Revisiting the constructivist turn in political representation

Abstract

In recent times, representation theory has become one of the most productive and interesting sub-fields in democratic theory. Arguably, the most important theoretical innovation are the so-called ‘constructivist’ approaches to political representation. These approaches play a central role in Creating Political Presence: The New Politics of Democratic Representation and The Constructivist Turn in Political Representation, two impressive volumes that take stock of the state of the art in representation theory. I discuss the two volumes by focusing on three broader and interconnected themes: the problem that constructivism is meant to respond to, the tendency of representation theorists to expand the possibilities of representation as broadly as possible, and the normative aspects of political representation and how constructivists deal with them.

Although few would dispute that political representation is central to democratic politics, there is plenty of disagreement about what exactly representation involves, where it takes place, and how it can acquire democratic legitimacy. Recent times have seen a vibrant scholarly debate about these issues, which has made representation theory one of the most productive and interesting sub-fields in democratic theory.

Creating Political Presence: The New Politics of Democratic Representation (CPP) and The Constructivist Turn in Political Representation (CT) are two impressive volumes that take stock of the state of the art in this growing field. They provide an excellent overview of the main theoretical innovations and offer valuable new impulses for further developing the theory of political representation.

In this short piece, I want to discuss what I take to be the key ideas and arguments presented in the two volumes. Since I cannot possibly do justice to the full breadth of approaches presented in CPP and CT, I will structure my discussion around three broader and interconnected themes:

The problem that the new theories of representation are meant to be a response to, that is, globalisation and the crisis of representative democracy.

The tendency of representation theorists to expand the possibilities of representation and the signification of the term as broadly as possible.

The normative aspects of political representation and how the new representation theories deal with them.

I foreground these three themes for two reasons. First, they cut across the entire spectrum of contributions. Second, engaging with them helps us get a better sense of the distinctive virtues, as well as the limitations and blind spots, of the new theoretical approaches to political representation.

Constructivism in representation theory

The single most important theoretical development that the two volumes engage with is the recent ‘constructivist turn’ in representation theory. This is true not only for CT, which deals exclusively with constructivism, but also for CPP. It should therefore come as no surprise that this article will be primarily concerned with the promises and pitfalls of constructivism as presented and discussed in CT and CPP.

Put simply, constructivists foreground the performative and creative aspects of political representation. As Michael Saward, one of the pioneers of constructivism, puts it, constructivism is predicated on the idea that ‘representation’s political presence arises primarily by virtue of its being done – practiced, performed, claimed. Representative roles and relations gain a presence in our politics because … actors make claims to speak for others (and for themselves)’ (Saward, CPP, p. 271). In advancing such ‘representative claims,’ actors do much more than making the views or concerns of others present in the public sphere: they ‘solicit their objects’ of representation by actively conjuring up images or symbols of what or whom is to be represented, together with ‘images of the would-be representative’ (Disch, CT, pp. 5, 7).

Constructivism is not a unified paradigm: it has a long pedigree in ‘post-foundational’ thought, especially the work of Claude Lefort, Ernesto Laclau and Chantal Mouffe, and contemporary constructivist scholars draw on a wide range of different theoretical and historical resources (cf. Ankersmit and Hayat in CPP, Disch and Mulieri in CT). Still, there are a number of things that virtually all constructivists agree on. First and most fundamentally, they agree that ‘the identity, interests, or preferences of the represented are not given prior to representation but shaped through being represented’ (Fossen, 2019: 824). This has momentous theoretical implications, notably for the central normative question of how we should evaluate the legitimacy of acts of representation – the third theme I discuss below. For if the views the represented hold are in fact a product of acts of representation, widely accepted normative criteria like ‘responsiveness’ become meaningless. As we shall see, some Laclau and Mouffe-inspired constructivists respond to this challenge by suggesting to drop the subject of legitimacy altogether.

Second, constructivists widely agree that political representation should be conceived in much broader terms than is ordinarily the case. They argue that we should ask where and how it occurs beyond elections, parties and parliaments. This concern with ‘disentangling democracy from representation,’ as Castiglione and Pollak (CPP, p. 2) provocatively put it, is a logical consequence of assuming that representation can in principle be ‘practiced, performed, claimed’ (Saward, CPP, p. 271) by anyone within and without the familiar institutions of electoral democracy. This shift in thinking about representation promises to increase awareness for new and unconventional sites, contexts and practices of political representation; but as I will discuss in connection with the article’s second guiding theme, constructivists’ ‘expansionism’ (Wolkenstein and Wratil, 2021) also has drawbacks.

Third, at least in CPP and CT, constructivism is generally presented as a ‘political response’ (Disch, CT, p. 10) to the large-scale transformations established democracies have undergone in last 40 years or so. Many of the contributors indeed suggest that traditional conceptions of political representation that are centred on elections, parties and parliaments are unfit for an increasingly globalised world in which domestic legislatures are losing power to ‘transnational decision-making arenas’ (Castiglione and Warren, CT, p. 22), and growing numbers of citizens are alienated from established politics (Castiglione and Pollak, CPP, p. 1). And not least because it detaches the concept of representation from domestic electoral institutions and allows looking for alternative forms of political representation, constructivism is said to provide a meaningful theoretical alternative. This way of framing the relative merits of constructivism raises several questions, however. I address these in the next section.

Theme 1: The ‘problem’ that new representation theories respond to

Is the problem correctly framed?

One obvious question that arises is whether the just-mentioned analysis of democratic transformations is actually sound? Many contributors appear to think that national parties and parliaments have lost most, if not all, of their relevance. Disch echoes a widely shared view when she argues that ‘[w]ith the emergence of supranational political institutions such as the European Union, the increasing power of global corporations and global financial institutions such as the World Bank, and the urgency of global political issues such as migration, global trade, and environment,’ political representation now largely ‘takes place outside national parliaments and affects populations that are not formally organised as electorates’ (CT, p. 4). Again, this diagnosis is meant to underline the virtues of a constructivist approach to representation that opens the door to studying new, non-traditional ways in which representation makes itself visible.

Arguably, though, the diagnosis needs to be qualified. Without denying that the described developments put pressure on conventional understandings of political representation, one must be very cautious with overstating the irrelevance of nation state-based party politics and electoral representation. After all, it is still national political parties that organise elections and campaigns, and they continue to play a major role in domestic political debates (where they advance ‘representative claims’ on a daily basis), in addition to making collectively binding laws in parliament (for a thoroughgoing discussion, see Ignazi, 2017). Moreover, as the recent Covid-19 pandemic has shown, national governments and legislatures remain the principal wielders of power in multiple globally relevant policy domains – think of phenomena such as ‘vaccine nationalism.’

Of course, the conclusion that it would be rash to unquestioningly accept the notion that conventional domestic politics is rendered much less relevant by globalisation does not disallow constructivism. Quite the contrary, it is possible to endorse the broader constructivist agenda while rejecting (some of) the diagnostic arguments that are being put forward in order to justify that agenda.

This is nicely illustrated by the few constructivist contributions to CPP and CT that do not declare traditional electoral politics insignificant. Fossum, for example, criticises the tendency of representation theorists to ‘jump nimbly over political parties’ (CPP, p. 86) and goes on to argue that by conceiving representative relationships as dynamic and relational, constructivism allows us to ask whether the relationship between citizens and parties has changed in the way it did because citizens’ and parties’ expectations towards democratic representation have evolved over time and shifted away from the classic idea that good representation equals responsiveness (CPP, pp. 101–112).

More boldly, Geenens defends Lefortian constructivism (recall that the French political theorist Claude Lefort was a constructivist avant la lettre), according to which traditional representative institutions and procedures remain important because they ‘construct and uphold the point of view from where citizens can see and understand themselves as members of [a] specific society’ (CT, p. 96). A similar argument is advanced by Warren, who, criticising constructivists for thinking too little about the ‘systemic’ effects of representation, suggests that representative democratic systems as a whole can ‘support and develop capacities for democratic citizenship’ (CPP, p. 47).

Interestingly, this rather different way of framing constructivism has implications that run counter to how Disch and other influential constructivists present their approach. Instead of providing a theoretical language to describe, and underline the importance of, ‘alternatives to the established institutions of political representation’ (Disch, CT, p. 1), it offers normative reasons to preserve those institutions. This is especially true for Warren’s and Geenens’ chapters; Geenens in fact argues that his Lefort-inspired constructivism can help push back against currently popular proposals to replace the central institutions of representative democracy with (for instance) randomly selected assemblies (as in Landemore, 2020).

Can constructivism help us solve democratic deficits?

This second question addresses what may be called the remedial potential of constructivism. If constructivism is meant to be a valuable ‘political response’ (as Disch puts it) to the transformations that representative democracies have undergone, it might be said, it must also be able to offer some practical guidance for fixing the democratic deficits that have arisen in connection with those transformations.

Consider the present ‘crisis of representative democracy’ that both volumes take as their starting point (see Castiglione and Pollak, CPP, p. 1). The narrative is familiar: party memberships and public trust in parties are declining; parliaments are increasingly polarised and barely able to take collective decisions on key issues, from agreeing on a Brexit deal to effectively tackling climate change (e.g. Rosenbluth and Shapiro, 2018). Can constructivism contribute to solving some of these issues, for example, by informing institutional reform proposals that could help us overcome the dire status quo?

If we treat constructivism merely as a means to conceptualise and study new non-electoral forms of political representation, as many contributors do (see, paradigmatically, Montanaro in CPP), then it is doubtful whether it can be of much help in mitigating the crisis of representative democracy. For arguably, a major cause of that crisis is institutional malfunctioning: there is wide agreement that the difficulties facing representative democracy are rooted in the increasing incapacity of political parties to perform their representative functions (e.g. Ignazi, 2017; Rosenbluth and Shapiro, 2018; Wolkenstein, 2019). A shift of focus to unconventional representative practices is of little help when it comes to addressing these issues. If anything, it gives us reason not to preoccupy ourselves with the problems that beset parties and parliaments.

But of course, it is possible to think differently about constructivism, approaching it as a set of conceptual resources that can inform theory-building more generally, without committing one to a specific substantive research focus. Mansbridge’s contribution to CPP demonstrates how this might work. Starting from a constructivist understanding of representation, Mansbridge argues for intensifying ‘recursive’ two-way communication between citizens and elected politicians. A ‘radical communicative restructuring’ of the representative division of labour is one option she advocates. If ‘the elected representative did more communicating with both constituents and other legislators while the staff did more policy-crafting,’ for instance, this would ‘reposition the representative as interlocutor’ (Mansbridge, CPP, p. 309). Supplemental institutions such as specially designed deliberative assemblies could additionally help representatives and constituents to dynamically adjust their representative relationships to changing political circumstances (pp. 310–312).

Mansbridge’s approach is a rare outlier in the two volumes under discussion, but it persuasively shows that constructivism can inspire efforts to reform conventional electoral forms of political representation, thus potentially contributing to solving the on-going crisis of representative democracy. It thus seems safe to conclude that constructivism can indeed be a valuable response to the problems that it is meant to respond to – especially so when it is applied flexibly and with an eye to more specific institutional challenges.

Theme 2: The unlimited expansion of representation

The second and related theme that I want to consider is the ‘expansionism’ (Wolkenstein and Wratil, 2021) of many of the new theories of representation, that is, the already-mentioned tendency to expand the possibilities of representation and the signification of the term as broadly as possible. As noted, the main theoretical move that underlies constructivist expansionism is the ‘disentanglement’ (Castiglione and Pollak, CPP, p. 2) of representation from electoral democracy. Saward, a leading constructivist scholar, goes so far as to suggest that representation should be understood as a ‘liminal’ concept without any conceptual essence, and that its ‘liminality renders as fragile some efforts to fix and limit the concept’s meanings and range of reference’ (CPP, p. 276), such as the classic conception of democratic representation that sees the latter as closely bound up with the institutions of elections, parties and parliaments.

One objection to this that has been raised in recent representation scholarship is that such moves render representation a rather empty and useless concept. For example, Schweber (2016: 383) argues that ‘[w]hen any activity is considered representation, anyone can lay equal claim to the status of representation, and anyone (or anything) has a legitimate claim to be represented; then the idea of representative government has lost its normative and analytical force.’ Urbinati’s chapters in CPP and CT echo this criticism.

Schweber and Urbinati’s primary worry is that treating representation as a ‘liminal’ concept makes it virtually impossible to assess the legitimacy of representative practices. In their view, assuming some conceptual essence is necessary in order to formulate meaningful evaluative standards. How constructivists respond to this challenge will be the topic of the next section. First, however, I want to address another, potentially even more troubling, the implication of clearing the concept of representation of any conceptual essence. This is that it also makes it very difficult to distinguish between relevant and irrelevant instances of political representation.

Is constructivist expansionism power-insensitive?

In a nutshell, the problem is this: if political representation is not conceived as conceptually tied to particular representative sites (e.g. parliament), practices (e.g. what elected representatives say and do) or procedures (e.g. bargaining among parties), then it must be assumed, a priori, that all representative sites, practices and procedures are equally relevant, and by extension equally worthy of scholarly attention. At best, this assumption can increase awareness of marginal and unconventional representative practices. At worst, it will divert attention away from the most consequential forms of representation. A warning example is the fact that short-lived protest movements like the Spanish Indignados, failed ‘alternative parties’ like the German Pirate Parties, Syriza or the Movimento 5 Stelle (Disch and van de Sande, CT, Hayat, CPP) as well as all kinds of ‘self-appointed’ representatives (Montanaro, CPP) are throughout the two volumes presented as important objects of scholarly concern – while, as I have already noted, conventional electoral politics is barely considered relevant. ¹

The good news is that one need not restrict the ‘meanings and range of reference’ (Saward, CPP, p. 276) of the concept of representation in order to distinguish relevant from less relevant acts of political representation. It suffices to ask the simple question of whether representation is connected to power: to what degree can those who advance claims to represent others exercise power over others? Even accepting that the meaning of ‘power’ can vary quite a bit (see Flynn in CT), this question can guide us towards relevant forms of representation.

Asking the question of power can help constructivists to identify highly impactful representative practices and bring their conceptual apparatus to bear on them in a fruitful fashion. Incidentally, it might also reveal that their analytical devices are sometimes less useful than initially assumed (on this point, also see Rubinstein, 2014). Think in this connection of the numerous elite lawyers who, as Pistor (2019) shows in a critically acclaimed recent study, continue to play a central role in the construction and maintenance of the ‘the legal empire that sustains global trade, commerce, and finance’ (159). In a sense, these lawyers are powerful political representatives: their work is ‘central to the coding of capital and the distribution of wealth in society’ (162) – but the representative relationships they have entered with their wealthy and powerful clients is certainly not ‘constructed’ through representative claims. After all, legal representation is the quintessential form of representation by formal mandate (see Mulieri, CT, pp. 206–207).

Issue 3: The normative aspects of political representation

The third and final theme I want to discuss concerns the normative aspects of political representation. Nearly all contributors to CPP and CT address the central normative question of what makes representation democratically legitimate. And because most of them endorse constructivism in one form or another, they tend to reject traditional standards of democratic legitimacy.

Chief amongst the latter is responsiveness, the normative standard of choice for most empirical political scientists (Sabl, 2015). Political representation is typically thought to be ‘responsive’ to the extent that elected politicians implement policies that track voters’ policy preferences. Obviously, this sits uneasily with constructivism. Most of all, the concept of responsiveness presupposes that constituencies exist prior to acts of representation, and that political representatives should react to what those pre-existing constituencies want in terms of policies. Constructivists, in contrast, hold that constituencies are made by representatives, who advance ‘representative claims’ that evoke images of what or whom is to be represented (as well as shaping the preferences of constituents).

How can citizens critically engage with representative claims?

If, as most constructivists appear to think, old-fashioned responsiveness is a non-starter, what alternative normative standards are available?

Several contributors to CPP and CT converge on the notion that the democratic legitimacy of political representation depends on whether citizens can critically engage with the representative claims political actors put forward, and accept or reject those claims. As Montanaro (CPP, p. 194) writes: when assessing the legitimacy of representative claims, ‘[w]e want to know whether or not a representative is speaking for people in a manner that is acceptable to them and if he or she has enabled a constituency to exercise its judgment.’

It should be clear that this is rather demanding standard of legitimacy. It presupposes, on the one hand, that citizens are ‘actively engaged in discussing and judging … the claims made in their name,’ and on the other hand ‘requires institutions that assist them in acting as subjects and in putting their representative claims to the test’ (Hayat, CT, p. 135). Many, I suspect, will find the first expectation – that citizens should be disposed to constantly ‘use their agency to judge, criticise and transform’ representative claims (ibid.) – quite onerous. But here, I want to concentrate on the issue of which institutions may enable citizens to critically engage with representative claims, and how.

Democratic theorists seem uniquely well placed to offer relevant answers to this question. Unfortunately, though, the discussions of enabling institutions in CPP and CT remain rather abstract, and the arguments advanced tend to raise more questions than they answer. To cite just one example, Hayat argues that active membership in political parties allows citizens ‘to make their views heard in political arenas from which they were previously excluded’ (CPP, p. 144), thus facilitating the contestation and transformation of representative claims. But does this sort of partisan mobilisation actually involve political judgment, rather than being driven by the sort of ‘cue taking’ that is problematised by innumerable empirical studies (e.g. Bisgaard and Slothuus, 2018; Broockman and Butler, 2017)? And, even more importantly, how must parties (which are typically run in a top-down fashion) be designed in order for citizens to meaningfully make their views heard? These quite central issues are barely addressed.

The only contribution that tries to tackle some of these challenges is Mansbridge’s aforementioned chapter in CPP, which proposes to redesign traditional representative institutions in such a way as to enable ‘recursive’ deliberative communication between citizens and those who claim to speak for them. Future research in representation theory would do well to take inspiration from Mansbridge and engage more with institutional design questions, ideally informed by systematic and rigorous empirical research. For if one accepts the proposition that the democratic legitimacy of political representation is closely bound up with citizens’ capacity to contest and revise representative claims, arguably more thought needs to be put into how institutions can support and sustain that capacity.

Should we ignore the question of legitimacy?

Note that some constructivists deal very differently with the question of how political representation can acquire democratic legitimacy: they suggest that it is not a question worth attending to. Disch, in particular, argues that ‘constructivist theorists … need not answer the legitimacy brief’ (CPP, p. 163). Her thought-provoking counter-proposal is that representation theorists should instead embrace the ‘citizen standpoint,’ which ‘confronts theorists not with the problem of legitimacy but with that of hegemony’ (CPP, p. 164).

Following Laclau and Mouffe, Disch defines hegemony in terms of ‘the conditions for and peculiar challenge of modern politics: to institute politically the division of the social into groups and to build alliances among some in opposition against others’ (ibid.). There ‘exists no ground from which to adjudicate the legitimacy of such alliances’ (ibid.), only a set of political questions that citizens must take a position on: ‘What is this struggle to which I commit myself? Who or what is my opponent? Who might be an ally? And how might my struggle be transformed by choosing this alliance rather than that one?’ (CPP, p. 179).

Disch contends that this ‘radically democratic’ way of thinking about political representation most naturally aligns with constructivism’s ambition to ‘disclose representation’s capacity to remap society – its group identities, national boundaries and lines of conflict’ (CPP, p. 178). Yet one wonders whether a normatively richer approach that is equipped with (however minimal) standards of democratic legitimacy wouldn’t be equally compatible with constructivist commitments. The normative standard I discussed in the previous subsection certainly does not jar with anything constructivists stand for (see Saward, CPP, p. 288) – and it furthermore lends constructivist accounts of representation a welcome critical edge that Laclau and Mouffe’s work probably cannot provide.

This leads to a final worry: modelling a constructivist account of political representation on Laclau and/or Mouffe’s radical theories of democracy holds the risk of ‘inheriting’ the well-known shortcomings of those theories. One of these shortcomings is that their strong formalism, which is especially visible in the later work of Laclau, not only deprives them of normative resources to distinguish between democratically legitimate or illegitimate instances of representation but also makes them unable to explain why attempts to ‘remap society,’ as Disch puts it, sometimes misfire (Jäger and Borriello, 2020: 751–758) – as they arguably did in the case of recent left-wing protest movements like Syriza (on this, also see van de Sande, CT).

Another serious drawback is the ‘institutional deficit’ of Laclau and Mouffe’s radical theories of democracy, that is, the glaring ‘absence of considerations concerning the setup of the rules and procedures of political decision-making’ (Westphal, 2019: 188). Not least because of the continuing relevance of elections, parties and parliaments, an approach that has virtually nothing to say about formal political institutions appears ill-suited to inform a theory of representation. In other words, Laclau and Mouffe may be the ‘silent partners’ (Disch, CPP) of constructivists, but it is doubtful that they are useful partners. That said, there also exist promising efforts to connect radical democratic theory to constructivist theories of representation. Typically, however, these depart significantly from the primary texts of Laclau and Mouffe, as in Marchart’s thoughtful attempt to outline democratic ethics based on the insight that a ‘perfect fusion between represented and representative’ is impossible (CT, p. 151).

Conclusion

In sum, I suggest there are three main takeaway points. First, representation theorists would do well to pay more attention to conventional representative institutions, the specific institutional design challenges they face, and how they could support the contestation of representative claims. Second, theorists should also ask more often to what extent political representation is connected to power. Finally, and more positively, there can be no doubt that the ‘constructivist turn’ provides us with enormously useful theoretical resources for studying political representation.