The nature of the crisis: Losing touch with society

One of the most well-established findings in the empirical literature on political parties is that citizens vote for them less, participate less in their activities, and care less about the differences between them. This trend is perhaps clearest in the United States. It is well known, for instance, that electoral turnout levels in the United States have been consistently declining over the past few decades, leading to studies such as The Disappearing American Voter (Teixeira Reference Teixeira1992) and Why Americans Still Don't Vote (Piven and Cloward Reference Piven and Cloward2000). This is evidenced both in time-series data and in intergenerational research based on the U.S. national election survey (Miller and Shanks Reference Miller and Shanks1996). Moreover, data relating to party identification, participation in partisan activity, and levels of interest and information concerning partisan activities also shows a consistent pattern of decline (Putnam Reference Putnam2000; Putnam and Goss Reference Putnam, Goss and Putnam2002). But the decline of partisan activity is very marked also in other advanced western democracies. In a book evocatively entitled Parties Without Partisans, published in 2002, Dalton and Wattenberg have generalized this into a broader “partisan de-alignment thesis” drawing on evidence from 18 large and continuously democratic OECD countries, including most of Western Europe, over the past 40 years (Dalton and Wattenberg Reference Dalton and Wattenberg2001; also see Bartolini and Mair Reference Bartolini, Mair, Diamond and Gunther2001; Delwit Reference Delwit and Van Haute2011; Scarrow and Gezgor Reference Scarrow and Gezgor2010; Van Biezen et al. Reference van Biezen, Mair and Poguntke2012).

The reason this evidence corroborates the general impression that political parties are suffering from a crisis is that, as parties lose their foothold in society, they appear less and less capable of exercising the function of mediation between society and the state. In a highly influential article on this topic, for instance, Richard Katz and Peter Mair have suggested that parties have responded to the progressive erosion of their social base by seeking to find their sustenance from the state (Katz and Mair Reference Katz and Mair1995; also see Katz and Mair Reference Katz and Mair2009). This has encouraged them to adopt collusive practices which increase their distance from society and therefore undermine their capacity to serve as “intermediary bodies” between society and the state (on this point, see also Detterbeck (Reference Detterbeck2005) and Panebianco (Reference Panebianco1988)).

While several aspects of the “cartel party hypothesis” have since been called into question (Dalton, Farrell, and MacAllister Reference Dalton, Farrell and McAllister2011; Kitschelt Reference Kitschelt2000; Krouwel Reference Krouwel2012; Scarrow Reference Scarrow2006), it seems hard to deny that the progressive erosion of the traditional social base of political parties has gone hand in hand with a transformation of their organizational structure from mass organizations to “electoral machines” (Crouch Reference Crouch2000). This has in turn reinforced the public perception that political parties are self-referential political enterprises, interested only in the pursuit of office, and without any real link to the broader interests of society. It is this development that forms the core of what we shall refer to as the “crisis of party democracy.”

The roots of the crisis: “Bowling alone” or cognitive mobilization?

What needs to be explained in order to understand the roots of the present crisis is why parties are losing their foothold in society. A large part of the existing literature on this topic has focused on what is sometimes referred to as the “supply side” of political participation; that is, the forms and venues for such participation that are on offer, and their relative degree of potential impact (Hay Reference Hay2007; Scarrow Reference Scarrow2015). In this regard, scholars have highlighted the importance of the emergence of new media as alternative vectors of political information and socialization (Graber Reference Graber2003; Kriesi et al. Reference Kriesi, Bochsler, Matthes, Lavenex, Bühlmann and Esser2013; Mazzoleni and Schulz Reference Mazzoleni and Schulz1999); the impact of globalization (and Europeanization), which is thought to challenge the capacity of national political parties to address the most important problems of present-day societies (Bickerton Reference Bickerton2012; Held Reference Held1995; Kriesi et al. Reference Kriesi, Bochsler, Matthes, Lavenex, Bühlmann and Esser2013; Papadopoulos Reference Papadopoulos2007); as well as the collapse of “actually existing socialism” and the consequent crisis of ideologies, which is taken to reinforce the perception that “there is no alternative” to the presently dominant form of neoliberal governance (Anderson Reference Anderson2005; Crouch Reference Crouch2000; Fukuyama Reference Fukuyama1992; Offe Reference Offe2013).

Without denying the importance of these studies, for the purposes of the present analysis, we intend to look more closely at the “demand side” of political participation, that is, at the factors that affect individual motivation to participate in party politics. Two broad schools of thought seem to exist in the literature on this issue. One rather widespread idea is that individuals are becoming less willing to participate in party politics because they have become more absorbed in “private” or “individualistic” pursuits, and are consequently less interested in politics as such. This thesis is for instance at the core of Robert Putnam's influential book Bowling Alone, which seeks to explain an observed decline in what he calls “social capital” (measured inter alia in terms of partisan participation) with reference to a series of factors that all point in the direction of a more individualistic and self-absorbed society: what he calls the “constraints of time and money,” the “decline of the traditional family structure,” and the diffusion of “television” as a mass medium of both information and entertainment (Putnam Reference Putnam2000). The same basic idea is also at the heart of Peter Mair's book Ruling the Void, where he writes that:

We shall refer to this idea as the depoliticization hypothesis, since the key claim is that declining levels of partisan participation can be explained in terms of a more general decrease in interest for political participation as such. In sociological terms, this may be interpreted as a version of a specific strand of modernization theory (familiar at least since Tocqueville) according to which rising levels of social complexity and differentiation go hand in hand with the emergence of a specific form of “individualism” that is characterized by a growing concern for private pursuits and material benefits, at the expense of more traditional political virtues (Tocqueville Reference Tocqueville1835 [2006]).

An alternative explanation for declining rates of participation in party politics ties it to a different version of modernization theory. In their book entitled Modernization, Cultural Shift and Democracy, for instance, Ronald Inglehart and Charles Welzel argue that socioeconomic modernization should be understood as a process of “cognitive mobilization,” which “reduces the external constraints on human choice by increasing people's material, cognitive, and social resources” (Inglehart and Welzel Reference Inglehart and Welzel2005, 2–3). For them, this goes hand in hand with a growing emphasis on “self-expression values” and “civil and political liberties” (ibid., 5). “Cultural emphasis,” they write, “shifts from collective discipline to individual liberty, from group conformity to human diversity, and from state authority to individual autonomy” (ibid., 7).

The reason why this alternative strand of modernization theory may provide an explanation for the observed decline in participation in party politics is that political parties have traditionally been highly bureaucratic and elite-driven organizations. This is something that is not always made explicit in the existing literature on party decline (which, as several commentators have noted, often displays an implicit tendency to posit the “mass party” model that prevailed in the early and mid-twentieth century as the normative ideal; see Van Biezen and Romée Piccio Reference van Biezen, Piccio, Cross and Katz2013), but it is nonetheless a central theme in the earlier literature on parties from Michels (Reference Michels1917) to Duverger (Reference Duverger1954). If we assume that modernization involves a process of social acculturation that translates into a greater demand for “individual autonomy” and “self-expression,” it seems plausible that at the most advanced stages of this process both the need for, and the willingness to, participate in political parties as they were traditionally organized would decline. This hypothesis was spelled out very clearly by Dalton and Wattenberg in their already-mentioned book Parties Without Partisans:

We shall refer to this alternative explanation for the decline in partisan participation as the cognitive mobilization hypothesis. This contrasts with the depoliticization hypothesis in that it explains the observed decline in participation in traditional party politics not in terms of an overall decrease in the demand for political participation as such, but in terms of a transformation in the nature of this demand.

To be sure, the cognitive mobilization and the depoliticization hypotheses are not necessarily inconsistent with one another, since a decrease in the overall demand for political participation can certainly coexist with a transformation in the nature of the same demand. However, these two hypotheses hold very different implications with respect to the prospects for addressing the underlying motivational causes of the present crisis of party democracy from a normative point of view. If the decline in partisan activity is due primarily to a decrease in the interest for political participation as such, the prospects for reviving political parties as participation- or membership-based organizations are bound to be very small. If, on the other hand, the current crisis of political parties is due more to a transformation in the nature of the demand for political participation, the scope for addressing this crisis by reforming the organizational structure of political parties may be larger.

Advancing a set of normative prescriptions for addressing the underlying motivational causes of the present crisis of party democracy may therefore require taking a stand on whether the depoliticization hypothesis or the cognitive mobilization hypothesis is ultimately more plausible. Although this is far from an easy task, there is at least one well-established empirical finding that points decisively in the direction of the latter: the fact that the observed decline in participation in traditional party politics over the past few decades has coincided with a proliferation of “non-conventional” forms of political participation and engagement, amounting to what is generally referred to as the “rise of civil society” (Cohen and Arato Reference Cohen and Arato1992). This is a theme that has been developed in great detail by Pippa Norris in her book Democratic Phoenix: Reinventing Political Activism, where she explicitly takes issue with Putnam's version of the depoliticization hypothesis, arguing that “whereas the conventional wisdom suggests that in the late twentieth-century many post-industrial societies experienced a tidal wave of citizen withdrawal from the traditional channels of political participation,” data from the same countries reveals that “multiple newer channels of civic engagement, mobilization and expression are rapidly emerging to supplement the traditional modes” (Norris Reference Norris2002, 3–4).Footnote ³

The reason why this insight provides reason to prefer the cognitive mobilization hypothesis over the depoliticization thesis is that if the recent decline in participation in traditional party politics were due to a generalized decline in the demand for political participation as such, it would seem hard to explain why over the same time period we have also witnessed a veritable explosion of new forms of unconventional political participation. The cognitive mobilization hypothesis, on the other hand, appears perfectly consistent and capable of explaining both results. We therefore maintain that the cognitive mobilization offers a more plausible explanation of the underlying motivational causes of the observed transformation in the nature of contemporary political participation compared to the depoliticization hypothesis.

The case for making parties more democratic and deliberative

On the basis of the analysis carried out in the previous section, we now move on to the normative task of developing a model of party organization that appears more suited for addressing the underlying motivational causes of the present crisis of party democracy than political parties as they exist today.

If, as we have suggested, cognitive mobilization offers a more convincing explanation of the underlying motivational causes of the present crisis of party democracy than depoliticization, reforms in the structure of intraparty organization emerge as a plausible target of intervention aiming to stem the present tide of membership decline. For if the main reason why individuals are increasingly unwilling to participate in party politics is that the channels of participation currently offered by contemporary parties are perceived as too bureaucratic and elite-dominated to meet the transformed nature of the demand for political participation, a change in the organizational structure of such parties may well provide renewed incentives for participation.

This is in fact something that has already been intuited by several existing parties and their observers. In a recent piece devoted to several parties’ organizational responses to the large-scale sociological process she calls “individualization” (which overlaps in several ways with what we have called “cognitive mobilization”), Anika Gauja has noted that “party rhetoric and public comments made by leading party figures stress that organizational change is necessary” (Gauja Reference Gauja2015, 95). “In this discourse”, she adds, “reform, renewal and modernization are entwined with a concern to ‘open up’ the policy process to greater individual involvement from within and outside the party” (ibid., 96). In a similar vein, Susan Scarrow records in her recent survey of the response strategies adopted by contemporary parties to stem the tide of declining membership that “many political parties have made important and highly visible changes in the opportunities they offer members to influence party affairs” (Scarrow Reference Scarrow2015, 178). “These efforts,” she notes, “have taken two main forms: giving interested members more opportunities to shape party policy priorities, and giving members the right to vote on important party decisions” (ibid., 179).

To be sure, several commentators have also noted that the success of these strategies has so far remained rather modest. For instance, William Cross and Richard Katz have recently described the observable effects on membership of moves towards more intraparty democracy as “disappointing” (Cross and Katz Reference Katz, Cross and Katz2013, 6). Our contention, however, is that this is because most actual attempts at incentivizing participation through enhancing intraparty democracy have remained tied to what it sometimes called an aggregative conception of democracy that is, a conception of democracy that sees the aggregation of individual preferences as the essence of democratic activity, and brackets out procedures of debate and preference formation (cf. Gutmann and Thompson Reference Gutmann and Thompson2004). This is reflected in the fact that such attempts have for the most part consisted in the extension of “voting rights” to the membership, through the adoption of procedures such as “primary elections” and “membership ballots” (Gauja Reference Gauja2015; Hazan and Rahat Reference Hazan and Rahat2010; Scarrow Reference Scarrow2015).

What we want to suggest instead—or rather in addition—is that to adequately address the underlying motivational causes of the ongoing decline in partisan participation, parties ought not merely adopt aggregative forms of intraparty democracy, but also become more internally deliberative. By this we mean that they need to institutionalize fora of discussion and debate amongst members as a key component of the internal decision-making process. In the remainder of this section, we first advance three arguments in support of this proposition and then sketch the shape a more deliberative model of intraparty democracy might take.Footnote ⁴

Deliberation as a participatory incentive. The first reason why more internally deliberative parties may be better able to stem the tide of declining membership than merely aggregative forms of intraparty democratization emerges if we dig deeper into the actual reasons why individuals decide to participate in party politics to begin with. This topic has already been the object of a large literature, which has for the most part adopted a rational choice approach, focusing on the individual incentives to join political parties and participate in their activities. A first generation of studies concentrated primarily on what Peter Clark and James Wilson (Reference Clark and Wilson1961) called “instrumental” or “purposive” incentives, referring to the personal gains one can expect from party membership in terms of career opportunities (Seyd and Whiteley Reference Seyd and Whiteley1992; Sundberg Reference Sundberg1986), political patronage (Hopkin and Paolucci Reference Hopkin and Paolucci2006; Müller Reference Müller, Pelinka and Plasser1989), and specific policy outcomes (Whitley, Seyd, and Billinghurst Reference Whiteley, Seyd and Billinghurst2006). When discounted for the probability that one's involvement could actually be decisive in obtaining the desired benefits, however, these studies encountered a problem already familiar from collective action theory more generally: that from the point of view of a purely instrumental cost-benefit analysis, it would seem to be almost always irrational for anyone to take part in any partisan activity at all (see Katz (Reference Katz, Cross and Katz2013, 51–2) and Olson (Reference Olson1967)).

A later body of studies has therefore been led to analyze individual decisions to participate in party politics in terms of a different set of factors, which are usually referred to as “expressive,” “process,” or “political” incentives (cf. van Haute and Gauja Reference van Haute and Gauja2015; Katz Reference Katz, Cross and Katz2013; Scarrow Reference Scarrow2015). Acknowledging that “within the strictly individualistic framework of the classical rational choice model, the question is . . . why anyone would join parties at all,” Katz, for example, has recently suggested that party membership should be understood as primarily “expressive, rather than instrumental.” “One joins a party,” he writes, “because one agrees with its general philosophy or specific policies and wants to demonstrate this agreement” (Katz Reference Katz, Cross and Katz2013, 52–3). Similarly, in their recent summary of the latest empirical findings on party membership motivation, Anika Gauja and Emilie van Haute note that “party members everywhere are generally motivated by political incentives and values, rather than by private benefits . . . Process incentives take second place [only] to ideology as a motivating factor” (van Haute and Gauja Reference van Haute and Gauja2015, 191).

This is an interesting result because it challenges an assumption that had been at the core of most previous studies of partisan mobilization: that the process of participating itself must necessarily be understood as a cost of political mobilization. If the opposite is rather the case—if, that is, the individuals that do join political parties and participate in their activities do not see this as a cost but rather as an incentive in itself—it follows that merely providing members with opportunities to vote once in a while may not be sufficient to attract many additional ones to join and participate. Instead, the fact that more deliberative parties would necessarily demand a higher degree of commitment might provide a more powerful reason for doing so.

This idea is consistent with, indeed follows logically from, an aspect of the cognitive mobilization hypothesis. If the demand for political participation is undergoing a transformation such that individuals now seek greater scope for “self-expression” and “autonomy” (cf. Inglehart and Welzel Reference Inglehart and Welzel2005, 2–3), it seems plausible to expect that more deliberative parties would prove more attractive for prospective members. For deliberation clearly provides a more meaningful form of self-expression compared with procedures in which preferences are merely aggregated. This is because it does not reduce the individual to a more or less anonymous vote but allows her to give and demand reasons before a decision is taken. This point is stressed by virtually all theories of deliberative democracy: deliberation shows more respect for people qua self-determining agents, precisely because it treats them as individuals capable of giving and demanding reasons for what they do (e.g., Cohen Reference Cohen, Hamlin and Pettit1989; Dryzek Reference Dryzek2000; Elster Reference Elster1998; Gutmann and Thompson Reference Gutmann and Thompson2004; Habermas Reference Habermas1996).

Deliberation as a way of shaping individual preferences. A second reason why a deliberative model of intraparty democracy is likely to prove more effective in addressing the underlying motivational causes of the present crisis than a merely aggregative one is that it does not need to rely on the assumption that individual political preferences are given prior to actual involvement in the political process. This presumption might have been appropriate at an earlier stage of societal development, during which individuals could identify rather unproblematically with specific social groups, such as the “working class” or the “religious” constituency (Katz and Mair Reference Katz and Mair1995; Kriesi et al. Reference Kriesi, Bochsler, Matthes, Lavenex, Bühlmann and Esser2013; Lipset and Rokkan Reference Lipset and Rokkan1967). Cognitive mobilization implies, however, that individuals increasingly have both the desire and the capabilities to develop their own preferences, reflexively abstracting from their sociological or material circumstances (cf. Beck and Gernsheim (Reference Beck and Gernsheim2002) and Giddens (Reference Giddens1991)).

If they want to cater to this new aspect of the demand for political participation, political parties cannot conceive of themselves merely as vehicles for the aggregation of preformed preferences, but must rather provide concrete opportunities for individuals to develop and refine their political views. This is exactly what intraparty deliberation can achieve: it provides a mechanism for individuals to jointly define and sharpen their political views through a process of reciprocal exchange with each other, out of which there can emerge a political platform they can all stand for. In this sense, intraparty deliberation can function as the central mechanism of political mediation in the sense we propose to conceive it here: effectuating the transition from individual preferences to a shared conception of the common good. This essential role played by political deliberation in both “forming” and “refining” collective political preferences is also a key trope in the existing normative literature on deliberative democracy: it was first noted by Joshua Cohen in his pioneering essay “Deliberation and Democratic Legitimacy” (Reference Cohen, Hamlin and Pettit1989, 32), and frequently recurs in the writings of such authors as Habermas (Reference Habermas1996), Dryzek (Reference Dryzek2000), and Gutmann and Thompson (Reference Gutmann and Thompson2004).

The perils of a merely aggregative conception of intraparty democracy. Finally, a third reason to suppose that a deliberative conception of intraparty democratization is likely to prove more effective than a merely aggregative one in stemming the tide of party decline is that the latter alone may only lead to an apparent democratization of the party structure, while in reality strengthening the control of the leadership over the base. This problem is well-documented. In a recent comment on the strategies adopted by political parties to counter the tendency towards membership decline, for instance, Kenneth Carty has pointed out that the kind of political participation offered by merely “aggregative” models of intraparty democracy is “atomistic” in the sense that “individuals are isolated from one another and engaged in direct communication only with the party center, in a fashion that inhibits their ability to act in common with each other” (Carty Reference Carty, Katz and Cross2013, 19). This, he adds, “provides the party in public office with the ability to manipulate a formally popular decision-making process, by ensuring that members’ choices are constrained and limited to alternatives acceptable to the existing elite” (ibid.). A similar refrain is also palpable in a number of recent empirical accounts of aggregative forms of intraparty democracy (Hopkin Reference Hopkin2001; Katz and Mair Reference Katz and Mair2009, 759; Mair Reference Mair1997).

The reason why the same problem would not arise—or at least arise to a much lesser extent—in a deliberative model of intraparty democracy is that deliberative settings require individual participants to exchange arguments for and against the courses of action proposed by those convening them, while also providing an opportunity devise alternative proposals to be held against them (see Gutmann and Thompson Reference Gutmann and Thompson2004, 16–7). This is likely to be a much more effective mechanism to check the power of the leadership over the base than merely voting on their proposals. Thus, moves towards more deliberative forms of intraparty democracy are more likely to promote real, rather than merely formal, democratization. Note that this too is a standard trope of both theoretical and empirical work on deliberative democracy. To speak with James Fishkin, “in the setting of a deliberative microcosm” a group ordinary citizens “tends to become more thoughtfully empowered, rather than manipulated” (Fishkin Reference Fishkin2009, 129). “After all, a disengaged and uninformed public is more easily manipulated than one that has firm opinions based on extended thought and discussion” (ibid., 130).

Towards a deliberative model of intraparty democracy

To the extent that intraparty democratization holds any promise at all for countering the current decline of parties, the arguments above suggest that a deliberative rather than merely aggregative model of intraparty democracy is more likely to be suited for achieving this goal. There remains, however, the question of how existing political parties might best institutionalize more inclusive and meaningful forms of democratic deliberation within their internal structure. While it is beyond the scope of this article to develop a full-blown deliberative model of intraparty democracy, we indicate some potential avenues for reform in this direction.

Empower local party branches. Local party branches are the “basic unit” of intraparty deliberation, being the venues where active and committed party members meet regularly to discuss local issues, or debate politics more generally, on a face-to-face basis (see Clark Reference Clark2004; Wolkenstein Reference Wolkenstein2016). Their accessibility makes them an important “entry point” for citizens into parties: lacking barriers to participation, party branches are a natural place to start if one wants to become politically active in a party. Because of this, and because of their uniquely wide scope of internal deliberation (since branch meetings recur on a regular basis, those who engage in them usually over time address a wide range of political challenges as well as the trade-offs between possible collective responses to these challenges), party branches are particularly well suited as vehicles for re-establishing the connection between citizens and parties, and promoting deliberation in a partisan context. The forward-looking task is then to make them more attractive as sites of partisan engagement. The most straightforward way of achieving this is to raise their influence within the party, for example by giving them more decision making power and allowing them to address more weighty deliberative tasks.

Reform the organizational structure of the party around the concept of “deliberative representation.” What we mean by this is that the deliberative processes that take place within separate party branches ought to be connected with each other by an institutional structure that generates deliberative relationships between them. Using the language of deliberative theory, one might say that such an institutional structure conceives the party as a “deliberative system” consisting of a number of deliberative sites which interact with one another (on deliberative systems, see Mansbridge et al. Reference Mansbridge, Bohman, Chambers, Christiano, Fung, Parkinson, Thompson, Warren, Mansbridge and Parkinson2012; also see Dryzek Reference Dryzek2012; Felicetti et al. Reference Felicetti, Niemeyer and Curato2016; Goodin Reference Goodin2008). In practice, this means that the outcomes of deliberative processes within local branches ought to be discussed by representatives of each branch in “executive committees,” which should themselves be internally deliberative and stand in a deliberative relationship to the individual branches. Moreover, branch representatives should upon return from their meetings with the executive committees justify the outcome of their deliberations vis-à-vis the members and respond to their concerns.Footnote ⁵ The idea is that in this way the decisions of the party as a whole emerge out of a series of interconnected deliberative interactions, ensuring that power is exercised within the party from the “bottom up” as well as from the “top down.” This structure can then be replicated in order to generate a territorial party structure, depending on the size and type of the polity.

Introduce function-specific fora for intraparty deliberation. In addition to the just-sketched institutional structure, intraparty deliberation can also be fostered by the creation of “function-specific” fora of deliberation within the party, which deal with specific issues that come up in a community or polity. For example, prior to a membership ballot, a forum could be established in which members from different branches and executive levels within the party exchange reasons for and against the issue on which the ballot is held (Wolkenstein Reference Wolkenstein2016). We stress however that function-specific fora ought not replace party branches as main sites of partisan deliberation. They can be useful additional means when issues that lie beyond the ambit of the branches’ deliberations arise—but they cannot substitute for the continuous and comprehensive deliberations that occur in branches and other executive committees, in which members develop shared positions on a broad range of political questions and so are compelled to address the politically important trade-offs between them.Footnote ⁶

Make use of new technologies. Finally, the vastly improved range of information and communication technology (ICT) can also powerfully aid the deliberative democratization of parties, in the sense that it can provide additional means for joining in partisan deliberation for those who are potentially interested in it but, say, lack the time to attend face-to-face meetings, or simply prefer the simplicity of participating online (see Bader Reference Bader2014, 367; Scarrow Reference Scarrow2015, 138–41). Parties could for instance experiment with online discussion fora as well as “wiki” methods for the joint formulation of strategic policy documents, as some parties have recently done (see Gauja Reference Gauja2015). Again, however, it is important to emphasize that ICT-assisted deliberative fora ought not replace party branches as the central venues for deliberation. The main reason is that the evidence on the ability of online discussion fora to promote reasoned and meaningful exchanges among citizens is very mixed, ranging from very promising (Polletta and Lee Reference Polletta and Lee2006; Price Reference Price, Davies and Noveck2006) to rather disappointing results (Coleman Reference Coleman2004; Smith et al. Reference Smith, John and Sturgis2013; for a comprehensive review of the recent literature on online deliberation, see Steiner Reference Steiner2012). Thus, parties should be cautious with overemphasizing ICT-driven deliberation, and see it more as a complement rather than a substitute for face-to-face deliberation.

Although mainly indicative, these four institutional design proposals already suggest that quite far-reaching reforms are needed. Rendering parties more internally deliberative requires considerable effort, and depends on political will insofar as party elites would have to renounce some of their powers to the benefit of the members. But given the severity of the current crisis of parties, we submit that the benefits of institutional reform on balance outweigh its costs—even for the party elites. For without party organizations that enjoy a level of legitimacy among the citizenry, party elites will find it increasingly hard to garner the electoral support needed to hold onto political power.

Response to some objections

Before drawing any conclusions from the above analysis, it might be useful at this stage to address some possible objections to our proposal. This will help to further clarify several aspects of our argument, as well as provide an occasion to put forward some additional empirical evidence in support of it.

Is deliberation really an incentive? A first possible line of objection might challenge whether a move towards more internally deliberative parties would indeed provide sufficient reason for prospective members to join. In an article entitled “Should We Believe that Improved Intra-Party Democracy Would Arrest Party Decline?” for instance, Richard Katz takes up a question that comes close to the key argument we have been seeking to advance in this article, but counters that parties have in fact already been offering more opportunities for membership involvement, and these have generally not been taken up as effectively as one might have expected (Katz Reference Katz, Cross and Katz2013, 59). This argument chimes with an extensive broader literature suggesting that individuals in contemporary societies do not see political participation as something inherently valuable, but are rather are only willing to get involved if the perceived costs of not doing so are greater than those of taking part (e.g., Hibbing and Theiss-Morse Reference Hibbing and Theiss-Morse2002; Posner Reference Posner2003; Verba, Shlozman, and Brady Reference Verba, Schlozman and Brady1995). From this point of view, is it not naïve to assume that giving individuals even more opportunities to get involved in the formulation of party platforms through deliberative means will provide an incentive for increased participation?

Several points may be advanced in response to this question. First of all, it is worth noting that the claim according to which actual or prospective party members are not taking up the opportunities for greater participation that are being offered to them has not yet been subjected to rigorous empirical scrutiny—and is in any case is far from having been confirmed by the studies that do exist (Scarrow Reference Scarrow2015, 193). Secondly, we should not forget that the kinds of extensions of intraparty democracy authors like Katz have in mind when doubting their positive impact on participatory incentives concern mainly the extension of “voting rights,” and therefore still refer to an essentially aggregative conception of intraparty democracy. This means that evidence suggesting that citizens do not take up the participatory opportunities parties provide them with says little, in fact nothing, about deliberative forms of intraparty democracy, and for the reasons given above it seems plausible to assume that deliberative participatory opportunities will not suffer from the same fate as aggregative ones.

Finally—and perhaps most importantly—it is worth emphasizing that the broader idea that citizens are largely uninterested in political participation, remains hotly contested in the literature too. An at least equally widespread thesis is that the degree of interest citizens display for political participation is “endogenous” to the kinds of opportunities they are actually offered (cf. Thompson Reference Thompson2008). In an influential article published in this journal, for instance, Michael Neblo and his colleagues test the hypothesis that “disaffection with modern mass democracy” stems mostly from “feelings of disempowerment and disillusionment” by relying on both survey data and extensive qualitative interviews concerning the “separability” of preferences for participation from different kinds of judgments concerning the state of contemporary politics (Neblo et al. Reference Neblo, Esterling, Kennedy, Lazer and Sokhey2010). This leads them to draw the following three main conclusions:

These conclusions are particularly relevant for the argument we seek to advance in this article because they can serve to both explain the negative results obtained by previous studies on the willingness to engage in political participation, and “invert their normative implications,” so to speak. For if the individuals who are most likely to be interested in participating in deliberation are also the ones that are most disillusioned by “politics as usual,” the reasons for their current lack of interest for participation would seem to be closely related to the kinds of opportunities they are offered. Thus, providing more opportunities for deliberative—as opposed to merely aggregative—participation would seem to be a plausible way of providing greater incentives to get involved.

Are parties still relevant? A second objection to our proposal can also be evinced from Katz's just-mentioned article. This objection relies on some of the same premises we have used to build our argument, but points to a different possible implication. At several junctures in his article Katz suggests that a more cognitively mobilized citizenry is bound to find the kind of participation political parties are capable of offering frustrating, since the nature of political parties is that they have to provide “comprehensive” policy platforms and therefore strive for “compromises” both with other parties represented in the legislative body and with other institutions involved in the policy-making process (Katz Reference Katz, Cross and Katz2013, 62). This leads him to contend that present-day citizens are more likely to seek “alternative forms of involvement,” through venues that can tailor their goals more closely to individual preferences, such as single-issue pressure groups or other kinds of nongovernmental organizations in civil society. Katz refers to this as political participation “à la carte” (ibid., 63).

Responding to this objection allows us to clarify several important aspects of our argument. First, nothing of what we have stated in this article is meant to undermine or challenge the utility (or indeed the value, from a democratic point of view) of forms of political engagement that occur outside of party politics. Indeed, we think that the mobilization of “opinion” through the mechanisms available in civil society is an essential complement to institutionalized party politics (cf. Urbinati Reference Urbinati2014). Secondly, however, we contend that parties perform an irreplaceable function within the framework of democratic regimes, precisely because they are required to formulate comprehensive political visions for society in large. For insofar as different individually desirable policy goals may be at odds or in tension with one another, politics cannot be done entirely “à la carte.” While there is no doubt room for single-issue initiatives and pressure groups, feasible and sustainable strategies for governing a country require a policy platform that integrates a great number of individual goals into a greater whole, addressing the various kinds of trade-offs that may emerge between them. Thus, independently of how other kinds of political organizations are structured (some of which may be more attuned than others to the transformed nature of the demand for political participation), political parties remain necessary for democracy and should therefore strive to become more internally democratic and, as we have suggested, deliberative in order to address the underlying motivational causes of the present crisis.Footnote ⁷ That participation “à la carte” might on its face exercise a greater appeal on many citizens thus ought not lead one to conclude (like Katz) that parties may reasonably be abandoned as venues of democratic participation.

Isn't there a tradeoff between intraparty and polity-level democracy? A third possible objection to our argument asks whether the specific kind of individuals who would be attracted and empowered by more deliberative forms of intraparty democracy would not pose a problem from the point of view of the party's overall functions. This is a version of a standard objection raised against intraparty democracy in general, according to which there is an inevitable “trade-off” between intraparty and polity-level democracy, inasmuch as those willing to take part in intraparty deliberations are likely to hold more “extreme” views compared to the party's electorate at large (see Katz and Cross (Reference Katz and Cross2013), Strom (Reference Strom1990), Wilson (Reference Wilson1962), and Wright (Reference Wright1971)). Applied to our specific proposal, the concern is therefore that empowering party members through mechanisms of intraparty deliberation might make parties less effective at competing on the broader electoral market, and thereby undermine the quality of democracy at the level of the polity as a whole.

Our contention is that this objection—and indeed the whole idea of a “trade-off” between intraparty and polity-level democracy—rests on at least two misconceptions which are worth dispelling here because this will enable us to provide further empirical and theoretical support for our key proposal. The first such misconception is the idea that people willing to take part in intraparty deliberations are by default likely to be unrepresentative of the party's broader electorate. This point is typically raised with reference to a presumed “law of curvilinear disparity,” which postulates that individual political views are likely to become more “extreme” in proportion to one's willingness to mobilize for them (May Reference May1973). The available empirical evidence that such a law exists is however very mixed and tends to end up pointing in the opposite direction. Whereas almost all recent studies on party membership have highlighted important sociological differences between those who are active in parties and the electorate in large—in that they tend to be disproportionately “male,” “older,” “more educated,” and “wealthier” compared to the general electorate (van Haute and Gauja Reference van Haute and Gauja2015; Scarrow Reference Scarrow2015; Scarrow and Gezgor Reference Scarrow and Gezgor2010)—there is very little evidence that their political views are unrepresentative of the wider electorate. In fact, several studies devoted specifically to this question have shown that almost the whole spectrum of political views present in society are more or less proportionately represented amongst members of parties (e.g. Norris Reference Norris1995; Scarrow and Gezgor Reference Scarrow and Gezgor2010; Van Holsteyn, Ridder, and Koole Reference Van Holsteyn, Ridder and Koole2015). This provides reason to suppose that, even if they are not sociologically representative of the broader electorate, party members and activists can still offer a fair representation of their political views.

The other misconception implicit in the idea of a trade-off between intraparty and system-level democracy is that an oligarchically run party would be more effective at tracking the preferences of the electorate than an internally democratic one (Strom Reference Strom1990; Wright Reference Wright1971). First of all, it should be made clear that this is at core a normative claim, resting on what are ultimately contestable beliefs about the superior efficacy of nondemocratic procedures compared with democratic ones (see Katz Reference Katz2014, especially 188–9). Even if we grant that the objection has some empirical force, therefore, there would still be scope for questioning it on the grounds of internal coherence. For, if we are to assume that elite-driven procedures are more effective at tracking the preferences of the general electorate than democratic ones, it is not clear why we should think that party democracy is desirable at the polity level to begin with. After all, why couldn't an enlightened oligarch at the polity-level perform the same function party elites are assumed to be more effective at carrying out at the level of the individual party? From the normative perspective we assume in this article, the only possibly answer is this: if preferences are not given prior to the decision-making process, but at least in part molded through it, decision-making processes have to be as inclusive and participatory as possible in order to track substantive preferences. If this argument is accepted, it should be clear that the idea that there may be a “trade-off” between intraparty and system-level democracy doesn't make much sense: the two must rather be seen as separate but complementary stages in the process of construction of a common will.

Are deliberative parties any different from old-school “mass parties”? A fourth possible objection to our argument might be to say that we are recommending ultimately amounts to little more than a return to the mid-twentieth century “mass party” model, which the empirical literature on parties has shown to be “obsolete” today (e.g., Van Biezen and Romée Piccio Reference van Biezen, Piccio, Cross and Katz2013). After all, the internally deliberative party we have canvassed has a territorial organizational structure comprising several hierarchical levels from the local branch up to the national level; it considers the membership organization to be the core of the party; and it is organized in a “bottom-up” fashion, with the membership being represented by delegates at each organizational level. All of these features are standardly attributed to the mass party model (see Duverger Reference Duverger1954; Katz and Mair Reference Katz and Mair1995; Krouwel Reference Krouwel2012; Panebianco Reference Panebianco1988).

But this objection misunderstands both the nature of traditional mass parties and the substantive content of our proposal. First of all, it should be noted that deliberation has never historically been a key component of the internal organizational structure of mass parties. On the contrary, it seems to play hardly any role. This emerges most clearly when we consider that the internal organization of the mass party is structured around a rather simplistic delegate conception of representation, which implies that the branch members’ representatives are authorized to speak on behalf of them for a fixed period of time, with the only mechanism of accountability being re-nomination procedures (Duverger Reference Duverger1954, 40–1; cf. Katz Reference Katz2014). By contrast, the proposed deliberative model of intraparty democracy requires representatives not only to receive mandate from the deliberative bodies they are supposed to represent, but also to continuously be accountable to them, defending and justifying their actions and decisions in a process of two-way communication. This is what is intended by the concept of “deliberative representation” we have introduced above; and in this important sense, our model of intraparty democracy differs starkly from the traditional mass party model.

From this perspective, it also doesn't seem so surprising that almost all observers of actually existing mass parties concluded that such parties were “democratic in appearance and oligarchic in reality” (Duverger Reference Duverger1954, 133). For it is far from clear that a purely delegative model of representation provides sufficient mechanisms for ensuring effective democratic accountability (see Pitkin Reference Pitkin1967; Urbinati Reference Urbinati2008). We believe that deliberative parties would fare much better than the traditional mass parties on the same count, since deliberation provides a more effective mechanism for ensuring democratic accountability than mere delegation.