Setting soft power in motion: towards a dynamic conceptual framework

Abstract

The concept of soft power is widely regarded as unfit for academic service. But instead of dismissing it, this article grapples with its unresolved conceptual problems in an attempt to reveal the hidden value of soft power. Arguably, the proposed conceptual revision can prove its merit to International Relations scholars by structuring research around the study of mechanisms and reconciling rationalist and constructivist approaches for the analysis of power. While the revised conceptual framework builds on Nye’s original account and on the achievements of soft-power scholarship, it is the intellectual import from the contentious politics research programme that enables the crucial adjustments setting soft power in motion. The article proposes three conceptual interventions: the first draws a twofold difference-in-kind between soft and hard power, the second develops an understanding that is both relational and diachronic, while the third shifts the focus from an actor-centred to a process-centred analysis. The three interventions coalesce into a common approach to tackle the unresolved problems that takes an agnostic stance towards paradigmatic debates and prioritises inquiries into the mechanisms of interaction between the various forms of power.

Keywords

Soft power hard power mechanisms process International Relations dynamics of contention

Introduction

The concept of soft power has been heavily criticised by International Relations (IR) scholars. ‘Ambiguous’ (Kearn, 2011: 66), ‘elusive’ (Feklyunina, 2016: 774), ‘fuzzy’ (Callahan, 2015: 217), ‘maddeningly inconsistent’ (Layne, 2010: 54), ‘shallow’ (Bohas, 2006) and ‘vague’ (Hayden, 2017: 34) are only a few of the terms used to express the widespread scepticism towards the concept. But if the concept is indeed so discouragingly unfit for academic service, why not dismiss it as some authors suggest (Roselle et al., 2014)? Why save soft power? This article proposes that behind the curtain of obscurity lies a conceptual framework integrating rationalist and constructivist approaches to the study of power in IR, which can provide avenues for thinking beyond the paradigmatic clashes of old. To be sure, the debate between rationalists and constructivists has been highly productive in generating knowledge about the different modalities of power (Baldwin, 2016; Barnett and Duvall, 2005; Guzzini, 2005); however, it is becoming increasingly difficult to make sense of the multitude of approaches and how they hang together. By offering an agnostic stance towards paradigmatic debates, the soft-/hard-power conceptual framework could inform a pragmatic approach that prioritises inquiries into the interplay between the various forms of power. Moreover, setting soft power in motion would feed into the promising endeavour to structure IR research around the study of mechanisms (Bennett, 2013; Checkel, 2015; Goddard and Nexon, 2016) by grappling with the workings of the primary force in world politics: power (broadly conceived). But in order to reveal the hidden value of soft power, its deep-seated problems first need to be addressed.

This article aims to contribute to the discussion of what soft power is and how it works. The argument builds on Nye’s original account and the growing scholarship on soft power in an effort to propose a revised conceptual framework. While these sources provide the groundwork, it is the intellectual import from the contentious politics research programme (McAdam et al., 2001; Tilly and Tarrow, 2015) that sets what is currently a static conceptualisation in motion. More than simply borrowing specific terms and fitting them to a different context, this interdisciplinary exchange proposes an earnest commitment to a more comprehensive approach for studying International Relations. The claim is that the scholarship on soft power can find solutions to the unresolved problems in the field by employing a clear-cut, dynamic and process-centred conceptual framework. Arguably, this would promote a more nuanced understanding of power relations in world politics and enable more fine-grained explanations of outcome trajectories in IR.

As demonstrating an application of the proposed approach goes beyond the limitations of a single article, the text relies on Frank Schimmelfennig’s (2003) acclaimed book on the drivers of European integration as a repository of empirical illustrations for key points. Conceptually, The EU, NATO, and the Integration of Europe is a suitable reference point for the dynamics discussed here, as it shares the underlying ambition of reconciling rationalist and constructivist approaches to the study of IR. Schimmelfennig presents a compelling case for the accommodation of the two logics within the framework of rhetoric action, while this article seeks to explore how this might apply to other modalities of power and what mechanisms condition the reconciliation. Empirically, the enlargement of NATO and the EU, as demonstrated by Schimmelfennig, offers abundant evidence for the use of both soft and hard power, which is highly instrumental in the discussion of their interplay. Taken together, these considerations make the book a particularly well-suited source of illustrations, even though its author has not subscribed to the conceptual framework proposed here.

The article begins with a specification of three unresolved problems of soft-power research, outlining their conceptual origins and methodological implications before introducing the promise of the contentious politics research programme. The discussion proceeds with three consecutive interventions addressing core aspects of the conceptualisation. The first concerns the categorisation of soft and hard power, the second introduces an understanding that is both relational and diachronic, while the third centres the analytical focus on processes of agent–target interaction. The conclusion re-addresses the three problems from the perspective of the revised conceptual framework, outlines its practical implications for the study of IR and highlights the remaining challenges.

Unresolved problems and unspoken promises

There are at least three unresolved problems undermining soft-power research: (1) a differentiation problem; (2) a unit-of-analysis problem and (3) a fixity problem.¹ Firstly, the differentiation problem concerns the indeterminate demarcation between soft and hard power. It is rooted both in the attributes and type of difference that are usually employed in the analysis. As regards definitional attributes, Nye’s ambiguity on the position of resources and behaviours in the distinction between soft and hard power complicates the development of a clear conceptualisation (Nye, 1990, 2004, 2011b). While the practice of bracketing resources and retaining only behaviours as a marker of difference has become commonplace in soft-power research (Gallarotti, 2011; Kearn, 2011; Kroenig et al., 2010; Mattern, 2005; Rothman, 2011), the ramifications of this decision are rarely considered. As to the type of difference, scholars of soft power have unanimously adopted a classification based on a difference-in-degree, whereby instances of power are measured on a spectrum defined by a hard and a soft pole. In the absence of a clear demarcation line, overlaps in the middle section of the spectrum have significantly complicated the distinction between the two types of power. Consequently, the differentiation problem has led authors to conclude that ‘[u]ltimately, tangible resources can deliver both hard and soft power’ (Gallarotti, 2011: 35). This paradox of disparate effects, where a single type of resource generates opposite types of power exercise, undermines the utility of Nye’s taxonomy, particularly with respect to measurement tasks, as it obscures the relations between the categories in the classification.

Secondly, the unit-of-analysis problem refers to the precarious position of the concept of soft power between structure and agency. Narrow actor-centred conceptualisations of soft power have been exposed as insensitive to the complexities of interaction in world politics (Bilgin and Elis, 2008), but the available alternatives (Cheskin, 2017; Kiseleva, 2015) pose significant difficulties in the specification of a unit of analysis for soft-power research. In a framework of productive power (Barnett and Duvall, 2005), the ownership of resources and the behavioural agency play a marginal role, or none at all, as agents themselves are constituted through underlying structures of power relations. According to this view, soft and hard power are seen ‘as interrelated aspects of discursively-realised forms of reality’ (Cheskin, 2017: 4). On the one hand, this proposition encourages an expansion of the scope of soft-power analysis beyond actor-centred understandings, but, on the other, it subverts the relevance of the distinction between soft and hard power, as both are fused together in the crucible of productive power structures. The tension between these two opposing lines of argumentation defines the unit-of-analysis problem and precludes the conceptualisation of the relation between structure and agency in the study of soft power. As a result, scholars have faced difficulties in grappling with the dual nature of attraction: one as a structural condition resembling a resource of power and another as an action representing an exercise of power (Kearn, 2011; Mattern, 2005; Solomon, 2014).

Thirdly, the fixity problem denotes the limitations of static analytical models in the study of dynamic interactions in world politics. Scholars of soft power usually introduce a situation in which a unitary actor with fixed characteristics exerts influence over another such actor in a pursuit of goals defined by a static notion of interest (Goldsmith and Horiuchi, 2012; Holyk, 2011; Köse et al., 2016; Szostek, 2014). Such conceptual frameworks provide little analytical leverage in the study of changes occurring in the composition and characteristics of actors, interests and identities, thereby narrowing the scope of soft-power analysis and limiting its explanatory potential. Static analytical models face a daunting challenge when applied to the analysis of a process, in which sequences of interactions between its elements produce transformative effects in the actors, resources and behaviours of a power relationship. Even if outcomes in terms of changes are analysed, results-as-attitudes (as measured by opinion polls) and results-as-behaviours (as measured by behavioural changes) are often lumped together, raising concerns about measurement errors and endogeneity bias (Pestsov and Bobylo, 2015). As such, the fixity problem highlights the complications of inferring causation in soft-power research.

The three problems share common features and are deeply intertwined. While pertaining to an abstract level, each has concrete methodological ramifications. The differentiation problem induces the paradox of disparate effects, the unit-of-analysis problem obfuscates the dual nature of attraction and the fixity problem impedes causal inference. Furthermore, the three problems are strongly interrelated. For instance, the static understanding engendering the fixity problem obstructs the observation of the complex interplay between structure and agency in world politics, which in turn compounds the unit-of-analysis problem. Equally, the indeterminate demarcation between soft and hard power, denoted by the differentiation problem, undermines the attribution of causal weight to either of the two forces (hard and soft) in the study of world politics, thereby exacerbating the methodological implications of the fixity problem.

The three unresolved problems can be traced back to some widespread understandings in the field, yet at the same time soft-power scholarship already carries many of the intellectual resources necessary for their alleviation. Feklyunina’s (2016: 775–776) nuanced distinction between relational and individual aspects of soft power offers an insightful perspective on the differentiation problem. The development of discrete mechanisms of soft power extend the available knowledge on the workings of the phenomenon (Daßler et al., 2019; Gallarotti, 2011; Kroenig et al., 2010; Roselle et al., 2014; Rothman, 2011), which bears implications for both the unit-of-analysis and fixity problems. Further still, the appeal to abandon essentialist understandings (Bilgin and Elis, 2008; Hall, 2010; Keating and Kaczmarska, 2019; Kiseleva, 2015) facilitates a more critical and context-sensitive approach to the study of soft power, which impinges on aspects of the differentiation and fixity problems. Scholars who employ a more rigorous research design (Goldsmith and Horiuchi, 2012; Holyk, 2011; Hudson, 2015; Köse et al., 2016; Szostek, 2014) provide salient examples of how to deal with the methodological issues related to each of the three problems. Finally, Pestsov and Bobylo (2015) make a crucial contribution by proposing to replace the actor-centred framework with a process-centred one, thereby addressing some of the root causes of the fixity problem.

Notwithstanding the achievements accrued in soft-power research, the three problems persist because: (1) the advancements remain separated by intra-disciplinary divides and (2) some crucial problematic aspects go unnoticed. Firstly, the conceptual accomplishments of many soft-power scholars have remained scattered across the literature and isolated by paradigmatic and linguistic fault lines. By implication, their compatibility has not been explored in the context of a unitary analytical framework. Secondly, despite the inconsistencies of the categorisation based on a difference-in-degree between hard and soft power, it has not been problematised in juxtaposition to an alternative difference-in-kind distinction. Equally, soft-power scholarship has kept faith in static analytical models that rely on unchanging elements bound by linear processes, thus falling short of engaging in an earnest discussion of diachronic processes of interaction.

As solutions within the soft-power literature are scarce, the recourse to other bodies of knowledge becomes a viable and, indeed, a crucial alternative. Soft-power research can benefit immensely from a transfer of knowledge from the contentious politics research programme (McAdam et al., 2001; Tilly and Tarrow, 2015). This is a stream in scholarship on social movements launched by Doug McAdam, Sidney Tarrow and Charles Tilly (MTT) with their seminal book The Dynamics of Contention (DOC). The contentious politics research programme offers a coherent and well-developed analytical framework for the study of phenomena that share many core features with those arousing the interest of soft-power scholars. Firstly, both the contentious politics and soft-power research agendas focus on the interaction among actors. While the units of analysis vary, both bodies of literature investigate encounters between actors and their transformative effects. Secondly, both fields of research concentrate on events that are political. Even if the actors are different, the factors conditioning the event-producing processes are similar, i.e. interests, identities and power. Thirdly, both academic communities are concerned with the availability of resources, the exercise of actions and the generation of outcomes. Whether or not the types of resources, actions (performances) and outcomes diverge, both research programmes share the objective of understanding the interplay between these elements. Finally, borrowing insights from MTT is nothing new to IR (Goddard and Nexon, 2016). This article aims to build on existing contributions and move from the meta-theoretical discussion to the resolution of conceptual problems.

The import from the contentious politics research programme offers two promises to soft-power scholarship: (1) integrating the accrued achievements into a common approach and (2) introducing dynamics to the soft-power analytical model. DOC provides a framework for analysis that has been undergoing refinement for many years now. Arguably, it can fit with the conceptual adjustments that were identified as advances in the soft-power literature and offer insights on how to blend them together into a coherent framework. Furthermore, it devotes special attention to the inquiry into the dynamics of interaction, which is notably lacking in soft-power research. MTT offer a powerful example of integrating relational, diachronic and process-centred aspects in the study of political events that can be used to set soft power in motion.²

In the remainder of the article, the three unresolved problems are targeted with the following interventions:

Categorisation: both resources and behaviours are included as measures of difference and a difference-in-kind approach is applied to the soft/hard power classification;

Dynamics: mechanisms and iterative sequences are introduced to explain the interactions between actors, levels of analysis and types of power;

Process: the analytical perspective is shifted from static actors to dynamic outcome trajectories.

Notably, the delimitation of three separate interventions serves descriptive purposes only. In fact, they are deeply interdependent and coalesce into a common approach. No single intervention can resolve any of the persisting problems individually; it is only through a unitary conceptualisation that they can deliver on their promises to soft-power research.

Categorisation

While the concern about the differentiation problem is shared by many authors (Gallarotti, 2011; Kearn, 2011; Kroenig et al., 2010; Lukes, 2007; Mattern, 2005; Rothman, 2011), few have addressed its sources (Gallarotti, 2011; Rothman, 2011). Arguably, the problem flows from the commonly adopted measures of difference (behaviours only) and the type of difference (difference-in-degree) in the structure of the categorisation. In contrast to the standard reading, this intervention proposes a categorisation based on a difference-in-kind between soft and hard power, drawing on both resources and behaviours as attributes of the definition.

Resources and behaviours

The decision to move beyond resource-centred analytical models is an advancement of soft-power research compared to previous approaches, as it expands the limited study of power potential (as measured by resources) by enabling an inquiry into the actual exercise of power and its effects (Nye, 1990). But discarding resources altogether carries the risk of steering research into another analytical dead-end, where the importance of structural forces is discounted. In this sense, excluding either resources or behaviours from the conceptualisation of soft power would constitute an unnecessary limitation of the analytical framework. Variation in both should be recognised as consequential for the study of power in concert with the view that includes both resources and behaviours as attributes in the distinction between soft and hard power (Nye, 2004: 7). This suggestion comes with a caveat, however, as a more meticulous interrogation of the nature of resources and behaviours is thus far missing in the literature. The ambiguity of Nye’s original account undermines the utilisation of resources and behaviours in the conceptual framework of soft power and underscores the need to revisit the conceptualisation of both terms.

The classification of resources in the soft-power literature is usually understood to revolve around the notion of tangibility:

In general, the types of resources associated with hard power include tangibles such as force and money. The types of resources associated with soft power often include intangible factors such as institutions, ideas, values, culture, and the perceived legitimacy of policies. (Nye, 2011b: 21)

Nye’s specification exhibits a twofold indeterminacy that makes it a tenuous basis for differentiation. Firstly, Nye leaves the relation between the type of resource (tangible vs intangible) and the type of ensuing behaviour (hard-power exercise vs soft-power exercise) imperfect (Nye, 2011b: 21) by repeatedly clouting it in indefinite phrasing such as: ‘tends to’ (Nye, 1990: 168), ‘general association’ (Nye, 2004: 8) and ‘often include’ (Nye, 2011b: 21). This results in an indeterminacy of the connection between resources and behaviours (i.e. what type of resource enables which type of behaviour) undermining the structure of the categorisation. Secondly, the unspecified nature of tangibility itself is a source of indeterminacy in the conceptualisation of soft power. In an illustrative example of this complication, Nye takes for granted the tangible nature of money in the quotation above, although it cannot straightforwardly fit into a category of tangibility that is understood to be defined by an immediate accessibility to the senses. Money, while often possessing a material representation, is a contractual relation of trust that is difficult to reconcile with the idea of tangibility thus conceived.

To avoid these pitfalls, it would be instructive to explore additional attributes to complement the standard reading (‘immediate access to the senses’) in the demarcation between hard-power and soft-power resources. A promising qualification can be found in a key resource attribute referring to the level of access to and control over a resource by an actor in its possession, described by Edwards and McCarthy (2004: 130–131) as ‘proprietarity’. The concept can be utilised in opposition to a relational status (denoting a low level of access and control), so as to sharpen the specification of the kind of variation captured by the differentiation between tangible and intangible resources. Proprietary resources have an existence independent of a target actor’s acknowledgement, i.e. they function on a material level. In this sense, their transfer is straightforward and agents possess immediate control over their utilisation (Edwards and McCarthy, 2004: 130). Relational resources, on the other hand, are existent only under the condition of a target actor’s recognition and emerge as a product of meaning construction, i.e. they operate on the ideational level. As such they are difficult to transfer or convert and are not under the immediate control of the agent.

The argument is that some resources can emerge only to the extent that target actors attribute meaning to them. This can be illustrated through Schimmelfennig’s observation that

[i]n NATO and the EU, the actors interested in enlargement used the liberal identity, values, and norms of the Western international community to put moral and social pressure on the reluctant member states and shamed them into acquiescing to the admission of CEECs [Central and East European Countries]. (Schimmelfennig, 2003: 193, emphasis added)

In this example, the enlargement proponents would not have been able to achieve the desired effect if their targets (enlargement opponents) did not recognise the Western ‘liberal identity, values, and norms’ as a meaningful reference point, thereby transforming them into an intangible resource. In contrast, the existence of tangible resources is not contingent on the recognition of target actors. As such, when ‘the new members increased the absolute power resources of NATO in terms of population, territory, economic output, and defense capabilities’ (Schimmelfennig, 2003: 43) they did so without requiring any external certifying authority, i.e. NATO was in a position to use these tangible resources whether or not their existence was acknowledged by others. In this sense, the distinction between tangible and intangible resources can be recast through recourse to their character status (proprietary vs relational), so that: (1) tangible resources are independent of target recognition and (2) intangible resources are dependent on target recognition.

The additional qualification can offer a better insight as to why financial resources fall into the category of tangible resources in Nye’s example. Money is quantifiable and ‘quite proprietary’ (Edwards and McCarthy, 2004: 130) just as any tangible resource. It can be stored or transferred in and between locations, it can be accumulated or depleted over time and it can be straightforwardly converted into other resources. While it does represent a formalised institutionalisation of relations of trust between actors, in a development that stretches back to the beginning of written history, money has established itself as an integral part of social practice to an extent that its relational origins have been substituted with an ‘as-if’ material existence. The clear proprietary character combined with the many key features shared with material (i.e. accessible to the senses) resources offer a basis to suggest that money indeed belongs to the category of tangible resources.

Regardless of the conceptual attributes utilised in their distinction, the nature of resources generally remains the same, i.e. they denote the conditions in and through which actors operate. The availability of resources creates opportunities for actors to pursue desired outcomes, while their deficiency constrains the scope of possible action. Resources can be both the vehicles through which actors arrive at their objectives, and the structures conditioning an actor’s trajectory towards those objectives. The double character of resources transpires the distinction between their (in)tangibility. In this sense, tangible resources can be both vehicles (e.g. aircraft carriers) and structural factors (e.g. nuclear weapons) in power relationships, just as intangible resources can represent both the tool (e.g. narrative) and environment (e.g. identity) of power interaction.

The classification of behaviours receives more attention in Nye’s work, which is understandable given that it is considered the primary marker of difference between soft and hard power. In The Future of Power, Nye presents his most complete account:

The use of force, payment, and some agenda-setting (. . .) I call hard power. Agenda-setting that is regarded as legitimate by the target, positive attraction, and persuasion are the parts of the spectrum of behaviors I include in soft power. Hard power is push; soft power is pull. Fully defined, soft power is the ability to affect others through the co-optive means of framing the agenda, persuading, and eliciting positive attraction in order to obtain preferred outcomes. (Nye, 2011b: 20–21)

Building upon the insistence on a differentiation between power capacity and its exercise, which his critics arguably overlook (Nye, 2011b: 20), Nye discerns hard-power (‘use of force’, ‘payment’ and ‘some agenda-setting’) and soft-power behaviours (‘persuading’, ‘framing the agenda’ and ‘eliciting positive attraction’). Hard-power behaviours are grouped together under the label ‘command’, whereas soft-power behaviours revolve around the notion of ‘co-optation’ (Nye, 2011b: 21).

Nye frames behaviours as a pre-given exhaustive list of actions available to actors, but this view could be misleading. Innovations in observed behaviour become problematic for the analysis, if agents are expected to engage in a known set of behaviours only. There is a tension between agent actions on the ground and the list of available behaviours, as the former may not fully conform to the latter. By conflating these two aspects, the concept of behaviours does not provide the tools to make the distinction. In this sense, there are advantages to conceiving behaviours in terms of ‘performances’ and ‘repertoires’. Tilly and Tarrow define performances as ‘relatively familiar and standardized ways in which one set of political actors makes collective claims on some other set of political actors’ (Tilly and Tarrow, 2015: 14) and repertoires as ‘arrays of performances that are currently known and available within some set of political actors’ (Tilly and Tarrow, 2015: 14). A performance is the action or inaction that occurs in practice, while a repertoire represents the set of performances that are typically associated with a particular scenario. In DOC, the authors highlight that ‘[p]erformances within repertoires do not usually follow precise scripts to the letter’, as ‘small-scale innovation modifies repertoires continuously’ (McAdam et al., 2001: 138). Hence, repertoires include both the structure of available possibilities for action and the improvisations that adapt them to particular situations.

Following this approach, the development of the Partnership for Peace (PfP) initiative in the 1990s could be interpreted as an innovative performance within NATO’s repertoire of cooperation (agenda-setting). As Schimmelfennig (2003: 186) notes, the PfP ‘deepened and institutionalized cooperation with all CEECs, involved them in the task of peacekeeping, and burdened them with the costs of making their forces interoperable with those of NATO’ – as any standard membership agreement would – ‘without, at the same time, creating any formal obligations to guarantee their security, antagonizing Russia, or weakening the cohesion and effectiveness of the alliance’. The latter qualification describes a performance that adapts the existing repertoire to the requirements of the particular interaction, thus departing from previous practice without being independent from it. Applied to the analysis of power, repertoires enable a shift from the category of behaviours, which is associated with a tradition in the social sciences emphasising the analytical autonomy of rational, self-propelled actors, to a term, which is well-suited to illuminate the interplay between structure and agency. Employing the concept of repertoires challenges researchers to think about performances in terms of their relation to structural opportunities and restraints, rather than in terms of strategic considerations alone.

Difference-in-kind

The question about the type of distinction between soft and hard power is intertwined with the one about its markers of difference. Soft-power scholarship generally takes only behaviours as a determinant, which generates a difference in degree (Gallarotti, 2011; Kearn, 2011; Kroenig et al., 2010; Mattern, 2005; Nye, 2011b; Rothman, 2011). A spectrum, as the one proposed by Nye (2011b: 21), is a unidimensional differentiation defined by variation on a single factor (Baldwin, 2016: 166). If both resources and repertoires are taken into account, then the construction of a categorical distinction becomes feasible. A difference-in-kind would in this case represent a two-dimensional differentiation defined by variation on two factors.

Nye’s categorisation is often juxtaposed to the classification that developed out of the power debate in social theory in the 1960s and 1970s, perhaps best encapsulated in the work of Steven Lukes (2005). Lukes builds a difference-in-kind taxonomy that clearly distinguishes between three dimensions (or faces) of power: the first focusing on overt power interaction on the decision-making arena (Lukes, 2005: 16–19); the second – on the covert factors that condition which issues enter the decision-making process and which ones fall aside (Lukes, 2005: 20–25); and the third – on the latent forces that elicit willing consent to domination (Lukes, 2005: 25–29). Despite the discrepancies between Nye’s and Lukes’ approaches, there has been a recurring attempt to square the circle by fitting a unidimensional spectrum scale onto a multidimensional categorical one by equating hard power to the first face of power and soft power – to the second and third faces of power (Bilgin and Elis, 2008; Gallarotti, 2011). The two designs, however, do not match, as they rely on distinct markers of difference and employ a different number of definitional attributes.

By arguing that ‘soft power fits with all three faces or aspects of power behaviour’, Nye (2011b: 90) suggests that the two typologies are complementary, rather than equivalent. According to Nye, each of the faces has a soft and a hard aspect (see Table 1). Nye specifies further that the first face of soft power relates to persuasion, the second – to framing the agenda, and the third – to eliciting attraction (Nye, 2011b: 91–94). He is less concrete about the exact hard-power performances that fit into each of the three faces, but suggests that they exist (Nye, 2011b: 40–42, 70–79).

Table 1.

Compatibility grid of power typologies (based on Nye [2011b: 91, Table 4.1]).

Lukes’ typology	Nye’s typology
Lukes’ typology	Soft power	Hard power
First face	‘A uses attraction/persuasion to change B’s existing preferences’ (Nye, 2011b: 91) (persuasion)	‘A uses force/payment to change B’s existing strategies’ (Nye, 2011b: 91) (?)
Second face	‘A uses attraction or institutions so that B sees the agenda as legitimate’ (Nye, 2011b: 91) (agenda-framing)	‘A uses force/pay to truncate B’s agenda’ (Nye, 2011b: 91) (?)
Third face	‘A uses attraction and/or institutions to shape B’s initial preferences’ (Nye, 2011b: 91) (eliciting attraction)	‘A uses force/pay to shape B’s preferences’ (Nye, 2011b: 91) (?)

Source: Reproduced with permission from Public Affairs, an imprint of Hachette Book Group Inc., Nye JS Jr (2011) The Future of Power.

Nye’s combination of the two typologies is instructive, as it identifies repertoires that can be grouped together under the categories of soft and hard power, thereby facilitating the systematisation of their analysis. There remains, however, an important caveat. The faces typology rests on multiple markers of difference, one out of which directly relates to the repertoires of power: the type of conflict between actors (overt vs covert vs latent) (Lukes, 2005: 29). Nye’s classification, on the other hand, is defined by a single determinant – behaviours (repertoires) – when conceived as a difference-in-degree. The common marker of difference allows for the juxtaposition of the two frameworks. However, there needs to be another factor that specifies the demarcation between hard and soft types of repertoires that fall into each of the three faces of power; otherwise the distinction between the frameworks would collapse. As it stands, Nye implicitly employs the opposition between ‘force/payment’ and ‘attraction/persuasion’ to account for the variance, but does not explicate the difference between them. As such, there must be some shadow marker of difference that underlies the typology but is not made apparent.

Recognising resources as the second attribute in Nye’s classification offers a tentative solution to this conceptual inconsistency. As already suggested, resources come in different types depending on their (in)tangibility. The diverse nature of resources gives rise to categorically different repertoires: (1) intangible resources enable soft-power repertoires, while (2) tangible resources enable hard-power repertoires. This intervention makes the combination between the faces’ typology and Nye’s categorisation tenable by exposing resources as the shadow marker of difference. In this sense, force and pay are different from persuasion and attraction, as the former pair grows out of tangible resources, while the latter is rooted in intangible resources. Employing two attributes in the distinction transforms the type of classification from a continuum, based on a difference-in-degree, to a 2 × 2 grid, based on a difference-in-kind (see Figure 1).

Figure 1.

Difference-in-kind categorisation of soft and hard power.

The categorisation of power resulting from the intervention covers varying levels of analysis. Both types of resources (tangible and intangible) are compounded by structures and vehicles. The structural conditions broadly define the realm of what is possible, while the vehicles allow agents to navigate the landscape in pursuit of their objectives. Similarly, both soft-power and hard-power repertoires comprise of pre-existing performance templates pertaining to different dimensions of abstraction (first, second and third face). Whether agents decide to follow or rewrite the available script depends on negotiations about what is desirable (likely outcomes) and what is possible (available resources), yet this type of strategic interaction is invariably embedded in a panoply of previous decisions, ideas and outcomes (McAdam et al., 2001: 141). In any case, the emphasis of the relation between structure and agency falls on the maintenance of equilibria. An overly structuralist view of power raises the problem of unattributable agency, which obviates the distinction between soft and hard power (differentiation problem), while an excessive focus on a single actor renders the analytical framework unfit to study interactions (unit-of-analysis/fixity problem). Instead of adopting a priori assumptions about the pre-eminence of either structural or strategic factors, the proposed approach turns the tables by centring the analysis on investigating the interplay between the two.

Dynamics

Introducing a difference-in-kind categorisation based on resources and repertoires does not in itself resolve the differentiation problem, as it provides a static picture that cannot capture the dynamic interdependence between the observable referents of the categories. Furthermore, it remains unclear how the constitutive link between the type of resource and the kind of power exercise can be explained. And for that matter, if tangible resources are indeed constitutive of hard-power repertoires alone (and intangible resources enable soft-power repertoires only), then how can the paradox of disparate effects be accounted for? As it stands, the proposed categorisation is at odds with the typical example: ‘[a] tangible hard power resource like a military unit can produce both command behavior (by winning a battle) and co-optive behavior (attraction) depending on how it is used’ (Nye, 2011a: 19). The first intervention also touches upon the unit-of-analysis problem by identifying resources on different levels of analysis (vehicles and structures) and by replacing the actor-centred concept of behaviours with the polyvalent category of repertoires. The problem, however, remains unresolved, as the mechanisms that drive the interaction between the various levels of analysis have not yet been specified. Finally, the previous intervention offers no solutions that can alleviate the fixity problem, as it proposes a static perspective on the distinction between soft and hard power.

Some of the persisting complications are rooted in the stationary understanding of soft power. While the origins of the static model can be traced back to Nye’s (2011b) original conception of how soft power works, its establishment as a standard in the literature has been marked by an unchallenged re-iteration by soft-power scholars. Even authors who expand Nye’s soft-power model by identifying complex mechanisms (Daßler et al., 2019; Gallarotti, 2011; Kroenig et al., 2010; Roselle et al., 2014; Rothman, 2011) acquiesce to the idea of an unchanging process connecting the resources and effects of soft power. Nye describes two types of process: direct and indirect (Nye, 2011b: 94–95). The model of direct effects works in an unmediated way between an agent activating resources and the government of a target actor (resources → government elites → attraction → elite decision), while the model of indirect effects takes a detour through the general public of the target actor before reaching its government (resources → publics → attract/repel → enabling/disabling environment → elite decision). In both cases, the scheme represents a one-way street connecting unchanging soft-power resources to singular outcomes.

While the static model (in both its direct and indirect versions) is valuable in terms of identifying important elements of a power relationship that should serve as a point of departure for research, it is limited in what it can reveal and explain. Firstly, the static model is ill-equipped to capture crucial transformations occurring within its boxes. Taking elements such as ‘resources’ and ‘government elites’ as given (i.e. static), it reproduces the fallacies of resource-centred theories that Nye originally sought to overcome. To remedy this situation, the analysis needs to be sensitised to the particularities of both the actors and the sites of interaction. Power is neither uniform across relationships, nor autonomous from history and location (Baldwin, 2016; Guzzini, 2005). A more pronounced relational approach would develop an expectation for transformation that challenges the unchanging status of the elements in the model and centres analysis on the processes that condition the contents of the boxes in the model.

Secondly, the standard soft-power model neglects the relation between hard and soft power. To be sure, Nye does consider the interplay between the two (Nye, 2004: 25–30), yet these considerations are conspicuously absent from the specifications of his two models (Nye, 2011b: 94–95). Similarly, scholars of soft power generally recognise the interdependence between the two types of power in empirics, but, with rare exceptions (Daßler et al., 2019), focus their analytical models on soft power alone (Gallarotti, 2011; Holyk, 2011; Köse et al., 2016; Kroenig et al., 2010; Roselle et al., 2014; Rothman, 2011). But if hard and soft power are indeed considered to be ‘inextricably intertwined in today’s world’ (Nye, 2004: 30), the model used for their study would certainly have to capture their interplay. This does not mean that soft and hard power share a common existence, but that they operate in parallel in the same context. Providing the tools to capture the nuances in the relation between the two types of power is crucial for the development of a framework that lends itself to application in empirical studies.

The reinforced relational emphasis on the transformation between elements in the model and on the interplay between soft and hard power provides a foundation for a dynamic understanding of how soft power works. But in order to set the conceptual framework in motion, the relational emphasis needs to be supplemented by a diachronic account of the interactions between agent and target. Soft-power scholarship can benefit in this endeavour from the insights informing MTT’s criticism of a similar ‘static, cause-free single-actor model’ (McAdam et al., 2001: 18). MTT present a framework that ‘shows a clear move away from static variables to dynamic mechanisms’ (McAdam et al., 2001: 50) and that substitutes the ‘objective accounting’ of categories with a ‘dynamic analysis’ of the transformative processes that shape them (McAdam et al., 2001: 50). MTT substantiate the departure from the static model by emphasising two key aspects of their object of study – interactive sequences and mechanisms – which arguably hold the key to setting soft power in motion.

The static difference-in-kind categorisation (see Figure 1) can be expanded into a dynamic model (see Figure 2) by introducing activation and feedback mechanisms to propel its boxes. Tilly and Tarrow define mechanisms as ‘a delimited class of changes that alter relations among specified sets of elements in identical or closely similar ways over a variety of situations’ (Tilly and Tarrow, 2015: 29). Firstly, activation mechanisms are transmissions mobilising the capacity accumulated as resources and enabling agent performances. For example, ‘boundary activation’ (Tilly and Tarrow, 2015: 36) can be understood to explain an important development in NATO’s Eastern enlargement. The introduction of membership criteria in the early 1990s served to establish a hierarchy among aspiring states, which had previously shared a common identity of states ‘returning to Europe’ (Schimmelfennig, 2003: 230). This structural development activated dormant resources for Poland, Hungary and the Czech Republic, thus enabling them to perform persuasion: ‘the best-qualified candidates and the most likely new members of NATO (. . .) called for differential treatment of the CEECs: they did not want to see their admission delayed by being put in the same category as the more problematic candidates in the region’ (Schimmelfennig, 2003: 234). That is, these three countries ‘were more advanced than other CEECs’ (Schimmelfennig, 2003: 94), but this intangible resource acquired relevance only when the differentiation of CEECs was conceived in terms of membership criteria, thus triggering the boundary activation mechanism.

Figure 2.

Diachronic categorisation of soft and hard power.

Secondly, feedback mechanisms represent a loop connecting performances to the availability of resources in subsequent interactions. Extending the above example, the exercise of persuasion could be seen to bear implications on the agent’s inventory of intangible resources in a next iteration of agent–target interaction. As such, through a mechanism of ‘certification’ (McAdam et al., 2001: 121) the claim to excellence of Poland, Hungary and the Czech Republic was acknowledged by NATO member states, who recognised them as ‘the forerunners and paragons of liberalization and democratization in the region’ (Schimmelfennig, 2003: 95), thereby enhancing their image, which was crucial for their early accession to NATO. On the other hand, the successful performance of persuasion required the commitment of more tangible resources on the trio’s side in order to cover the costs of integration, thus triggering a mechanism of depletion: ‘[s]tudies on the cost of enlargement calculated that this upgrading [of military equipment] would come to between US $10 and $42 billion to be paid mainly by the new members’ (Schimmelfennig, 2003: 39). Of course, these costs were then converted to military resources (through accumulation) that became available in the next stage of the process and improved their subsequent capacity to engage in performances within the hard-power repertoire.

These are no more than illustrative examples that should not be taken as exhaustive accounts, but rather as indications of how the categories relate to each other in successive iterations. What should be noted is that activation mechanisms are conditioned by specific resource endowments, i.e. a particular assemblage of intangible resources may enable the triggering of a soft-power mechanism, but not of a hard-power one. Nonetheless, soft-power performances enabled by that same assemblage of intangible resources can, through a feedback mechanism, condition the structure of interaction in a way that a hard-power activation mechanism is triggered at a later stage. Equally, tangible resources condition mechanisms that enable hard-power repertoires only, yet through a feedback mechanism command performances could affect the agent’s endowment with both intangible and tangible resources.³

A diachronic model of soft power that accounts for successive iterations is better placed than a static one to make sense of the paradox of disparate effects. In a reference back to Nye’s example, both static and diachronic perspectives can capture how the hard-power resource (military unit) conditions command performances (use of force). Yet, in the other case (a military unit eliciting attraction), only a diachronic account can explain how a tangible resource enables a soft-power performance without abandoning the clear-cut categorisation of soft and hard power. By adopting this view, it can be revealed that the static understanding distorts the observation by obscuring a transformation process that includes an interactive sequence comprising at least the following stages: (1) command performance (use of force); (2) feedback mechanism (e.g. certification); (3) enhanced intangible resources (e.g. positive image); (4) activation mechanism (e.g. boundary activation) and (5) co-optive performance (e.g. persuasion). Recognising this winding trajectory shows that the paradox of disparate effects conceals a chain of time-bound interactions between actors (agent and target) and between powers (hard and soft). The chain appears convoluted when displayed as a static snapshot but can be disentangled if shown as a motion picture.

Moreover, the dynamics introduced with this intervention can help address the methodological challenge posed by the dual nature of attraction (Bilgin and Elis, 2008; Kearn, 2011; Kiseleva, 2015; Mattern, 2005; Solomon, 2014; Szostek, 2014). Recognising the two forms tangible and intangible resources take – vehicles and structures – allows for a reading that can accommodate the notion of attraction as a feature of the environment in which an agent operates (i.e. a structure), while the performances comprising soft-power repertoires are well-suited to capture actions directed at eliciting attraction. The activation and feedback mechanisms conceptualise the transformations between the two. Attraction, indeed, has a dual nature – one as condition and another as action – but this need not impede research, nor should it necessitate its bifurcation (compare Mattern, 2005; Solomon, 2014). Instead, the analysis can trace the iterative process of transformation from attraction-as-condition to attraction-as-action (and back again), thereby embracing the dual character of the phenomenon.

Process

While the previous intervention touches upon the fixity problem on the conceptual level by setting the categorisation of soft power in motion, its key methodological challenge is still standing. The research task of attributing causation to either soft- or hard-power exercise requires that the proposed model traces the trajectories of both forces through the entire process of outcome generation. As it stands, however, the mechanisms of interaction between soft and hard power remain underspecified and the concept of outcome is yet to make a formal appearance.

Completing a workable model of the soft-power process would require that outcomes comprise a link in its chain, as they denote the important junctures which delimit one sequence of interaction from the other. However, the diachronic categorisation of power (see Figure 2) omits the category of outcomes altogether and appears to assume that feedback mechanisms take their place. Such an assumption, however, would constitute a misconception of the nature of mechanisms. Mechanisms are illustrated with arrows in the model, not with boxes. They are distinct from resources, repertoires and outcomes, which define the stepping stones in the process and which can be captured in snapshot descriptions. A resource endowment can be observed in the context of a time and a location. The same goes for performances and outcomes, but not for mechanisms, as they are transmissions between elements in a relationship that are elusive to observation. While evasive, mechanisms can be identified through the transformations they denote, which are ‘identical or closely similar’ across cases (Tilly and Tarrow, 2015: 29). For instance, the feedback mechanism of certification as a rule describes the same kind of transformation, regardless what activated it and whether or not the generated intangible resource was utilised in a next stage. In this sense, a mechanism is not a box to be filled with contents from the studied episode, but a cogwheel performing the same function regardless of the process it sets in motion. Therefore, a mechanism should not be confused for an outcome in the model, even if it immediately follows an action, as a mechanism designates a change, while an outcome – a state.

Hence, the model proposed in Figure 2 would be unfit for the study of power, unless it included a separate element in the process that captures outcomes. It was already suggested that soft and hard power retain a separate existence, but operate on the same site to produce outcomes in isolation from or in combination with each other. This qualification would imply that the mechanisms connecting repertoires to outcomes are different from the ones linking resources to repertoires. The latter (resources to repertoires), it is argued, denote a straightforward transformation, whereby intangible resources are activated to produce soft-power performances or, alternatively, tangible resources are mobilised to engender hard-power performances. The former (repertoires to outcomes), however, need to interact in some way to produce outcomes that combine soft- and hard-power effects, as world politics is replete with outcomes carrying traces of both types of power (Nye, 2004: 25–30). To pass the test of empirics, the conceptual framework requires a category of outcome that reconciles the effects of soft and hard power, as well as a mechanism that connects repertoires to the outcomes that they cause.

As already suggested, feedback mechanisms can connect soft- and hard-power repertoires to the availability of either type of resource (tangible and intangible). The same process can be assumed to operate between repertoires and outcomes: soft- and hard-power mechanisms concatenate in different combinations to produce the outcomes observed in world politics. These action-formation mechanisms connect repertoires to outcomes in a variety of ways, so that signing an EU membership agreement can be the result of: (a) hard-power performances alone: economic incentives compelling a target actor to accept an agent’s terms (Schimmelfennig, 2003: 52); (b) soft-power performances only: a narrative of cultural proximity convincing a target country to sign the treaty (Schimmelfennig, 2003: 90) or (c) both: combining the resonance of a narrative of cultural proximity and the incentive of economic gains (Schimmelfennig, 2003: 279).

But if the same process is operational between repertoires and resources as the one between repertoires and outcomes, where does the difference between resources and outcomes lie? This question raises concerns about endogeneity in the model and exacerbates the problem of causal inference. Resources condition the interactions that cause outcomes and as such precede them in the causal chain. If a resource is mistaken for an outcome, then the conceptual order in the causal inference would be compromised, which in methodological terms would amount to endogeneity bias. Equally, if an outcome is mistaken for a resource, it would mean that there is a temporal inconsistency in the research design, which would undermine any claim to causal inference.

The third conceptual intervention represents a commitment to and a revision of the process-centred approach proposed by Pestsov and Bobylo (2015) and can best be illustrated in the treatment of the causal inference problem as specified by the unclear distinction between resources and outcomes. Pestsov and Bobylo (2015: 111–112) develop a model that goes beyond the standard account by shifting the analytical focus from the agent to the process of agent–target interaction. However, their model reproduces key shortcomings by disregarding the transformative dynamics of iterative sequences and soft-hard power interaction.

Somewhat controversially, a more committed process-centred framework would suggest there is no difference between resources and outcomes, at least not in the strict sense. Conceiving of power relations in terms of iterative sequences of interactions carries the implication that every outcome completing a process represents the springboard for the next one. The sum of all intended and unintended effects produced by an agent’s interactions (with structures and targets) represents that agent’s resource endowment in the next sequence of interaction. Adopting this view would mean there is no ‘final outcome’ (Pestsov and Bobylo, 2015: 112); however, it does make sense to employ certain outcomes as borderlines isolating a distinct process within a broader episode for closer analysis (McAdam et al., 2001: 27–28). In the example where an agent influences a target actor to sign a treaty, there can be multiple intermediary outcomes along the way (positive changes in public opinion in the target country, budget deficits, etc.), but the outcome that matters for the particular study is the signed treaty. Understood like this, any outcome can assume the status of a resource in the next iteration, but the signing of the treaty is the outcome completing a process that defines a particular research design. By implication, the demarcation between resources and outcomes becomes an analytical task exercised in every individual study.

Assigning the labels ‘resource’ and ‘outcome’ carries important methodological implications. In a research design centred on tracing the causes of effects (Mahoney and Goertz, 2006; Mahoney, 2010), designating a phenomenon to the outcome category absolves it from the requirement to undergo a categorisation as a resource. As the focus of analysis is concentrated on the process leading up to the outcome, it need not be disambiguated into tangible and intangible resources.⁴ For instance, a signed treaty is a representation of neither type of power individually, as it can be comprised of building blocks which pertain to both categories: (1) it is an indication of a shared value system like that of the Western liberal community of states (i.e. an intangible resource) and (2) it can define a certain quantity of tangible resources to be obtained by a party to the treaty like under the Structural and Cohesion Funds of the EU (i.e. a tangible resource). This classification would concern the analysis if the membership treaty would represent a starting or an intermediate point in a process leading to another outcome of interest, but for a study investigating its underlying causes the treaty is a fusion of soft- and hard-power effects. Either type of effect may be prevalent in the outcome of interest and since it is not subjected to a disambiguation into different types of resources, a difference-in-degree approach would be applicable. This would not, however, absolve the analysis from the requirement to trace the causes (soft and/or hard power) explaining that effect.

By introducing outcomes and action-formation mechanisms (see Figure 3), the move to a process-centred analytical perspective can be completed by shifting the focus of analysis onto several necessary features: (1) an agent, (2) a target actor and (3) successive interactions between the two. The move also implies that the interactions between: (1) agent and target, (2) soft and hard power and (3) different levels of analysis, represent a recurring theme relevant to all elements in the model (resources, repertoires, outcomes and mechanisms).

Figure 3.

Dynamic process-centred model of soft power.

Almost none of the nuances discussed above would be accessible from a vantage point that centres on a single unitary actor. Intangible (i.e. relational) resources and their associated mechanisms would be inconceivable, at worst, and indeterminate, at best; agent interactions with targets and structures would be entirely discarded, at worst, and occasionally interesting, at best; and the difference between soft and hard power would be irrelevant, at worst, and obscure, at best, to name a few of the ramifications related to preserving an actor-centred model. In contrast, a clear-cut, dynamic and process-centred model would provide analytical leverage to trace the parallel outcome trajectories of soft and hard power in world politics.

Conclusion

The proposed model arguably addresses each of the three identified problems of soft-power research on the conceptual level. The differentiation problem is alleviated with the introduction of a sharp distinction between soft and hard power, based on a difference-in-kind enabled by two definitional attributes and propelled by successive iterations. The requirement to decide for either structure or agency imposed by the unit-of-analysis problem is recast as a research task to investigate their relative importance in different settings using the tools of the dynamic categorisation. The fixity problem is rectified by the interventions that set the analytical model in motion and instil a transformative expectation in all of its elements.

Tentative solutions are also suggested for the methodological implications of the three problems. The differentiation problem’s paradox of disparate effects was revealed to be a parallax, which can be remedied by observation through the perspective of interactive sequences. Rather than being dismissed as an impediment that stifles research, the dual nature of attraction compounded by the unit-of-analysis problem has been embraced as an opportunity to enrich analysis by interrogating the relation between structural and strategic aspects of attraction. Finally, the problem of causal inference based on the existing actor-centred models is remedied by a shift in analytical focus from static actors to dynamic processes that trace the interplay between soft and hard power.

The proposed conceptual framework is rooted in Nye’s original terminology, incorporates breakthroughs achieved in the literature on soft power and relies on DOC as the scaffolding enabling its reconstruction, while borrowing material from Schimmelfennig for illustration. But the resulting conceptualisation is more than a sum of its parts. A synergetic effect sharpens each of its building blocks. Nye’s model of soft power is reinforced with a setup accommodating the transformative effects within and between its elements. Pestsov and Bobylo’s process-centred approach is set in motion to bring it up to speed with dynamic interactions in world politics. The contentious politics analytical framework is embedded into a clear-cut categorisation to enable more nuanced accounts of the types of trajectories that shape power relations among actors. Finally, the conceptual framework proposes to apply Schimmelfennig’s example of reconciling rationalist and constructivist logics (in the context of rhetorical action) to a wider range of power modalities and to structure the analysis around its underlying mechanisms.

The practical merit of the proposed approach lies in providing an opportunity to overcome some of the existing paradigmatic limitations to the study of power in IR. By encouraging an engagement with the complex interplay between different forms of power and levels of analysis, the dynamic framework puts research on a more pragmatic basis that prioritises comprehensive explanations over paradigmatic purity. As such, it offers a means to bridge divides between scholars of IR. This is achieved through the promotion of mechanisms to the centre stage of the analysis, which allows the sequencing of explanatory logics without sacrificing the conceptual consistency of the research design. In this way, scholars of different theoretical backgrounds can contribute to each other’s work by uncovering diverse mechanisms that can then connect into complex causal chains explaining outcomes of mutual concern.

Nonetheless, at least two crucial caveats remain. Firstly, the conceptual framework represents a simplification. All empirical processes are embedded in contexts of overwhelming complexity that cannot be fully grasped by the framework without sacrificing its intelligibility. But equally, it is imperative that the proposed concise list of mechanisms is both expanded and expounded, if the conceptual framework is to be of use to scholars across the discipline. Secondly, the proposed conceptual framework does not represent a theory. It identifies key elements of power relationships in IR and establishes how they fit together, but it does not specify any patterns of interaction between the categories, neither does it stipulate any model explanations of outcomes. Further effort is required in order to infuse the conceptualisation with theoretical expectations about the ways in which causal effects are transmitted through the boxes and arrows in the model. But demanding as these challenges are, they also serve as an indication that the concept of soft power can offer a viable research agenda and thus merits further attention.

Footnotes

Acknowledgements

I am deeply grateful to Jan Matti Dollbaum, Valentina Feklyunina, Juan Masullo, Peter Mayer, Heiko Pleines and Gergana Vaklinova for their thoughtful and constructive criticism on earlier versions of this article. I would like to thank audiences at the BIGSSS doctoral colloquium, the FRRESH 2018 Summer School in Orilampi and the panel on the repertoires of power politics at the 2018 EISA Pan-European Conference in Prague for providing useful comments. I am grateful to the anonymous reviewers at the European Journal of International Relations for their feedback. All remaining shortcomings and errors are entirely my own.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the H2020 Marie Skłodowska-Curie Actions under grant agreement No [713639].

ORCID iD

Ivan Bakalov

Notes

Author biography

Ivan Bakalov is a COFUND Marie Skłodowska-Curie PhD Fellow at the Bremen International Graduate School of Social Sciences, University of Bremen | Jacobs University. The focus of his research is on studying the concept and theory of power in international relations. His empirical work concerns foreign policymaking in Eastern Europe.

References

Bakalov

(2019) Whither soft power? Divisions, milestones, and prospects of a research programme in the making. Journal of Political Power 12(1): 129–151.

Baldwin

(2016) Power and International Relations: A Conceptual Approach. Princeton: Princeton University Press.

Barnett

Duvall

(2005) Power in international politics. International Organization 59(1): 39–75.

Bennett

(2013) The mother of all isms: causal mechanisms and structured pluralism in international relations theory. European Journal of International Relations 19(3): 459–481.

Bilgin

Elis

(2008) Hard power, soft power: toward a more realistic power analysis. Insight Turkey 10(2): 5–20.

Bohas

(2006) The paradox of anti-Americanism: reflection on the shallow concept of soft power. Global Society 20(4): 395–414.

Callahan

(2015) Identity and security in China: the negative soft power of the China dream. Politics 35(3–4): 216–229.

Checkel

(2015) Mechanisms, process, and the study of international institutions. In: Bennett

Checkel

(eds) Process Tracing: From Metaphor to Analytic Tool. Cambridge: Cambridge University Press, 74–97.

Cheskin

(2017) Russian soft power in Ukraine: a structural perspective. Communist and Post-Communist Studies 50(4): 277–287.

10.

Daßler

Kruck

Zangl

(2019) Interactions between hard and soft power: the institutional adaptation of international intellectual property protection to global power shifts. European Journal of International Relations 25(2): 588–612.

11.

Edwards

McCarthy

(2004) Resources and social movement mobilization. In: Snow

Soule

Kriesi

(eds) The Blackwell Companion to Social Movements. Malden: Blackwell Publishing Ltd., 116–152.

12.

Elster

(2005) A plea for mechanisms. In: Hedström

Swedberg

(eds) Social Mechanisms: An Analytical Approach to Social Theory. New York: Cambridge University Press, 45–73.

13.

Feklyunina

(2016) Soft power and identity: Russia, Ukraine and the Russian world(s). European Journal of International Relations 22(4): 773–796.

14.

Gallarotti

(2011) Soft power: what it is, why it’s important, and the conditions for its effective use. Journal of Political Power 4(1): 25–47.

15.

Goddard

Nexon

(2016) The dynamics of global power politics: a framework for analysis. Journal of Global Security Studies 1(1): 4–18.

16.

Goldsmith

Horiuchi

(2012) In search of soft power: does foreign public opinion matter for US foreign policy. World Politics 64(3): 555–585.

17.

Guzzini

(2005) The concept of power: a constructivist analysis. Millenium: Journal of International Studies 33(3): 495–521.

18.

Hall

(2010) An unclear attraction: a critical examination of soft power as an analytical category. The Chinese Journal of International Politics 3(2): 189–211.

19.

Hayden

(2017) Scope, mechanism, and outcome: arguing soft power in the context of public diplomacy. Journal of International Relations and Development 20(2): 331–357.

20.

Holyk

(2011) Paper tiger? Chinese soft power in East Asia. Political Science Quarterly 126(2): 223–254.

21.

Hudson

(2015) ‘Forced to friendship’? Russian (Mis-)understandings of soft power and the implications for audience attraction in Ukraine. Politics 35(3-4): 330–346.

22.

Kearn

(2011) The hard truths about soft power. Journal of Political Power 4(1): 65–85.

23.

Keating

Kaczmarska

(2019) Conservative soft power: liberal soft power bias and the “Hidden” attraction of Russia. Journal of International Relations and Development 22(1): 1–27.

24.

Kiseleva

(2015) Russia’s soft power discourse: identity, status and the attraction of power. Politics 35(3–4): 316–329.

25.

Koopmans

(2003) A failed revolution – but a worthy cause. Mobilization 8(1): 116–119.

26.

Köse

Özcan

Karakoç

(2016) A comparative analysis of soft power in the MENA region: the impact of ethnic, sectarian, and religious identity on soft power in Iraq and Egypt. Foreign Policy Analysis 12(3): 354–373.

27.

Kroenig

McAdam

Weber

(2010) Taking soft power seriously. Comparative Strategy 29(5): 412–431.

28.

Layne

(2010) The unbearable lightness of soft power. In: Parmar

Cox

(eds) Soft Power and US Foreign Policy: Theoretical, Historical and Contemporary Perspectives. London and New York: Routledge, 51–82.

29.

Lukes

(2005) Power: A Radical View. 2nd edn. Basingstoke: Macmillan.

30.

Lukes

(2007) Power and the battle for hearts and minds: on the bluntness of soft power. In: Berenskoetter

Williams

(eds) Power in World Politics. London: Routledge, 83–97.

31.

Mahoney

(2010) After KKV: the new methodology of qualitative research. World Politics 62(1): 120–147.

32.

Mahoney

Goertz

(2006) A tale of two cultures: contrasting quantitative and qualitative research. Political Analysis 14(3): 227–249.

33.

Mattern

(2005) Why soft power isn’t so soft: representational force and the sociolinguistic construction of attraction in world politics. Millenium: Journal of International Studies 33(3): 583–612.

34.

McAdam

Tarrow

(2011) Introduction: ‘dynamics of contention’ ten years on. Mobilization 16(1): 1–10.

35.

McAdam

Tarrow

Tilly

(2001) Dynamics of Contention. New York: Cambridge University Press.

36.

Nye

Jr (1990) Soft power. Foreign Policy 80: 153–171.

37.

Nye

Jr (2004) Soft Power: The Means to Success in World Politics. New York: Public Affairs.

38.

Nye

Jr (2011a) Power and foreign policy. Journal of Political Power 4(1): 9–24.

39.

Nye

Jr (2011b) The Future of Power. New York: Public Affairs.

40.

Pestsov

Bobylo

(2015) Myagkaya Sila v Mirovoy Politike: Problema Operatsionalizatsii Teoreticheskogo Kontsepta [Soft Power in World Politics: The Problem of Operationalisation of the Theoretical Concept]. Vestnik Tomskogo Gosudarstvennogo Universiteta [Journal of the Tomsk State University] 2(34): 108–114.

41.

Roselle

Miskimmon

O’Loughlin

(2014) Strategic narrative: a new means to understand soft power. Media, War and Conflict 7(1): 70–84.

42.

Rothman

(2011) Revising the soft power concept: what are the means and mechanisms of soft power? Journal of Political Power 4(1): 49–64.

43.

Schimmelfennig

(2003) The EU, NATO and the Integration of Europe: Rules and Rhetoric. Cambridge: Cambridge University Press.

44.

Solomon

(2014) The affective underpinnings of soft power. European Journal of International Relations 20(3): 720–741.

45.

Szostek

(2014) Russia and the news media in Ukraine: a case of “soft power”? East European Politics and Societies and Cultures 28(3): 463–486.

46.

Tilly

Tarrow

(2015) Contentious Politics. 2nd edn. New York: Oxford University Press.