Is Populism a Political Strategy? A Critique of an Enduring Approach

The political-strategic approach is one of the most employed frameworks within the methodologically heterogeneous subfield of populism studies. In the last two decades, it has contributed to the analysis of populism both in Latin America and the United States and, more recently, in Western and Eastern Europe. That being said, a close inspection of its axioms and its conceptualization of the phenomenon shows that it is built on ill-conceived premises. This article intends to be a comprehensive critique of the approach that can contribute to the methodological progress of the field. It criticizes the three main dysfunctions of the approach: selective rationalism, leader-centrism, and normative bias.

by researchers: the ideational approach (whereby populism is understood as an ideology, although as a particular one), the sociocultural approach (whereby populism is understood as a particular form of political relationship between leaders and followers), and the political-strategic approach (Hawkins and Rovira Kaltwasser et al., 2017). 1 This article intends to formulate a comprehensive critique of the latter in order to show its conceptual unfitness for the analysis of populism.
The strategic-political approach has Kurt Weyland as its main proponent. He has a long record as an analyst of Latin American populism, although more recently he has also examined forms of populism in Europe and the United States. According to him, "populism is best defined as a political strategy through which a personalistic leader seeks or exercises government power based on direct, unmediated, uninstitutionalized support from large numbers of mostly unorganized followers" (Weyland, 2001). He developed this new definition partly as a reaction against the conceptual flaws of the state of the literature on populism during the late twentieth century, which was mainly based on socioeconomic structuralism and historicism and was not able to integrate new phenomenon such as the existence of personalistic leaders who conducted market-oriented reforms in terms of definitional extension.
That being said, there are other proponents of similar versions of this view, even if they do not necessarily identify themselves as followers of Weyland's works. Alan Ware (2002: 104) defines populism as "a political strategy deployed by a wide range of politicians," while Hans-Georg Betz (2002) presents it as "primarily a political strategy, whose political rhetoric is the evocation of latent grievances and the appeal to emotions provoked by them." Robert R. Barr (2009) for his part considers populism to be "a mass movement led by an outsider or maverick seeking to gain or maintain power by using anti-establishment appeals and plebiscitarian linkages." Although the article will focus mainly on Weyland's analytical framework, it will also pay attention to these "thin" versions of the political-strategic approach, as they are considered to reproduce some of its basic axioms.
What do these approaches have in common? Their underlying assumptions are mainly three. First, populist leaders are portrayed as power-seeking politicians who act rationally in order to maximize their profit, measured in popular support, which leads them to engage in strategies of deideologization and other tactics as a means to gain political prominence. Second, it is assumed that populism can be conceptualized and analyzed by focusing on the behavior of its leaders, which implies that both the demand-side and institutional and historico-cultural dynamics tend to be overlooked. And third, there is a clear normative content inasmuch as both populist politicians and their policies are systematically (and aprioristically) condemned, even when the research objective of the paper or the book is in principle unrelated to normative considerations.
This article seeks to criticize those starting premises. In order to do so, it is organized thematically in three parts. The first examines how some of the (implicitly or explicitly outlined) axioms of the strategic-political approach are in fact shared with primitive versions of the rational choice theory as it was applied to political science. Needless to say, such theory has not and should not be methodologically rejected from the very bottom, but its application to certain subfields (such as populism studies) carries analytical issues. The second section analyzes to what degree the leader-centric approach defended by Weyland and others is suitable for a conceptualization of a rich and complex phenomenon such as populism, which is generally analyzed at a movement or party level. Finally, the third section points to the normative biases that run through the strategic-political approach and how they can become a research limitation. While rejecting the idea that a theoretical framework for the analysis of populism needs to be completely value-free, it points to how in this case the normative damps the analytical.

Selective Rationalism
What is rationality? In both political science and microeconomics, the word clearly has many different uses (Hodgson, 2012). While some understand it merely as "acting for reasons" and others as "doing the best in particular circumstances," this is not the traditional connotation that the word has for political scientists who employ it. Instead and since the arrival of the rational choice theory to the political science field in the 1950s, acting rationally means acting in a way that one's "objectives, whether selfish or unselfish, should be pursued by means that are efficient and effective for achieving these objectives" (Olson, 1965: 65).
Authors such as Mancur Olson, Kenneth Arrow, and Anthony Downs paved the way for the mainstreamization of this connotation among political scientists. The latter, for example, considered that political leaders "seek office solely in order to enjoy the income, prestige, and power that go with running the governing apparatus" and that "political parties in a democracy formulate policy strictly as a means of gaining votes" (Downs, 1957). According to this view, a political leader would want to maximize utility, which means that "when confronted with an array of options, he picks the one he believes best serves his objectives" (Green and Shapiro, 1994: 14).
Some versions of the rational choice theory applied to politics see rational actions as those based on calculated self-interest (Klosko et al., 1987). I argue that this latter notion of political rational action is embedded in the strategic-political approach to populism, even if this is not explicitly noted by Weyland and other authors. Such notion has particular and fruitful applications for both economic and political analyses, but the conceptualization of populism is not one of them. I am not considering the strategic-political approach as being part of the broad family of rational choice political theories, but rather arguing that some of its claims inform the theories of its proponents and thus lead to the same kind of analytical issues as the former.
This view of populist leaders as power-seeking rational actors can easily be identified on the works of the major proponent of the political-strategic approach. Weyland defines populists as "individual leaders [who] seek or exercise government power based on support from large numbers of followers" (Weyland, 2001), and as politicians who "seek to muster backing from a diverse range of people with diffuse, often emotional, appeals" (Weyland, 2003a) and who "lack firm commitment to ideologies and principles and concentrate on the quest for personal power" (Weyland, 2013). In short, populist leaders are politicians who use a series of tactics in order to concentrate power and boost their personal leadership (Weyland, 1999), a view defended by Weyland during more than two decades now.
The logical consequence of this approach is that political action coming from populist politicians is seen as a means to gain power. Therefore, "elections, plebiscites, mass demonstrations, and most recently opinion polls" are presented as "instruments with which populist leaders mobilize and demonstrate their distinctive power capability" and populist leaders as rational actors who "whip up support from largely unorganized masses to win office" and "constantly invoke their broad mass support to boost their own influence and overpower their opponents' institutional bastions" (Weyland, 2001). For example, populists who engage in neoliberal reforms are not driven by ideological motives (a notion that could have fostered an in-depth analysis of the subcategory of "neoliberal populism") but by their personal ambitions: "[neoliberals] ally with neopopulist leaders, who seek to boost their own autonomy and power [. . .] neopopulist leaders can use market reform to give their own power hunger a rational, modern justification" (Weyland, 2003b). Donald Trump for his part "has [irrationally] used exclusionary ethno-nationalist appeals to solidify backing from some of his most loyal white supporters, but at the cost of lower overall approval ratings" (Weyland and Madrid, 2019: 5).
This view is shared by Barr (2009), who qualifies mass demonstrations as "standard tools of populist leaders" and notes how "in contexts of high public discontent with politics, some political actors find it advantageous to cultivate this attitude for political purposes through an anti-establishment discourse." Later he referred to how "in the right circumstances, populists can foster group identity, cultivate anger, and thus influence voter choice" (Barr, 2017: 7). Betz (2002) for his part states that "populist rhetoric is designed to tap feelings of ressentiment and exploit them politically," while Ware (2002) pointed to how in the United States "populism has become one component of the rhetoric deployed by politicians in the major parties, and has thereby become a strategy that may be deployed against their opponents." The problem is that the idea that politicians (populist or not) are rational actors who only seek political power by selecting ad hoc strategies is not only a very robust assumption but also one that has long been contested. In 1983, Donald Wittman showed how a model that included policy preferences in the equation was more realistic and empirically consistent. This is because contrary to what Anthony Downs and other proponents of his version of the rational choice theory thought, "in the real world candidates are both office oriented and issue oriented" (Wittman, 1983). Weyland (2017) tries to circumvent this reality by drawing a differentiation between populists (who want nothing but personal power) and ideocrats (who want power as a means to implement ideologically based policies), which leads him to define politicians like Hugo Chávez or Alberto Fujimori as unidentifiably in terms of ideological content, a dichotomy that will be criticized later on.
A few years later, Kaare Strom empirically showed how the distinction between office-seeking, vote-seeking, and policy-seeking parties (or political leaders, for the matter) was problematic since "pure vote seekers, office seekers, or policy seekers are unlikely to exist" (Strom, 1990). Instead, he suggested analyzing political behavior from "an understanding of the interrelations and trade-offs between different objectives" (Strom, 1990). Needless to say, this included every kind of party and politician, regardless of their political tactics or their allegiance. Furthermore, he argued that party leaders were generally constrained by a concrete institutional, historical, and party competition context. Weyland tries to avoid this by insisting on how populists are personalistic leaders who operate aside political organizations and try to debilitate them when they exist. This is dubious even for Latin American ''presidentialist'' populists. As examples, let us think about the importance of the PSUV (Partido Socialista Unido de Venezuela) and the MAS (Movimiento al Socialismo) for Hugo Chávez and Evo Morales, respectively, but also about how such claim is simply impossible to apply to European populist parties such as Rassemblement National, Syriza, or Lega, which are robust (and sometimes relatively old) political parties that could change their leadership without necessarily being undermined as key political actors.
Moreover, there is a fundamental epistemological issue: how can we know when a politician is a power-seeking actor or someone who wants to gain power in order to implement certain ideology-based policies? How can we know if a politician's policies and discourse are or not a mere way to attain power? As Cas Mudde and Cristobal Rovira Kaltwasser noted while briefly assessing the explanatory power of the politicalstrategic approach, "whether populism is a strategic tool is largely an empirical question, which is often almost impossible to answer conclusively without getting into the populist's head" (Mudde and Rovira Kaltwasser, 2012: 9). Since Weyland does not provide any clue on how to differentiate between rational power-seeking politicians and genuine ideologically driven candidates (except the fact that he considers them to be so), he and those who follow a variant of his approach can freely select who is in and who falls out of the category by abductive reasoning. But, in the last instance, it is not clear how we can be sure of whether Martin Luther King, Charles de Gaulle, George Washington, Benito Mussolini, or Angela Merkel were moved by power hunger or sincere, ideological commitment.
The problem is therefore not that this view cannot be corroborated by the behavior of populist politicians but the opposite: any behavior conducted by populist politicians can fit the approach, and so the explanatory power is difficult to appreciate. Instead of an axiom that frames empirical analysis, there is one that can integrate any new set of data. The idea that populist politicians consistently act in such way does not have truth-conditions (factual conditions under which the idea would be true) and thus cannot have truthvalue (it cannot be true or false). In other words, it is invulnerable to any empirical claim, which is not a good attribute for a political science theory. Since the idea that populist leaders only wish political power and that their public actions are subservient to that goal is an unfalsifiable hypothesis, it should not be taken for granted, let alone be included as a key element of a conceptual definition.
Let's now focus on the dichotomy between "ideocrats" and populists. According to Weyland, "populist movements are notorious for not espousing a clear, systematic, and comprehensive worldview" to which he opposes "organized, truly empowered mass movements such as European Social Democracy" or fascism (Weyland, 2017: 62). Since there is not much information about these "ideocrats" (who are defined apophatically), we shall understand that any non-populist is to be considered part of the category. Neither it is clear whether ideology acts as a distortion (which would lead non-populist leaders to misidentify the strategies for attaining power) or as a source of honesty (which would make non-populist leaders more "sincere" without that implying a tactical deficit). Be that as it may, this binary opposition between ideology and populism leads to some questionable assumptions in Weyland (2017: 54): The driving force behind populism is political, not ideological. Prototypical populist movements are practically impossible to define in ideological terms. Argentine Peronism for decades spanned the full arch from fascist right to radical left. And who could define the Bolivarianism of Hugo Chávez, who took advice from reactionary Norberto Ceresole as well as Marxist Heinz Dieterich?
Both conclusions are dubious, and the dichotomy between "the political" and "the ideological" when referring to driving forces and political leaders (or movements) seems misleading. Peronism is surely difficult to define, but experts tend to identify it either as a form of third positionism (McLynn, 1984) or as a political identity (not an ideology or a form of populism) that has varied through the decades depending on the specific national context (Fair, 2016). Bolivarianism, on the contrary, is certainly a populist movement, but there is consensus on describing it as a twenty-first-century form of leftism or socialism (Buxton, 2019;Panizza, 2005;Wilpert, 2007). In fact, Weyland himself has denounced "Chávez's efforts to trace a 'Bolivarian' alternative to neoliberalism," along with his "ideological purges and his populist politicization" (Weyland, 2009).
But this bimodal conceptualization goes beyond Latin America and is used to differentiate between both left-wing and right-wing "ideocrats" and populists of any kind and from any continent. For example, when referring to the relatively recent rise of the European radical right, Weyland, (2017: 68) states the following: Ideational definitions attribute too much importance to extreme rightists' rhetorical appeals to "the people." By contrast, the strategic approach emphasizes that ideological radicalism confines leaders to the political margins and inherently limits their ability to win support from a broad cross-section of the people. Therefore, ideocratic leadership differs from the opportunistic personalism that is typical of populism.
The idea clearly starts from a non-sequitur fallacy: being radical does not imply being confined to the political margins in terms of electoral success. Apart from that, once again we find echoes from Downs (1957), who also considered that the rational strategy for a party or leader who wants to maximize votes and power was to satisfy the preferences of the median voter and reject radicalism. This has been empirically contested by many authors (Green and Shapiro, 1994), but it suffices to observe the behavior of contemporary populist politicians to reject it. Marine Le Pen, for example, probably understands that her party and her program (and herself) find serious obstacles in second-round presidential elections due to their radicalism, yet that has not led to ideological transformation or the abandonment of the "political margins," even though she is a populist according to Weyland (2017) and her discourse is different from her father's.
Both Podemos and Vox are other examples of populist parties that despite a very selfconscious use of populist-based discourse have never renounced to propose reforms that are seen by the electorate as too extreme (Asuar Gallego, 2019). Hugo Chávez himself, who seems to be Weyland's paradigm of a populist politician, maintained his reformist (and ideologically driven) ambitions and announced a constitutional referendum in 2007 even though his project was rejected by many citizens including his own supporters, which made him lost the vote (Romero, 2007). This is once again a consequence of using a framework that by overemphasizing the power-seeking element downplays the empirically proven importance of policy-seeking.
Both the ideational approach and the Laclauian approach prevent this impasse by defining populism as a political category (be that a thin-centered ideology or a way of constructing the political) that is always and necessarily accompanied by additional ideological elements (Laclau, 2005;Mudde, 2004). In such way, populism never exists in the abstract but is always intertwined with concrete ideas, even though appealing to a heterogeneous mass of voters leads to some surplus of transversality, in contrast with non-populist movements and parties. The ideational approach is particularly valuable in this sense, since, as showed by the work of Mattia Zulianello (2020), it allows for a nuanced and empirically supported classification of populist parties in function of the ideology to which they are "attached." Weyland, on the contrary, is led to an analytical deadlock due to the axiom that populist politicians are rational profit-maximizers that understand the political arena and act in consequence to their preferences, which are aprioristically fixed and purely personal, although impossible to empirically know. As we have seen, this leads him to state that some populist leaders who are conventionally identified as belonging to a concrete ideological tradition are actually "outside" the political spectrum or barely connected to it, which could not explain why many populist leaders (Chávez, Morales, Salvini, Iglesias, Farage, etc.) engage in political alliances with "ideocrats" (although this could be easily accommodated as yet another personalistic strategy). 2 This view also implies a tendency to build a reality/appearance dichotomy whereby it is posited that instead of analyzing discourses or ideological traits, the observer should "go beyond" the surface. Ware (2002: 108), for example, states that "populist rhetoric may often conceal the politics of self-interest on the part of fairly prosperous social groups" and later that "the slogan of 'government versus people' provides a useful camouflage for what is actually a conflict between different sections of society" (Ware, 2002: 109). Weyland (2017: 56), on the contrary, considers the political-strategic approach to focus "not on what populists say, but on what they actually do, especially how they pursue and sustain political power." How can we empirically identify this "underlying reality"? And isn't it problematic to draw an exclusionary distinction between discourse and political practice, especially since many experts on populism engage in different forms of discourse analysis which reject such distinction axiomatically? As with the idea of personalistic rational power-maximization, we can only conjecture about what we cannot empirically confirm or disprove while we assist to a potential new source of post hoc rationalizations.
Finally, there is the additional problem that the notion of rational action is employed only partially. Few can doubt that the analogy between economic and political behavior leads to many issues, but additionally to that the political-strategic approach commits the mistake of making a partial, atheoretical, and seemingly unconscious use of it. This is evident when one notices that Weyland and others attribute rationalism and profit maximization only to populist leaders, leaving other political actors (including voters) 3 aside from such quality, which contrasts with rational choice theorists, who start from the premise that "all agents act always to maximize their well-being as they understand it, based on their beliefs, preferences, and strategic opportunities" (Ferejohn, 1991: 281). Indeed, models starting from the basis that political actors act rationally in order to pursue powermaximization would never introduce an asymmetric version of the approach, since it would imply abandoning some of the key anthropological premises that inform its intellectual foundations.
This raises the question of what could be the consequences of such an arbitrary theoretical decision. To begin with, Weyland cannot explain differences of success between populist and non-populist leaders and parties. If populist leaders are free from ideological biases and can focus on developing strategies to mobilize masses of (nonrational) voters, avoiding extremism, and formulating ambiguous, unmediated discourses, how come they are often displaced by their competitors in both institutionalized and inchoate party systems? Moreover, where does this asymmetric view leave structural dynamics? Proponents of the theory of the rational actor applied to the study of political behavior tend to emphasize that their approach only "works" under certain circumstances (e.g. there is only a certain number of competitors and issues and voters have limited information), but this is nowhere to be found in the works that are being scrutinized in this article. It is difficult not to see that this depiction of populist politicians is arbitrarily formulated, since it does not make part of any comprehensive theory (although it retains partial axioms from others), as this section has tried to show.

Leader-Centrism
As abovementioned, Weyland developed his framework partly as a critique of other approaches (which were based on socioeconomic-centered premises), as it is often the case in theory-building. This, on one hand, undoubtedly contributes to the debate and keeps a certain field dialectical, but, on the other hand, it has the risk to produce new conceptualizations that are biased by the fact that they were articulated partly in a negative way, by distinguishing themselves from others. Such is the case of Weyland (2001: 11), who presented his definition of populism as a theoretical solution to a conceptual impasse: [. . .] A political definition of populism is therefore preferable. It conceptualizes populism as a specific way of competing for and exercising political power. It situates populism in the sphere of domination, not distribution. Populism first and foremost shapes patterns of political rule, not the allocation of socioeconomic benefits or losses. This political redefinition captures best the basic goal of populist leaders, to win and exercise power, while using economic and social policy as an instrument for this purpose. Thus, this reconceptualization is most attuned to the opportunism of populist leaders and their weak commitment to substantive policies, ideas, and ideologies.
I argue that this will to formulate a purely political notion of populism might be at the origin of one of the key dysfunctional elements of the political-strategic approach: leadercentrism. This element is probably also related to the fact that Weyland's original theorizations are closely associated with Latin American politics (permeated by a presidentialist political culture), as he himself stated: "[my] redefinition of populism in political-organizational terms is useful in analyzing contemporary Latin American politics" (Weyland, 2001). This becomes clearer when we notice that his concept of populism is built in opposition to South-American-related categories such as caudillismo or clientelism. Be that as it may, what matters is to show why such element is problematic in analytical terms, inasmuch as its origin is (logically) not sufficient to invalidate it.
First of all, this top-down view underestimates the importance of both the demand-side of politics and the organizational and identity-building levels. Weyland (2001) states that in contrast with political parties, "the relationship between populist leaders and their mass constituency is uninstitutionalized and fluid," which would make leader-centrism a more pertinent way of analyzing populism. This goes against the consistent tendency among scholars who study populism to take a comprehensive approach whereby the unit of analysis is either the political party or movement (Canovan, 2006;Laclau, 2005;Mudde, 2007;Ostiguy, 2017) or the attitudes of its voters (Norris and Inglehart, 2019;Rooduijn and Pauwels, 2011;Wuttke et al., 2020), or a combination of all of these. In fact, it also goes against the general tendency among political scientists (including the proponents of the rational choice) to focus on parties or movements when analyzing macro-level phenomena such as populism, nationalism, communism, democratic socialism, fascism, and whatnot. Overall, Weyland's conceptual exceptionalism toward populism does not seem justified and unnecessarily reduces the scope of populism studies. 4 The consequences of overlooking the demand-side of populism are evident. Of course, both Weyland and other proponents of similar versions of the political-strategic approach make reference to voters, but these tend to play a very secondary, passive role. Weyland (2001) refers to how "populist aspirants whip up support from largely unorganized masses to win office" and, more recently, to how "populist leadership [is] sustained by amorphous mass support that is inherently unsteady, fickle, and unpredictable" (Weyland, 2017: 54). It follows that aspects as fundamental as the content of popular demands, issues of electoral sociology (such as why men are more likely to vote for some populist parties), or electoral swing (where do populist voters come from in electoral terms?) are overlooked.
Actually, those who have been voting for Kaczynski, Le Pen, Salvini, Iglesias, Morales, or Tsipras in the last years are no less ideologically and organizationally consistent than other voters. For example, Trump's vote in 2016 was not very different from Romney's in 2012 (Carnes and Lupu, 2020), and the voters of Jean-Marie Le Pen and his daughter do not differ that much (even though the former is not considered a populist by Stockemer and Amengay, 2015). Here again, we see the limitations of focusing on the populist leader, especially when he or she builds support from an institutionalized party (sometimes one that preexists them). It would have sufficed to observe with which concrete parties populists compete for votes in order to clearly see that there is no "amorphous" mass but rather citizens with certain ideological dispositions (Rama and Santana, 2020). Voters of populist parties are in fact not that "abnormal": they come from ideologically similar parties and sometimes have them as second choices (Akkerman et al., 2013), in the same way that non-populist voters can have a populist party as their second preferred option when it is congruent with their ideology (Chiaramonte et al., 2020;Crépon et al., 2015).
Moreover, downplaying the demand-side can also have the consequence of downplaying the offer side. What do populists have to offer? What popular identities do they articulate? And are these created "from the top" or is it possible that they are created "from below," as the result of unresolved political inputs, for example? How can we explain that their voters, as mentioned, are not necessarily more ideologically heterogeneous than others? Since, according to Weyland, most populists are personalistic leaders who are not interested in policies, strong organizations, and ideologies, they can only articulate a precarious electoral base. The question is, how can the seemingly contradictory facts that populists lack ideological consistency and that they have a strong mobilizing power be theoretically conciliated? The answer is charisma. According to Weyland (2001: 14), "To compensate for the fragility of their mass support, populist leaders seek to create a particularly intense connection to their followers. Such intensity requires charisma." The problem is that charisma has been discarded as a necessary element of populism (Mudde and Rovira Kaltwasser, 2014). Indeed, populist leaders can be more charismatic and pivotal than their counterparts, but that does not imply that such element is the "secret formula" for their success. In addition to that, Weyland does not provide a clear definition of charisma but limits himself to pointing toward it as a personal element that "can solidify the quasi-direct relationship of personalistic leader and supporters that constitutes the core of populism" (Weyland, 2017: 56). Inasmuch as he focuses on populist leaders and downplays the attitudes of voters or party members, he misses the key element of charismatic leadership as already identified by Max Weber (1978): its recognition by those "subjected" to it. Considering the latter and starting from a concise definition of the phenomenon, Duncan McDonnell (2015) has shown that the relationship between populist leaders and their voters is not necessarily related to charisma. Barr (2009) himself notes that "certainly the most successful populists have had tremendous personal charisma, but there have been notable non-charismatic populist leaders as well" and that charisma "is a useful resource for any politician, helping him gain support no matter the type of appeal." Populist leaders such as Alberto Fujimori and Alexis Tsipras do not seem to be very charismatic, and non-populist leaders such as Barack Obama and Ronald Reagan certainly do. Be that as it may, the main problem with attaching so much importance to charisma is that we are yet again dealing with a category that cannot be easily measured and thus lacks conceptual suitability. 5 Furthermore, since the definition proposed by Weyland is grounded in a South American context (both the case studies and the theoretical debate revolve around the continent), it could be surmised that there could be a case of what Giovanni Sartori (1970) called "the traveling problem," that is, the idea that conceptualizations that are based on particular case studies and are used for comparative politics might be unfitted for the analysis of new cases. Introducing an element based on particular regional case studies as a characteristic of a generic, "universal" definition is certainly problematic. This seems to be the case with the leader-centric element, which might be related to a context in which presidentialist institutional arrangements favor this type of strong leaderships in different types of parties (Aslanidis, 2017;Rhodes-Purdy and Madrid, 2019). This does not necessarily apply to Europe, where populist leaders are part of a political culture and an institutional system in which strong and personalistic leaders are rare (Mudde and Rovira Kaltwasser, 2013). This is also why the centrality of personal charisma in populism has been contested in the light of recent political developments in Europe (Mudde, 2007: 260-262).
In addition, a leader-centric approach to populism has the risk of being applied to other organizations that also share a similar kind of leadership, paving the way for potential false positives. As Kirk A. Hawkins notes, "religious or labor-based parties and millenarian movements also have charismatic leaders and/or low levels of institutionalization early in their organizational life cycle and may seek to change the political system, yet we do not necessarily consider them populist" (Hawkins, 2010: 168). Weyland's conceptual stretching can perfectly lead to identify as populists many leaders from different regions and eras, including some of the most "ideocratics" that one could think of, such as nationalist, communist, anti-colonial, religious fundamentalist, or far-right aspirants. The proof of this is that he considers populist leader figures such as Vladimir Putin (Weyland, 1999) and Robert Fico (Weyland and Madrid, 2019), which are not classified as such by the proponents of the other main approaches.
Finally, leader-centrism leads Weyland to serious problems of operationalization. In order to quantitatively calculate degrees of populism and noticing that the distinction between ideocratism and opportunism is problematic, he employs the fuzzy-set theory, an approach to mathematics that has been applied to social sciences as well (Ragin, 2000). By introducing conceptual gradations, the fuzzy-set theory allows to study complex and hybrid social realities. The starting premise is the idea that political phenomena are different from the subject matters of other sciences: "A cat is a cat and a dog a dog-but in politics, contextual complexity and human ingenuity soften or even erase boundaries; there can be partial and mixed types that fixed conceptual categories do not fully capture" (Weyland, 2017: 65). This is indeed a very interesting and pertinent point that should be noted by other experts on populism, and its inclusion probably has to do with Weyland's awareness of the limitations of his own approach. Unfortunately, he uses it to focus solely on the attitudes of political leaders, thereby losing the opportunity to scrutinize party literature, party platforms, identity-building dynamics, or public or voter attitudes. In addition to that, this "degreeism" does not lead to any radial categorization in which the importance of ideology is integrated in the analysis of populism, probably due to the fact that such move would compromise Weyland's theoretical apparatus, inasmuch as it could imply acknowledging the existence of "ideocratic populists," which could mean accepting some of the premises of the ideational approach he rejects.
Using a scale from 0 to 1, Weyland (2017) classifies several political leaders in order to see how populist they are. The methodology is not outlined, although in each case he schematically deals with the reasons whereby a political leader might score 0 (non-part of the category), 0.33 (more outside than inside the category), 0.66 (more inside than outside the category), or 1 (fully representing the category). Some categorizations are not surprising even if one can question the taxonomy created by Weyland: Evo Morales (a left-wing populist) scores 0.33, Alberto Fujimori (a right-wing populist) scores 1, and Michelle Bachelet (a center-left politician) ranks at 0. Yet the explanations are once again too schematic. For example, "full-scale populists" are simply defined as those "enjoyed tremendous personal autonomy, founded their own flimsy electoral vehicles, and were never constrained by them" (Weyland, 2017: 66), a description that could fit many authoritarian leaders that create anocratic regimes.
At the other end of the "spectrum," the non-populists are qualified as such because "their organizational and institutional insertion has reliably constrained personal ambitions, for instance precluding constitutional reforms to facilitate their reelection" (Weyland, 2017: 66). Therefore, there seems to be a shift toward the institutions, and so suddenly being a populist is more a matter of institutional constriction than personalistic will, a new indication that the approach is irregular and fragmentary since it tries to overcome the obstacles that it keeps coming across. The problem is, as this article intends to make explicit, that the methodological problems of the political-strategic approach are located at an axiomatic level, and thus they will not be circumvented by ad hoc corrective measures.
A new symptom of the issues afflicting this approach emerges when we move to Europe, especially when Jean-Marie Le Pen and his daughter Marine Le Pen (both leaders of the French far-right party Front National, today known as Rassemblement National) are compared. Whereas Marine Le Pen has a score of 0.66 (which, again, seems pertinent even though the methodology is far from being carefully built), her father scores 0. How is this possible? Even if it is true that Marine Le Pen changed some aspects of the party's discourse (Eltchaninoff, 2018), her leadership is not very different from her father's, who is generally considered a populist by the literature (Mouffe, 2005;Mudde, 2007). This is due to the alleged fact that "for decades [Jean-Marie] put ideological radicalism ahead of popularity and mass appeal" (Weyland, 2017: 67), which is strange since their respective vote and their ideology did not vary much until recently (Stockemer, 2017).
Moreover, Jörg Haider (a right-wing populist) scores 0.66, "due to his transition from ideocratic dogmatism to opportunistic personalism" and the fact that he "pursued vote maximization, and used flashy appeals and telegenic appearances for this purpose" (Weyland, 2017: 68), even if Jean-Marie Le Pen, his French ally, did the same and had similar electoral results at the time. Berlusconi, on the contrary, scores 1, whereas Pablo Iglesias scores 0.66. This is surprising because while the former engaged to some extent in populist discourses, the latter very self-consciously articulated a populist party drawing directly from Laclau and other pro-populism thinkers (Valdivielso, 2017). This is a great example of a researcher ignoring the Sartorian maxim that concept formation stands prior to quantification.
The heuristic advantages of leader-centrism are clear: clarity (by extremely narrowing down the object of study), simplicity (by focusing on a unitary actor and avoiding complex organizational or ideational matters), and operationalizability (by reducing the variables to personal attitudes, goals, and characteristics). The problem is that these are advantages from which the analyst, not the analysis, can profit. When applied to a complex and variable phenomenon such as populism, in which it is essential to take into account ideas, organization, participation, demands, and party competition, it can be fatal in terms of conceptualization.

Normative Bias
It might be idealistic to pretend that it is possible to conduct a completely value-free research in political science (Morrice, 1996;Taylor, 1985). There are certainly some methods and approaches that are less susceptible of being normatively biased, but it is clear that unless one understands political science as mere data collection (which is not), we have to accept certain degree of "value contamination." In fact, this article itself is partly driven by the normative goal of contributing to the methodological progress of a field which object of analysis has gained prominence both in Europe and the United States and whose existence raises concerns for many people.
That being said, there are many ways in which this normative influence can take place. It is one thing to select your object of study or think on the potential impact of your research out of normative concerns and a very different one to let those concerns imbue your methodological approach and your remarks about it, thereby hampering your own analysis. In other words, normative concerns are legitimate and likely to be found both a priori and a posteriori, but they should be avoided at an analytical level. One can perfectly imagine an analysis of populism that is conducted out of concerns about how it can represent a threat against liberal democracy and whose author intends to use to make a certain impact in the public debate but that is methodologically pristine.
Just as the analysis of fascism, democracy, economic inequality, or communism, the study of populism is an activity especially prone to suffer from normative biases. This is probably due to the fact that it focuses on a phenomenon whose nature is highly contested and that exerts an important influence in contemporary politics. Already in 1981, Margaret Canovan pointed to how "interpretations of populism have been deeply influenced by the fears of intellectuals" (Canovan, 1981: 11). About two decades later, she still lamented the fact that "recent students of politics have tended to treat populist movements as pathological symptoms of some social disease" (Canovan, 2006: 1). Today, the state of art looks certainly better, but normative bias still lurks researchers who study the phenomenon and are inevitably part of a media and civil sphere in which the concept tends to be used in an illocutionary, non-descriptive way. As Aslanidis (2015) notes, "more often than not, scholars of populism tend to write as if they are loyal opponents or supporters of a political cause, rather than objective observers." In fact, the current populism studies field can be roughly divided into two camps: one in which populism is seen as a potential threat against liberal democracy (Kaltwasser, 2012;Müller, 2016;Urbinati, 1998) and another in which populism is seen as either a possible revitalization of democracy or a legitimate political strategy (Eklundh and Knott, 2020;Laclau, 2005;Mouffe, 2019). That being said, both the former and the latter often display a nuanced view whereby populism is seen as an ambiguous phenomenon, since they tend to acknowledge that it can be either a threat or a corrective for contemporary democracies.
The proponents of the political-strategic approach are undoubtedly part of the first, since for them populism is based on a personalistic desire to attain power which leads to the undermining of competing adversaries and institutional democratic mechanisms. But they do not share the mentioned nuance that one can find in many other authors: according to Weyland (2013), "populism, whether of the left or the right, is a threat to democracy." 6 The nuance is instead introduced at an institutional level (proving once again that there is no lack of analytical capacity, but rather problems of conceptual misformation), for they carefully analyze which party systems and civil societies are more likely to resist the populist threat, concluding that Western Europe and the United States are better equipped than Latin America and Eastern Europe (Weyland and Madrid, 2019).
Weyland's tone when referring to populist leader, especially when they are Latin American, should make anyone aware of possible normative biases. He refers to them as "opportunistic" (Weyland, 2001), specialists in "strategies of confrontation and polarization" (Weyland, 2019) who use "diffuse, emotional appeals" (Weyland, 2003a), and gain popularity "by attacking dangerous enemies and mobilizing the followers for heroic missions" (Weyland, 2017: 50) and by exploiting "nationalist skepticism about globalization, fierce repudiation of the established political class, and a questioning of pluralist, representative democracy" (Weyland, 2009). Ware (2002: 104), on the contrary, defines populism as "abnormal politics," whereas Georges- Betz (2002: 198) sees it as a strategy "whose political rhetoric is the evocation of latent grievances and the appeal to emotions provoked by them" and Barr (2009) considers populists to be clientelistic by nature.
Hugo Chávez, for example, is contemptuously defined by Weyland (2009) as "a radical populist" who won in 1998 "because realistic assessments would have yielded frustrating, painful conclusions, [so] many Venezuelans seem to have felt a psychological need to be strikingly hopeful" (Weyland, 2003a). We also learn that his death "triggered outbreaks of collective hysteria" (Weyland, 2017: 58). It seems that his selective methodological individualism (although a leader-centric one, as we saw) can also fit the purpose of "personalizing" his invectives. Unless we look into Richard Hofstadter's (1955) The Age of Reform, we probably will not find any equivalent of these attitudes among scholars who specialize in populism. Moreover, even if Weyland (2013) sees both leftwing and right-wing populism (suddenly accepted as legitimate categories) as threats to democracy, he considers the former to be at least potentially more dangerous, although the reasons of why that is the case are not outlined: The political demise of Fujimori-who is now serving jail time for corruption and human-rights abuses-brought back full democracy, with ample public debate and free and fair elections. Thus, right-wing populism did not ruin democracy in Argentina, Brazil, or Colombia, and in Peru democracy's destruction and temporary replacement were followed by a quick resurrection. By contrast, left-wing populism has a more negative balance sheet.
It is difficult not to appreciate the ideological charge of Weyland and other thinkers who consider populism to be a personalistic strategy designed to seize power and undermine democratic mechanisms and whose analyses often reveal more about their ideological views than about populism. The conceptualization itself is in fact normatively biased, since it fixes the nature of populists as self-interested, power-maximizing cynical actors, thereby prompting the analyst to automatically "pick sides." This normative-driven definition prevents us from rationally tackling an already slippery conceptual question, inasmuch as it hinders attempts to differentiate between types of populism, explain the relationship between populists and non-populists, and analyze their concrete relationship with democracy and liberalism.
As mentioned, this problem goes well beyond Weyland and other proponents of the political-strategic approach. The negative connotation of the word, though pertinent for partisan politics or normative views, inevitably obstructs clarity and problem-driven conceptualization, and it prevents scholars to engage in what George Mosse (2000: 9) called "methodological empathy." This is the idea that researchers who focus on "stigmatized" movements or ideologies need to avoid their normative bias and "enter" the world of its protagonists. It has been used successfully among experts in fascism, for example, in order to explore the reasons and ideas behind the movement as to better understand their actions and projects (Griffin, 2007: 351). 7

Conclusion
This article has shown that the political-strategic approach to populism, developed by Kurt Weyland and partly replicated by other authors, has serious methodological and conceptual flaws. Needless to say, this does not imply downplaying its merits. At the beginning of the twenty-first century, a time when many analysts of Latin American populism considered it to merely be a series of economic policies that were related to national particularisms, Weyland propounded a redefinition whereby populism was to be understood as a political category which could engage with different kinds of economic programs. This has now become a conventional axiom for scholars who specialize in populism studies, but it was far from being part of the theoretical doxa at the time.
But this shift has now proven to have conceptual problems, mainly due to the blossoming of populism (both right-wing and left-wing) in different parts of the world and the resulting expansion of the object of study. Weyland's framework is clearly grounded in Latin American political dynamics, it arbitrarily concedes a type of self-serving rationality to populist leaders, it overlooks many aspects of the demand-side of populism, and it starts from a series of problematic normative premises that are evident for any scholar who examines it critically. Moreover, it is a source of both false positives (via conceptual stretching) and false negatives (via conceptual shrinking). Today, when populism can be found in any continent with many different ideological "shapes" and scholars have formulated ideational and discourse analysis frameworks with more explanatory power, this approach seems outdated.
As political scientists, our normative feelings about Donald Trump or Hugo Chávez cannot prevent us from seeing that they do mobilize and articulate a popular identity (and sometimes also a strong political party) and create what Laclau (1990: 64) calls "an imaginary horizon," that is, a set of ideological collective expectations with respect to the political future. They also propose concrete policies and can be located in the political spectrum (in fact, contemporary populists seem to be easier to identify ideologically than some of their "centrist" competitors), and so they are fundamentally no different from other political leaders who apart from gaining power seek to implement certain policies and normalize certain ideologically informed discourses. This does not imply that there are not populist politicians that are inclined to act as rational power-seeking rather than policy-seeking actors, but they do not necessarily do so in a greater degree than nonpopulist politicians.

Declaration of Conflicting Interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.