Towards a Renovation of the Laclauian Paradigm

This article provides a renovation of Ernesto Laclau’s theories of political competition and collective identity-building for their better application at an empirical level. It does so not as a critique but as a way to make the Laclauian approach more operational and rigorous from a methodological perspective. The key goal is to make available a form of discourse analysis that centers on the political sphere and has two key characteristics. The first is an emphasis on the role of what Antonio Gramsci called “organic ideologies” in political identity-building. The second is the differentiation between a descriptive and a subjective level of analysis. The article intends to initiate a debate on the prospects of the Laclauian approach both with its advocates and its detractors.


Introduction
It can be argued that one of the essential characteristics of a good theoretical framework is its capacity to endure changes. The opposite of that would be a series of rigid assumptions unable to adapt to different contexts, both theoretical and empirical, benefiting the most intransigent believers but damaging the purposes of rigorous research. The task of any social scientist at a methodological level is therefore to use the elements that have value at a descriptive and explanatory level and reject those that have been overridden by either new theoretical developments or the transformations of social reality itself.
The approach formulated by Ernesto Laclau, produced first and foremost in close collaboration with Chantal Mouffe but also with a series of members of the Essex School of Discourse Analysis (ESDA), can be considered to be going through something similar to what Thomas Kuhn described as a scientific crisis. 1 Not only it has received criticisms and amendments from its adversaries but also from its advocates, for whom on the other hand there does not seem to exist a conventional and shared precise methodological toolbox. It has been described as "extremely abstract" (Mudde & Rovira Kaltwasser, 2012, p. 7), too reliant on a few examples of populist movements (Melo, 2011;Carlés, 2013), too ahistorical (Borriello & Jäger, 2020), too leader-centric (DeÍpola, 2009), and indifferent to extra-discursive (particularly economic) elements (Townshend, 2003). Those who employ it have tried to revise it during the last decade, either by slightly modifying Laclau's definition of populism Mouffe, 2019) or by trying to amend some of the theoretical axioms by adding components from other traditions (Aslanidis, 2015;Kim, 2017).
In the face of such situation there are two alternatives: rejection or renovation. The first would imply accepting the fact that the approach has too many flaws, such as those pointed out by its critics, and thus should be abandoned. This would imply admitting the predominance of methodologies like Critical Discourse Analysis (CDA) or certain forms of Gramscian discourse analysis (if the Gramscian component is deemed worthy to be rescued) for the study of how the social and the discursive are intertwined. Despite the undeniable value of other forms of discourse analysis, it is here considered that Laclau's works can still provide us with pertinent ways of approaching how discourses operate at a political level. His framework therefore needs to be renovated instead of discarded.
This article starts from the premise that the Laclauian approach was originally formulated from a theoretical perspective and therefore can allow for different translations at a practical level. 2 It will respect the theoretical core of Laclau's approach, which can be summarized as the articulation of Gramscian and post-structuralist theory for the analysis of political competition and identity-building from a constructivist perspective, and focus on developments at a methodological level. The goal is thus not to create a new type of discourse analysis from scratch but rather a variation of the Laclauian framework, which will be referred to as "Laclauian Discourse Analysis" (LDA). The resulting interpretivist methodology is considered to be of interest not only for members of the ESDA but also for other discourse analysts focusing on the objects of study that the approach can cover.
The article is divided in three parts. The first seeks to clarify what exactly the Laclauian framework can offer from a political science perspective and how it can be more operational for researchers, as both issues have not been yet addressed by those who use it. 3 It locates Laclau's key ideas in the context of the study of contemporary political phenomena and ascribes them to particular ontological (post-foundationalism) and epistemological (interpretivism) traditions. Moreover, it emphasizes the importance of organic ideologies, a concept borrowed from Antonio Gramsci which can make the approach more intelligible and effective. The second part outlines a methodological program consisting of two moments that will have to be the starting point of any LDA work, both of qualitative nature: a first one which describes a particular discursive space and a second one that focuses on how political actors behave within such space. Finally, the third part focuses on data collection and inferences. The second and the third part seek to provide clear guidelines for those interested in following the program, which is considered to be replicable both in terms of time and space as long as enough empirical material (here, mostly texts) is available.

Discourse, Identity-Building, and Competition
What is it that Laclau intended to understand when he formulated his theoretical framework? The question is certainly too broad, as during his intellectual career he addressed not only political topics but also issues concerning psychoanalysis, language, and continental philosophy (Vergalito, 2017). In order to narrow things down, it can be argued that from a political perspective his works revolve around how collective identities (such as the ones emanating from social movements or political parties) are articulated and maintained. In the final analysis his goal is to understand, just like Donatella Della Porta and Mario Diani, "how ideas, individuals, events, and organizations are linked to each other in broader processes of collective action, with some continuity over time" and "how do certain social actors come to develop a sense of commonality and to identify with the same 'collective we' in order to attain certain goals" (Della Porta & Diani, 2006, p. 5). In other words, his theories can provide a methodology for the study of collective action from a constructivist perspective.
The epistemological aspects of the approach emphasize the importance of discourse, which is here understood as a force that constitutes social meanings and thus goes beyond purely linguistic elements. 4 The conceptualization comes to a large extent from the works of Michel Foucault, for whom discourse had to be regarded as "a violence that we do to things, or in any case a practice that we impose on them" (Foucault, 1981, p. 50). At the core of this notion, lies the Saussurian idea that "far from the world determining the order of our language, our language determines the order of the world" (Sturrock, 2003, p. 37). This idea sounds counter-intuitive because "Western tradition of thought about language has tended to privilege referential discourse and to imagine that discourse (at least ideally) reflects the pre-existing world" rather than one that sees the (social) world as at least partially shaped by discourse (Johnstone, 2002, p. 24). The unobservable nature of the effects of discourse (which will be addressed later on) undoubtedly also contributes to the counter-intuitive character of the approach.
Be that as it may, what matters is how social reality is shaped by intersubjective understandings of it. If discursive practices have such an important creative force that is because the theoretical framework is also post-foundationalist, that is, it starts from the premise that frameworks that assume an essence or an ultimate ground of society (be that metaphysical, sociological, or economic) are flawed. The primary implication of this is that political actors are able to constitute and transform their own terrain, as discourse is seen as capable of shaping realities (rather than passively reflecting them). 5 Overall, the idea is to formulate a dynamic and anti-essentialist view of politics (Errejón & Mouffe, 2015). As shall be explained, this does not mean that in the political arena everything is precarious or that existing structures can easily be overcome. As Oliver Marchart summarizes, post-foundationalism "does not assume the absence of any ground; what it assumes is the absence of an ultimate ground, since it is only on the basis of such absence that grounds, in the plural, are possible" (Marchart, 2007, p. 14). These assumptions can only lead to an interpretivist perspective that focuses on how discourse "constructs" the social contexts in which political actors participate. That being said and contrary to what some critics of Laclau have assumed this does not imply thinking that "material reality" does not exist, but rather and as other constructivists propose that "reality has no social role/causal power independent of the agent's/ group's/society's understanding of it" (Furlong & Marsh, 2010, p. 191). 6 After all, for Laclau every object is constituted as an object of discourse, which simply means that "all objects and actions are meaningful" (Howarth 2000, p. 101).
Here it is crucial to clarify an important element of the methodology. If this approach includes the description of a certain political context, does that imply that it will analyze extra-discursive elements? After all, political contexts are not only about discourses (even if we define them from a very broad perspective) but also include elements that are difficult to describe as "discursive." Despite what is generally thought, Laclau does not downplay the importance of non-discursive variables. It is in fact impossible, to the best of my knowledge, to find an example of a discourse analysis inspired by Laclau that does not refer to economic or historical aspects to explain continuities or discontinuities in particular discursive systems. 7 The question is thus what the role of those elements is exactly and how is that role compatible with a constructivist approach that focuses on discourses. The only way to include them is to see them as non-discursive elements which importance only emerges once they affect the discursive system by triggering social changes that inevitably or discretionally are incorporated by political actors into their discourse. An economic crisis, a natural disaster, or a pandemic, to take a few examples, will only be part of a LDA analysis as long as they have an impact on the discursive sphere (which does not necessarily happen automatically or inevitably, as mechanistic analysis would expect). To put it bluntly, non-discursive events will need to be integrated into the discursive arena in order to be relevant for our analysis. In this way, both discursive and non-discursive variables can be considered, as avoiding falling into economic determinism cannot lead to falling into discursivism, which is nothing but yet another form of reductionism.
The study of collective action requires, as mentioned, an analysis of how political actors come to exist and persist. Around that question lies the core of the Laclauian analysis: an encounter between the works of Antonio Gramsci and post-structuralism (or, rather, a post-structuralist reinterpretation of Gramsci). Both are selectively incorporated, inasmuch as elements of both Gramscianism and post-structuralism are excluded, deliberately or not.
From post-structuralism, Laclau and Mouffe took ideas of diachroneity and dynamism that were arguably absent among orthodox structuralist thinkers, who privileged synchroneity and staticity (even though in principle structures were seen as susceptible of changing), as well as a language-reality isomorphism approach (whereby linguistic categories could be used to explain non-linguistic social phenomena). For post-structuralists, structures cannot have any essential center (an ordering principle) but are rather in a continuous "play," that is, in a sort of decentered anarchy in which the components of the system were always at least potentially adrift. This dynamism explains the fact that whereas structuralists tend to use spatial and architectonic metaphors that refer to a sense of stability post-structuralists prefer metaphors which imply a sense of instability, generally liquid ones ("fluid," "floating," "slippage," "flowing," etc.). The Derridean notion of undecidability helped Laclau to underline the contingency that characterizes political identity-building, while Lacan's nodal point ("empty signifier") introduced the possibility of at least partially fixing those identities. 8 The key concept borrowed from Gramsci is that of hegemony. The word has had different meanings for different authors, but it always referred to a form of having or exerting power. For Gramsci, hegemony requires a hegemonic actor (for him, a class or a movement representing a class) who builds ideological consensus in order to integrate different sectors of the population, implying that power is not always related to coercion. 9 From a revolutionary perspective this is part of what he called "wars of position" (long-term ideological struggles over values and meanings, also referred to as "trench warfare"), in opposition to "wars of maneuver" (direct and frontal assaults against the enemy, as in the Bolshevik revolution). Laclau and Mouffe borrowed from him the idea that hegemonic actors "create" their movements (which are therefore not a mere reflection of a pre-existing social structure) and the social contingency that is implied by the idea that different groups can be articulated in different blocs and which can easily be connected to post-structuralism. 10 This framework, which underlining of contingency can be particularly helpful in post-industrial societies in which political allegiances are less solid, takes three elements as its bedrock to explain how collective identities emerge. The first is, of course, the existence of a hegemonic actor: a party, leader, or movement capable of formulating a certain discourse which appeals to a plurality of demands. The second is the use of empty signifiers, signifying elements in a broad sense that "can be attached to the most diverse social contents and [cannot have] a content of their own" (Laclau, 2005, p. 76). These are signifiers that renounce to their own particular meaning to represent a plurality of demands. Finally, social demands are the minimal unit of analysis for Laclau, who rejects focusing on already constituted groups or individuals in order to avoid both structuralist and methodological individualist simplifications (Laclau, 2005, p. IX). In this way, we have three levels of analysis: a hegemonic subject, the empty signifiers it formulates, and the demands those signifiers articulate through the creation of equivalential links.
Yet there is a fourth element which appears to be implicit in the Laclauian approach that goes beyond empty signifiers and social demands. In fact, when Laclau (2005) and Laclau and Mouffe (1985) point to examples of discursive formations it seems that there is something additional that encompasses both of those elements. For example, when analyzing Ronald Reagan's and Margaret Thatcher's hegemonic projects they refer to their "neo-conservative discourse" based on "conservative reaction and neoliberalism" (Laclau & Mouffe, 1985, p. 170). Laclau (2005) for his part refers to "Boulangism" when analyzing the discursive formation created by Georges Ernest Boulanger in the late 19th century. This tendency to point to encompassing ideologies can also be seen in empirical analyses conducted by advocates of the ESDA framework: Adamson (2000) explores the role of social-democratic and neoliberal ideas in the discursive construction of Romanian social democracy in the late 20th century while Stavrakakis and De Cleen (2017) refer to "radical right populism," just to name a couple of examples. Those "discourses" or "ideologies" are thus clearly important from a Laclauian perspective, but they are neither discursive formations, empty signifiers or social demands and therefore they escape the original Laclauian conceptual toolkit.
It is here argued that there is a missing yet implicit element that can be key at a methodological level: organic ideologies, that is, encompassing worldviews that shape both empty signifiers and social demands (which are both always molded by particular ideologies) and act as overarching "bonding agents" in collective identities. The concept was coined by Antonio Gramsci and is mentioned a few times in Hegemony and Socialist Strategy (1985) but not in On Populist Reason (2005). It is this element that differentiates LDA from other approaches inspired by Laclau's original framework when it comes to analyze how collective identities come to be. Organic ideologies will be defined as the "cement" that gives consistency to discursive formations and that shapes both empty signifiers and social demands. In fact, every empty signifier has inevitably a particular ideological charge: its "emptiness" has to do with its ability to connect with different demands rather than with its connotation at a normative level (in practice there is no context in which a leader or words such as "nation" or "sovereignty" have meaning outside ideological assumptions, even in cases of high discursive vagueness). Understanding which organic ideology is formulated by a particular hegemonic actor is a precondition for exploring the way its hegemonic project is articulated. (Figure 1) An example can help to clarify this part of the methodology. Let us imagine a communist party that intends to articulate a series of demands (going beyond those coming from the industrial working class, as it has historically been the case) and has as its ideological core ideas around class struggle, progress, anti-capitalism and collective liberation. A communist organic ideology will permeate empty signifiers such as "the nation" (which will be seen through class lenses as a popular subject), "freedom" (which will be seen as freedom from wage labor and socioeconomic misery), or "equality" (which will be seen as equality in the workplace as well as in the political sphere, for example) and also the demands that are articulated by them: a struggle for national autonomy will be presented as a struggle for proletarian liberation while demands around gender equality will be connected to class struggles, for example. It would be misleading to focus on the empty signifiers as the only or main cohesive force, as any analysis of them would have to include an explanation of their ideological traits (which implies, once again, the existence of a qualitatively different level of analysis).
To summarize the approach, at a methodological level and when it comes to analyze the constitution of a particular discursive formation it will be necessary to identify the hegemonic actor, then the overarching organic ideology and the key empty signifiers and finally the social demands that are articulated. To use another example, if we are analyzing Donald Trump's hegemonic project, we will first have to analyze its ideological foundations (what has been called "Trumpism": a blend of militant masculinity, nationalism and illiberal conservatism), then the main empty signifiers that he used (such as "America," "freedom," or "security") and finally the way in which demands (such as anti-immigration or tax cuts) are shaped by both the organic ideology and (more directly) the empty signifiers. The traces of an organic ideology can thus be seen both in the empty signifiers and the social demands, in the same way that for Juri Lotman "the fragment of a semiotic structure or text preserves the mechanisms for the reconstruction of the whole system" (Lotman, 1984, p. 10). The focus on organic ideologies and the formulation of this step-by-step process both enriches and simplifies Laclauian analyses at a methodological level. This part of the framework will be explored more in detail in the next section.
If one key aspect of collective action in Laclau's framework is the creation of collective identities, it is important not to leave out competition. Indeed, the resulting discursive formations not only have to face internal issues but also obstacles when it comes to compete with other hegemonic actors to attain their objectives. A close read of Laclau's framework allows to see that it starts from the tacit premise that political actors act strategically, in the sense that their behavior can be explained as means to achieve certain goals in the most effective way. That being said, this does not imply that they act rationally (in the way "rationality" has often been understood in the social sciences) but rather in a relational way with the discursive system they are part of, in a similar way than Colin (2002) puts it: "their ability and capacity to act strategically is mediated and filtered through perceptions (and indeed mis-perceptions) of the context they inhabit" (57). In this sense and following Pierre Bourdieu (1992), it can be said that they act reasonably (their behavior is adjusted to a particular context) rather than rationally (as they do not engage in pre-discursive forms of reason or interest).
Hegemonic actors have as their key goal to attain power positions and maintain and expand their social bases. Needless to say, here gaining power does not only imply reaching decision-making positions, but also (and arguably mainly) being able to "shape perceptions, cognitions and interests" (Lukes, 2021, p. 15). 11 In order to do so, and as it is clear by reading Laclau's main works, they have to compete with each other. They do so in two ways: first, by engaging in hegemonic struggles to represent particular demands. For example, if Euroscepticism becomes a central demand, it is likely that more than one hegemonic actor will try to integrate it in its discourse (we can imagine right-wing and left-wing parties presenting different kinds of criticisms against the European Union). Needless to say, some organic ideologies are more prone to connect with particular demands, as these can represent specific values even when we think about them in the abstract. For example, it is unlikely that a progressive party would compete for the representation of a demand for harsher criminal punishments, even if such demand becomes extremely popular. 12 Second, hegemonic actors compete for the meaning of certain signifiers, which can become what post-structuralist literature calls "a site of struggle," that is, "a terrain in which dominant discourses compete for ideological hegemony in an endless quest to fix meaning" (Chandler & Munday, 2011, p. 121). As Laclau notes, it is possible that some signifiers (and, more specifically, those which operate as empty signifiers) are disputed between two or more hegemonic actors. For example, "the people" can be contested both by left-wing and right-wing populist actors, which would associate it a different signified. This is what Laclau called "floating signifiers" (they "float" in the sense that their meaning is "suspended"), which tend to arise "when the same demands receive the structural pressure of rival hegemonic projects" (Laclau, 2005, p. 132). Those who employ the LDA will have to identify which actors are competing with each other and what is exactly at stake at a discursive level. This requires analyzing their particular discourses (e.g., whether there is a signifier that seem to be disputed by different actors) but also the discursive status quo (i.e., which signifiers seem to have enough value to be contested by different forces). 13 The dispute around those signifiers is not necessarily part of routinary politics (as political forces tend to keep their meaning on a leash, particularly in moments of virtuous competition); it is in fact in moments of political upheaval or change when a particular system is contested that sedimented meanings become polysemic and these kind of signifier arise: "the 'floating' dimension becomes most visible in periods of organic crisis, when the symbolic system needs to be radically recast" (Laclau, 2005, p. 132).

Two Methodological Moments: The Descriptive and the Subjective
It follows from the previous section that it is possible and pertinent to divide the Laclauian approach at a methodological level into two moments. Indeed, it is clear by now that the LDA focuses on both particular discourses and discursive contexts, which means that it is heuristically appropriate to divide our analysis in two steps: a descriptive one (seeking to "objectivize" a particular discursive context) and a subjective one (focusing on how political actors act in that context). Before describing both, it is important to wonder whether the distinction implies seeing hegemonic actors as being determined by a particular discursive context. The answer is negative: what is tacitly offered by Laclau is actually a dialectical approach in which "structures constrain and enable agents while agents interpret structures and in doing so change them" (Marsh (2010), p. 216). We cannot understand why the idea of "sovereignty" can become salient at a certain time by looking at the decisions made by a particular political party, but neither we can understand how the semantic dispute around it plays out without looking at the individual level. From this perspective, structure and agency are thus not seen as incompatible points of departure but rather as different methodological moments. Even though they are inseparable they need to be studied separately for the same reasons that we cannot see the two sides of a coin at the same time.
Needless to say, neither of those steps intends to be exhaustive: what the LDA allows us to do is an analysis at a discursive level from a constructivist perspective, which is only a part of what a political analyst can do. There are other elements both of political systems (such as the voting system, the impact of international relations, or the role of lobbies) and political actors (such as party funding, the difference between party activists and voters, or the importance of think-tanks) that are not covered. This is because political realities are too complex to be explored by only one methodology and therefore researchers need to make decisions which will always imply exclusions. To put it bluntly, in the context of the LDA when we talk about describing a certain political context or a political actor it does not mean that those descriptions are comprehensive. Other aspects are inevitably left aside, in the same way that other forms of analysis leave aside what is being covered here. The starting point here is that the elements outlined here are more prominent than others, at least in contemporary political systems, and thus deserve a privileged treatment.
The descriptive step has to begin with identifying the key political actors in the particular discursive system, which will always require a spatial and temporal delimitation. 14 Those who have employed forms of discourse analysis that draw from Laclau's works have focused on political organizations like Green parties (Stavrakakis, 2000), Vlaams Belang (De Cleen, 2016), Syriza (Stavrakakis & Katsambekis, 2014), or Podemos (Rueda, 2020) but also on social movements such as the environmentalist movement in Manchester (Griggs & Howarth, 2000), Post-Peronism (Barros & Castagnola, 2000) and the 15-M movement (Errejón, 2011). Laclau himself covered a broad range of political actors including both parties like Richard Nixon's Republican Party and movements like Boulangism (Laclau, 2005). It is of course possible that a particular discursive system includes both parties and movements.
Once the nature and number of political actors involved is mapped it is possible to examine the most important discursive aspects of the resulting structure. First of all, it is essential to identify the political frontiers that divide the identified political actors. From a Laclauian perspective, these are constructed and precarious, which distinguishes it from classical notions of cleavages that focus on how they reflect pre-discursive realities. There are also plural: different political actors can engage in frontier-building struggles in order to impose certain dichotomies in detriment of others (e.g., some parties will accept the left-right divide while others will build frontiers opposing the people and the elite, women and men or the patriots and the anti-patriots) and make the other actors accept them. There can be an implicit consensus between competing parties on how discursive agonism takes place, which would make them wary of any challenges regarding "the rules of the game". 15 Second, the researcher needs to locate the main demands that are the terrain of dispute for the political actors, who will compete for their meaning and representation. This part tackles questions such as which issues are at the center of the agenda at a certain time and how different discourses can reinterpret them. Finally, there needs to be a compilation of the key signifiers at stake. As already mentioned, Laclau emphasized the role of both empty and floating signifiers for the constitution of collective identities. Those are not unique to particular discursive formations but have a public dimension, and thus can be considered to be part of the discursive structure.
Finally, and as previously mentioned, it is important to take into account the role of extra-discursive elements. Despite what is generally thought, Laclau does not downplay the importance of non-discursive variables. It is in fact impossible, to the best of my knowledge, to find an example of a discourse analysis inspired by Laclau that does not refer to economic or historical aspects to explain continuities or discontinuities in particular discursive systems. The question is thus what the role of those elements is exactly and how is that role compatible with a constructivist approach that focuses on discourse. The only way to include them is to define them as non-discursive elements which importance only emerges once they affect the discursive system by triggering social changes that inevitably or discretionally are incorporated by political actors into their discourse. An economic crisis, for example, only will be part of a LDA analysis as long as it has an impact on the discursive sphere (which does not necessarily happen automatically, as materialist analysis would expect). To put it bluntly, non-discursive events will need to be integrated into the discursive arena in order to be relevant for our analysis.
Once the researcher has delineated the main elements of a particular discursive system it is time to focus on the particular actors that interact within it. In contemporary liberal democracies these are likely to be political parties, but as already suggested it is perfectly possible that other types of political entities (a military junta, social movements, or unions, for example) are an important part of a given structure. In order to examine a particular hegemonic actor, it is important to understand its discursive composition. Some of these aspects have already been outlined in the previous section, but it is important to specify this task even at the risk of being slightly redundant.
As we already know a hegemonic actor (which could also be referred as a discursive formation) has three key components from a LDA perspective: an organic ideology, a series of empty signifiers, and a plurality of demands.
An organic ideology is, according to Antonio Gramsci, a worldview that provides consistency, direction, and continuity to a group of people (Jones, 2006). The concept of "organic" should be seen, in the light of his works, as a synonym of "systemic" or "structural." Organic ideologies (but also organic intellectuals) are seen by him as functional elements that every collective identity needs. Indeed, we can imagine a series of individuals or groups coming together for a certain purpose without sharing an overarching worldview, but only in the short term and for specific purposes. If a political actor is supposed to exist and persist its components will need to share a series of stances on what social problems need to be solved and how they can be solved. One of the key tasks of a hegemonic actor is the formulation of an organic ideology that can connect with different demands. It is possible to affirm that hegemonic actors lacking an organic ideology simply do not exist, regardless of their actual size in terms of popular support. As it is evident, for an ideology to become organic it has to actually perform its organic function, which means many ideologies exist at a non-organic level (e.g., in the form of a manifesto which has not connected with any political formation). Examples of organic ideologies are social-liberalism, fascism, or ethnic nationalism, but also specific articulations (which tend to survive only for a short period of time) such as Boulangism, Legionarism, or Chavism.
The relationship between an organic ideology and the group or individual who supports it is to some extent direct (in the sense that supporters of a particular political formation share a certain set of diagnostic and prognostic elements) but there is an intermediate element that acts as a bridge between the two: the empty signifiers. As already explained, these are elements which can structure a chain of signifiers in order to partially fix the meaning of its elements. The problem with Laclau's use of the concept is that it is twofold: sometimes it refers to a signifying element which has renounced its particularity to represent a whole chain of equivalences (in these cases it is seen as a singular entity, the empty signifier, and often refers to a political leader) while on other occasions it seems to refer to a series of signifiers that together can give consistency to a chain of demands. As it should be clear by now the LDA opts for the second connotation and thus will focus on identifying which empty signifiers lie in between the organic ideology and the articulated social demands. 16 In this sense it will be necessary to understand both the extent to which they are permeated by a certain organic ideology and the way in which they connect with different demands. For example, in a social-democratic party the empty signifier "freedom" will be related to the overarching ideology ("freedom" will imply a positive conception of liberty requiring material conditions and state intervention, for example) and connect with certain demands (e.g., it will articulate demands like same-sex marriage, increasing leisure time, safety of women at night in public spaces, or self-expression values around a positive conception of "freedom"). Within the same discursive formation other empty signifiers (like "sovereignty," "equality," or "progress") will be linked to other demands.
The third element is that of the social demand, which can be defined as a political request made by a certain group, which does not imply the consistency of such group (it can exist as an aggregate of disconnected individuals). Examples of demands are women rights, anti-feminism, energy transition, democracy, anti-immigration stances, animal rights, fighting for a salary raise, deurbanization, and reindustrialization. As mentioned, according to Laclau the unit of analysis when examining a social or political movement should be the demand (instead of individuals or social classes). Demands are always undecidable: they can never be fully fixed into a particular discursive formation, which means they are integrated only in a precarious way (they are never "assured"). Hegemonic actors articulate different demands in order to create chains of equivalence which will always be threatened by what Laclau and Mouffe called "the logic of difference" (the tension between the particularity of a specific demand and its "submission" to a particular discursive formation) (Laclau & Mouffe, 1985, p. 127-134). This is why he concluded that one the key tasks of hegemonic actors was "the formation of the equivalential chain and its crystallization in a unified entity through the production of empty signifiers" (Laclau, 2005, p. 153). Because there is a third element in our approach, we will say that the main task of a hegemonic actor is the formation of an equivalential chain of demands through the production of empty signifiers which are always permeated by an overarching organic ideology.

Data Collection and Inferences
The two key aspects of the LDA at a methodological level are thus the introduction of organic ideologies (overarching worldviews that give consistency to particular discursive formations) and the division of the process into a descriptive and a subjective moment (focusing on discursive systems and particular discourses, both in constant interaction with each other). The elements to be analyzed and the steps to take are clear, but the question remains of how this framework operates at a data collection level, an issue overlooked by both Laclau and the members of the ESDA. If a scholar came to accept this approach, which evidence should be of her concern at a level of data collection? How could it be employed systematically rather than illustratively? How should she organize it and categorize it in order to draw the appropriate conclusions?
First of all, it is essential to clarify what is the nature of the empirical material examined by any researcher using the methodology that is being outlined here. Since, as it is clear by now, the Laclauian approach is holistic both from an ontological perspective (as discursive systems are seen as distinct from and non-explicable in terms of its parts) and from an epistemological one (since it is understood that the way the parts of a system behave is regulated by the characteristics of the system), it is easy to assume that it will focus on a broad range of empirical material. 17 In reality, and as it is the case with virtually any other form of discourse analysis, the LDA focuses on texts.
That being said, and once again this can be found in other forms of discourse analysis as well, every textual examination will need to be accompanied with a considerable degree of knowledge on the political system that is being studied. This is due to the fact that "discourse," as it has already been explained, is seen as an entity that is intertwined with social reality rather than as a purely linguistic one.
What kind of "texts" does the LDA focuses on? Indeed, the notion of "text" has been semantically inflated in the last decades, to the point that for some researchers it has come to refer to anything that can be "read" or interpreted. The LDA does not go that far, as it focuses mainly on the enunciative aspect of political discourses. This means that it examines the way in which political actors construct (and contest) representations of social reality through language, focusing mostly on written texts as a source of evidence. As already mentioned, the starting premise of the approach is that the social world is subjectively (or rather inter-subjectively) created and that the main task of hegemonic actors is to impose particular meanings on it. This implies that the LDA will focus on those texts that are representative of particular political discourses, such as party manifestos, posters, campaign slogans, journal interviews, declarations, tweets, or speeches. 18 In this way, the approach privileges the analysis of political leaders (who can be heuristically considered to be representative of the organization's discourse), but it is perfectly possible to formulate a more bottom-up perspective. 19 Data can be either primary (collected by the researcher) or secondary (collected by others) and can consist on interviews, observations and document analysis, even though the approach privileges the latter. Those researchers who employ it will thus have a plurality of data to choose from, and they can either be selective or indiscriminate on this issue. All of that data will be of a qualitative nature, which means that it is non-numerical and privileges detail over generality.
Those texts should allow the researcher to identify the organic ideology, the empty signifiers and the social demands of the discursive formation. Indeed, the LDA focuses on identifying those three elements. For example, if Thatcherism is being examined the researcher will have to carefully select the empirical material that she considers can be representative of this particular discursive formation and then locate the overarching ideas that give it consistency (such as traditional popular Toryism and neoliberalism), the empty signifiers that serve to unite different demands (such as "family," "nation," "efficiency," or "the individual") and then those demands (such as centralism, tax cuts, anti-unionism, or the fight against unemployment). In order to do so, she might use party manifestos, parliament speeches and interviews as data sources. As already mentioned in addition to this it will be essential to delimitate the research to a particular context (the national political arena in the United Kingdom from 1979 to 1990, for example) and to analyze the discursive system that constructed such context. Both the organic ideology, the empty signifiers and the demands they represent need to be clearly identified within the selected material. If a researcher working within this paradigm cannot connect her conjectures on how a particular discourse is articulated with actual textual manifestations of it then those conjectures will have to be reviewed.
A key question arises at this point: what are the causal mechanisms that connect discourse (understood as performative enunciations coming from certain actors) and social reality, including the attitudes and beliefs of those who are interpellated by different discourses? Indeed, this methodology starts from the premise that discourse and reality are intertwined and yet can be separated in a way that should allow to examine their interaction. But how can we establish causalities by interpreting qualitative data? The physical sciences have introduced in our conceptions of causality ideas of simple mechanisms which effects are easily traceable, but this is not how causality should be understood from a social science perspective (Rendueles, 2013, p. 45). On the other hand, from a positivist perspective our approach can be seen as problematic, as we are dealing with non-observable entities from which it is difficult to draw clear inferences (not to mention statistical generalizations) and establish causalities. From an interpretivist perspective, on the other hand, it is understood that political analyses have to focus "on understanding the meaning that social behavior has for actors" (Gibbons, 2006, p. 563), and therefore it is accepted that robust inferences are somewhat more difficult to attain, as they will always require a certain degree of interpretation coming from the researcher. The causal mechanisms that can be explored by LDA are those employed by constructivist researchers and outlined by Charles Tilly called "cognitive mechanisms," that is, "those which operate through alterations of individual and collective perception" (Tilly, 2001). The examination of these, which cannot be based on direct observation, is different from positivist analyses due to the different nature of their object of analysis.
It is also important to note, regarding causalities between discourse and reality, that in any case LDA focuses on how discourses are articulated independently of their success, which means its object of analysis is the enunciation and interpellation coming from hegemonic actors. For example, if a right-wing populist party seeks to integrate demands coming from the working class it will be considered that those demands are part of their discourse even if working class voters remain mostly indifferent to the interpellation. 20 To put it more clearly the focus is on the attempts to construct social reality even when those attempts fail, in the same way that a racist discourse can be analyzed independently of whether it is popular among citizens of a particular country. A focus on triumphant discourses alone would leave too many discourses aside (those coming from marginal political parties, for example) and prevent examining them before they reach the mainstream (if they ever do). On the other hand, since discourses are always contested by different actors trying to find those which have completely permeated social reality would prove almost impossible. 21 That being said, claims on how hegemonic actors actually represent particular demands can be made (and this can be an important part of a LDA analysis), but they will depend on methods (such as polls and surveys) that go beyond this methodology.
Regarding the selection of the empirical material there is a potential issue that must be prevented: cherry-picking. Indeed, it could be tempting for a researcher employing the LDA to focus on those texts that are most representative of the elements that she is trying to identify. For example, if she is analyzing the way nationalism can shape a populist movement, she could only focus on those texts in which such movement expresses nationalist stances. The problem is that those texts might not be representative of a particular discursive formation, and therefore there can be a bias in their selection. It is possible that, for example, in a particular speech a political leader emphasizes a particular empty signifier which is actually anecdotal if one analyses his discourse synoptically. Because the LDA focuses on discursive formations (seen as structured totalizations, regardless of their precariousness) it is problematic to explore a characteristic that might not be an important part of the structure itself. In order to avoid this bias, the researcher will have to base the selection of the material on two premises: selecting a broad range of empirical material (ideally covering a certain time period) and basing that selection on the importance of the texts (focusing on those which can be representative of the discourse as a whole). In this way, she will focus on texts that are representative of a particular discourse. A good selection will allow to identify every element (organic ideology, empty signifiers, and demands) of the discursive formation.
Needless to say, even though a series of texts will be the core of the discourse analysis, they can (and naturally will) be accompanied by peripheral texts as long as they can confirm and enrich the findings drawn from the main empirical work. The idea is that the elements that can be identified in the main corpus of data will reverberate in other minor interventions, which can be used as complementary material. That is, if the study was conducted correctly then it will be possible to find the main elements of analysis in other interventions. Be that as it may it is important to note that the core elements that articulate a discourse (from our methodological perspective, the organic ideology, the empty signifiers and the social demands) cannot necessarily be found on each and every expression of that discourse, which means that the fact that we can find an example of one of a party leader in which the key aspects of his or her discourse cannot be identified does not invalidate our analysis (which is another reason for selecting a broad range of texts, if possible).

Conclusions
This article has intended to reformulate Laclauian discourse analysis in a way that it makes it both more manageable and better adapted to the current state of the art regarding discourse analysis. As it will be clear for anyone familiar to Laclau's works, it is not an alternative to his framework but rather a methodological translation of it, one among others. The type of methodology that has been labeled as LDA has three key elements. First, it is a form of discourse analysis in which "discourse" is understood from a social constructivist way as an element that shapes political realities. Analyzing discourses will therefore mean focusing on "texts" while at the same time exploring their interaction with particular contexts. Second, the LDA is split in two "moments": a descriptive one that analyses a particular discursive system and a subjective one that focuses on how political actors come to exist and compete within that system. And third, it emphasizes the role of organic ideologies, a Gramscian concept which will be employed as an overarching worldview that permeates both empty signifiers and social demands. The task of those who will use it is both to have a good understanding of the political context that will be explored and carefully select the empirical material in order for it to be representative of certain discursive practices.
This methodological framework cannot answer every research question posited by those who want to analyze collective identities. Important elements like the significance of economic resources and organizational skills or the tendency to the emergence of party oligarchies are overlooked. At a more systemic level, the approach does not cover issues like the equilibrium conditions of party systems, the effects of electoral laws or the mechanisms of assignment of responsibility for policy outcomes. As suggested in the introduction, this is not a problem inasmuch as the LDA does not intend to be a grand theory able to solve every aspect of political behavior. Its goal is rather to focus on the emergence and persistency of political collective identities from a constructivist perspective. Such was one of the main objectives of Laclau's intellectual career. His works can certainly be criticized, and it is clear that the LDA distances itself from some of his premises, but they are here considered to be too valuable to be either dismissed or poorly applied to methodological frameworks. This article proposes a new Laclauian methodology, but it also seeks to draw attention to the significance of a certain theoretical perspective. In this sense, whether advocates of the methods of the ESDA and other scholars concerned accept the proposal or not is secondary as long as the article can at least trigger a debate around the issues it covers.

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

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD
Daniel Rueda  https://orcid.org/0000-0001-5169-2631 Notes 4. In this way, it is part of what Paul Gee called "Discourse with a capital 'D'", in opposition to "discourse" seen as a purely linguistic concept (Gee, 2011, p. 29). 5. In the last instance, this should be seen a vindication of politics as an autonomous and creative force, a view that was also partly adopted by Antonio Gramsci and that might be traced back to Machiavelli, a great inspirer of the Italian thinker. 6. Laclau and Mouffe (1985) use the example of an earthquake to explain this stance. An earthquake certainly exists as a "physical reality" which we perceive with our senses (a series of seismic waves make the surface of the Earth shake, an event that can be experienced both visually and bodily), but its meaning (a divine punishment, a geological event, a result of the political decision of creating a city in a dangerous location…) will depend on the existence of particular discourses (outside which there is no "ultimate truth" about the earthquake). 7. Just to mention a few examples: Laclau and Mouffe (1985) refer to how economic crises particularly affect the young in consumer societies, which makes them more prone to be the spearhead of new antagonisms (164), David Howarth (2000) points to geopolitical and economic variables to explain the crisis of the apartheid regime in South Africa (169-170), Barros and Castagnola (2000) implicitly take into account Argentina's socioeconomic distress to explain the rise of Perón, Laclau (2005) mentions the economic crisis of the early 1970s to explain the crisis of the Italian Communist Party (186) and Stavrakakis and Katsambekis (2014) point to the 2008 economic crisis to explain the rise of Syriza and other forms of populism. 8. The point de capiton literally refers to an upholstery button. Lacan uses this analogy to point to a linguistic element that is able to stop the chain of signifiers from slipping, that is, to fix at least partially the signified of one or more signifiers. It has also been translated as "anchoring point" and "quilting point." See Laclau (2005, p. 230-232). 9. There are two forms of hegemony which are sometimes not differentiated. One refers to a general situation in which certain discourse has triumphed (e.g., Keynesianism or neoliberalism) while the other refers to particular "individual" hegemonic blocs (such as the alliances articulated within a political party). This distinction is very clear in Errejón and Mouffe (2015) and can be appreciated when they refer to Spanish politics (18-25) and Thatcher's Britain (37) and then to Podemos (62-69). 10. They left aside the revolutionary and economistic elements of the approach. For a critique of the way in which Laclau and Mouffe incorporated Gramscianism to their approach see Sánchez Berrocal (2019). 11. In fact, it could be argued that Laclau would have agreed with Steven Lukes' threedimensional conception of power (covering decision-making, agenda-setting, and preference-shaping) although he gave a privileged position to the third dimension and had less interest in the others. 12. Here lies a great problem of the Laclauian framework, which is seeing demands as empty elements that can be incorporated in any discourse, missing the fact that some of them are already ideologically charged. 13. This is why it might be pertinent to create the concept of "valued signifiers," as only a few prominent signifiers are actually contested in a given discursive system and they are connected to shared public perceptions (it is not surprising, e.g., that in a liberal democracy words like "democracy," "equality," or "freedom" are in high demand). Those signifiers are important because they are spread in a particular society and thus have the capacity to connect with a plurality of demands (paraphrasing Marx: they have value because they circulate). 14. As a qualitative methodology which seeks to formulate thick descriptions it is likely that those employing the LDA (or other variants of Laclauian discourse analysis) focus on one or a few case studies. 15. As Laclau (2005) noted, "the forces which have constructed their antagonism on a certain terrain show their secret solidarity when it is that very terrain which is put into question. It is like the reaction of two chess players to somebody who kicks the board" (140). 16. That being said, it is possible for a leader to become the prominent empty signifier (such has often been the case in presidential systems), but that does not change the fact that the leader itself will be part of a particular organic ideology (even though it might be presented in vague terms) and he will have to formulate a series of empty signifiers (which might be, at least for a period of time, secondary in comparison with his name). 17. For a differentiation between individualism and holism from a methodological perspective, see Halperin and Heath (2017, p. 36-40). It is important to note that the fact that the LDA is a holistic approach does not mean that it does not take into account both the micro and the macro level, as the relationship between the two (here, between political discourses and discursive systems) is seen as dialectical. 18. There is an exception to this "text-centric" stance and it affects empty signifiers. As we know, these are signifiers that renounce to their own particularity to represent a plurality of demands. These can be words that can be retrieved from carefully selected material, but they can also be human beings (such as a leader) or visual entities (such as a flag or a particular symbol). 19. For example,  and Errejón (2011) explore social movements by analyzing the different discourses emanating from anonymous participants. 20. This is why opinion polls and social surveys can be important (in order to see if a hegemonic actor is being successful) but are not essential. 21. For example, it would be impossible to ascribe a demand for hard Euroscepticism to any political party unless it could be proven that every voter that participates in such demand supports the political party in question.