Taking stock of the field of populism research: Are ideational approaches ‘moralistic’ and post-foundational discursive approaches ‘normative’?

This article sets out to examine two claims that have increasingly come to define the dividing lines between the ideational and the post-foundational discursive approaches to populism: namely, that the former is moralistic and the latter is normative in orientation. The article considers the conceptual merits of both critiques while using them to further examine some of the implicit assumptions and pitfalls within Cas Mudde’s and Ernesto Laclau’s paradigmatic conceptualizations of populism. It is argued that ideational scholars’ attribution of a moralistic particularity to populism runs the risk of pathologizing the latter for characteristics that are arguably constitutive of all politics, while the danger of a certain crypto-normativity can be seen in Laclau’s tendency to equate populism with the political and simultaneously emphasize its emancipatory effects. The key difference between the two approaches ultimately consists in the location that they assign to populism within the wider topography of politics itself.


Introduction
Recent years have seen not only a veritable boom in the study of populism but also an increasing crystallization of different 'schools' of populism research, as evidenced by the publication of volumes such as The Ideational Approach to Populism  or articles that have contributed to formalizing 'discourse theory in populism research' (Stavrakakis, 2017a) or 'a discourse theoretical framework for the study of populism' (De Cleen and Stavrakakis, 2017). In this context of growing differentiation, numerous lines of demarcation have emerged between the various research perspectives: on the most basic level, in terms of the conceptual status ascribed to populism as a discourse, frame, ideology, strategy, or style; in addition, and more subtly, in terms of the conceptual and normative presuppositions underlying the different definitional approaches to populism. This has especially been the case when it comes to staking out the dividing lines between two of the most influential approaches in the literature: ideational approaches, based on Mudde's (2004) 'thin-centred ideology' conception of populism, on the one hand and post-foundational discursive approaches, based on Laclau's (2005a) conceptualization of populism as a 'political logic', on the other. While Mudde and Rovira Kaltwasser themselves have referred to 'ideational' approaches in a broad sense as encompassing all those that conceptualize populism as a 'set of ideas' (Mudde and Rovira Kaltwasser, 2013: 150) or 'as a discourse, an ideology, or a worldview' (Mudde and Rovira Kaltwasser, 2017: 12), recent exercises in mutual self-demarcation between Laclau-and Mudde-inspired approaches warrant deeper scrutiny. This article sets out to isolate and critically examine two claims, in particular, that have increasingly gained currency in recent years: first, the contention that ideational approaches to populism are moralistic (Jörke and Selk, 2018;Katsambekis, 2020;Stavrakakis, 2017b;Stavrakakis et al., 2017;Stavrakakis and Jäger, 2018); second, the notion that post-foundational discursive approaches to populism are normative Mudde, 2017;Peruzzotti, 2019;Rovira Kaltwasser, 2012).
What should be noted here from the outset is the author's own positionality as a researcher committed to a post-foundational discursive approach to populism as well as a broad and inclusive understanding of the productive intersections and differences among ideational approaches in the wider sense. The goal of this article is not least to build bridges while inviting critical self-reflection for both 'schools' of populism research, in addition to taking stock of the field of populism research as a whole through the lens of specific debates. In the following, I first undertake a brief overview of the ideational turn in populism research within which both the ideational (in a narrower sense) and the post-foundational discursive approaches can be situated. I then examine the critiques of moralism and normativity in turn, keying in on some of the implicit assumptions and pitfalls within Mudde's and Laclau's conceptualizations of populism in the process. The article closes with a concluding section that discusses both the challenges and opportunities that the preceding discussion points to for the ideational and post-foundational approaches alike.

Defining the terrain: The ideational turn in populism research
The field of populism research has come a long way since the debates in the 1960s and 1970s, for which the Ionescu-Gellner volume of 1969 -based on an international conference held at the London School of Economics -remains highly illustrative. The various contributions to this volume, for all their differences, ultimately converged in 'ascribing to populism some particular social content' (Laclau, 2005a: 8) and reducing populism to an epiphenomenal expression of underlying socio-structural conflicts. Seen this way, the 'people' invoked by populists could always be taken to have a determinate socio-structural basis, whether this was defined as 'a predominantly agricultural segment of society' (McRae, 1969: 163) or some form of 'alienation' that can be 'racial', 'geographical', or 'urban', but is 'always social' (Wiles, 1969: 167). Ionescu and Gellner (1969: 4) summarized this consensus when they wrote in the introduction that populism worshipped the people. But the people the populists worshipped were the meek and the miserable, and the populists worshipped them because they were miserable and because they were persecuted by the conspirators. The fact is that the people were more often than not identified in the peasants who were and are, in underdeveloped societies especially, the most miserable of the lot -and the more miserable they were the more worshipped they should be.
The early field of populism research is characterized by what Sartori ([1968] 1990) referred to as an 'objectivist bias' that treats political identities as mere 'artifacts' epiphenomenal to the underlying 'facts' of society. While this can be seen, in exemplary fashion, in the move of reducing the 'people' of populism onto specific socio-structural group categories, there is also a parallel tendency to locate populism outside the assumed normality of politics (especially party politics) -from the characterization of populism as 'a-political' (McRae, 1969: 163) to the notion that populism is 'moralistic rather than programmatic' and 'loosely organized and ill-disciplined' (Wiles, 1969: 167) as opposed to 'highly-structured parties' (McRae, 1969: 156-157). Implicitly visible here are the presuppositions of an objectivist political sociology, for which Lipset's and Rokkan's understanding of party politics takes on a paradigmatic character: political parties have 'an expressive function' of 'crystalliz [ing] and mak[ing] explicit the conflicting interests, the latent strains and contrasts in the existing social structure' (Lipset and Rokkan, 1967: 5;emphasis in original). Understanding populism as the reaction of 'a predominantly agricultural segment of society' to 'some kind of modernization, industrialism, call it what you will' (McRae, 1969: 163-164) entails essentially following the same objectivist logic, yet it can already be seen here that populism possesses a certain conceptual elusiveness insofar as it resists straightforward classification in terms of Lipset's and Rokkan's theory of four cleavages and their institutionalized 'freezing' into place over time. The concept of populism poses a special challenge to objectivist understandings of politics insofar as the group that populists claim to represent is no less than 'the people' as such; faced with this problem, social scientists can either conceptually reduce 'the people' of populism onto more determinate locations within the social structure such as the peasantry (what populists actually invariably mean when they speak of 'the people') -or they can turn the apparent indeterminacy of 'the people' into part of the definitional basis of populism itself: namely, as a necessarily contingent construction that can take on wide variations of meaning.
In the face of these two possibilities, the ideational turn in populism research entails a passage from the first to the second. What ideational approaches (in a broad sense) have in common is an understanding of the 'people' in populism as a construction, rather than an expression of an underlying socio-structural group; following from this is the sheer variability of populist phenomena, from left to right, agrarian to urban, radical democratic to autocratic, socially progressive to reactionary. Understood thus, all ideational approaches to populism are ultimately grounded in social-constructionist premises; what is foreclosed in earlier objectivist approaches -namely, the question what populists actually mean when they say 'the people' -becomes the main object of inquiry for ideational ones. The definitions of Mudde (2004) and Laclau (2005a) take on a paradigmatic character for this literature insofar as they explicitly and systematically invest the concept of populism with a set of constructionist theoretical underpinnings: Freeden's (1996) morphological conception of ideology in the first case and a 'post-foundational' (Marchart, 2007) theory of discourse and hegemony in the second. The two differ markedly, however, in the location that they assign to populism within their respective understandings of politics: Mudde (2004) refers to populism as a 'thin-centered ideology' -a designation that Freeden (1998) originally applied to nationalism and later rejected for populism (Freeden, 2017) -in other words, a less-than-fully-fledged ideology with a limited conceptual core; Laclau (2005a: 67), on the other hand, argues that populism is the quintessential political logic, 'the royal road to understanding something about the ontological constitution of the political as such'. Whereas populism falls short of a normal -ism for Mudde (2004: 544) -it 'does not possess "the same level of intellectual refinement and consistency" as, for example, socialism or liberalism' -it represents for Laclau the pinnacle of the political, or perhaps the proverbial tip of the iceberg that points to a dimension inherent to all politics. As will be seen in the following sections, this difference is crucial for understanding the lines of (self-)demarcation between the two approaches.

Are ideational approaches 'moralistic'?
There is a paradoxical double movement in Mudde's ideational approach insofar as populism is ascribed a high degree of indeterminacy and, at the same time, a specifically moralistic character. 1 If all ideologies, following Freeden (1996), can be understood as morphological systems of political concepts recurring in certain patterns of signifying relations to each other, populism, according to Mudde (2004Mudde ( , 2017Rovira Kaltwasser, 2013, 2017), is a 'thin-centred ideology' insofar as its conceptual core is limited to 'the people', 'the elite', and the general will of 'the people' -all of which can take on a wide range of thicker ideological inflections, albeit as specifically moralized categories, with moralism being 'the essence of the populist division' of people versus elite (Mudde, 2017: 29). Indeed, Mudde (2004: 543) stipulates in his minimal definition from the outset that populism is specifically about '"the pure people" versus "the corrupt elite"'; he even argues in this vein that [p]opulism is moralistic rather than programmatic. Essential to the discourse of the populist is the normative distinction between 'the elite' and 'the people' [. . .]. Opponents are not just people with different priorities and values, they are evil! (Mudde, 2004: 544;emphasis in original) Moralism as a definitional characteristic of populism has been prominently taken up in subsequent ideational approaches -from the likes of Hawkins (2009Hawkins ( : 1042, who defines populism as a specifically 'Manichaean discourse that identifies Good with a unified will of the people and Evil with a conspiring elite', to Müller (2016: 38), who posits as the defining feature of populism 'a claim to exclusive moral representation'. It is worth noting that, with his moralistic/programmatic distinction, Mudde cites Wiles' (1969) understanding of populism as a diffuse form of social 'alienation', pointing to a line of continuity from earlier objectivist approaches that situated populism outside the assumed normality of politics. Unlike the latter's assumptions of cleavage-based politics, however, Mudde's (2004: 544) approach is consistently ideational (rather than objectivist) in conceptualizing populism's abnormality not in terms of the ambivalence of its social base but in terms of its status as an ideology that simply 'does not possess "the same level of intellectual refinement and consistency" as, for example, socialism or liberalism' as well as its reliance on 'moralistic rather than programmatic' appeals.
This definitional emphasis on moralism has been criticized by post-foundational scholars in particular who argue that moralism is neither specific to populism nor even a consistent feature of discourses that otherwise fit the people versus elite criterion for populism (Katsambekis, 2019;Stavrakakis and Jäger, 2018). This critique is of an empirical, practical-political but also conceptual nature. On one level, Stavrakakis and Jäger (2018: 13) cite examples -ranging from Allende to Thatcher -for 'morality as a discursive and affective resource [that] can be put to very different uses', while Katsambekis (2019: 23) refers to left-wing populist discourses from Latin America to Greece that openly acknowledge that 'the people', far from being morally virtuous and pure, 'have their own vices and flaws', while attributing this in populist fashion to 'their subordinate and marginalised position' vis-à-vis the elite. In addition, Stavrakakis (2017aStavrakakis ( , 2017b points to the double hermeneutics of a morally charged anti-populism that has a long intellectual tradition (most notably in the work of Richard Hofstadter) and carries important practical-political effects, not least in discrediting challenges to the established sociopolitical order in the name of 'the people' as irrational and outside the boundaries of 'normal' politics. In the context of the Eurozone crisis, and in conjunction with his analyses of anti-populist media and partypolitical discourses in Greece (Stavrakakis, 2014;, Stavrakakis (2017a: 3) emphasizes the hegemony-stabilizing effects of the conceptual choice to define populism as a pathology of some kind or another: By un-reflexively adopting an exclusively pejorative definition of populism, a large part of populism research has also adopted the normative, if not axiomatic and stereotypical fallacies of Hofstadter, and has, by default, placed itself in the service of a normalizing, disciplinary technology of domination defending at all cost the post-democratic mutations of the established order [. . .] against all challengers irrespective of their ideological belonging, democratic credentials, discursive genealogies and political agendas.
In this manner, Stavrakakis' double hermeneutics argument comes full circle, suggesting that those who conceptualize populism as moralistic are themselves characterized by a moralizing normativity. In effect, the simple view of populism as a moralistic phenomenon reproduces the moral economy of post-democracy, which sets narrow boundaries on what can be considered rational, legitimate politics; if, as Dahrendorf (2003) argued, the 'accusation of populism can itself be populist', the attribution of moralism can itself be a moralizing gesture.
To what extent, however, is this critique of moralism a purely practical-political oneand is there something to be said for its specifically conceptual merits (without, however, losing sight of the practical-political dimension with which they are ultimately intertwined)? To begin with, it is worth revisiting the argumentative steps with which the likes of Mudde arrive at their designation of populism as moralistic. As mentioned previously, Mudde contrasts the 'moralistic' character of populism to 'programmatic' politics; yet there is a certain slippage (for lack of a better term) in Mudde's (2004: 544) subsequent description of how populism is moralistic: on the one hand, populism entails the specific accusation that the Other is 'evil'; on the other hand, '[p]opulism presents a Manichean outlook, in which there are only friends and foes'. The latter point, however, points to a more formal dimensionthat of an antagonistic division of society into opposing camps -independently of whether 'friends' are actually constructed as morally virtuous and 'foes' as morally evil. In the work of Hawkins (2009Hawkins ( : 1063, this distinction becomes explicit, with populism being ascribed a 'Manichaeanism' that comprises the two separate dimensions of moralism and dualism: '[populism] ascribes a Manichaean vision of the world, that is, one that is moral (every issue has a strong moral dimension) and dualistic (everything is in one category or the other, "right" or "wrong," "good" or "evil")'. Understood thus, dualism is conceptually distinct from moralism, with the latter being one possible way of coding a dualist opposition of us versus them, while populism according to Hawkins specifically entails both dimensions of moralism and dualism. The question remains, however, why a dualism of us versus them has to go hand in hand with moralism in the strong sense of the pure, virtuous 'us' versus the evil, conspiring 'them' in the case of populism; what is it, in other words, about speaking for a 'people' against an 'elite' that entails a Manichaean conspiracist view of the world? Conversely, if moralism is understood in a weaker sense and conflated with dualism, it is equally unclear how this makes populism an abnormal and non-programmatic form of politics. Stavrakakis and Jäger (2018: 13) already point to the slippage between the two possible understandings of moralism in noting that while '[i]t is difficult to find a clear definition of moralization' in Mudde's and Rovira Kaltwasser's (2017) work, 'it would be fair to assume that it refers to the supposed tendency of populism to simplify political antagonism [. . .]'. Either way, in packaging dualism together with moralism in a strong sense when it comes to populism -and barring a more systematic understanding of what role different forms of dualism play in politics -ideational definitions run the risk of throwing out the baby with the bathwater and disqualifying all forms of adversarial us-versus-them politics as pathological aberrations from neatly structured politics-as-usual. There are clearly many different guises that dualism and polarization can take, from civil war to agonistic pluralism (see also Stavrakakis, 2018); Mouffe's (2000) theory of the latter even makes the case that a tamed, 'agonistic' form of dualism is not only compatible with but indeed a precondition for pluralism in a democracy. The conceptual reduction of populism onto a pure, virtuous 'us' versus evil, conspiring 'them' short-circuits this diversity and, at worst, blurs the lines between Manichaean conspiracist thinking and the formal dimension of dualism, which is hardly unique to populism nor even a particularly abnormal feature of politics as such.
The problem thus becomes magnified at the conceptual level insofar as at least part of the 'moralism' that is ascribed to populism arguably points to a dimension inherent to all politics. One need not embrace a Schmittian conception of the political (with a war of annihilation as its ontological horizon) to recognize that politics in a democracy presupposes the possibility of social division: this has been a common theme in democratic theories ranging from Lefort's (1986) understanding of democracy as the form of society that recognizes its own permanently divided character to Mouffe's (2000) theory of agonism (which she develops in contradistinction to a Schmittian understanding of antagonism). If this is the case, politics as such is hardly conceivable without moments of a dualistic opposition of us versus them, which presumably also entails some kind of positive (or 'moralistic' in a weak sense) affirmation of the 'us' in negative demarcation against the 'them'. The question then becomes how this opposition is specifically articulated, but to assume that there is a standard mode of politics outside of moralism and dualism altogether would be problematic; Mouffe (2005a) has criticized as 'post-politics' the notion that politics can exist outside of the political (i.e. beyond the register of us vs. them distinctions) and be essentially reduced to rational consensus-based administration. There is thus a danger of not only reproducing the assumptions of a hollowed-out understanding of what can or cannot be considered 'normal' democratic politics -once again pointing to the hegemonystabilizing effects of the intellectual anti-populism criticized by Stavrakakis -but also, on a conceptual level, of lacking a systematic baseline understanding of politics that allows for some kind of distinction between moralistic and non-moralistic politics. Canovan's (1999Canovan's ( , 2002 theory of populism, for instance, arguably avoids this pitfall by providing a general account of politics whereby all politics in a democracy has a 'pragmatic' and a 'redemptive' dimension -and populism arises as a product of the unbridgeable gap between the two, taking up the name of the 'people' as an unredeemed (an ultimately unredeemable) subject of democratic representation.
What has been referred to in other contexts as the 'moralism trap' (Stegemann, 2018) can thus be said to consist in a certain temptation to pathologize populism on the basis of characteristics without which politics as such is hardly conceivable. There is a moralizing thrust to every discourse that purports to stand above something that it itself cannot escape from; a prime example can be seen in the 'post-political' discourses criticized by Mouffe (2005b), which profess commitment to a conflict-free, rational consensus-based form of politics but themselves resort to a moralized dualism in positioning themselves as 'the good democrats' against 'the evil extreme right'. (This is not, of course, to say that those who want to defend democracy against the far right are disqualified from doing so, but that the assumption that a politics of 'us' vs. 'them' is the exclusive domain of the far right is a hardly tenable, and indeed self-defeating, gesture.) The paradox of anti-populist discourses is that they often claim to oppose a 'politics of division', but they themselves resort to division (and construct a dangerous Other in the form of 'populism') in doing so. This kind of moralism that insists that 'x is always the others' is, to be sure, far from unique to discourses about populism; the same can be said for problematic labels such as 'identity politics' -especially when it is contrasted to 'class' or 'interest'-based politics, as if only certain kinds of identities are politically constructed and others are somehow naturally given (Laclau, 2000) -or 'politics of emotion' as a foil to the assumed normality of rational, dispassionate politics that does not actually exist (Eklundh, 2020). If this is the case, the critique of moralism must maintain a critical selfreflexivity -as enjoined in a similar vein by Stavrakakis (2017a) -in accepting that some form of moralism, affect, dualism, identification with a collective identity, the construction of an Other, and the like are part and parcel of all politics (including one's own) and not simply problems unique to the Other.

Are post-foundational discursive approaches (crypto-) 'normative'?
If ideational approaches that ascribe a moralistic quality to populism run the risk of overemphasizing populism's particularity, Laclau's understanding of populism as the political logic par excellence faces the reverse danger of conflating populism with the political as such -a point that has been widely criticized across the ideational literature (Arditi, 2010;Müller, 2014;Rovira Kaltwasser, 2012;Stanley, 2008). This critique has often been accompanied by the specific (and often unelaborated) claim that Laclau's theory is essentially normative, especially from those seeking to demarcate it from an ideational approach to populism Mudde, 2017;Peruzzotti, 2019;Rovira Kaltwasser, 2012). To those well acquainted with Laclau's work, the claim that his theories are 'normative' might come across as a puzzling one; Laclau has indeed been criticized -not least by his own students -for an excessive 'formalism' in his conceptualization of populism in particular (Arditi, 2010;Borriello and Jäger, 2020;Lluis, 2020;Stavrakakis, 2004). Yet is there nonetheless a crypto-normative strain in Laclau's work -an unacknowledged normativity lurking behind all the formalism?
There is a paradoxical double movement in Laclau's theorization of populism as well: on the one hand, populism is understood as a logic inherent to all politics to some extent or another and, therefore, as a highly variable phenomenon that can take on inflections ranging from the radical left to the far right (as seen in his examples ranging from Titoism to 'ethnopopulism'); on the other hand, there is a tendency in Laclau's work to (over-)emphasize the emancipatory effects of populism as a politics of the underdog. Zicman de Barros (2019: 15) has noted this paradox and argued that Laclau exhibits a certain 'bias towards the democratic potentialities of populism' -a bias that, as the author goes on to show, runs all the way down to Laclau's use of psychoanalytic concepts: Laclau equates 'the people' as an empty signifier in populism with the logic of sublimation, which, however, can be contrasted with the logic of fantasy characteristic of most right-wing populist discourses (see Stavrakakis, 1999: 131-134, for a discussion of this distinction in the context of the 'left Lacanian' literature). The paradox here is that 'the people' is understood as a formal category and, at the same time, conceptualized in terms of a specifically post-fantasmatic understanding of politics. Such tendencies have facilitated the notion widespread in the ideational literature that for Laclau, '[p]opulism incarnates the normative ideal of a radical democratic project' (Rovira Kaltwasser, 2012: 191; emphasis in original) -even though there are clear tensions between populism and radical democracy in his work, as the subsequent reception has emphasized (Howarth, 2015;Kim, 2019;Nonhoff, 2019;Thomassen, 2016) and Laclau (2005b) himself has suggested. Clearly, any normative tendencies in Laclau's work are not explicit, but rather located, if at all, in some of the implicit assumptions within his complex theorization. Mudde (2017: 39) summarizes the differences between Laclau's perspective and ideational ones in a narrower sense as follows: Laclau's approach is essentially a highly abstract, normative, universal theory in which 'the people' has no specific content. In contrast, most of those who adhere to the ideational approach define populism in a specific manner, in which the key opposition is moral [. . .].
If Mudde simultaneously attributes a high degree of abstraction and a normative thrust to Laclau's work, it is worth asking whether there are indeed traces of (perhaps unintended) normativity within the aporias of Laclau's formalistic approach to populism -the 'antinomies of formalism' (Stavrakakis, 2004) -and what implications this has for the theory.
To begin with, Laclau (2005a: 74) conceptualizes populism in terms of three interrelated definitional components: '(1) the formation of an internal antagonistic frontier separating the "people" from power'; '(2) an equivalential articulation of demands making the emergence of the "people" possible'; and (3) 'the unification of these various demands' around 'the people' as an empty signifier. Laclau starts out from the category of demand as 'the elementary unit of politics' (Marchart, 2018: 111): the most basic relation in politics is established by a demand addressed to some locus of power that is thus called upon to accept or reject it. Implicitly visible here is the Lacanian notion of the subject of lack: every subject is constituted in terms of a dependence on the symbolic order (Lacan's 'big Other') and has no choice but to articulate its lack in the form of demands, beginning with the mother-child relation. The key move that Laclau makes here is with the assumption that every demand is directed to a locus of power by virtue of the latter's presumed capacity to decide on the demand; what follows from this is that once the demands are no longer articulated in isolation, or following the logic of difference, but rather in common antagonistic demarcation against the locus of power deemed unable or unwilling to fulfil them, that is, following the logic of equivalence, any equivalential articulation of demands will be directed against 'power' and thus follow a populist (and arguably emancipatory) logic of generating an antagonistic frontier of people versus power. What is overlooked here, however, is the possibility that the addressees of demands are interpellated not (or at least not primarily) as instances of power but rather as cultural Others or indeed as subjects supposed to obey what they are told to do from above -as seen in examples such as 'Act normal or leave' (Mark Rutte addressing immigrants in an open letter in 2017) or 'If you come to Hungary, you must respect our culture' (Fidesz government poster addressing would-be refugees in 2015). As soon as such a distinction is made between different types of addressees of demands -and, following from this, different types of antagonism, such as the 'horizontal' one of national versus non-national in nationalism as opposed to a 'vertical' one pitting underdog against power in populism (De Cleen and Stavrakakis, 2017) -it becomes clear that populism cannot simply be identical with the political as such, but constitutes, 'quite simply, a way of constituting the political' among others (Laclau, 2005a: xi).
There is thus a slippage in Laclau's (2005a: 154) work between populism as 'synonymous' with the political, on the one hand, and populism as one of multiple possible ways of constituting the latter. The charge that Laclau's theory is a normative one centres particularly on this first tendency to equate populism and the political; while Laclau makes noises in both directions, a semi-formal reading following De Cleen's and Stavrakakis' 'architectonics' approach makes it clear that there are other possible ways of deploying the logic of equivalence and drawing political frontiers besides that of people versus power. From the standpoint of this second interpretation sensitive to the limitations of a purely formalistic approach, Laclau's (2005a: 60) 'royal road' argument can be taken to mean that populism simply takes on a metaphorical character for the political insofar as the ingredients of the latter -logic of equivalence, antagonistic frontier, empty signifieremerge in exemplary fashion in populism, taking on the name of 'the people' as such against constituted forms of power that are supposed to represent it in a democracy. Populism thus emerges as a specific form of the political that is constitutive for democracy (as argued in a similar vein by Canovan, 1999Canovan, , 2002 -insofar as democracy, with Lefort, is understood as the form of society that recognizes the permanent gap between 'the people' as its symbolic legitimizing instance and the forms of power constituted in its name. The possibility of politics in a democracy presupposes the possibility of claiming that those in power do not, in fact, represent 'the people'; it is in this (weaker) sense that Laclau's (2005c: 48) argument that 'the end of populism coincides with the end of politics' might be understood.
If these conceptual distinctions are not made explicit, however, there is a certain risk of crypto-normativity -one that is compounded by some of Laclau's (2006) and Mouffe's (2018) practical-political interventions that directly advocate a populist strategy of the left while suggesting that the only alternative is the perpetuation of neoliberal post-politics. For Laclau and Mouffe, however, this line of argument is a strategic rather than a normative one -as was the case already in their joint book Hegemony and Socialist Strategy, in which they framed radical-democratic politics as the strategically most viable option for the left (rather than in terms of normative desirability) in the context of Thatcherism and the New Social Movements (Laclau and Mouffe, 1985). Mouffe (2018: 81) similarly argues in her recent work that populism is 'particularly suited' as a 'political logic adapted to the [current] conjuncture' and that a strategy of revitalizing the left versus right frontier as she had previously argued 'was no longer the most adequate path under the current conditions' (Errejón and Mouffe, 2015: 112). In other words, Mouffe's argument is ultimately that of a strategic -and very much contingent -choice of one form of politics over others. In this context, the charge of (crypto-)normativity might serve as a warning of the potential pitfalls involved in simply deriving practical-political imperatives from a theory of the political: those calling for a left-populist strategy from a Laclauian perspective must avoid reproducing the post-political mantra of 'there is no alternative' in assuming that populism, by virtue of its conceptual status as a 'royal road', is the only path to re-politicizing post-democratic constellations; on the contrary, the question why populism, of all the possible forms of politics, is the most suitable one in a given context requires explicit justification.

Conclusion
The ideational and the post-foundational discursive approaches to populism share a basic constructionist thrust insofar as they understand 'the people' in populism as a contingent construction -and, following from this, populism as a highly variable phenomenon depending on how 'the people' and 'the elite' are constructed. Approaches following the ideational turn in this broad sense have produced significant advances in the study of populism vis-à-vis earlier objectivist ones that ascribed to populist phenomena a determinate socio-structural character. Where the ideational (understood in a narrower sense) and post-foundational approaches crucially diverge, however, is in the location that they assign to populism within the wider topography of politics: as a less-than-fully-fledged ('thin-centered') yet specifically moralistic form of politics or, on the contrary, as the quintessential political logic ('the royal road') that points to a dimension inherent to all politics. Both the paradigmatic approaches of Mudde (2004) and Laclau (2005a) entail a paradoxical double movement in this regard: while both of them ascribe to populism a high degree of indeterminacy in how 'the people' and 'the elite' are ultimately constructed, Mudde simultaneously reins in this indeterminacy by conceptualizing populism as specifically 'moralistic rather than programmatic', while Laclau foregrounds the emancipatory and post-fantasmatic uses of purportedly formal categories such as 'the people'. It is here that scholars working from these two perspectives have increasingly sought to stake out the boundaries between them: on the one hand, with a critique of moralism and its implicit reproduction by ideational approaches; on the other hand, with the attribution of normativity to Laclau's post-foundational conceptualization in spite of, or perhaps even in conjunction with, all its formalism.
The preceding sections dug deeper into these claims as well as the populism definitions in question. The most basic conclusion that can be drawn is that both prominent 'schools' of populism research are far from immune from the pitfalls of moralism and normativity, respectively -both of which take on a 'crypto'-character due to their implicit, unintended nature and, in the case of ideational approaches, the distinct danger of reproducing a certain pathologizing moralism through the very move of ascribing a distinctly moralistic quality to populism. Yet both points of criticism should also be understood as an opportunity -namely, for both approaches to reflect on and further clarify their baseline understandings of politics: the question for ideational approaches is to what extent moralism and dualism are specific to populism or more generalizable properties of politics in a wider sense, while a semi-formal reading of Laclau's theory leads to a weaker but analytically more productive version of the 'royal road' argument by allowing for distinctions between different types of antagonism and manifestations of the political. Here, again, a basic divergence between ideational and post-foundational approaches emerges: the latter extends the constructionist underpinnings of the ideational turn onto an understanding of all political identities as ultimately contingent and based on the construction of political frontiers, for which the 'people' in populism can be understood as the proverbial tip of the iceberg that points to constitutive features of the political condition. The twofold challenge here for populism research after the ideational turn more generally is to avoid over-generalizing populism's metaphorical character for the political to the point of blurring the distinction between populism and other forms of politics -while, at the other extreme, refraining from over-determining (and pathologizing) populism's particularity to the point that it becomes unclear what baseline understanding of politics we are ultimately left with.