The Study of Intraparty Frictions: Conceptual Reflections on Preference Heterogeneity, Disagreement, and Conflict

Abstract

Every complex organization is sometimes marked by preference heterogeneity, disagreement, and conflict. Within political parties, such frictions are traditionally viewed negatively, while recent research has started to perceive them more positively. How might such contradictory evaluations be explained? Through a three-step conceptual analysis we (1) identify two analytical perspectives on intraparty friction, one rooted in a primarily structural conception of parties, one in a primarily behavioral conception; and (2) specify a minimal definition of intraparty friction, which underpins a hierarchical concept structure to (3) suggest a way to resolve contradictions in the consequences attributed to intraparty frictions. Structuralist accounts often view frictions as negative due to a more demanding conceptual threshold, suggesting different types and levels of risk taking by conflict partners. Conversely, behavioralist perspectives see friction more often as beneficial because they focus on expressed disagreement without necessitating an organizational response. Our conceptual tools have important implications for research on membership organizations generally.

In research on political parties, the unitary actor assumption still features prominently. However, it is now widely recognized that political parties are complex organizations that need to cope with internal frictions (e.g., Bawn et al. 2012; Eldersveld 1964; Grossmann and Hopkins 2015; Ichino and Nathan 2013; Noel 2016; Scarrow, Webb, and Poguntke 2017). Virtually every party is at times marked by a pluralism of internal preferences, disagreement, and conflict (e.g., Rahim 2002), with fundamental consequences for democracy in terms of these organizations’ ability to operate as representatives, legislators, and governors. Traditionally, these phenomena are conceived of as negative (e.g., Blondel 1978; Jung and Somer-Topcu 2022; March and Simon 1958; Mikkelsen and Clegg 2018; Sartori 1976). However, more recent research has started to view internal frictions in a more positive light, highlighting parties’ enhanced mobilization and representational capacity alongside strategic advantages when governing in coalition (e.g., Boucek 2009; Campbell et al. 2019; DiSalvo 2012; Meyer 2012; Wagner, Vivyan, and Glinitzer 2020).

What might account for these diverging views on the consequences of intraparty friction, defined as (different forms of) organization-relevant preference heterogeneity between at least two internal party actors? We suggest that one central answer to these contradictions lies in the use of different conceptualizations of the phenomenon studied, different answers to the initial “what is” question (Mair 2008, 72). If equivalent phenomena are studied using different labels while different phenomena are studied under the same heading, it obscures the conditions under which internal frictions might have harmful—or, alternatively, beneficial—effects.

With the expansion of new data sources and methods in recent years, the study of intraparty friction, its sources, and its consequences has started to receive much scholarly attention from a diverse set of fields in political science, such as legislative studies (e.g., Haber 2015; Proksch and Slapin 2012; Sieberer 2006; Slapin et al. 2018; Spirling and Quinn 2010), electoral behavior (e.g., Greene and Haber 2015; Klingelhöfer and Müller 2023; Lehrer, Stöckle, and Juhl 2024; Plescia, Kritzinger, and Eberl 2021; Pyeatt 2015; Wagner, Vivyan, and Glinitzer 2020), coalition and executive governance (e.g., Bäck, Debus, and Müller 2016; Giannetti and Benoit 2009; Laver and Schofield 1990; Vercesi 2016), and party research (e.g., Ceron 2012; Cross and Katz 2013; Grossmann and Hopkins 2015; Haughton and Deegan-Krause 2020; Ibenskas and Sikk 2017; Noel 2016). Consequently, we find an extensive scholarly treatment of intraparty friction or closely related phenomena such as “preference heterogeneity,” “disagreement,” “conflict,” “dissent,” “incongruence,” and so on. These notions are partially overlapping but also show important differences—for example, differences related to a friction’s severity and the nature and level of risk taking it entails for conflict partners. Both the similarities and the differences between these often-used terms in the literature have been largely overlooked so far, which is one central reason why “[t]he various ways in which intra-party dissent manifests itself empirically are neither properly theorized nor well understood” (Lehrer, Stöckle, and Juhl 2024, 220).

Addressing these issues, we engage in this paper in a three-step conceptual analysis. We first identify two different analytical perspectives on intraparty friction in existing research, one rooted in a primarily structural conception of political parties, one in a primarily behavioral conception. Structuralist accounts are predominantly “inward oriented,” emphasizing the party as a social system in which frictions typically occur because intraparty actors, given their respective roles within the organization, hold different political goals (e.g., Michels 1915; Pedersen 2010a; Schumacher, de Vries, and Vis 2013). In contrast, behavioral accounts primarily conceive of the party as a vehicle for external goal attainment in which frictions are largely the result of actors’ differences in policy, ideology, or issue orientations (e.g., Gherghina, Close, and Kopecký 2019; Polk and Kölln 2018; Strøm 1990; Willumsen and Öhberg 2017).

Based on the two perspectives’ commonalities, we then propose a minimal definition of intraparty friction—the presence of organization-relevant preference heterogeneity between at least two internal party actors—which underpins a hierarchical concept structure. This concept structure has decreasing empirical scope, as subconcepts are more narrowly defined—that is, they require more defining features to be present than required by our baseline concept (Sartori 1970). These subconcepts of intraparty disagreement and intraparty conflict emerge directly from the literature: the former is more strongly connected to behavioralist accounts, the latter to structuralist accounts distinguished in the first step. Intraparty disagreement requires not just the presence but also the verbal or behavioral expression of preference heterogeneity (translating mere friction into open disagreement), while intraorganizational conflict additionally requires the party’s official response and thus a formal acknowledgement of disagreement. It is thus intraparty conflict’s reciprocal manifestation of internal differences through organizational roles and channels that marks the last and most severe stage of frictions in our hierarchy of concepts.

In a final step, we argue that our hierarchy of concepts helps to resolve existing contradictions in how research views the consequences of intraparty frictions, particularly crucial when trying to explain similar empirical phenomena. Structuralist accounts, which tend toward our notion of “conflict,” usually see friction as negative because the conceptual threshold to consider phenomena as “conflict” is more demanding to start with; it involves a higher level of risk taking by the internal actors involved. Behavioralist accounts, on the other hand, have started to consider the benefits of internal friction because they tend toward a concept of friction that merely requires expressed “disagreement” but no official organizational response. Indeed, studies of voter responses to intraparty friction that (implicitly or explicitly) focus on the consequences of conflict overall detect negative voter responses, while those studying the effects of disagreement find positive ones.

Our findings have important implications for the literatures on political parties—as societal actors and as governors—and their adjacent fields, such as legislative studies, coalition politics, and electoral behavior. The shared conceptual foundation of structural and behavioral accounts of intraparty friction in the form of a baseline concept clarifies the common ground of existing studies. This helps to overcome divides across subfields and allows for theories from either account to be integrated into one overarching framework, which to date is virtually nonexistent. The identified subconcepts, in turn, are specific enough to establish which findings on phenomena labeled as conflict, friction, division, disagreement, incongruence, disunity, or lack of cohesion speak to each other and which do not, as well as making clear which caveats the field has produced and why. They also allow researchers of intraparty friction to clearly distinguish related yet distinct empirical phenomena that might affect the same outcome variables differently. Simultaneously, the subconcepts are sufficiently generic to be useful for studying membership organizations such as interest and service-orientated groups, which, like parties, need to continuously reconcile tensions between internal actors such as leaders, members, or paid managers (Bolleyer 2024). In the conclusion we specify how our conceptual tools can advance research on not only intraorganizational but also intrainstitutional dissent.

Taking Stock: Structural and Behavioral Perspectives on Intraparty Friction

Reviewing the political science literature, we can broadly distinguish between a primarily structural and a primarily behavioral conception of political parties that shape how intraparty friction (and related phenomena) are understood in different subfields. Their main differences are systematized in table 1.

Table 1 The Structural and Behavioral Account of Intraparty Friction

Starting with the structural account of intraparty friction, respective studies are primarily “inward orientated.” The bulk of research that uses the term “conflict” is located here and goes back to classical work in organizational sociology and public administration. Following Smith (1966, 505–6), “[I]ntra-organizational conflict has its source in the nature of the organization as a social system, in the way it is structured and in the manner in which component subsystems are interrelated.” Preference divergences underpinning conflict—rather than being nourished by ideological or policy disagreement—stem from “basic differences of interests between participants occupying different positions in the organizational hierarchy” (505–6). This view directly aligns with Michels’s (1915) famous “iron law of oligarchy,” predicting, for example, divisions between leaders and members, a perspective that has fundamentally shaped the literature on party organization and intraparty dynamics ever since.

This perspective on intraparty friction is prominent, for instance, in research on intraparty reform and leadership selection (Gauja 2017; Wolkenstein 2020). Friction that can result from the increasing inclusiveness of intraparty decision making tends to be approached from the perspective of diverging interests attributed to distinct intraparty strata (e.g., elites versus activists) rather than differences in decision makers’ or leadership contenders’ substantive or ideological preferences. Following the same rationale, intraparty conflict has been analyzed as rooted in the constraints and pressures generated by the different arenas in which parties simultaneously operate, which as a consequence generate conflicts over how to set priorities (Katz and Mair 1995; Pedersen 2010b). This is, among other strands, evident in recent research on the relations between politicians, staffers, and paid consultants (parties’ paid personnel), on the one hand, and (unpaid) volunteer leaders and activists, on the other (Bolleyer 2024; Karlsen and Saglie 2017; Moens 2024).

Related to the question of which actors feature in intraparty frictions is the issue of the substance of friction (see table 1). As already alluded to in the above discussion, friction in the structure-based account is predominantly conceptualized as rooted in actors’ positions and roles within and their relationship to their organization, rather than their specific ideological, policy, or issue preferences. Consequently, the substance of conflicts between actors holding different intraparty positions or roles relates to goals, priorities, tactics, or strategies, more so than the content of particular issues or ideologies.

Subsequent work employing structure-based perspectives has demonstrated that ideological or organizational preferences vary indeed by intraorganizational functions or roles and that their existence and interaction also affect a party’s behavior (Schumacher, de Vries, and Vis 2013). Likewise, affiliated organizations such as youth or women’s organizations can at times find themselves in conflict with the rest of the party qua their intraorganizational role, and thus can be viewed as institutionalized channels for tolerated disagreement (Poguntke 2006). Accordingly, the currency of friction—what conflict partners predominantly aspire to when willingly risking a clash with an actor or unit belonging to their own organization—is about intraparty organizational control. The main drivers of intraparty friction in structural accounts are situated within the organization and its structures. By implication, different intraparty structures or their reform could either reduce or enhance such frictions (Bolleyer 2013; Gauja 2017; Wolkenstein 2020).

The second perspective, a behavioral account of intraparty friction, is more concerned with the microfoundation of parties than their structural underpinning (see table 1). Prominent terms describing friction in this strand of the literature are disagreement, incongruence, lack of cohesion, disunity, or dissent—in short, terms that all stress intraparty actors’ individual and diverging attitudes and behaviors. This is prominently illustrated by a recent special issue in Party Politics dedicated to “intraparty conflict.” While this issue shows that the term “conflict” also appears outside structure-based accounts (Gherghina, Close, and Kopecký 2019, 650), it also displays a plurality of terms used to study the phenomena of interest, including ideological incongruence (Kukec 2019), disloyalty (de Vet, Poletti, and Wauters 2019), or differing intraparty viewpoints (Ceron and Greene 2019).

Overall, the behavioral perspective is particularly common in legislative studies and electoral behavior research and typically does not view the party organization as a social system but rather as a vehicle for goal attainment (Strøm 1990). Therefore, it views the drivers of intraparty friction as predominantly rooted in concerns and pressures related to external goal attainment, such as winning elections, entering government, or implementing a particular legislative or governmental agenda. The substance of intraparty friction is thus defined by differences in policy, ideology, or issue orientations, meaning intraparty actors’ disagreements are ultimately about gaining or maintaining policy control over the party (i.e., currency of conflict). This can indeed be seen in numerous places in the literature. Recent work in legislative studies, for example, draws attention to “ideological differences” (Willumsen and Öhberg 2017) but also to individual policies (Bhattacharya 2023; Willumsen 2023). Similarly, disagreement within the party organization outside parliament or government is also often seen as rooted in divisions over specific issues, such as EU integration, economy, culture, or immigration (Jolly et al. 2022), or as incongruence of ideology (Budge et al. 2012). Other individual-level work adopting a behavioral perspective shows that party candidates or members regularly hold different ideological or issue-position attitudes from the rest of the party (Carey and Shugart 1995; Kölln and Polk 2017; Polk and Kölln 2018; Van Haute and Carty 2012). For instance, candidates or legislators sometimes strategically use their ideological or issue-based disagreement with the party to boost their electoral performance (André and Depauw 2013; Pedersen and Rahat 2021).

While much of the literature on intraparty friction tends toward one of these two perspectives, the literature on factions—party subgroups that “engage in collective action to achieve their members’ particular objectives” (Boucek 2012, 37)—endorses both, as echoed by the distinction between policy-based and spoils-based factionalism (Boucek 2009). Factions can represent internal policy differences, with control over the policy positions adopted by the organization being the main currency (e.g., Ceron 2019; Webb and Bale 2021). Alternatively, factions can also be “vehicles in a struggle over the allocation of intra-organizational spoils such as financial resource, career advancement or leadership positions” (Ichino and Nathan 2013; Kölln and Polk 2023, 1553), with intraorganizational control being the main currency. These two perspectives are not always easy to separate, as the fight over the party’s policy position might be motivated by considerations of vote maximization and, relatedly, office seeking, which if successful (and the party enters government) multiplies the material spoils to be allocated to the dominant faction as compared to opposition status. In other words, factions might adopt policies driven by the desire to implement the latter, or based on strategy and the desire to maximize the party’s appeal to advance their careers—policy being an end in the former scenario and a means in the latter. Still, both align with behavioral accounts of intraparty conflict and contrast with factionalism portrayed as an expression of internal struggles over organizational positions, whereby the conflict partners involved try to take control of the party organization in terms of the spoils it directly provides or the spoils it can provide access to.

Intraparty Friction, Disagreement, and Conflict: A Hierarchy of Concepts

The above systematization of research on intraorganizational frictions broadly defined allows us to specify a hierarchy of concepts capturing interconnected notions of friction with decreasing empirical reach as subconcepts become more narrowly defined (cf. Sartori 1970). In the following subsection, we identify three criteria that constitute a shared conception of intraparty friction across the accounts. This shared conception describes a notion of friction underpinning a strand of research in its own right, while helping to integrate existing research. Next, we identify two more demanding concepts that are both embraced by our “umbrella concept” of friction, one systematically aligning with the structural perspective on parties, the other with the behavioral perspective. To develop our distinctions, we reach beyond politics research to draw on the literature on conflict in organizational sociology and management studies. Integrating these literatures helps us to detail hierarchically ordered criteria that underpin our central varieties of intraparty friction.

Toward a Baseline Definition of Intraparty Friction

Based on the above synthesis, we define intraparty friction as the presence of organization-relevant preference heterogeneity between at least two internal actors, a phenomenon that existing literatures have studied under different labels such as conflict, dissent, disagreement, disunity, preference incongruence, or heterogeneity.

Specifying the three constitutive criteria of our baseline definition one by one, both the structural and the behavioral perspectives require the involvement of at least two intraparty actors for frictions to emerge. These actors can be two organizational units/organs, two individual actors with a formal association to the organization, or a mix of the two. Organizational units or organs encompass collective or corporate actors such as the party executive, local or regional branches, or functional units such as youth wings (e.g., Bolleyer 2012; Poguntke 2006). Individual intraparty actors are organizational members with a formal affiliation to the party that finds expression in an exchange of privileges and duties inviting loyalty (Bolleyer 2009, 563–64; Polk and Kölln 2018; Scarrow 2014, 30–31), while constituting an interdependence between potential conflict partners who benefit from the continuation of the relationship (Kochan, Huber, and Cummings 1975; March and Simon 1958). This is not the case in looser forms of affiliation because they are characterized predominantly by a one-way relationship in which affiliates gain benefits but have no duties toward the party (Scarrow 2014; Gauja 2015).

Furthermore, friction in either account necessitates preference heterogeneity between these intraparty actors. ¹ Some work on organizational conflict more generally is very specific and requires incompatible preferences between intraorganizational actors (see, e.g., Coleman, Deutsch, and Marcus 2014; Fisher 2014; March and Simon 1958; Smith 1966). Though incompatibility might make conflicts particularly intense (e.g., Rahim 2002), we consider diverging preferences (“mere” heterogeneity) as enough because they can trigger a degree of intraparty friction that is of substantive importance, as some party politics literature suggests (e.g., Klingelhöfer and Müller 2023; Kölln and Polk 2017; May 1973; Polk and Kölln 2018; Van Haute and Carty 2012). In these instances, compromise is in principle still an option because preferences are not incompatible per se, but the willingness of those involved to reconcile and deviate from their position can still be limited. Accordingly, severe conflict might manifest itself nonetheless.

As the third element of our baseline definition, the preference heterogeneity concerned needs to be “organization-relevant.” This means it either transcends purely personal differences between the internal actors involved, or if the source of frictions is purely personal, it risks having organizational repercussions because of actors’ prominent roles in the party. This rests on the prominent distinction between affective or relationship conflict and substantive conflict in organization and management research (Jehn 1997; Rahim 2002). The former relates to emotional or interpersonal issues, the latter to goals, contents, issues, tasks, or roles relevant to the operation and activities of an organization. Both accounts of political parties are thus concerned with substantive conflict, acknowledging the same minimum requirement because frictions have to have—at least potentially—implications for the organization, either in terms of self-maintenance or goal attainment (e.g., Michels 1915; Panebianco 1988).

Taking the three criteria together, works that share this notion of friction stress the importance of theorizing and studying preference heterogeneity within parties, whether expressed or not, in its own right (e.g., Krehbiel 2000; Willumsen 2013). They are concerned with attitudinal diversity within the party organization or party units (e.g., parliamentary group) as such. This strand studies heterogeneity that remains unexpressed, directly aligning with our baseline definition of intraparty friction. Prominent examples are studies of parliamentary cohesion through member of parliament (MP) surveys (e.g., Freire et al. 2021; Lisi and Serra-Silva 2021) or research on “ideological misfits” in party membership studies—party members who express their ideological disagreement with the party but nevertheless remain loyal (e.g., Cross and Young 2002; Kölln and Polk 2017; Van Haute and Carty 2012). In contrast, most work on parties’ ideological blurring, ambiguity, or broad-appeals strategies does not fulfill our baseline criteria for friction because it typically assumes that “party leaders” or “officials” unilaterally adopt such strategies (Koedam 2021, 663; Rovny 2012, 271). Only some of these studies explicitly consider such strategies to be underpinned by several internal actors’ diverging policy preferences, scenarios therefore qualifying as instances of friction (Somer-Topcu 2015, 843).

However, most studies to date only start caring about intraparty frictions once differences are expressed. For them, our baseline definition constitutes a building block to which definitional attributes are added (e.g., Klingelhöfer and Müller 2023), an issue we turn to now.

Distinguishing Varieties of Intraparty Friction: Disagreement and Conflict

Having pinned down our baseline definition as a “conceptual umbrella,” we can now detail five hierarchically organized attributes that distinguish three varieties of intraparty friction, visualized with an example in figure 1. Following a hierarchical logic, these five criteria shed light on two central conceptual boundaries relevant to the study of phenomena associated with what we refer to as intraparty friction.

Figure 1 Conceptual Hierarchy of Intraparty Friction, Disagreement, and Conflict

One of these boundaries relates to the demarcation between our baseline concept and its more narrowly defined variants. The first three attributes constitute our baseline definition itself. Another two allow us to distinguish two more specific subconcepts thereof: disagreement and conflict. To transcend our baseline definition, concerned with organization-relevant attitudinal diversity as such, internal differences must be articulated by at least some of the actors and not be withheld—that is, preference divergences need to become verbally or behaviorally manifest. When “friction” becomes visible to others beyond those immediately involved (e.g., in party congresses, the media, or through dissenting votes; Klingelhöfer and Müller 2023, 2; Lehrer, Stöckle, and Juhl 2024, 221–22; Polk and Kölln 2018), it translates into “disagreement.” While preference divergence is considered an early and thus necessary “stage of a conflict episode” (Pondy 1967, 300), this is distinct from expressed disagreement, a stage in which friction becomes manifest and thus consequential (Kochan, Huber, and Cummings 1975, 10; Mikkelsen and Clegg 2018). To illustrate this important distinction (sometimes referred to as latent versus manifest conflict [Pondy 1967]), party members may have diverging preferences about their party’s official election manifesto but stay silent for loyalty’s sake or for strategic reasons. This qualifies as intraparty friction but as neither disagreement nor conflict.

Our first conceptual boundary is essential for theorizing both the drivers and the consequences of (different types of) friction, as both disagreement and conflict involve not only risk taking by intraparty actors who express their “disagreement” to start with, but also those actors who decide to either tolerate such disagreement or officially respond to it (Klingelhöfer and Müller 2023; Wagner, Vivyan, and Glinitzer 2020). As alluded to earlier, the political science literature starts getting interested in preference heterogeneity between intraparty actors if it can have repercussions for the party, which is why the first three criteria constitute the minimum or baseline definition (see table 2). Consequently, a fight between two people who dislike each other personally and happen to be members of the same local party branch would qualify as a manifestation of an interpersonal conflict but not of an intraparty conflict, because their disagreement does not relate to their party membership. That said, it is conceivable that such articulation of private issues or personal dislike can generate substantive intraparty friction. For example, it can have direct repercussions for the party organization given individuals’ prominent positions within the organization. Imagine a leading party figure airing personal matters concerning the party leader in the media that—though private in nature—negatively impact on the party leader’s public standing and cast doubts about their suitability for such a prominent role in the party. Such “spillover” from interpersonal to substantive disagreement occurs when private tensions between individuals impact on the goal attainment of the (party) organization as a whole (positively or negatively)—for example, by discrediting central party figures and so weakening the party’s electoral prospects. Indeed, classics in organizational sociology consider interference with organizational goal attainment to be a central cause of conflict behavior between interdependent intraorganizational actors with diverging preferences (Kochan, Huber, and Cummings 1975; March and Simon 1958). Table 2 summarizes our hierarchical concept structure along several central dimensions.

Table 2 Summary of a Hierarchical Conceptualization of Intraparty Friction

Note:^† Here participants might express deviant opinions in anonymous candidate surveys. But as long they remain silent in party and public fora, repercussions of such friction differ from those of expressed disagreement.

A second conceptual boundary lies between the subordinate notions of friction: disagreement and conflict (see table 2), which are anchored in the differences between the two accounts of political parties identified earlier. Behavioral accounts focus on the microfoundation of social behavior rather than structural aspects. Accordingly, much work in this area concerns frictions that are articulated by (at least) one conflict partner, which, however, remain unilateral. This subtype of friction we label “disagreement” because one actor has acknowledged and expressed the preference heterogeneity but has done so without triggering an organizational response, which would be required for “conflict.” To illustrate, MPs or electoral candidates might publicly take positions on critical issues different from the official party line, signaling low attitudinal cohesion to outside audiences, which might impose reputational damage (Sieberer 2006; Tromborg 2021). The party might decide to tolerate this, preventing “disagreement” from translating into “conflict.” The bulk of the current literature taking a behavioralist perspective deals with either friction or disagreement, which is why this line of research is also particularly good at explaining when and why individuals choose to voice disagreement (Bøggild et al. 2021; Dingler and Ramstetter 2023; Haber 2015; Proksch and Slapin 2012; Slapin et al. 2018). As indicated earlier, although in this scenario the party does not respond officially, it still involves risk taking for both conflict partners: the actor who expresses the disagreement risks being sanctioned by the party, but by remaining silent, the party risks inviting others to express their disagreement more openly in the future. Such individuals might hope to score points with some parts of the electorate (e.g., Campbell et al. 2019; Wagner, Vivyan, and Glinitzer 2020), without suffering from any organizational sanctions.

If, however, the party does respond, mere “disagreement” becomes “conflict” (see table 2). Consequently, our most demanding, or exclusive—to use Sartori’s (1970) terminology—notion of intraparty friction, “conflict,” requires five individually necessary and jointly sufficient conditions to be met: (1) the interaction of at least two intraparty actors, (2) preference heterogeneity between those actors, which (3) directly relates to the party as an organization or to the relevant actors who belong to the party, and which is (4) articulated and (5) officially responded to by the party or opposing intraparty actors. ² Structural perspectives tend toward using this notion and provide insights into when—and if so, how—a party officially responds to expressed disagreement, which can materialize in open discreditation in the media; in the imposition of sanctions related to the allocation of speaking time in parliament or of seats in committees and party offices; or, at worst, in expulsion from the organization (Andeweg and Thomassen 2011; Bolleyer, von Nostitz, and Smirnova 2017; Rossner 2014). In those instances, conflict manifests itself through several intraparty actors jointly acting upon their differing preferences (either verbally or behaviorally), including a party unit (e.g., the parliamentary group, the executive, a party tribunal) issuing an official response. Consequently, in the case of conflict a division within the organization occurs that openly affects or risks affecting organizational performance and goal attainment.

We label this most demanding form of friction “conflict” because as compared to “disagreement,” it presupposes a higher level of escalation, signaling to outside audiences the inability of the party to address differences informally. An official party response signals that a divide is more severe than a unilateral expression of disagreement that the party officially decides it can tolerate instead. Easily portrayed as a “fight” and thus emotionalized by the media (Strömbäck 2008, 233–34), conflict is likely to attract more and a different type of media attention and be amplified as a consequence (Kölln and Polk 2023, 1563), generating an additional layer of risk for all the actors involved. To be clear, a party member or voter who observes an MP or candidate expressing merely disagreement, as we define it, is mostly asked to form an opinion about the content of the disagreement. In contrast, when the party counteracts and turns the disagreement into conflict, the party member or voter is also tasked to evaluate that reaction and its proportionality, and is thereby pushed to side with one conflict partner over the other. It is this dynamic, created through intraparty conflict as we define it, that risks creating deeper and long-lasting divisions within a party and producing negative electoral consequences.

Conflict versus Disagreement: Resolving Diverging Evaluations of Intraparty Frictions

The qualitative differences ascribed to disagreement and conflict as distinct phenomena are useful to explore contradictions within existing research that is interested in the same outcome variable. This can be illustrated by briefly assessing the growing literature on voter responses to different types of intraparty friction. Research on voter responses to MPs’ rebellious behavior—that is, disagreement—shows that voters generally react positively and reward such behavior (e.g., Besch and López-Ortega 2023; Bøggild and Pedersen 2020; Campbell et al. 2019; Vivyan and Wagner 2012). A similar finding emerges from the literature on intraparty competition at primary elections in the US and Europe. This research also shows that more ideologically extreme candidates tend to be electorally rewarded for displaying disagreement with their more moderate party leaderships (e.g., Broockman et al. 2021; Isotalo, Mattila, and von Schoultz 2020; Leimgruber, Hangartner, and Leemann 2010; Nielson and Visalvanich 2017; Rehmert 2022; Stone and Simas 2010). In contrast, Duell and colleagues’ (2023, 87) recent study on MPs’ rebellious behavior explicitly incorporates the leadership’s reaction, which “upgrades” friction from disagreement to conflict according to our conceptualization and finds more negative voter reactions. Similarly, work studying the electoral consequences of a party being disunited or incohesive—for example, if a party is described as such in experiments (e.g., Johns and Kölln 2020; Lehrer, Stöckle, and Juhl 2024) or is perceived as more or less cohesive by voters (e.g., Greene and Haber 2015)—shows broadly negative effects. The latter studies stay silent on how a party’s disunited or incohesive state presents itself to voters (e.g., which party actors—MPs, members, and/or leaders—express their diverging preferences and thereby create the friction), which would allow us to clearly distinguish disagreement from conflict. However, they clearly do assume that a party’s divided state is apparent to voters, which is most likely if conflict manifests itself between various party actors.

Essentially, by applying our conceptual distinctions to these two sets of studies—one broadly aligning with our notion of disagreement, the other with conflict—we imply that voters consider manifestations of party-level conflict to be more problematic than the ostentatious disagreement of individual party actors, echoing the substantive differences attributed to our subconcepts and rationalizing the contrasting findings each set has tended to generate.

Returning to the longer-term evolution of social science research on intraorganizational differences we started out from, the conceptual insights presented help to systematize the roots of different evaluations of intraparty friction and recent shifts therein. Traditionally, friction within the study of organizational settings has been approached as something dysfunctional that needs to be controlled, avoided, and eliminated (March and Simon 1958; Mikkelsen and Clegg 2018). Nowadays, organization and management research is more nuanced, associating distinct types of frictions with different (positive or negative) effects on intraorganizational relations, functioning, and performance (Ben-Hador 2017; Mikkelsen and Clegg 2018, 190). Similarly, political science research has been focused on the adverse effects of division and disunity, strongly influenced by The Federalist Papers (1787–88) and Michels’s (1915) work. Indeed, both structuralists and behavioralists have long shared a negative perspective (e.g., Andeweg and Thomassen 2011; Bolleyer, von Nostitz, and Smirnova 2017; Greene and Haber 2015).

Recently, however, some behavioralists in particular have started to move in a different direction, attempting to theorize and examine the potential benefits of internal frictions, especially in arenas visible to external audiences (e.g., Campbell et al. 2019; Klingelhöfer and Müller 2023; Kölln and Polk 2023; Wagner, Vivyan, and Glinitzer 2020). To be fair, we also find arguments considering the benefits of friction in the structuralist camp: think of research on parties’ youth or women’s wings (Poguntke 2006) effectively constituting institutionalized venues for tolerated disagreement. Other examples are research on intraparty democracy and its implications for member commitment, or on party families, such as the Greens, endorsing intraparty pluralism ideologically, indicating that internal friction is “not always a bad thing” (Bolleyer 2013; Cross and Katz 2013; Kitschelt 1994, 212). Nonetheless, the costs of intraorganizational diversity—especially when spilling over into the public domain—have remained the prevailing concern, particularly in this camp.

Our conceptual differentiations reveal that one reason why we see a turn toward the benefits of friction predominantly in recent behavioralist work is that behavioralists more frequently study phenomena that align with our notion of “disagreement” and which conceptually sit “in between” (more demanding) conflict and (mere) friction. Unlike friction, disagreement involves the expression of organization-relevant preference heterogeneity, which tends to be visible to outside audiences. It does not require an organizational response—which can range from articulating a counterposition to imposing sanctions on those with deviant positions. This is different from the notion of “conflict” more prominent in the structuralist camp. The presence of an organizational response—a reaction of a counterpart—that our notion of conflict presupposes is critical. Such a response is likely to be triggered by an articulation of attitudinal difference of a nature or with an intensity that the organization finds difficult to ignore or tolerate. This might be the case because such a display impacts too much on the credibility or authority of the party for it to be “excused” or, indeed, “sold” as a display of internal pluralism, tolerance, and diversity.

This suggests that structuralist accounts tend to see friction as more negative because the analytical threshold to consider something as “conflict” is more demanding to start with (as reflected in a larger number of criteria—see table 2). It presupposes not just the display of preference heterogeneity but also a counterresponse likely to invite an explicit divide in the organization. Implicit conceptual underpinnings thus fundamentally impact how we study phenomena, without this necessarily doing justice to the phenomenon studied—as the earlier turn toward a more nuanced perspective on the costs and benefits of friction in organization and management research tells us (e.g., Ben-Hador 2017; Mikkelsen and Clegg 2017; 2018).

Conclusion

To date, political science research has remained divided over how to evaluate the implications of intraparty frictions for a variety of phenomena, such as organizational stability and commitment, intraparty democracy, and electoral performance. One reason lies in conceptual differences resulting from parallel but overall disconnected developments in subfields—such as electoral and legislative studies, coalition and executive governance, and party research—that deal with “party actors” broadly defined, but nevertheless speak too little to each other. Following Lehrer, Stöckle, and Juhl’s (2024, 220) recent call to treat intraparty friction as a multidimensional phenomenon, we have proposed a hierarchy of concepts able to capture different variants of intraparty friction—that is, different manifestations of organization-relevant preference heterogeneity between at least two internal party actors, which can help to bring together scholars working in different traditions and in different subfields (see table 2). Distinguishing an encompassing minimum definition as a baseline concept from two (more specific) subconcepts rooted in different theoretical accounts on political parties leads not only to higher conceptual clarity but also to a clear anchoring in the literature, which are both critical to assure an accumulation of knowledge on the broader phenomenon of interest.

Applying our hierarchy of concepts (figure 1) can help to separate out and overcome at least two barriers to such accumulation. One emerges when scholars explain the same phenomenon with different notions of friction and accordingly generate conflicting results, without conceptual divergences becoming explicit. As illustrated by research on the electoral consequences of (essentially different types of) friction, separating studies on the consequences of disagreement from those on conflict helps to overcome such contradictions.

The other barrier to such accumulation, amplified by rather than rooted in conceptual discrepancies, refers to distinct evaluations of intraparty friction due to a diversity of different outcome variables studied. The alternative conceptions of parties that underpin structuralist and behavioralist accounts, which respectively understand parties as social systems or as vehicles for goal attainment, not only influence how each camp theorizes friction (see table 1) but also influence which of the consequences of friction each camp is particularly interested in. This, in turn, rationalizes the types of “blind spots” we find in current research, caveats that future research ought to tackle. For instance, there is relatively little work on intraorganization conflict—as we define it—in the behavioralist literature. Questions such as when and how parties officially respond to MPs who express divergent views publicly or how voters in turn react have so far received too little attention beyond rare cases of outright MP expulsion (Duell et al. 2023; Klingelhöfer and Müller 2023; Lehrer, Stöckle, and Juhl 2024). Likewise, as behavioralists tend to focus on individuals rather than organizational units, some questions—such as “when does a party’s youth or women’s wing express disagreement and under what conditions does a party respond to or tolerate this disagreement?”—are less likely to be asked.

Linking structuralist and behavioralist scholarship on friction will not only avoid such caveats and help to ameliorate them. By using our baseline concept as the foundation, it also invites more encompassing perspectives on friction as a phenomenon. Applying our concepts dynamically, for instance, allows us to analyze the conditions under which different party actors and units transcend two critical boundaries demarcating different stages of how frictions manifest themselves: the expression of deviant views as opposed to remaining silent (thereby translating friction into disagreement); and issuing an organizational response to such expression as opposed to tolerating it (thereby translating disagreement into conflict). Each potential outcome at these two critical stages suggests different levels of escalation linked to different types of risk taking by those actors who hold diverging views, which future research should not only theorize but examine in conjunction.

Finally, our conceptual tools are directly applicable to other fields. This concerns not only those looking at the nexus between intraparty and interparty frictions but also those concerned with intrainstitutional dissent. In terms of the former, research on government coalitions and cabinet governance increasingly explores the intersection between intraparty conflict and coalition formation, governance, and termination (e.g., Bäck, Debus, and Müller 2016; Bergman, Ilonszki, and Hellström 2023; Laver and Shepsle 1999; Meyer 2012; Vercesi 2016). Strikingly, research on the actual working of the mechanisms of coalition governance has remained limited, although governments’ ability to contain internal conflict is no doubt critical for their stability as well as their policy-making productivity (Müller, Bäck, and Hellström 2024, 17–18, 20). The containment of frictions between parties collectively forming a government can be usefully approached in terms of the prevention of disagreement as well as of conflict. Mechanisms of coalition governance can be directed toward preventing coalition partners from publicly expressing diverging views, allowing for “nondecision” to prevail, while sparing them an official response, of which government termination would be the most drastic (Vercesi 2016, 188–92). Mirroring our earlier discussion on the distinct repercussions of disagreement and conflict, existing typologies of coalition conflict management mechanisms stress the critical difference between handling frictions within government and transferring them to external areas, such as party summits or parliamentary committees (Moury and Timmermans 2013, 119). Intense disagreement expressed by a partner that cannot be contained within intragovernmental fora is often moved into arenas involving outside actors, such as parliamentary MPs or party leaders (Vercesi 2016, 189–90). Once disagreement becomes more visible to members and voters as a consequence thereof, coalition partners are pressed to take a stand. Strategic posturing and signaling to outside audiences become more likely when disagreement within a government coalition turns into fully fledged intragovernmental conflict, making conflict resolution more difficult. Whether and when government parties might be able to strategically benefit from the display of disagreement as compared to conflict (as MPs might benefit from expressing deviant views, e.g., Slapin et al. 2018) is an open question that our conceptualization can help to theorize.

Concluding with research on intrainstitutional dissent, the basic logic of our hierarchy of concepts might be usefully adapted to the study of different types of friction within bodies that perform important functions in political systems, including courts. Courts often encompass members with heterogenous preferences but are pressed to adopt joint positions and signal unanimous agreement to maintain their legitimacy (Sunstein 2015). An individual judge may hold preferences at odds with the majority of the court but remain silent (friction). Alternatively, they may write a dissenting opinion and express these preferences (disagreement). Finally, they may in response be disciplined by other members of the court (e.g., by being publicly called out by other members of the court for their opposition to a particular judgment or by getting overlooked for promotions within the court, such as appointments as chief judge) (conflict). Whether disagreement leads to conflict is critical for whether judges are likely to express disagreement in the future, which is considered essential to not only sustain judicial independence but also to assure the evolution of the law (Vitale 2014, 88–89). The transition from friction to disagreement in the first place is critical for how outside audiences perceive court judgments and ultimately for the wider acceptance of these judgments, with ongoing research exploring when disagreement might help or hinder such acceptance (Bentsen 2019, 593–94; Salmone 2014). Though either transition has important repercussions, more research is needed not only on the conditions that make each transition more or less likely but on their interplay over time as well. The conceptual tools proposed here might help to build the foundation for this.