Paradigms and Practice

Abstract

Thomas Kuhn's concept of paradigm has long been a part of ordinary parlance in political science. Aside from its role in metatheoretical debate, scholars have enlisted the paradigm concept to explain policy change, particularly in the international political economy (IPE) literature. In this context, policy paradigms are defined primarily in ideational terms and with respect to a specific domain of policymaking. We argue that this stance overstates the ideational coherence of policymaking and runs a risk of reification. We re-evaluate the paradigm concept by drawing a link to the recent literature on norm change that emphasizes the importance of practice and process. This analysis highlights theoretical difficulties in using the paradigm concept, as the relation of ideas to practical logics elides the distinctness of paradigmatic frameworks. Without clear boundaries, paradigms lose much of their analytical purchase. While the paradigm concept initially proved useful in highlighting the role of ideas, it is time to recognize its limits in explaining stability and change in policymaking.

Desde hace tiempo, el concepto de “paradigma” de Thomas Kuhn forma parte del lenguaje habitual de la ciencia política. Aparte de su papel en el debate metateórico, los estudiosos han recurrido al concepto de “paradigma” para explicar el cambio de políticas, sobre todo en la literatura de la economía política internacional (EPI). En este contexto, los paradigmas políticos se definen, principalmente, en términos ideológicos y en relación con un ámbito específico de elaboración de políticas. Sostenemos que esta postura exagera la coherencia ideológica de la elaboración de políticas y corre el riesgo de dar lugar a la reificación. Volvemos a evaluar el concepto de “paradigma” estableciendo un vínculo con la bibliografía reciente sobre el cambio de normas que hace hincapié en la importancia de la práctica y el proceso. Este análisis destaca las dificultades teóricas para utilizar el concepto de “paradigma,” ya que la relación de las ideas con las lógicas prácticas elude la distinción de los marcos paradigmáticos. Sin límites claros, los paradigmas pierden gran parte de su capacidad de análisis. Si bien el concepto de “paradigma” resultó inicialmente útil para destacar el rol de las ideas, es hora de reconocer sus límites para explicar la estabilidad y el cambio en la formulación de políticas.

Le concept de paradigme de Thomas Kuhn a longtemps fait partie du jargon courant en sciences politiques. Outre son rôle dans le débat métathéorique, les chercheurs ont fait appel au concept de paradigme pour expliquer le changement de politique, en particulier dans la littérature sur l’économie politique internationale. Dans ce contexte, les paradigmes de politiques sont principalement définis en termes idéationnels et par rapport à un domaine spécifique d’élaboration des politiques. Nous soutenons que cette position surestime la cohérence idéationnelle de l’élaboration des politiques et encourt un risque de réification. Nous réévaluons le concept de paradigme en établissant un lien avec la littérature récente sur le changement de normes qui met l'accent sur l'importance de la pratique et du processus. Cette analyse met en évidence les difficultés théoriques de l'utilisation du concept de paradigme tandis que la relation entre idées et logiques pratiques élude la distinction des cadres paradigmatiques. Sans limites claires, les paradigmes perdent une grande partie de leur avantage analytique. Bien que le concept de paradigme se soit initialement avéré utile pour souligner le rôle des idées, il est temps de reconnaître ses limites lorsqu'il s'agit d'expliquer la stabilité et le changement dans l’élaboration des politiques.

Scholars have deployed Thomas Kuhn's The Structure of Scientific Revolutions across many disciplines, and political science is no exception. Its arguments are ripe material to stoke debate, as it touches on issues of epistemology and the possibility of progress. One of the book's central contributions is the articulation of the concept of “paradigm” to describe a set of exemplary problems and their solutions that informs the basic orientation of a community of scientists toward their object of study (Kuhn 2012 [1962], 10–11). In this paper, we discuss how scholars have applied this concept to understand changes in policymaking, particularly in IPE.

We argue that the recent “practice turn” and the emergence of a processual and relational sensibility in international relations (IR), particularly in the theorization of norms, reveal important shortcomings in the current discussion of policy paradigms. While the paradigm concept helpfully brought to light the role of ideas in policy development, the literature's continued focus on policy paradigms as ideational frameworks obscures ongoing processes of change and puts scholars at risk of reifying particular ideational structures. Furthermore, the ideational reading of paradigms actually deviates from Kuhn's own more practical understanding of the concept. We develop a critique of the policy paradigm concept based on the recent turn to practices in IR to illustrate the problems inherent in thinking of them as distinct and separable ideational frameworks that govern action. In light of this analysis, we conclude that policy paradigms provide little analytical leverage on the issue of change. Rather than develop another variation on the theme of paradigm, which is already subject to numerous interpretations, we recommend that scholars carefully qualify their use of the concept. We conclude by discussing several theoretical approaches that address policy change in more practical, processual terms.

Paradigms in Kuhn and IR

Political scientists were early adopters of Kuhn's concept of paradigm. Not long after its elaboration by Kuhn, it was introduced to IR scholarship to explain the development of the field (Lijphardt 1974). While scholars have frequently mobilized Kuhn's work for metatheoretical debate of this kind, our concern here is with how they have used it to explain dynamics of policymaking. Starting with Peter Hall's landmark 1993 paper, scholars have applied Kuhn's framework to make sense of historical patterns in foreign and economic policymaking. According to Hall's development of the concept, a “policy paradigm” is “a framework of ideas and standards that specifies not only the goals of policy and the kind of instruments that can be used to attain them, but also the very nature of the problems they are meant to be addressing” (Hall 1993, 279). Thus, a policy paradigm is defined primarily in ideational terms and with respect to a specific domain of policymaking. In principle, the policy paradigm concept is broadly applicable to policymaking. Aside from some discussion in security-related contexts, however (e.g., Legro 2005; Krebs 2015), scholars have most frequently applied it to economic policymaking, and it is a recurring reference in the broader IPE literature. Examples of policy paradigms in this field that scholars frequently refer to include Keynesianism and neoliberalism, among others. Although Hall's characterization of the emergence of neoliberalism as a paradigm shift is broadly accepted, there is less agreement on how new paradigms emerge or what counts as a paradigm.

Alongside its appropriation in other fields, Kuhn's work has been subject to substantial critique, and he refined and elaborated many aspects of his initial theoretical edifice. As a result, his theoretical legacy is complex. In an essay otherwise sympathetic to Kuhn, Margaret Masterman notes twenty-one different ways that he used the term “paradigm” in his foundational work (Masterman 1970, 61–65). She groups these uses into three categories that characterize paradigms in varying degrees of breadth: (1) as an instrument or apparatus that can be used for puzzle-solving; (2) as a set of scientific habits that lead to successful puzzle-solving; and (3) as akin to a worldview (Masterman 1970, 66–67). Kuhn himself subsequently amended his understanding of his central notion of “incommensurability.” In his original work, this referred to the proposition that two competing paradigms were not directly comparable with one another. In later years, however, Kuhn developed the notion of “local incommensurability” (Kuhn 2000, 52,189). Whereas the strong notion of incommensurability implies paradigms that are wholly distinct from one another across significant theoretical and empirical domains, local incommensurability suggests that different paradigms actually have much in common outside of a few core concepts. He also came to emphasize the possibility of communication across paradigms, arguing that the process was similar to translation (Kuhn 2012 [1962], 197–203).

The ways in which scholars apply the paradigm concept in political science today parallels this theoretical ambiguity. On its introduction as a tool to analyze policymaking, Hall conceived of policy paradigms as mutually exclusive worldviews that significantly circumscribed actors’ horizon of action (Hall 1993, 280). Thus, there was no common ground on which the relative value of different paradigms might be adjudicated, and even communication between adherents of different paradigms was difficult. Scholars have built on and extended Hall's understanding of paradigms and their distinctness (e.g., Babb 2013), but the positions they have taken have varied widely. Some applications posit less rigid understandings of paradigms and suggest the possibility of paradigm “hybrids” (Ban 2013) or “inter-paradigm borrowing” (Hay 2011). Others expressly emphasize the incommensurability of competing policy paradigms (Blyth 2013). Scholars also differ over whether paradigms refer to specific policy positions or broader foundational understandings (Brooks 2005; Becker and Hendriks 2008), and whether or not paradigms are necessarily coherent (Becker and Hendriks 2008). In the broader literature, the use of the term “paradigm” shades off to briefer mentions, often without further reference to either Kuhn or Hall's foundational works, which we take as the substantive theoretical basis for the policy paradigm concept. Some of these works (such as Brooks and Kurtz 2012) are clearly indebted to the policy paradigm concept while the provenance of others is less certain given the lack of references.¹ The contemporary discussion of policy paradigms also does not mine the depths of Kuhn's work or the controversies surrounding it in much detail. For example, a search on “local incommensurability” turned up only one IR paper (Jackson and Nexon 2009), and it was concerned with the use of the paradigm concept in a metatheoretical sense rather than as a tool for understanding policy development.

Given this fairly superficial engagement with Kuhn, the contemporary discussion in IR, particularly on issues such as the appropriation of ideas across paradigms or their potential lack of coherence, has developed a dynamic of its own. Kuhn remains a common reference, but the conversation around paradigms is no longer determined or constrained by his conceptual apparatus. Indeed, Steve Fuller has commented on how the easy adoption of Kuhn in disparate fields is due in part to the “holographic qualities” of the text. For Fuller, Kuhn's seminal book is a relatively shallow interdisciplinary work characterized by “nonthreatening prose” and a lack of technical references that is more the product of Kuhn's pedagogical duties at Harvard rather than archival research (Fuller 2000, 31–32).

Despite this variability of the paradigm concept in contemporary IR literature, there is one striking manner in which all of these usages are similar: the marked ideational nature of the concept. According to Bentley Allan, the policy paradigm approach “has a narrow ideational ontology on which economic knowledge dominates policy” (Allan 2019). Rooted in Hall's work (1993), this understanding is borne out by how policy paradigms are frequently described as “foundational” (Drezner and McNamara 2013, 160) or “widely shared, dominant” ideas (Becker and Hendriks 2008, 832). The centrality of ideas is even salient among some critics of the paradigm framework (Morrison 2012). This importance of ideas reads very much like the literature on norms and norm change as it emerged in the 1990s and early 2000s—what Simon Frankel Pratt has called the “first wave” of norm research (Pratt 2020, 62–63). In fact, discussions of paradigms frequently reference this literature (e.g., Brooks 2005; Schmidt 2011; Chwieroth 2014; Güven 2018). They also mobilize similar concepts, such as internalization (Güven 2018), diffusion (Brooks 2005), the taken-for-granted nature of appropriate action (Hall 1993; Brooks 2005), and editing, which resembles norm localization (Ban 2013).²

The borrowing between these literatures is understandable in view of their significant conceptual overlap. Norms, understood as discursively shared standards of appropriate behavior, bear a strong resemblance to policy paradigms as defined by Hall and subsequent works. Like norms, policy paradigms frame certain actions as appropriate and others as less so. Scholars have operationalized policy paradigms in several different ways as outlined above, some of which function more or less like norms. Brooks and Kurtz (2012, 99), for instance, note the importance of the “social construction of appropriate behavior” in the adoption of pension privatization, an orientation which they identify as a policy paradigm. Yet the key similarity is that both paradigms and norms consist of rules known to a relevant group of actors, and that these rules prescribe specific behaviors under a specified set of circumstances.

This focus on the impact of ideas in the policy paradigm literature and its similarity to norms is somewhat ironic, given how Kuhn himself understood paradigms to operate. In Kuhn's formulation, paradigms have a significant practical dimension; they are not defined by a set of discursively grasped rules and they “need not even imply that any full set of rules exists” (Kuhn 2012 [1962], 44). Scientists become part of the community of scholars working within a paradigm not by learning concepts from textbooks but through problem solving and experimentation—that is, through practice. According to Kuhn, “one is at liberty to suppose that somewhere along the way the scientist has intuitively abstracted rules of the game for himself, but there is little reason to believe it (…) If they have learned such abstractions at all, they show it mainly through their ability to do successful research. That ability can, however, be understood without recourse to hypothetical rules of the game” (Kuhn 2012 [1962], 47). The knowledge that informs scholarly work therefore has a significant unarticulated and practical dimension. This posed a problem for Kuhn: how was a paradigm to cohere in the absence of a set of explicit rules? Here, Kuhn reached for Ludwig Wittgenstein's concept of “family resemblance” to describe how “what guides a scientist is not a set of rules but the acquired ability to recognize resemblances between new research problems and the fields’ established achievements” (Bird 2012, 878; Kuhn 2012 [1962], 45). For Kuhn, practice is fundamental in defining the field of inquiry and the elaboration of knowledge structures, and the work of scientists is oriented by concrete problems, not abstractions.

Although Kuhn develops these themes in his discussion of how scientists carry out “normal science” within a paradigm, he does not extend this thinking to other aspects of his framework, such as paradigm change. Given that the publication of Kuhn's seminal book in 1962 predates the broader turn to practice in sociological theory (Schatzki, Cetina, and von Savigny 2001), the necessary tools might not have been available. Kuhn's later work, which focused for a time on the issue of exemplars, turned more toward issues of language, perhaps in light of the critiques of his interlocutors, who were primarily philosophers (Bird 2012).³ However, his original proposition that knowledge is in part practically constituted and his references to thinkers like Wittgenstein and Michael Polanyi resonate with the current interest in practice theory in IR. In fact, Kuhn's de-emphasis of the discursive representation of rules overlaps with how practices are understood in IR as “socially meaningful patterns of action” premised on a background knowledge that remains largely inarticulate (Adler and Pouliot 2011, 6). In the next section, we trace the development of the norm literature in IR in terms of this perspective on knowledge and with an eye to developing a critique of paradigms in IR.

Practice, Process, and Norms

Recent developments in the literature on norms and norm change have raised difficulties with the understanding of norms articulated in the “first wave” of norms scholarship. Norms in this earlier literature were conceived in straightforward terms as relatively unambiguous, discursive prescriptions of behavior. Initial work on norms in IR was focused on establishing their analytical significance for explaining outcomes in a field where the importance of material and strategic incentives held sway (e.g., Katzenstein 1996; Tannenwald 1999). Norm research in IR at that time also turned to the issue of how norms might change. This process was understood in unidirectional terms as the spread and adoption of new, progressive ideas of appropriate behavior. Some scholars focused on the role of activists and non-state actors. Champions of a new norm would promote its spread, and after it had been adopted by a critical mass of states, the norm would reach a “tipping point” after which it would spread rapidly across the globe (Finnemore and Sikkink 1998). Other scholars emphasized the persuasion and socialization of elites to new norms, particularly in institutional settings (Johnston 2001; Checkel 2001, 2005). In both cases, the new norm would be internalized and form part of actors’ taken-for-granted understanding of appropriate action.

Subsequent work soon elaborated this understanding of norms in light of an increasing appreciation of their complexity and of the importance of local context (Acharya 2004;Wiener 2009, 2014; Zarakol 2014; Dixon 2017). These works began to move away from the conceptualization of norms as uniform entities, highlighting instead their contested nature. The increasing interest in practices in IR theory and the emergence of a more elaborate relational and practical sense in some constructivist literature (McCourt 2016) reinforced this reappraisal of the nature of norms. Scholars grasped how the intersubjective nature of practices necessarily entails a normative dimension, as practices necessarily demonstrate how something is done (Schatzki 1996, 101). Recent works in IR have further developed the theoretical links between practices and normativity (Schmidt 2014; Hofferberth and Weber 2015; Adler 2019, 4; Pratt 2019, 2020).

This focus on practices downplayed the discursive representation of norms and further undercut the conceptualization of norms as having uniform and stable meanings. Practices by nature are not instantiated in identical ways across actors and contexts (Doty 1997; Bueger and Gadinger 2015). An appreciation of the normativity of practice therefore necessitates a more complex and unsettled conception of normativity than the uniformity implied in the notion of a “shared norm.” By this understanding, norms are not homogenous entities that determine action, but instead are carried along by—and negotiated through—their instantiation in practice. There is inevitably slippage in enacting normative prescriptions, even in the unlikely case that different actors somehow share the same understanding of a given norm. As a result, it becomes problematic to conceive of norms and structures of meaning as entities or static frameworks for action with clear boundaries.

To highlight this shift in thinking about norms (and other entities) in IR and to draw out its implications, it is helpful to consider a theoretical perspective that foregrounds processes and relations, as opposed to stable entities in the analysis of social life. This “processual-relational” perspective, as Patrick Thaddeus Jackson and Daniel Nexon (1999) have labeled it in IR, is a contemporary instantiation of a long-established alternative sociological tradition (Dewey and Bentley 1949; Emirbayer 1997; Elias 2000 [1939]; Abbott 2016). Its central tenet is to treat the entities that usually form the basis for sociological explanation, such as identities, norms, and agents, not as given objects with a suite of core characteristics, but as emerging through ongoing processes and transactions (Jackson and Nexon 1999; Epstein 2013; Hofferberth and Weber 2015; Sending, Pouliot, and Neumann 2015; Bucher and Jasper 2017). Rather than viewing process and interaction as happening between pre-constituted entities, these dynamics constitute and reconstitute the entities in question (Abbott 1995). In his earlier work, Alex Wendt emphasized the importance of process by noting that “structure has no existence or causal powers apart from process” (Wendt 1992, 395). In sum, structure is not a “thing” but exists only in process, in the actions of agents.⁴

An appreciation of process is closely aligned with a focus on practices, since the concept of practice, concerned with the “do-ings” of actors, is bound up with a notion of process (McCourt 2016). This kinship of practice and process is already clear in Freidrich Kratochwil's earlier work, including his discussion of norms. Kratochwil parted ways with a “static representation of norms” (Kratochwil 1989, 18) and argued that “actors are not merely programmed by rules and norms, but they reproduce and change by their practice the normative structures by which they are able to act, share meanings, communicate intentions, criticize claims, and justify choices” (Kratochwil 1989, 61). Rules and norms do not stand apart from and determine or constrain action; rather, they are taken up in the practices of actors—“one of the most important sources of change”—and are always in process (Kratochwil 1989, 61).⁵

In contrast, the more familiar theoretical approaches in IR in which the discussion on policy paradigms is couched are “substantialist” in that they conceive of the social sciences as the study of interactions between preexisting entities. They follow a methodology that isolates the entities of interest from ongoing social processes and endows them with essential and stable characteristics. These characteristics are taken to be extra-theoretical givens and remain untouched by the interactions under consideration. For example, the process of socialization understood in these terms consists of “adding a norm onto (or removing a norm from) a stable platform of identity” (Epstein 2010, 334). By this understanding, norms are effectively things, which are taken up by individuals, become part of the actor's identity, and cause norm-compliant behavior. Starting analyses with substances instead of processes is understandable, given that we see a world filled with things, individuals, and institutions that strike us as unproblematic, and even natural, entities (Jackson and Nexon 1999, 299). Yet, as Bernd Bucher (2017, 415) notes with respect to substantialism, “intuitive appeal does not point to ontological necessity or practical usefulness” in explaining phenomena.

The problem with a substantialist approach to normativity specifically is that it ascribes an unrealistic coherence and uniformity to norms. If norms diffused, were internalized by actors, and reduced contestation, they would need not only to have clear boundaries, but also to be uniform across time and space. This suggests an impaired notion of agency and invites structural determinism (Sending 2002). By contrast, the arc of IR theorization on norms toward a greater appreciation of their complexity is an implicit recognition of the problems with a substantialist stance. Practice-theoretical accounts that empathize the indeterminacy of practice as well as agents’ capacities for improvisation and creativity help recover a sense of agency (Cornut 2018). The potential creativity of action is particularly emphasized by the literature that builds on the insights of American pragmatist thought (Berk and Galvan 2009; Schmidt 2014; Avant 2016; Jabko and Sheingate 2018; Jabko 2019; Pratt 2020). The acknowledgement of indeterminacy and creativity necessarily implies the fluidity of boundaries, the ability to transcend structural constraints, and the syncretic (as opposed to coherent) nature of institutions. Conceived in processual-relational terms, norms are continually in process (Schmidt 2014), points of reference and orientation that are “constantly in the making” (Hofferberth and Weber 2015), or provisionally stabilized “normative configurations” (Pratt 2020).

An additional difficulty inherent in a substantialist approach is that references to purported stable entities may easily slide into implicit ontological claims that reify the entity in question (Jackson and Nexon 1999, 312). This concern for reification resonates with scholarship that is critical of certain aspects of the constructivist project in IR. Vincent Pouliot, for example, notes how much constructivist work is predicated on “acts of essentialization” by the analyst. These steps, taken for seemingly neutral methodological reasons, turn an analytical representation of the state or identity—or in our case norms as ideas—“into the reality itself and assume its existence in an a priori way” (Pouliot 2004, 328). The consequence is to endow the reified object with a natural and unquestioned givenness. Acts of scholarly reification set hard and fast limits to processes of social construction and undermine “the non-foundationalist starting point of constructivist and post-structuralist thinking” (Bucher and Jasper 2017, 2).

By not privileging certain entities, a processual-relational perspective helps avoid these pitfalls. It thus takes the constructivist project in IR to its logical conclusion, offering the possibility of moving beyond the reified substances that overestimate uniformity and set limits to how far change can be theorized. With this link between practice, process, and normativity in mind, we can turn again to the current literature on policy paradigms.

The Problem with Paradigms

While the turn to practice and process has made recent norm research sensitive to the contested and variegated meanings of norms and even questioned the relevance of “norms” as objects of analysis, the same has yet to happen in the discussion of policy paradigms. The similarity of the policy paradigms literature and earlier constructivist literature on norms means that similar problems arise. The discursive, ideational character of policy paradigms and the dynamics used to describe their fates, such as diffusion and internalization, constitute them as coherent and well-bounded entities in a substantialist sense. As with norms, the use of the paradigm concept in this manner predisposes analysts to see uniformity where there is little of it. The Washington Consensus, which is occasionally referred to as a paradigm, exemplifies this understanding, as the term was coined in a paper that actually laid out a series of policy prescriptions with which the Consensus came to be identified (Williamson 1990).

The ideational character of the policy paradigm concept has not prevented scholars from incorporating notions of practice into it. In an early example, Kathleen McNamara (1998, 67–69, 148–149) adopted Peter Hall's concept of policy paradigm, but then described the position of European central bankers in the run up to monetary unification as one of “pragmatic monetarism,” as opposed to a more purist adoption of Milton Friedman's monetarism. More recently, Ali Burak Güven (2018) notes the role of inherited organizational forms, expertise, and culture at international financial institutions in attenuating the implementation of the “post-Washington Consensus” (PWC) policy paradigm. However, he marks an explicit distinction between policy paradigm and policy practice, and he claims the actual implementation of the PWC has fallen short of the paradigm's ambitions: “the partial decline of the PWC constitutes an important change not in policy paradigm but in policy practice” (Güven 2018, 410). By this understanding, the paradigm is a collection of ideas that floats above the fray, distinct from the practices of policymakers that may significantly modify it in its execution. Güven judges the PWC, itself the result of a “paradigm expansion” rather than a paradigm change of the original Washington Consensus, to have never been fully realized.

This mode of analysis, however, puts a high burden on what the scholar decides is or is not a paradigm. The stance the scholar takes in these situations also necessarily impacts claims of paradigm change, expansion, or decline, and the different analytical consequences that follow these claims. It lends itself to Pouliot's (2004) critique of problematic “acts of essentialization” on the part of scholars. This scholarly approach thus not only grasps paradigms as an ideational framework—a given entity in substantialist terms—but also raises the issue of reification. The identification of paradigms recalls Pratt's (2020) discussion of norms and the inherently political act of identifying a norm, its bounds, and deviations from it. This stance puts scholars in the driver's seat in calling into being the objects of subsequent analysis. According to Pratt, the de-reification of norms, in turn, is an important step in analysis that allows scholars to minimize the potential bias introduced by their own perspective. It also tracks more closely the contested nature of concepts and how they are caught up in and transformed through ongoing practice.⁶ This is a view that builds on efforts to move constructivist work in a more post-foundationalist direction (Epstein 2013; Bucher and Jasper 2017), and it is directly applicable to how paradigms are often treated in the literature today.

The assumed fixity and stability of knowledge structures in the paradigmatic understanding of political economy arguably engages in precisely this process of reification. It results in discrete “snapshots” across historical time—now Keynesian, now neoliberal, for example.⁷ Rather than illuminating the process of change, the concept of paradigm in a sense obscures what is essentially an ongoing and contested process. These difficulties become apparent in the qualifications that scholars invested in the paradigmatic worldview offer to fit their account to the historical record. For instance, the time period stretching from the early 1970s to the mid-1980s—a not inconsequential length of time—is characterized as a period of “transition” (Hall 2013a, 133) and “electoral confusion” (Hall 2013b, 191) enabling the shift from a Keynesian to a neoliberal era. Yet Keynesianism had itself cleared the field of serious challengers only in the late 1940s. And even then, there is an argument that the dominant approach in economics was itself a “synthesis” between neoclassical and Keynesian economics that progressively “hollowed out” Keynesianism (Best 2005).⁸ So what is stability? What is transition? When periods of transition are as long as periods of purported stability, we should question the conceptual tools with which we are examining the empirical record.

Similar difficulties afflict discussions of more recent phenomena as well. While Güven characterizes the move from the Washington Consensus to the PWC as a paradigm expansion, Sarah Babb (2013) sees this merely as an “augmentation” of the consensus with apparently little consequence in terms of paradigmatic analysis. Meanwhile, Cornel Ban (2013) argues that we see the rise of a new paradigm in Brazil, which he dubs “neo-liberal developmentalism” and which he defines as a “hybrid” of neoliberal and developmentalist strategies. Significantly, he denies that the change to the new paradigm is revolutionary, whereas Babb (2013, 287) notes the lack of revolutionary change as evidence that a new paradigm has not arisen. Although such disagreements are part and parcel of any vibrant scholarly debate, they also in this case sow doubts about the strategy of retrospectively reconstructing generalized ideational frameworks. Historians of neoliberalism, in particular, are uncomfortable with the shorthand use of labels by social scientists to conveniently demarcate historical periods or institutional regimes as “Keynesian” or “Hayekian” (Slobodian 2018, 5).

Of course, it is clear that actors themselves engage in processes of reification—and the observation of these processes is in fact central to constructivist analysis (Pouliot 2004). Policymakers have a variety of preferences, for instance being more or less inclined to trust market mechanisms, and typically couch these preferences in terms of a particular label or shorthand. By deploying these labels—monetarism, neoliberalism, etc.—actors tend to reify them in their daily interaction as coherent worldviews. However, this does not mean that analysts should uncritically accept these reifications as the foundations of analysis, and it does not follow that these labels refer to stable and well-bounded structures of knowledge that determine preferences and possibilities for action. Taking for granted that categories like neoliberalism have clear meanings distinct from practical contexts and impose themselves upon actors erodes the agency of actors and assumes a coherence that actors themselves may not experience. Rather, we should heed the advice of Bruno Latour, whose work is a frequent reference in more relationally oriented constructivist theorizing. Instead of starting with heroic assumptions about the nature of social entities, we should “follow the actors themselves” and map out their transactions (Latour 2005, 12). In a similar manner, Pouliot's (2007) discussion of a postfoundational constructivist methodology appreciates the perils of analysts imposing their own understandings on historical phenomena. Recognizing that “theorization destroys meanings as they exist for social agents” (Pouliot,2007, 364), his approach relies on an inductive reconstruction of meanings as understood by the actors themselves and is aimed at recovering the intersubjective context of action.

Interpretations of Wittgenstein, whose work is a key reference to the “practice turn,” underscore the centrality of practical engagement in the categorization of objects in a way that reveals the fundamental ambiguity of the process. Rules for the categorization of objects—or of policies, in our case—do not operate independently from the specific practices of those engaged in the process and the concrete biological, psychological, and social contexts within which they find themselves (Bloor 1997, 19). Meaning is not the result of the interpretation of an abstract rule, such as the definition of what neoliberalism is. Rather, agents “create meaning as [they] move from case to case” (Bloor 1997, 19; see also Schatzki 1996, 51). Categories therefore have a temporal and historical dimension (Prien 2004) and are closely tied to practice, with all the attendant variability and contestation that this entails. As a result, concepts and classificatory schemes, such as those that are central to identify the content of paradigms, have inherently contested boundaries and change over time. For example, Jacques Rueff and Milton Friedman are both closely identified with the “neoliberalism” of the Mont Pelerin Society. Yet they held diametrically opposed views on the gold standard at a time when this issue was of central importance in the late 1960s. No one would suggest that this disagreement, significant though it was, meant that one was neoliberal while the other was not. Concepts and categories—like neoliberal—are artifacts of the historical process. Their boundaries remain ambiguous as actors confront new developments that must be apprehended. Drawing on the above example again, to claim that the gold standard is an important piece of neoliberal policies today would be foolish and plainly wrong, but in the early 1970s this was not so obvious. Taking concepts out of context and endowing them with stable characteristics that are then applied retrospectively distorts our understanding of policy development. In many cases, policymakers are unable to unambiguously apply a paradigm to a concrete case in ways that are clearly distinguishable from contingent considerations that have little or nothing to do with the paradigm.

Implications

The term paradigm has worked its way into the everyday language of policy analysis, and it is here to stay. A danger in the way it is often used, however, is that it may steer analysis away from context-sensitive historical reconstructions of policy development, leaving reified ideational frameworks in its wake. Paradigms are analytic constructs that scholars use to retrospectively make sense of past developments. As a result, they frame issues of stability and change in ways that actors did not confront, and their use is liable to create problems where there were none and to obscure those that practitioners actually dealt with. Ironically, Kuhn himself came to see paradigms as historical reconstructions, separate from the practical evolution of science (Kuhn 2000, especially 33–57, 86–89). The “discovery” of distinct ideational frameworks identifiable through a list of superficially stable characteristics is an act of reification we should all seek to avoid.

As currently deployed, the analytical significance of policy paradigms stands or falls on whether abstracted ideational frameworks have durable boundaries and reliably structure, or even determine, action. Recent theoretical developments and empirical investigations sensitive to the complexity of norms and the centrality of practice undermine these assumptions. As a result, acts like “inter-paradigm” borrowing are not analytically more significant than any other act of ideational appropriation. This is not to say that the term has no value, but rather that scholars should carefully qualify its use and avoid making it a linchpin of their analysis. For example, using paradigm as a shorthand to connote a general intellectual and practical orientation toward policymaking is a useful way to quickly relay information about a given actor's predilection for certain policies. This is, in fact, not far from the way some authors have used it, and it therefore has the advantage of familiarity. Scholars need to make clear, however, that such statements neither carry connotations of determinism or exclusivity vis-à-vis the broader universe of policy choices, nor say much of consequence with regard to particular policy prescriptions. This can be in part avoided by stressing the practical and inarticulate component of this orientation and thereby the inherent variability of its instantiation in specific policy contexts.

The broader significance of this adjustment is that, aside from acknowledging the indeterminate nature of paradigms, paradigms on their own do not help the analyst say much about policy stability and change. Moving away from policy paradigms when discussing such dynamics is not a step into the unknown, however. Recent works building more or less explicitly on practice-theoretical foundations and allied approaches provide analytical strategies to grasp stability and change in policy development without the problems inherent in the old paradigm concept. Jacqueline Best develops an account of change in global development finance since the 1980s around the concept of governing styles rather than paradigms. In explicitly drawing distinctions with the paradigm approach, she argues for an understanding of change in the spirit of Bruno Latour as the result of “a series of never-perfect translations as policy practices and ideas are borrowed, combined and transformed over time” (Best 2014, 68). Best also makes a point of not jettisoning ideas, but rather noting the importance of “the techniques that they enable and in which they are embedded” (Best 2014, 68). In a related critique of paradigms, Nicolas Jabko (2019) draws on the pragmatist tradition and recent scholarship in psychology to develop the concept of repertoire to explain policy development during the Eurozone crisis. In contrast to a paradigm, a repertoire is an open-ended discursive practice around which conflicting policy programs can converge and evolve. It therefore accounts for the seemingly contradictory persistence of neoliberal ideas of austerity alongside unconventional policy initiatives like the EU's new bailout mechanism and the European Central Bank's interventions.

Some works, while not specifically addressing the issue of paradigms, provide additional strategies for the analysis of stability and change. For example, Jean-Philippe Thérien and Vincent Pouliot operationalize the (Levi-Straussian) concept of bricolage to highlight “the improvisatory, haphazard and combinatorial nature of global public policymaking” in the articulation of the Sustainable Development Goals (Thérien and Pouliot 2020, 1; see also Carstensen 2011). Their contribution moves away from the importance of settled ideational frameworks and draws attention to the contingent nature of the policy process. As a result of this dynamic, global governance resembles a “patchwork” rather than the “long march toward Pareto-optimality described by public goods theorists” (Thérien and Pouliot 2020, 3). The recent introduction into IR of the concept of assemblage also offers a useful perspective from which to understand the dynamics of the world political economy. For example, rather than being characterized by frameworks of ideas or a set of institutionally stabilized processes, William Connolly conceives of the web of international economic and financial transactions in terms of “a temporal complex in which numerous coexisting elements are simultaneously interinvolved, externally related, and jostled by flows that exceed these two modes of connection” (Connolly 2008, 12). Such a perspective grasps a complexity that eludes conventional approaches that assume simplicity and coherence. Conventional modes of analysis, the tools of which Connolly likens to “methodological straightjackets,” also tend to foreclose possibilities of engagement, critique, and change (Connolly 2008, 12). These works are but a small slice of a diverse and growing literature that emphasizes contestation, practice, and process, and that questions the notion of clearly bounded ideational frameworks as guides to action.

Concepts are never neutral instruments; they often outlive their usefulness and may even have counterproductive effects. They simplify reality and thus are necessary for analysis and theory building. Yet they also carry a danger of reification, insofar they elide the complexity of reality (Hirschman 1985; Gunitsky 2019). We must therefore continually re-evaluate the conceptual tools at our disposal. Paradigms are one such tool that provided valuable insights on the role of ideas in the policy process, but its purpose is no longer so obvious. As Andrew Pickering argues from the perspective of science studies, characterizations of science in terms of paradigms “are undermined in studies of scientific practice which […] confront us with the multiplicity of scientific culture, with its disunity, its patchiness and scrappiness.” (Pickering 1995, 94). We believe it is time for political scientists to recognize—as most science studies scholars have done—that the concept of paradigm may have become more often misleading than helpful, and that they should rethink its use for a better understanding of policy development.

Footnotes

Notes

Author's note: We would like to thank the ISQ editors, the anonymous reviewers, Vincent Pouliot, David Steinberg, Jazmin Sierra, Victoria Paniagua, and panel participants at the 2018 Annual Meeting of the American Political Science Association for helpful comments on earlier versions of this article.

Nicolas Jabko is an Associate Professor of Political Science at Johns Hopkins University. His current research interests include neoliberalism, sovereignty, crisis politics, and constructivist and pragmatist approaches in political science.

Sebastian Schmidt is an Assistant Professor of Political Science at Johns Hopkins University. His research interests include the application of pragmatist thought to international relations theory and the sociological underpinnings of security strategies.

References