
Secondly, given that we aim at assessing the extent to which primaries could be used by parties in order to
counter the pervasive antiparty rhetoric, it could be pertinent to collect individual level data on primary
voters’ attitudes towards the degree of competitiveness of primary elections. These data could contribute in
explaining whether primary voters, either party members or the electorate at large, perceive primary
19
elections as being real and transparent competitions, or as simple coronations of nominees already chosen
by party elites. If the degree of competitiveness of primary elections is perceived as satisfactory by primary
voters, this could lead members and voters in general to consider primaries as an effective instrument of
intra-party democratization. Thirdly, it could be useful to take into account not only voters’ attitudes
towards the competitiveness of primary elections but also to look at the actual evolution of the degree of
competitiveness of primary elections. A pertinent indicator is thus represented by the difference in the
share of votes of candidates in (open or closed) primary elections. Moreover, with regard to this point, it
could be useful also to consider the political communication adopted by candidates in primary elections as
a pertinent indicator of the degree of competitiveness of the contest. On the basis of quantitative and
qualitative analyses of the content of candidates’ speeches and official statements during primary elections
campaign and of their relative degree of negative rhetoric it is possible to assess the degree of “negativity”
in the political communication strategies of the primary campaign.
Finally, one of the most important messages presented by parties in order to counter antiparty arguments
concerns the renewal and turnout of political elites. The idea is that primary elections could incentive
renewal processes within parties. These party organizational changes triggered by primaries concern thus
their public image and not only the internal distribution of decision-making power as we have explored in
the previous section of this paper. The main indicator for assessing the degree to which primaries trigger
organizational renewal is represented by party central elites’ turnout rate.
6. Conclusions.
Since the mid-nineties, a substantial body of scholarly literature on primary politics outside the US especially
within European and Israeli parties - has been developed. Some studies focused on the main
rationales for adopting such intra-party democracy instrument, other explored its functioning in various
national and cross-country contexts, and other investigated its consequences on parties themselves in
terms of electoral performance both in primary and general elections and in terms of internal organization,
cohesion, power distribution and unity. In this paper, we do not try to elaborate new or original approaches
to the subject, either theoretical or methodological or empirical. Instead, we started from the existing
literature on European primaries, we selected the more common and relevant analytical elements and then
attempted at providing a synthesis of the main dimensions of primary politics analyzed in previous studies.
In particular, we focused on the potential organizational changes and electoral consequences triggered by
the adoption of such inclusive methods of leadership and candidate selection, rather than on main
rationales for using this instrument. This is because, given the enduring relevance of party organization in
the European (and non-US) political context, the question of potential decline of party structures entailed
by primaries appears to be still at the centre of the political and scholarly debate.
Moreover, we consider that, unlike the century old, burgeoning and extremely well developed literature on
the American experience of primary elections, the “younger” literature on primaries outside the US might
not have yet reached a consensus on some fundamental definitions and methodological aspects – what is a
primary election? Which are the main dimensions to be explored when investigating the consequences of
primaries on political parties? – given the peculiarity of each case, both at party and country political
system level. The aim of this paper is thus to attempt at formulating a (at least partial) systematization of
the main elements analyzed by previous studies and to supply a (very) general overview of the subject.
Hence, we have tried here to provide a preliminary, exploratory set of suggestions for elaborating a
20
comprehensive analytical framework for investigating the consequences of main dimensions of primary
politics within the European context. We tried to identify a few key pertinent variables and indicators that
could be developed for analysing at the same time, and using similar research design, both open and closed
primaries and both procedures for selecting leaders and candidates. Some indicators remain relevant or
useful only for some specific types of primaries (i.e. leadership duration in office could be conceptually
linked only to the measure of the potential consequences of primaries for selecting the leader), but the
great majority of analytical dimensions suggested in this paper refers to a general, comprehensive
framework for studying this phenomenon as a whole. The idea was to launch the discussion, on the one
hand, on the main elements to be taken into account when studying primaries and, on the other hand, to
set the grounds for identifying some common definitions of the most relevant concepts and variables of
this area of research.
21
References (to be completed).
Adams, J. and Merrill, S., Candidate and party strategies in two-stage election beginning with a primary, in
«American Journal of Political Science», LII, 2, 2008, pp. 344-359.
Atkeson, L. R., Divisive Primaries and General Election Outcomes: Another Look at Presidential Campaigns,
«American Journal of Political Science», XLII, 1998, pp. 256-71.
Baras, M.; Barbera, O., Teruel, J. and Barrio, A., Representation within political parties: Spanish party
delegates’ opinions on intra-party democracy. Paper presented at the XVII ISA World Congress of Sociology,
Goteborg, 11-17 July 2010.
Bardi, L., Anti-party sentiment and party system change in Italy, in «European Journal of Political Research»,
XXIX, 1996, pp. 345-363.
Barnea S., Rahat G., Reforming Candidate Selection Methods: A Three-Level Approach, «Party Politics», XIII,
no. 3, 2007, pp. 375-394.
Bernstein, R. A., Divisive Primaries Do Hurt: U.S. Senate Races, 1956-1972, «American Political Science
Review», LXXI, 1977, pp. 540-45.
Bolleyer, N., Inside the Cartel Party: Party Organisation in Government and Opposition, «Political Studies»,
LVII, 2009, pp. 559-579.
Bolleyer, N., New Party Organization in Western Europe: Of Hierarchies, Stratarchies and Federations,
«Party Politics», forthcoming 2011.
Born, R., The Influence of House Primary Election Divisiveness on General Election Margins, 1962-1976, «The
Journal of Politics», LXIII, 1981, pp. 640-61.
Bosco A. and Morlino L. (eds), Party Change in Southern Europe, London, Routledge, 2007.
Blondel, J. and Thiébault, J.L., Political leadership, parties and citizens, Routledge, London and New York,
2010.
Cermel M., Dalle primarie alle liste bloccate (e ritorno?). Riflessioni sulla nuova legge elettorale e sulle
elezioni primarie, in «Teoria politica», XXIII, no. 1, 2007, pp. 99-121.
Cohen, Marty, David Karol, Hans Noel, and John Zaller. (2002). Beating reform: The resurgence of parties in
presidential nominations, 1980-2000. Paper presented at the annual meeting of the American Political
Science Association, Boston.
Cohen, Marty, Hans Noel, David Karol, and John Zaller. (2003). Polls or pols: The real driving force behind
presidential nominations. Brooking Review 3 (Summer): 36-39.
Cohen, Marty, Hans Noel, and John Zaller. (2004). Constant rules, changing game: A general model of
presidential primaries, 1972 to 2000. Paper presented at the annual meeting of the American Political
Science Association, Philadelphia.
Cohen, Marty, Hans Noel, and John Zaller (2008).The Party Decides: Presidential Nominations Before and
After Reform. Chicago: University of Chicago Press.
Cross W. and Blais A., Who selects the Party Leader? A Cross National Analysis, Paper presented at the
PSA’s Elections, Public Opinions and Parties Annual Conference, Manchester, UK, 12-14 September 2008.
Cross W. and Blais A., Who Selects the Party Leader? Motivations for Change and Implications for Party
Democracy, Paper presented at the ECPR Joint Sessions of Workshops, Lisbon, 14-19 April 2009.
22
Dalton R.J., Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies,
Chatham, Chatham House Publications, 2002.
Dalton, R.J. and Wattenberg M.P. (eds), Parties Without Partisans. Political Change in Advanced Industrial
Democracies, Oxford, Oxford University Press, 2000.
Dalton, R. J., Cognitive Mobilization and Partisan Dealignment in Advanced Industrial Democracies, «The
Journal of Politics», XLVI, 1984, pp, 264-84.
De Luca, R. and Venturino, F. (eds.), Mobilitazione e partecipazione elettorale. Un’analisi delle ‘primarie’ per
l’elezione del primo segretario del PD, Roma, Aracne, 2010.
Djupe, P.A. and Peterson, A.M., The impact of negative campaigning. Evidence from the 1998 senatorial
primaries, in «Political Research Quarterly», LV, 4, 2002, pp. 845-860.
Downs, A., Teoria economica della democrazia, Il Mulino, Bologna, 1988.
Duverger, M., I partiti politici, Milano, Edizioni di Comunità, 1961.
Faucher-King, F. and Treille, E., Managing intra-party democracy: comparing the French Socialist and British
Labour conferences, «French Politics», I, 2001, pp. 61-82.
Gallagher, M. and Marsh, M. (eds.), Candidate Selection in Comparative Perspective. The Secret Garden of
Politics, Londra e Newbury Park, Sage, 1988.
Hacker, A., Does a «Divisive» Primary Harm a Candidate’s Election Chances?, «American Political Science
Review», LIX, 1965, pp.105-10.
Haynes A.A., Flowers J.F. and Gurian P., Getting the Message Out: Candidate Communication Strategy
During the Invisible Primary in «Political Research Quarterly», September 2002, vol. LV, no. 3, pp. 633-652.
Hazan R.Y. and Rahat G., Democracy within parties, Oxford University Press, 2010.
Hazan R, Candidate selection, in Le Duc, Niemi and Norris, Comparing democracies 2. New Challenges in the
Study of Elections and Voting, London, Sage, 2002, pp. 108-126.
Hazan R., Rahat G. (2006). Candidate Selection, in Richard Katz and William Crotty (eds), H
Hazan R.Y., The 1996 Intra-Party Elections in Israel: Adopting Party, «Electoral Studies», XVI, 1996, pp. 95-
103.
Heidar, K. and Saglie, J., A Decline of Linkage? Intra-Party Participation in Norway 1991–2000, «European
Journal of Political Research», XLII, 2003, pp. 761–86.
Heidar, Knut 2006. ‘Party membership and participation’ in Handbook of Party Politics. Edited by Richard S.
Katz and William Crotty, Sage, London, pp. 301-315..
Hogan, R. E., The Effects of Primary Divisiveness on General Elections Outcomes in State Legislative
Elections, «American Political Research», XXXI, 2003, pp. 27-47.
Hopkin, J., Bringing the members back in? Democratizing candidate selection in Britain and Spain, in «Party
Politics», VII, 2001, pp. 343-361.
Johnson, G.B., Petersheim, Meredith-Joy and Wasson and Jesse T. Divisive Primaries and Incumbent
General Election Performance:Prospects and Costs in U.S. House Races, «American Politics Research»,
XXXVIII, 2010, pp. 931–955.
Katz, R.S. and Mair, P., Changing models of party organization and party democracy. The emergence of the
Cartel Party, in «Party Politics», I, 1995, pp. 5-28.
Katz, R.S. and Mair, P., How Parties Organize, Sage, 1994.
Katz, R.S. and Mair, P., The Ascendancy of the Party in Public Office: Party Organizational Change in
Twentieth-Century Democracies, in «Political Parties», IX, 2002, pp. 113–136.
23
Katz, R.S. and Mair, P., The Cartel Party Thesis: A Restatement, in «Perspectives on Politics», IV, 2009, pp.
753–766.
Katz, R.S. and Mair, P., The evolution of party organizations in Europe: Three faces of party organization,
«American Review of Politics», XIV, 1993, pp. 593-617.
Katz, R.S., The Problem of Candidate Selection and Models of Party Democracy in «Party Politics», VII; 3,
2001, pp. 277 -296.
Kaufmann, K.M., Gimpel, J.G. and Hoffman, A.H., A promise fulfilled? Open primaries and representation, in
«Journal of politics», LXV, 2, 2003, pp. 457-476.
Kenig, O., Democratization of party leadership selection: do wider selectorates produce more competitive
contests?, «Electoral Studies», XXVIII, 2009, pp. 240–247.
Kenney, P. J., Sorting Out the Effects of Primary Divisiveness in Congressional and Senatorial Elections,
«Western Political Quarterly», XLI, 1988, pp. 765-77.
Kenney, P. J. and Rice, T. W., The Effect of Primary Divisiveness in Gubernatorial and Senate Elections, «The
Journal of Politics», XLVI, 1984, pp. 905-15.
Kirchheimer, O., The transformation of the Western Europe Party System, in J. La Palombara e M. Weiner
(eds.), Political parties and Political development, Princeton, Princeton University Press 1966, pp. 177-200.
Laurent, A. and Dolez, B., Une primaire à la française. La désignation de Ségolène Royal par le parti
socialiste, «Revue française de Science Politique», LVIII, 2007, pp. 133-161.
LeDuc, L., Democratizing party leadership selection, in «Party politics», VII, 3, 2001, pp. 323-341.
LeDuc, L., Niemi, R.G. and Norris, P. (eds.), Comparing democracies 2: new challenges in the study of
elections and voting, London, Sage Publications Ltd, 2002.
Lisi, M., The Democratization of Party Leadership Selection in Comparative Perspective, Paper prepared for
presentation at the ECPR general conference, Potsdam 10-12 September 2009.
Mair P. and van Biezen I., Party memberships in twenty European democracies, 1980-2000, «Party Politics»,
I, 2001, pp. 5-22.
Mair, P., Party Organizations: From Civil Society to the State, in R.S. Katz e P. Mair, How Parties Organize,
Sage Publications Ltd, 1994, pp. 1-22.
Mair, P.; van Biezen, I.; Poguntke, T. (2011). Going, going….gone? The decline of party membership in
contemporary Europe, European Journal of Political Research, forthcoming, 2011.
May J., Opinion Structure of Political Parties: the special law of curvilinear disparity, «Political Studies», XXI,
pp. 135-51.
Mayer, W.G.; Bernstein, J. (2011). The making of the Presidential Candidates 2012. Lanham: Rowman &
Littlefield.
Makse, T. and Sokhey Anand E. Revisiting the Divisive Primary Hypothesis: 2008 and the Clinton−Obama
Nomination Battle, «American Politics Research», XXXVIII, 2010, pp. 233-265.
Marsh M., Introduction: Selecting the party leader, in «European Journal of Political Research», XXIV, 1993,
pp. 229-231.
Mavrogatos G. Th., Greece, in «European Journal of Political Research», XXIV, 2005, pp. 229-231.
Massari, O., I partiti politici nelle democrazie contemporanee, Bari, Laterza, 2004.
Meirowitz A., Informational party primaries and strategic ambiguity, in «Journal of Theoretical Politics»,
XVII, 2005, pp. 107-136.
Miller, P. M.; Jewell, M. E. and Sigelman, L., Divisive Primaries and Party Activists: Kentucky, 1979 and 1983,
«The Journal of Politics», L, 1988, pp. 459-70.
24
Morton, R. and Gerber, E. Primary Election Systems and Representation, «Journal of Law, Economics, and
Organization», XIV, pp. 304-324.
Neumann, S., Towards a comparative study of political parties, in Neumann S. (ed.), Modern political
parties, Chicago, Chicago University Press, 1956, pp. 395-421.
Norrander, B., Ideological representativeness of presidential primary voters, in «American Journal of
Political Science», XXXIII, 3, 1989, pp. 570-587.
Obler, J., Intraparty Democracy and the Selection of Parliamentary Candidates: The Belgian Case, «British
Journal of Political Science», IV, 1974, pp. 163-185.
Palmer, N., The New Hampshire Primary and the American electoral process, Praeger, Westport,
Connecticut, 1997.
Panebianco, A., Modelli di partito, Bologna, Il Mulino, 1982.
Pasquino, G. and Venturino, F. (eds.), Il Partito Democratico di Bersani. Persone, profili e prospettive,
Bologna, Bononia University Press, 2010.
Pasquino, G. and Venturino, F. (eds.), Le primarie comunali in Italia, Bologna, Il Mulino, 2009.
Pennings, P. and Hazan, R.Y., Democratizing candidate selection. Causes and consequences, in «Party
Politics», VII, 3, 2001, pp. 267-275.
Peterson, D. A. and Djupe, P. A., When Primary Campaigns Go Negative: The Determinants of Campaign
Negativity, «Political Research Quarterly», LVIII, 2005, pp. 45-54.
Piereson, J. E. and Smith, T. B., Primary Divisiveness and General Election Success: A Re-Examination, «The
Journal of Politics», XXXVII, 1975, pp. 555-62.
Pogunkte, T. and Webb, P. (eds.), The presidentialization of politics. A comparative study of modern
democracies, Oxford University Press, Oxford, 2007.
Poguntke, T. and Scarrow, S., The politics of anti-party sentiment: Introduction, «European Journal of
Political Research», XXIX, 1996, pp. 251-262.
Poguntke, T., Anti-party sentiment - Conceptual thoughts and empirical evidence: Explorations into a
minefield, in «European Journal of Political Research», XXIX, 1996, pp. 319-344.
Ranney A., Turnout and Representation in Presidential Primary Elections, «American Political Science
Review», LXVI, 1972, pp.21-37.
Raniolo F., Introduzione: partiti politici e cambiamento partitico in F. Raniolo (a cura di), Le trasformazioni
dei partiti politici, Messina, Rubettino 2004, pp. VII-XXXIV.
Raniolo F., Un'analisi organizzativa dei partiti politici, in L. Morlino e M. Tarchi (a cura di), Partiti e caso
italiano, Bologna, Il Mulino 2006, pp. 19-51.
Rapaport, R.B. and Abramowitz A.I., McGlennon J. (eds), The Life of the Parties. Activists in Presidential
Politics, Lexington, Kentucky University Press, 1986.
Rahat, G., Which Candidate Selection Method is the Most Democratic?, «Government and Opposition»,
XLIV, 2009, pp. 68-90.
Rahat, G., Hazan, R. and Katz, R., Democracy and Political Parties. On the Uneasy Relationships between
Participation, Competition and Representation, «Party Politics», XIV, 2008, pp. 63-683.
Rahat, G. and Hazan, R. , Participation in Party Primaries: Increase in Quantity, Decrease in Quality, in T.
Zittel and D. Fuchs (eds), Participatory Democracy and Political Participation. Can Participatory Engineering
Bring Citizens Back In?, Routledge, ECPR Studies in European Political Science Series, 2007, pp. 57-72.
Rahat, G. and Hazan, R.Y., Candidate selection methods. An analytical framework, in «Party Politics», VII, 3,
2001, pp. 297-322.
25
Rahat, G. and Sher-Hadar, N., The 1996 Party Primaries and Their Political Consequences, in Asher Arian and
Michal Shamir (eds), The Elections in Israel 1996, New York, State University of New York Press, 1999, pp.
241-268.
Scarrow, S., Politicians against parties: Anti-party arguments as weapons for change in Germany, in
«European Journal of Political Research», XXIX, 1996, pp. 291-311.
Scarrow, S., Parties and their members: Organizing for victory in Britain and Germany, Oxford, Oxford
University Press, 1996.
Scarrow, S., Parties and the expansion of direct democracy. Who benefits?, «Party Politics», V, 1999, pp.
341-362.
Scarrow, S., Parties without members? Party organization in a changing electoral environment, in R. Dalton
e M.P. Wattenberg (eds.), Parties without Partisans, Oxford, Oxford University Press, 2000, pp. 79–101.
Scarrow, S. and Kittilson, M., Political Parties and the Rhetoric and Realities of Democratization, in Bruce
Cain, Russell Dalton, Susan Scarrow (eds.), Democracy Transformed?, Oxford University Press, 2003, pp. 59-
80.
Scarrow S., Webb P. and Farrell D., From social integration to electoral contestation, in R. Dalton e M.P.
Wattenberg (eds.), Parties without Partisans, Oxford, Oxford University Press, 2000, pp.129-153.
Seddone, A.; Venturino, F. (2011). Choosing the Leader. The Italian Democratic Party at Polls, 2007 and
2009, Paper presented at the ECPR Joint Session of Workshops, St. Gallen, April 12-17, 2011
Seyd P., New parties. New politics? A case study of the British Labour Party, in «Party Politcs», V, 1999, pp.
383-405.
Seyd P. and Whiteley P., British Party Members: An Overview, «Party Politics», X, no 4, pp. 355-366.
Stone, Walter, Ron Rapoport and Monique Schneider (2004). Party Members in a Three-Party Election:
Major Party and Reform Activism in the 1996 American Presidential Election. Party Politics, vol 10, no. 4, pp.
445-469.
Valbruzzi, M., Primarie. Partecipazione e leadership, Bononia University Press, Bologna, 2005.
Ware, A., “Divisive” primaries. The important questions, in «British Journal of Political Science», IX, 3, 1979,
pp. 381-384.
Ware A., The American direct primary: party institutionalization and transformation in the North,
Cambridge University Press, 2002.
Wauters, B., Explaining participation in intra-party elections. Evidence from Belgian political parties, «Party
Politics», XVI, 2009, pp. 237-259.
Wichowsky, A. and Niebler, S. E., Narrow Victories and Hard Games: Revisiting the Primary Divisiveness
Hypothesis, «American Politics Research», XXXVIII, 2010, pp. 1052–1071.