Illiberal, anti-liberal or post-liberal democracy? Conceptualizing the relationship between populism and political liberalism

ABSTRACT While most authors have attempted to conceptualize populism by exploring its relation with representative democracy, this note is concerned with a different, yet equally pressing issue: How is populism related to the other building block of contemporary democracies – political liberalism? Most commonly, populism has, in academic as political discourses, been described as ‘illiberal’. But the precise meaning of this ‘illiberal’ character is elusive, and often overlaps with other concepts such as ‘anti-liberal’ or even ‘post-liberal’. This conceptual haze is not only detrimental to political theorists, but to comparative political scientists as well. Therefore, the aim of this paper is to clarify these concepts, in order to cast light on the relationship between populism and political liberalism. It provides a review and critical discussion of the concepts of ‘illiberalism’, ‘anti-liberalism’ and ‘post-liberalism’; further, it offers to organize them in a coherent, and empirically productive, manner. In particular, it argues that ‘illiberalism’ should be treated with much caution; instead, it defines more workable concepts of ‘anti-liberalism’ and ‘post-liberalism’. So doing, it suggests how these concepts could be used in combination to fruitfully account for different dimensions of the study of populism, namely research on populist discourses, ideas and practices.


Introduction
While a definition of the 'chameleonic' populist phenomenon has been notoriously difficult to agree upon (Norris and Inglehart 2018, 65), most authors have attempted to apprehend it by exploring its relation with representative democracy (see, for very different perspectives, Moffitt 2020, 24-31;Mouffe 2018;Mudde and Rovira Kaltwasser 2013). Yet, these efforts have somewhat obscured an equally pressing issue: How is populism related to the other building block of contemporary democraciespolitical liberalism 1 ? The aim of this note is to cast light on the conceptualizations of this relationship.

Populism and (Il)liberalism
The literature has produced different accounts of the relationship of populism and political liberalism. Recently, Pappas (2019) and Moffitt's (2020) books, the former from the vantage point of comparative politics, the latter from that of political theory, have provided especially rich syntheses of populism research. Most importantly, both offer rigorous, and yet contrasting, analyses of the issue of '(il)liberalism'. As such, they can be used as convenient pathways into the debate.
Pappas highlighted the issue of populism and liberalism early on (Pappas 2012), and recently offered a systematic approach to the subject (Pappas 2019). In his view, the situation is clear: Populism is 'illiberal', and 'illiberalism' denotes a deep-seated opposition between populism and liberalism. This opposition is based on a Rawlsian understanding of liberalism (Pappas 2019, 42-43), which focuses on the institutions of post-war political liberalism rather than on a transhistorical definition. In this empirically oriented view, liberalism can be broken down into three dimensions: 'The inevitable acceptance of a plurality of divisions in society, the quest for political moderation and overlapping consensus, and respect for the rule of law and the rights of minorities' (Pappas 2019, 42, my emphasis). Conversely, populist politics involve a 'singular cleavage, adversarial politics, and majoritarianism' (Pappas 2019, 33). To that extent, populism is constructed as the 'negative pole' of political liberalismbut not necessarily of democracy. Populism in fact reflects a disjunction between democracy and liberalism. It can thus be defined as 'democratic illiberalism' (Pappas 2019, 33).
Others, however, have identified a less straightforward relationship of populism and liberalism. Moffit, in his analysis, 'considers liberalism in its ideological sense' (Moffitt 2020, 80). This ideational perspective allows him to introduce distinctions, not only between types of populism (left and right), but most importantly between aspects of liberalism: One centred on diversity (Lockean), one rather emphasizing autonomy (Kantian) and the last one self-expression (Millian). This distinction allows to get a more fine-grained grasp of the relationship between liberalism and populism, especially in its right-wing version. Indeed, Moffit astutely observes, populists in many cases claim to borrow from liberalism rather than straightforwardly opposing it. Typically, they profess to be in line with expressive (i.e. Millian) liberalism, i.e. 'a mode of 'hard' liberalism that places the ability to 'say whatever you like' and to express your 'true self' above all' (Moffitt 2020, 88). This for instance generates raucous advocacy of freedom of speech against the elites and the 'mainstream media'. Thus, drawing a clear-cut line between the two does not account for the empirical fact that populist discourses operate a selective reading of liberalism. Instead, Moffit points out that populism 'weaponizes' (elements of) liberalism (Moffitt 2020, 82). This can mostly be observed, he argues, in right-wing populist regimes; but a similar point was made about the practices of 'discriminatory legalism' in left-wing Latin American populism, too (Weyland 2013). All in all, the relationship of populism and liberalism is summarized in the somewhat paradoxical phrase of 'liberal illiberalism' (Moffitt 2020, 88).
Both Pappas and Moffit thus agree that populists are 'illiberal'. But they diverge when it comes to defining the meaning of 'illiberalism': While for Pappas it points at the clear-cut opposition between populism and liberalism, Moffit suggests that the line is more blurred. Interestingly, though, neither question the relevance of the concept of 'illiberalism' itself: They seem to take it for granted as a useful analytical category. A short historical detour, however, raises serious questions about the analytical added-value of the concept of 'illiberalism'.
The discourse of illiberalism 'Illiberalism' has, today, become standard in populism scholarship. But it has also been integrated into the populist discourse, that is, in the way populists present themselves in public speeches: It is now one of those concepts circulating between political analysts and actors, in a process of 'double hermeneutics' once theorized by A. Giddens. In particular, politically, it was introduced in a 2014 speech by Viktor Orban, in which he detailed his ideological agenda. There, he spelled out the reasons, and implications, of the alleged failure of liberalism. Following the economic and financial crisis, he claimed that 'the new state that we are constructing in Hungary is an illiberal state, a non-liberal state' (Orban 2014). But he was careful to signal that 'just because a state is not liberal, it can still be a democracy': His target thus unambiguously was the liberal element of liberal democracy. These declarations had an enduring impact. 'Illiberalism' has become a fairly common way of presenting not only Orban's regime, but other would-be populists as well (see Kauth and King 2020).
And yet, illiberalism is an elusive concept. Rather unsurprisingly, its trajectory paralleled that of liberalism. Its uses (in the English language) throughout the 18th and 19th centuries denoted a moral or intellectual limitation. For instance, the Earl of Chersterfield, in a treatise on polite education, warned against 'illiberal arts', such as the practice of music (in particular, the horrendous practice of pipes), which was below the dignity of a well-mannered person (Stanhope Earl of Chesterfield P 1802). In the 1919 issue of the Oxford Concise Dictionary, it was still defined as 'not befitting a free man; […] vulgar, sordid; narrow-minded; stingy' (Fowler and Fowler 1919). But throughout the twentieth century, the term lost much of its moral connotation, to slowly enter the political and legal vocabulary. It is to a 1997 article by the political scientist F. Zakaria that the recent good fortune of the term is to be traced (Zakaria 1997). There, it was used in a critical way: Reflecting on new democracies or transitional regimes, he cautioned against the weakening of constitutional democracy, that is, against the disjunction of democracy (based on free elections) and liberalism (rule of law, separation of powers, basic rights). While democracy might survive, in the long run, liberalism might not. That, he warned, might well be the major threat against the Western model. For instance, he observed that Yeltsin's Russia had increasingly been characterized by a 'super-presidency', which, he '[could] only hope, will not be abused by its successor', lest it become a truly illiberal regime (Zakaria 1997, 34).
In retrospect, it is easy to be struck by the adequacy between the picture he drew and Orban's project to separate democracy and liberalism. Of course, while to Zakaria, this was seen as a major threat, Orban was able to use the new contextthe crisis of economic and of political liberalismto revert the connotation of the term: What had once been presented as a problem now appeared as an alternative to the failure of liberal democracy. However, as Zakaria readily acknowledged, what was to be included in this 'illiberal' or 'non-liberal' category remained fairly vagueranging from near tyrannies to occasional infringements against the rule of law. For there are many ways of not being liberal. Indeed, the history of political liberalism is replete with 'illiberal' features of even the most acclaimed liberal theorists (Ryan 2014). Even among deeply rooted liberal countries, it is not rare to observe policies, or governmental actions, going against basic tenets of liberalism. In practice, constitutional courts regularly rule on just such cases involving infringements against freedom of the press, or basic rights. Yet, it remains dubious that this would suffice to label a regime as illiberal in any meaningful and operationalizable sense. Hence, if anything, this brief genealogy presents us with a case of conceptual stretching (Sartori 1970).
To turn 'illiberalism' into an analytically useful concept, a safer path has been suggested by the specialist of liberal thought, M. Freeden. According to him, 'the departure of what might, generally speaking, be a liberal polity from some generally acknowledged liberal ideas and practicesthat is to say, a liberalism with illiberal features' (Freeden 2015, 34) should be distinguished from more systematic attempts at undermining liberalism as such. While illiberalism covers a broad range of non-liberal, or non-purely liberal phenomena, systematic aggression against liberalism is more adequately described as 'anti-liberal', that is as reflecting 'an assault on liberal principles and practices by non-liberal systems'. This suggests drawing a line between illiberalism and anti-liberalism based on the critical criterion of the magnitude of attacks against liberalism. Crucial to this distinction, not only are anti-liberals occasionally disagreeing with liberal principles and practices; they actually target them.
It nevertheless remains that, in common academic and political parlance, it is 'illiberalism' that has gained currency. This empirical observation cannot be simply dismissed because 'illiberalism' is inadequately defined. Its very success also needs to be accounted for. To that aim, it is important to clearly distinguish between the term as an element of populist political discourses and as an analytical concept. While in the latter case, it is too vague to be useful, in the former, 'illiberalism' can fruitfully be regarded as a category, whose strategical use by political actors needs to be understood. Drawing on a rapidly expanding literature on the discursive practices of populism (Poblete 2015;Zienkowski and Breeze 2019), this might involve researching its occurrences in public speeches, parliamentary debates, social media or press coverage. For instance, the historian D. Gosewinkel noted that, immediately after the Second World War, strongly antiliberal movements 'toned down how they referred to themselves and their rhetoric by adopting the terms 'illiberal' or 'non-liberal'' (Gosewinkel 2015, 6). Likewise, using terms such as illiberal or non-liberal, both to be found in Orban's speech, is a convenient rhetorical way of affirming one's departure from liberalism, while avoiding the more negative undertones associated with systematically opposing it. Therefore, self-descriptions as 'illiberal' should be analysed as part of a discursive strategy whose stake for these regimes is to produce their public identity; yet, if conceptual precision is of any value, it should not be inadvertently and uncritically used to define their ideas and practices.

The principles of anti-liberalism
Now, applying Freeden's distinction, the principles and goals advocated by populists can much more adequately be labelled anti-liberal. As indicated, the concept of anti-liberalism emphasizes a deep rupture with liberal principles (Coman and Volintiru 2021). The one-off utterance of non-liberal discourses, or implementation of a non-liberal policy, is a nonsufficient condition for a regime to be described as anti-liberal. Instead, anti-liberalism denotes a systematic, that is, repeated (not occasional) and active (not incidental), contestation of political liberalism. Indeed, to put it with R. Koselleck, the concept of anti-liberalism is an 'asymmetric counter-concept' (Koselleck 2004), built to indicate a principled opposition to its counterpart. It is asymmetric, because it does not reflect an opposition that those labelled anti-liberals are likely to happily accept. As we just saw, a leader such as Orban precisely put his efforts into promoting a self-definition as 'illiberal', rather than as someone systematically opposed to liberalism. Hence, when using the concept of antiliberalism, the inquiry no longer bears on strategies of self-presentation, but moves on to investigate the ideas underlying these strategies and actions. In other words, in constructing the analytical category of anti-liberalism, we have moved from discourse analysis to political and social theory.
This more reflexive mode of construction of the concept lends itself well to empirical studies. It should be noted that anti-liberalism so definedthe systematic opposition to liberalismclosely corresponds to the account of populism as 'democratic illiberalism' given by Pappas. Indeed, in his own words, populism is, simply, 'the opposite of political liberalism' (Pappas 2019, 58) to the extent that populists systematically 'assaul[t] liberal democratic institutions' and conventions (Pappas 2019, 210). I thus propose that what Pappasand most othersterms 'illiberalism' should in fact be conceptualized more precisely as anti-liberalism, which can be operationalized negatively along the three dimensions of liberalism he distinguishes: It vehiculates ideas aiming at reducing diversity within society, at sustaining an adversarial politics, and at undermining the institutions designed to constrain power (see also Coman, Behr, and Beyer 2021, 4). A regime that systematically targets liberalism along these lines will qualify as anti-liberal.
This definition is in line with what other analysts have pointed out, especially (but not exclusively) in the case of right-wing populism. For, while anti-liberal ideological trends have been noted in the case of left-wing parties (Rone 2021;Dawson and Hanley 2019), right-wing parties and movements have provided their clearest illustrations. This suggests that the ideological anti-liberal aspect is more developed, or more prominent, among right-wing populistsas illustrated by the systematic violations of the rights of minorities they justify (Guinjoan 2022;Fitzi, Jürgen, and Turner 2018;Lacroix and Pranchère 2019), their attacks against typically liberal institutions such as courts (Halmai 2019) or the way populists construct the distinctions between social groups (Reid 2020). The various emphases of these attacks against one or the other dimension of liberalism will in turn allow to distinguish between different types of anti-liberalisma task that is beyond the scope of this paper.
It is worth noting that conducting the analysis in terms of anti-liberalism, rather than the vaguer notion of illiberalism, has the additional benefit of inscribing these regimes in a longer history of anti-liberal thought and practices (Behr 2021;Gosewinkel 2015). Far from appearing as radically new, ahistorical occurrences, they can be analysed in comparative historical perspective. However, insisting on the opposition in principle between populism and liberalism also leaves some issues unresolved. For, as systematically and purposefully as populists may attack liberalism, they do not always nor immediately reject the whole liberal apparatus. Rather to the contrary. As Pappas himself hints at, they 'colonize' liberal institutionsthat is, they take control of them, so as to empty them of their actual liberal meaning, rather than simply disbanding them (Pappas 2019, 210). This observation is central to Moffit's analysis of 'weaponizing' of liberalism by right-wing populist regimes. Likewise, it is also reminiscent of the 'illiberal constitutionalism' pointed at by Müller, and extensively analysed in the cases of right-wing populists in Europe (Canihac 2021; Castillo-Ortiz 2019; Müller 2016). But even more strikingly perhaps, it echoes analyses of Latin American populist movements. In this vein, for instance, Arditi warned against an absolute opposition of liberalism and left-wing populism, which can have a 'much more layered relation to the liberal tradition' and its institutions (Arditi 2008, 13).
Is it possible, then, to reconcile these contradictory insights on the relationship between liberalism and populism? The contradiction, I submit, disappears provided that we clearly distinguish between different objects of investigation: While one focuses on the aims and principles underlying populist discourses or policies, the other rather insists on their actual practices. To designate such practices, the concept of post-liberalism has opened a promising, if not fully explored, path. In the next part, I explain why this seems an appropriate concept, how it can be defined and how distinguishing between anti-liberal principles and post-liberal practices can prove empirically fruitful.

The practices of post-liberalism
The concept of post-liberalism, in the contemporary debate, can be traced back to a 1964 article by the Canadian political theorist C. B. MacPherson (1964). There, he pondered on the necessity of a 'post-liberal' theory of democracy. For, he recalled, liberalism, 'the theory of individual rights and limited government', had in the first place been distinctly 'anti-democratic' (Macpherson 1964, 488). In his view, the mid-1960s marked a concerning return to such a 'pre-democratic' liberal theory. Instead, a 'post-liberal' theory would be more democratic: It would largely moderate liberal individualism and ask 'whether meaningful liberty can much longer be had without a much greater measure of equality than we have hitherto thought liberty required' (Macpherson 1964, 498). Thus, the aim was not to give up on liberty and liberalism altogether, but to reconfigure it in relation to other principles, such as equality.
Following this path-breaking attempt, other political theorists gave further thoughts to the concept of post-liberalism, albeit in different directions. It has notably been taken up in the conservative tradition, to speculate on the regularly announcedor hoped forend of liberalism. For instance, John Gray, a self-identified 'liberal conservative' (Gray 1996, 9), concluded in a collection of essays to the death of liberalism. He nevertheless called on the preservation of some of its basic elements. In particular, he argued that 'what is living in liberalism is the historic inheritance […] of a civil society whose institutions protect liberty and permit civil peace' (Gray 1996, 284). Similarly, more recently, Adrian Pabst has developed a concept of post-liberalism trying to reconcile conservatism and social concerns, in the Christian-Aristotelian tradition (Milbank and Pabst 2016).
Another noteworthy attempt at constructing the concept has been provided, from a different political as much as scholarly perspective, by P. Schmitter. An empirically minded scholar, he too speculated, on the future of liberal democracy, and the possible disjunction of its two components. He argued that, given the complex interactions of processes at work in today's world, liberal democracies would eventually give way to a different model, that of post-liberal democracy. But, he noted, 'as is the case with all specters, that of post-liberal democracy suffers from a fuzzy, unstable and highly idiosyncratic configuration' (Schmitter 2018, 14). It is, in other words, not clear what exactly it will turn out to be. Nevertheless, he went on to identifying a series of criteria that would make post-liberal democracy desirable. Like MacPherson, a post-liberal democracy has to refrain from being 'contemptuous of what liberal democracy has accomplished and consciously seeks to build upon its strengths' (Schmitter 2018, 15). It would retain the importance of freedom as a crucial principle around which its institutions are organizedbut, instead of individual freedom, would encourage more collective freedom, in a somewhat republican fashion (Schmitter 2018, 19). In short, then, post-liberal democracy would protect the values and institutions of liberal democracy, while adjusting them to the new historical configuration.
All the uses of post-liberalism considered so far point towards a far-reaching redefinition of, rather than systematic opposition to, the liberal tradition; yet, they regard post-liberalism as a normative watchword, a more or less sketchy alternative political ideal to be achieved in the future, and thus disagree over what it would mean in practice. Nevertheless, another more recent strand of research has given an empirical twist to the concept. While 'until recently, writings on post-liberal democracy have focused on the "established liberal democracies" in Northern America and Western Europe and consisted mainly in normative sketches of what an "improved" democracy could look like', new research is 'interested in the experimentation with post-liberal politics that is "already happening" across Latin America', with left-wing populists attempts at transcending liberalism (Arditi 2008;Wolff 2013, 32). Jonas Wolff has provided a workable operationalization of post-liberalism in the Latin American context. It offers to break it down into five dimensions, encompassing various transformations of liberalism (Wolff 2013, 36-39): The electoral regime (e.g. creation of plebiscitarian modes of participation), political rights (e.g. extension of 'non-electoral politics'), civil rights (e.g. granting of collective rights), division of powers and accountability (e.g. strengthening control over non-majoritarian ones) and increasing citizens' 'effective power to govern' (e.g. control over multinational firms). These changes would pave the way for an increasing redefinition of liberalism that would, nonetheless, preserve and extend some of its basic institutions and practicesa redefinition somehow paralleled, in the economic area, by the concept of post-neoliberalism (Brand 2016;Hay and Benoît 2018).
This operationalized concept of post-liberalism works well with left-wing populist movements; yet, it might be too narrow to account for other contexts. For, originally concerned with conceptualizing Latin American populist movements, its very specificity precludes broader comparisonsin particular, with European right-wing populism: Although some practical convergences can be discovered (plebiscitarian tendencies, for instance), other transformations that take place in one case seem rather contrary to the policies implemented in the other (e.g. minority rights) (see Guinjoan 2022). It might of course be argued that this is precisely where the line between right and left-wing populism should be tracedi.e. between one defined by its attacks against liberal principles (anti-liberal), and another characterized by the transformations of its institutions (postliberal). Nevertheless, the line remains empirically blurry: As already observed in the case of constitutional courts, right-wing populist regimes, too, do 'weaponize' liberal institutions and rules. In other words, the observation that right-wing populists put more stress on challenging liberal principles, and that left-wing populists seem more prone on transforming liberal institutions is not an absolute difference, but rather, a matter of the preferred repertoire these movements use.
Thus, in order for the concept of post-liberalism to be empirically fruitful, it will be useful to operate with a slightly more general definition than proposed by Wolff. To that aim, it is illuminating to consider the similarly constructed concept of post-democracy. As C. Crouch has coined it, a post-democracy retains institutionalized democratic practices, but in an essentially formal way, meaning that they co-exist with non-democratic elements (Crouch 2004). Similarly, here, I propose to define post-liberalism as the coexistence of basic liberal practicessuch as the appointment of a constitutional courtand of distinctly anti-liberal goalssuch as evading institutionalized limits to power through the practice of plebiscites. Thus, in a nutshell, it allows to conceptualize a regime operating against the values of liberalism, but in a liberal framework. The specific added-value of the concept, once freed of its normative elements and of its too particularistic components, lies in the combination of apparently antithetical elements. Post-liberalism so defined provides a convenient and workable tool to study the slow transformation of liberalism into something else, which has not yet fully appeared.

Conclusion
This essay has offered to review several conceptualizations of the relationship between populism and political liberalism. Indeed, while it is generally accepted that there is a tension between liberalism and populism, the precise nature of their relationship is, I have argued, elusive. The central concept used to theorize this relationship, 'illiberalism', is in most current literature, simply taken for grantedhowever little agreement there is, in fact, on what it actually means, or on its links with other cognate concepts such as anti-liberalism or post-liberalism. I have contended that, given its political uses and its poor definition, it should be treated with caution. It would be more adequate to limit its uses to discourse analyses, that is, to regard it as a category strategically put forward by populist leaders in their discursive political battles.
In contrast, I have argued, the concept of anti-liberalism denotes more accurately the ideological goals they pursue, while the concept of post-liberalism opens a path to explore their actual practices. Far from contending that one perspective is inherently superior to the other, though, the point in drawing these distinctions is that combining these concepts allows for a comprehensive empirical and comparative study of populism as a discursive, ideological and practical phenomenon. It also allows to differentiate between various empirical manifestations of populist movements, by suggesting that right-wing populists are more likely to be characterized by their use of anti-liberal principles, while left-wing populists are more inclined to have recourse to post-liberal practices. However, the exact combinations of these different dimensions (discursive, ideological and practical) are a matter of empirical investigationwhich is exactly what this note would like to foster.
In conclusion, it should be noted that, according to some theorists, the very tension that was taken as our starting point is questionable. Put succinctly, there could be something as a 'liberal populism', since, it is argued, there are 'populist political projects that can be defended in conformity with the normative demands of political justification promoted by liberal democrats' (Wolkenstein 2019, 15). While the argument made is theoretically appealing, focusing only on political justifications provides a rather narrow take on liberalism and its challengers. It leaves aside, as the author himself acknowledges, the issue of the discrepancy between justifications and practices, which is precisely addressed here in terms of post-liberalism. Ultimately, the concept of post-liberalism does, at least, hint at an empirically observable process of transformation of political liberalism. Note 1. This paper will be limited to the discussion of political liberalism, understood as encompassing both certain ideas and certain institutions. However, the related issue of economic (neo)liberalism, to which an abundant literature has been dedicated, cannot be discussed here.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The research that led to this paper was funded by the FRS-FNRS (Belgian National Fund for Scientific Research).