Abstract
In this analytical essay, I aim to advance the debate on trust in International Relations (IR) by clarifying how trust can be understood in practice-theoretical terms. To this end, I build on recent IR scholarship that has shifted from viewing trust as a mental state to viewing it as an active process through which social actors relate to one another—that is, as trusting. To clarify what this shift entails and initiate a more systematic dialogue between the dynamic yet still largely unconnected literatures on trust in international politics and on international practices, I proceed as follows: first, I survey how IR scholars have thought about trust thus far; second, I identify three commitments that give coherence to the emerging practice-theoretical literature on trust, namely, a focus on relations, processes, and agency; and third, I outline three theoretical options for studying trust in practice-theoretical terms: conceiving of trust as a product of practices, as a background of practices, or as itself a practice.
Keywords
Introduction
This analytical essay aims to advance the debate on trust in International Relations (IR) by clarifying how trust can be understood in practice-theoretical terms. Over the last two decades, a growing body of literature has examined the conditions under which trust can be established in international politics and how, in turn, trust enables and shapes international cooperation. Much of the early work pondered how leaders of adversarial states could take the first step toward cooperation, often reflecting on the paradigmatic example of trust building at the end of the Cold War (Kydd, 2005; Larson, 2000; Rathbun, 2012; Wheeler, 2013, 2018). More recently, the scope of trust research in IR has expanded to include studies of additional scenarios, such as the continuous cultivation of trust and cooperation in multilateral settings (Davies, 2022; Versloot, 2022; Walker, 2022), as well as the dynamics that ensue when trust diminishes or is betrayed (Anderl and Hißen, 2024; Barnes and Makinda, 2022; Christian and Peters, 2025; Keating and Abbott, 2021; Wille and Martill, 2023). Some of the more recent work on trust highlights the dynamics of face-to-face diplomacy and their impact on international cooperation (Bramsen, 2022; Holmes, 2018; Holmes and Wheeler, 2020; Wong, 2016), contributing valuable insights to the broader literature on micro-interactions in IR (Solomon and Steele, 2017). 1
Like many of their colleagues in other academic disciplines, IR scholars have mostly conceived of trust as a mental state. For them, it is “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behaviour of another” (Rousseau et al., 1998: 395, my italics; adopted for IR by Hughes, 2024: 101; Walker, 2022: 5); a “mental state of expectation” in “contexts where betrayal is always possible” (Wheeler, 2018: 51, my italics); or simply the belief held in somebody’s mind either “that the other side prefers mutual cooperation to exploiting one’s own cooperation” (Kydd, 2005: 3) or “that one will not be harmed when his or her fate is placed in the hands of others” (Rathbun, 2012: 10). In the following, I will explore the potential of an alternative approach. What if we understood trust not as a mental state but as an active process through which social actors relate to one another—that is, in terms of trusting? To this end, I will build on some cursory reflections on trust that can be found in the early literature on international practices (Hopf, 2010; Pouliot, 2010) and on recent IR scholarship that has begun to introduce ideas from practice theory to the trust debate in IR (among others, Davies, 2022; Edwards, 2019; Eggeling and Versloot, 2023; Ku and Mitzen, 2022; Versloot, 2022; Walker, 2022; Wille and Martill, 2023). By doing so, I aim to bridge two highly dynamic fields: trust research in IR and research on international practices, which share a number of sensitivities but have had little interaction thus far. 2 As is often the case in the formative moments of a research agenda, key concepts and theoretical options for studying trust from a practice perspective must yet be clarified. This is the goal of my analytical essay, a format introduced by Cooperation and Conflict to integrate scholarship, clarify debates, offer new perspectives on research, and identify new directions for the field of IR.
First, I will provide a brief overview of how IR scholars have thought about trust thus far. Then, I will propose that the nascent literature on trusting, with its diverse positions and theoretical lineages, can be helpfully circumscribed by reference to three core commitments: a focus on relations, processes, and agency, respectively. Finally, I will outline three theoretical options and discuss how they can be used separately or in combination to study trust in practice-theoretical terms: conceiving of trust as a product of practices, as a background of practices, or as itself a practice.
Conceptualizations of trust in IR
Thus far, IR has firmly located trust in the minds of individual human decision-makers, conceptualizing it as either a rational calculation or a non-calculative attitude, that is, a mental state. 3 For rationalists, trust is the subjective probability with which an actor expects another actor to reciprocate cooperation. In the most basic model, trust is simply an expectation of future behavior (e.g. McGillivray and Smith, 2000). Offering a more sophisticated account, Andrew Kydd (2005: 6–12) defines trust as the subjective probability that another actor is trustworthy, meaning they have preferences that would allow for stable cooperation if the actors knew each other’s preferences. This conceptualization is appealing due to its parsimony and ease of integration into larger theoretical models. However, critics have correctly noted that rationalist accounts fail to fully grasp the essence of trust (Mercer, 2005; Michel, 2012). If we know, or can predict with a degree of probability, what others will do, then we do not really need to trust them. Proper trust involves unthinking acceptance, a leap that we take even though we do not know what the other person will do. A second strand of the debate has therefore theorized trust as a mental state in which an actor suspends calculation and simply acts as if they knew for certain what the other person will do, even though they do not (Wheeler, 2018; see also Möllering, 2001). Some of these accounts emphasize the particularized trust that people extend to others they perceive to be part of their social group (Mercer, 2005; Weinhardt, 2015), while other accounts focus on a generalized attitude of trust toward all others as a personality trait (Bayram, 2017; Fujita, 2024; Rathbun, 2012).
When considering trust as either a calculative or a non-calculative mental state, both lines of research face two challenges: one analytical and the other methodological. The analytical challenge is how to scale trust from the individual to the state level. Rational choice theorists in IR often assume that states—just as they assume that individuals—are unitary rational actors. 4 However, this assumption is fiercely contested. 5 Non-calculative approaches have an even harder time explaining how trust can be scaled up from individual minds to collective actors, such as governments or states. Consequently, most studies of trust in international politics remain at the level of individual behavior. This makes it difficult to explain how trust relationships can persist in contexts where decision-makers do not interact directly or rotate in and out of their political or bureaucratic roles (Ku and Mitzen, 2022: 804; for some recent attempts to overcome this problem, see Troath, 2022; Wilcox, 2023). The methodological challenge stems from the fact that mental states are not easily observable. Sociologists, economists, and social psychologists typically use surveys and experiments, like the trust game, to evaluate levels of trust (Lewicki and Brinsfield, 2015). However, IR scholars do not generally have sufficient access to decision-makers to use these tools. Instead, they must rely on interpreting public statements, diaries, and memoirs to assess trust as a mental state (e.g. Rathbun, 2012: 45–53; Wheeler, 2018: 67–72). Some also use the absence of hedging behavior as an indicator of such a mental state (e.g. Hoffman, 2006: 25–36; Keating and Ruzicka, 2014: 760–763). However, this research strategy makes it difficult to clearly disentangle the explanans of trust from the explanandum of cooperation.
The core contention of the present article is that practice theory offers trust research a promising, non-mentalist alternative that allows us to address the analytical challenge of scale and the methodological challenge of observation that mentalist approaches face. Practice theory is a broad field, and there is no agreement on its defining features or boundaries. 6 Nonetheless, the approaches I am concerned with here share an interest in “practices,” which they define as patterns of socially meaningful action through which discourses and implicit background knowledge are realized and reproduced (Adler and Pouliot, 2011b: 4; Bueger and Gadinger, 2015: 451–452; Neumann, 2002: 629–632; Reckwitz, 2002: 249–250; Schatzki, 1996: 89–90). A practice cannot be reduced to an individual’s mental state; rather, it encompasses shared meaning and its realization in concrete instances of action. Furthermore, practice approaches emphasize how this meaning is inscribed in bodies, objects, and spaces (Barry, 2013; Walters, 2002; Wilcox, 2015). Practice approaches can explain the continuity of social life with reference to habit and embodied experience (Hopf, 2010; Neumann and Pouliot, 2011), but they can also identify the conditions under which gradual or sudden change occurs (Hopf, 2018; Rösch, 2021; Schindler and Wille, 2015). From this theoretical perspective, international politics can be understood as the product of practices through which a state’s or other actors’ interests, policy options, and decisions are produced by drawing on various bodies of knowledge and different practical skills (e.g. Hofius, 2022; Lesch and Loh, 2022; Neumann and Sending, 2021; Wille, 2024).
Building on constructivist (Adler and Barnett, 1998; Wendt, 1999) and earlier practice-theoretical work (Hopf, 2010; Pouliot, 2010) that employed the concept of trust without ascribing it a central place in the analysis, recent years have seen a new wave of scholarship that has begun to theorize trust more explicitly in practice-theoretical terms. Among the texts already available in print are Larissa Versloot’s (2022) study of trust and information-sharing practices at the European Council and a co-authored article by Kristin Anabel Eggeling and Versloot (2023), which asks to what extent trust could be “taken online” during the COVID-19 pandemic. Hayley Walker (2022) proposes a complex model of trust building in multilateral negotiations that incorporates practice-theoretical elements. Benjamin Martill and I examined the Brexit negotiations to determine how experiences of betrayal affect trust, the latter understood as a shared practice of anticipation (Wille and Martill, 2023). Mathew Davies (2022) studies “performances of trust” in everyday diplomacy within the Association of Southeast Asian Nations (ASEAN). Similarly, Scott Edwards (2019) proposes the concept of practice-based trust to explain ongoing cooperation within ASEAN. Pieter Lagerwaard and Marieke de Goede (2023), as well as Pepijn Tuinier et al. (2023), emphasize the role of professional communities of practice in intelligence cooperation. Minseon Ku and Jennifer Mitzen (2022) explore how system trust in the international order provides the backdrop for diplomatic practices. Furthermore, Thierry Balzacq’s (2022) recent outline of a relational agenda for trust research shows strong affinities with practice thinking.
Although these contributions aim to accomplish different things, their practice perspective generally enables them to address the two aforementioned challenges. Regarding the analytical challenge of scale, these authors can think more consistently about trust at the level of collective actors by conceiving of trust in practice-theoretical terms. For them, collective actors are both constituted by and engage in social practices. If states are collective actors that can engage in practices, then they can also trust. There is no need to assume that they possess a mind or consciousness. Furthermore, regarding the methodological challenge of observation, practices are always shared and leave traces in the world that can be documented using standard social science methods. Studying practices does not require looking inside people’s heads. It is sufficient to reconstruct the meaning they attach to their actions and relate it to larger patterns in the world (Bueger, 2014; Pouliot, 2007). Admittedly, as the extensive literature on interpretive methods attests, this is no easy task. Nevertheless, accessing shared meanings is less daunting than accessing private mental states.
Three commitments: relations, processes, and agency
Recent practice-theoretical literature on trust in IR draws on a wide range of theoretical ideas and scholarly traditions. As a result, its contributors have arrived at different conceptualizations of trust and have employed them in a wide range of arguments. This raises the question of what unites these contributions. Can we really say that they all share one common agenda? In the following, I argue that what lends coherence to the emerging literature on trusting is that the contributions share three commitments characteristic of the broader literature on international practices, namely, their focus on relations, processes, and agency. 7 In this section, I will discuss the implications of each of these commitments for studying trust.
First, the emerging literature emphasizes the relational nature of trust. To some extent, the broader debate has always acknowledged that trust is a relationship between a trustor and a trustee that occurs in a specific social context (Forst, 2022; Lewis and Weigert, 1985; Rousseau et al., 1998). However, this acknowledgment sometimes sits uneasily with the prevalent conceptualization of trust as a mental state—that is, as an attribute of a specific actor. Some recent work has made substantial progress in overcoming this tension by spelling out the relational dynamics of trust building. For example, Nicholas Wheeler (2018: 51) conceptualizes trust as a “mental state of expectation,” i.e., a property of an individual that enables them to display “trusting behavior.” Despite this mentalist point of departure, in his groundbreaking work—both single-authored and co-authored with Marcus Holmes—he has illuminated how interaction rituals, emotional attunement, and relational co-construction lead to interpersonal trust (Holmes and Wheeler, 2020, 2025; Wheeler, 2018; Wheeler and Holmes, 2021, 2022). For Wheeler and Holmes, therefore, trust remains a property of an individual, while its antecedents are genuinely relational. Some practice scholars have taken a more radically constructivist position, locating not only the antecedents of trust but also trust itself within a social relation. In doing so, they draw on relational thinking in IR more generally (Adler-Nissen, 2015; Jackson and Nexon, 1999). Along these lines, Balzacq (2022: 6) has recently described trust relations as based on “the mutual imperatives to conduct oneself in a way that is appropriate to the relationship.” Importantly, this framing allows him to shift the focus from how rational calculation or emotions determine trust to how both elements can motivate actors within a relationship to trust and act in ways that honor that trust. Adopting a relational conceptualization of emotions (as well as rational calculation), a practice-based approach to trust offers an alternative to social-psychological work that also acknowledges the importance of emotions but reduces them to unidirectional determinants of behavior (Michel, 2012: 874–876; see also Head, 2012; Mercer, 2005).
Second, for practice theorists, trust is a social process. Thus, it is neither a stable characteristic of an individual nor a resource that can be accumulated and used variably for different purposes in the aggregate. As Eggeling and Versloot (2023: 641) write, “trust is not an end state but always ‘in-the-making’: it is a social process of trusting” (see also Wille and Martill, 2023: 2411–2412). In their theorization, they build on Guido Möllering’s (2001: 412, 414) groundbreaking conceptualization of trust as a “mental process of leaping” that shifts an actor from interpreting the available information to a state of positive expectation in which “our awareness of the unknown, unknowable and unresolved is suspended.” 8 In later collaborative work, Möllering complements this account with a more societal perspective, recognizing that “trust is not only a mental process of the trustor but also a social process involving the interaction of trustor and trustee with each other and with their social context” (Nikolova et al., 2015: 234). Practice theorists emphasize this shared social element of trusting even more. They also argue that practices of trusting are part of larger processes of social construction through which social reality is produced and reproduced. This claim resonates with constructivist positions within the broader trust literature (e.g. Sydow, 2006; Wright and Ehnert, 2010).
Third, practice theorists emphasize the importance of agency in processes of trusting, by which they mean that trust and responses to it are not entirely determined by external forces or social structures. To trust is an active choice and thus an expression of the trustor’s freedom (Mizrachi et al., 2007: 145–146; see also Möllering, 2005a: 26–27). Trusting someone, in turn, affirms their agency by acknowledging that their future behavior is not predetermined and that they can either honor or betray the trust placed in them (Möllering, 2005b: 287–288). Some constructivists in the broader trust literature have therefore framed trusting as a mediating element in the co-constitution of agents and structures (Sydow, 2006; Wright and Ehnert, 2010). Contemporary practice theorists, however, go beyond these still somewhat rigid notions of agents and structures, embracing a more dynamic understanding of agency as a relational achievement (Braun et al., 2019; see also Wille, 2025a). Expressing such a perspective, Vigdis Broch-Due and Margit Ystanes (2016: 24) note that “trust, as a noun, tends to emphasize an individual subject’s deliberation to enter a contract or take risk, while trust as a verb conjures up an intersubjective space of social anticipation binding subjects together”; this, in turn, requires us to “develop more performative approaches to trusting, which focus on various forms of agency.” Nevertheless, practice theorists acknowledge that agency is unevenly accessible to social actors, depending on their position in social space (Pingeot and Pouliot, 2024). The practices available to them thus enable and constrain whether and how any specific actor can trust.
Three ways to think about trusting
While all practice-theoretical accounts of trust in IR share the above-discussed commitments to relations, processes, and agency, there are a variety of options for how to conceptualize the role of trust in relation to international practices. In what follows, I will elaborate on three ways to frame trust in practice-theoretical terms: as a product of practices, as a background of practices, and as itself a practice. To outline these options, I will draw on the small but growing literature in IR as well as contributions to the broader interdisciplinary literature on trust. 9 My goal is to make their assumptions about the relationship between trust and practices explicit and to clarify the theoretical openings they offer.
Trust as a product of practices
The first option is to understand trust as a product of practices. In the development of the interdisciplinary literature on trust, this theoretical move represented an important step toward recognizing the processual character of trust. Möllering (2013: 287) used the analogy of the difference between an organism’s state of being alive and the process of living to illustrate this point. It is not that both cannot be part of the same theoretical account; rather, if we understand the process of living, we know much more than if we can only determine whether the organism is alive or dead. Similarly, determining whether an actor trusts someone is merely a snapshot of an outcome, while studying what Möllering (2013: 286) calls “trusting” reveals that “the ‘product’ of trust is always unfinished and needs to be worked upon continuously.” In his empirical work with colleagues, Möllering has therefore analyzed the development of trust as a “socio-cognitive-emotional process” consisting of different practices (Nikolova et al., 2015: 243). In IR, Walker (2022; see also Walker and Biedenkopf, 2020) has developed a sophisticated model of how trust in chairpersons emerges in multilateral negotiations. This model ascribes an important role to the competent performance of diplomatic practices. Tuinier et al. (2023) argue that the intelligence officer community of practice can generate trust based on, among other things, shared professional standards and a shared social identity, something rationalist theories cannot account for. In an interesting variation on this idea, Davies (2022) argues that ritualized “performances of trust” can still enable cooperative behavior even when interpersonal trust has broken down. According to Davies (2022: 11), these performances produce only “simulated trust,” which nevertheless still allows for considerable levels of cooperation. 10 These arguments resonate with the recent work of Holmes and Wheeler on interaction rituals, emotional attunement, and relational co-construction (Holmes and Wheeler, 2020, 2025; Wheeler and Holmes, 2021, 2022). However, the latter do not frame these relational and processual antecedents of trust as practices.
As IR researchers pursue this line of inquiry further, it may be worthwhile to distinguish more explicitly between practices understood by actors to build trust and those that create it inadvertently. The former kind of practice allows for a more focused study of trust because the actors are reflexively aware of its importance, as well as its absence in some cases. Obvious historical case studies include the rapprochement between the Soviet Union and the United States at the end of the Cold War (see Ku and Mitzen, 2022; Wheeler, 2018: 143–191) and the largely unsuccessful attempts to improve relations between Russia and the West through engagement and institutionalization in the 1990s and early 2000s (see Haukkala and Saari, 2018; Pouliot, 2010; Sarotte, 2021). Interestingly, practices that practitioners themselves understand as trust building not only produce trust but are also sites of contestation over what trust is or should be. In their actor-centered account of trusting, Nissim Mizrachi et al. (2007) demonstrate how managers creatively select which trust repertoires to use in a given situation. The notion of “trust work” also offers an interesting perspective, since it highlights how social actors continuously work on trust (Jackson, 2014; Versloot, 2025). Adopting a more decentered approach, Anna Weichselbraun et al. (2023) have demonstrated that new practices, technologies, and infrastructures that purport to generate trust also transform our very understanding of what it means to trust. All of these approaches seek to illuminate what happens in practice when individual or collective actors attempt to address what appear to them to be trust issues.
Inadvertent trust production that does not crystallize around a notion of trust important to the actors’ own shared understanding of their situation is much harder to capture with the tools of practice theory. One possible way to study these practices is to build on the theory of security communities—groups of states that share the expectation that political conflicts will be resolved peacefully (Adler and Barnett, 1998). Ted Hopf (2010: 554) argues that within such communities, habituation “produces perceptions, attitudes, and practices that already entail an unconscious confidence in [another] state’s trustworthiness.” 11 Vincent Pouliot (2010: 40) describes trust within security communities as an actor’s “background feeling” that they can “believe despite uncertainty,” based on reasons derived “from tacit experience and an embodied history of social relations.” In other words, over time, the practices of diplomacy within security communities create a shared expectation of trust. While some of this trust may result from explicit trust building, security communities also provide ample opportunity to study trust as a byproduct of everyday interaction in pursuit of other goals.
Trust as a background of practices
A second possibility to think about the relationship between trust and practices is to consider the former as a precondition or background of the latter. For ethnomethodologists, mutual trust that others are competent and willing to engage is a precondition for coordinated action. According to Harold Garfinkel (1963: 193–194), to trust is to share a definition of the situation in which one is interacting. In modern societies that lack rigid identity boundaries, such a shared orientation toward practice is an important source of social integration (Rawls and David, 2005). Garfinkel (1963: 190) further notes that when studying interaction and the trust underlying it, “there is no reason to look under the skull since nothing of interest is to be found there but brains,” thus fully situating trust within social relations (see also Watson, 2009: 489–490). Recent IR literature offers a similar theoretical perspective in the shape of Versloot’s (2022) analysis of “climates of trust” in European Council diplomacy, which can have different qualities and both enable and condition practices of information sharing. Similarly, Lagerwaard and de Goede (2023) conceptualize “circuits of trust,” which they define as “bounded social realm[s] with shared practices of meaning-making concerning obligations, worthiness, rights and symbols, and with [their] own media,” as the background of international financial intelligence information-sharing practices. Both of these accounts thus emphasize the boundedness of communities that are constituted by trust and share mutually intelligible practices.
A somewhat different conceptualization of trust as a background of practices can be found in the literature on ontological security. Drawing on Garfinkel, among others, Anthony Giddens (1991: 40) argues that “basic trust” in the stability of the social and material worlds enables “a bracketing, on the level of practice, of possible events which could threaten the bodily or psychological integrity of the agent.” This helps actors avoid paralyzing anxiety and continue with their daily lives. Thus, trust is analyzed here in relation to the psychological needs of human beings, albeit from a sociological perspective that does not reduce it to a mental state. Applying this concept to international politics, Mitzen (2006) contends that this bracketing of uncertainty provides states and their citizens with a sense of security. However, it can bind them into relationships of intractable conflict or enduring rivalry, which are actually detrimental to their physical security. Building on this idea and incorporating elements of Niklas Luhmann’s (1979) theory of trust, Ku and Mitzen (2022: 800) argue that “system trust,” defined as “trust in the continuity of the overarching, shared sociopolitical order” and, more specifically, as trust “in the existence of an international system, an anarchy of states that seem to act like persons” sustains the international order. Such system trust both alleviates anxiety by reducing complexity and provides orientation for action: “trust know-how is an emotional orientation of calm confidence toward the world, which is part of how we know who we are and how to ‘be’ ourselves” (Ku and Mitzen, 2022: 808).
One challenge in this line of research is clarifying the interplay between trust as a background and trust as a relationship between actors. Most authors seem to agree that background trust is more than, or at least different from, an aggregate of trust relations between actors. 12 According to Ku and Mitzen (2022: 809–810), for instance, system trust is a macro-phenomenon that supervenes on, but is not identical with, individual feelings and attitudes toward the world. Versloot (2022: 515) argues that a “climate of trust” emerges from, but is not the same as, “a configuration of trusting relations.” Two important points to note in this regard are: first, background trust in these accounts is not primarily understood as a precondition for horizontal trust between actors, but rather as a precondition for a wide range of practices, or even the possibility of practice itself. These practices, in turn, reproduce and transform background trust through a recursive process. Second, background trust enables not only horizontal relationships of trust but also of mistrust or distrust. As Ku and Mitzen (2022: 810) note, “even in an environment where interpersonal distrust can be rampant, such as [international] anarchy, day-to-day social interaction relies on system trust in the basic ingredients of the shared social order.” Thus, the connection between background trust and trust among actors appears almost accidental, or at least, less stringent than the use of the same term for both phenomena would suggest. Future work should clarify the relationship between these two kinds of trust. This can be achieved by identifying the various types of background knowledge that actors use to relate to the social world and to other actors.
Trust as a practice
The third, most radical, practice-theoretical option is to conceptualize trust as itself a practice of trusting, in other words, as “an inherently social activity of anticipating” (Wille and Martill, 2023: 2411–2412). One way to articulate this idea is to say that trusting is a habitual form of trust that develops over time when individuals have a mental state of trust, and it can persist even when that mental state is no longer present. Under such circumstances, “trusting behaviour becomes a natural response that is undertaken un-reflexively, and . . . trust becomes an essential part of the background knowledge that informs decision-making” (Edwards, 2019: 233). 13 A more radical move is to abandon, or at least bracket, the idea that trust is a mental state altogether and instead focus on the meaning social actors attach to their actions when they trust (Broch-Due and Ystanes, 2016). Articulated in Luhmannian terms, trusting then simply “inheres in anticipating the future actions of others without a corresponding weighing of risks and opportunities” (Wille and Martill, 2023: 2410). From an ethnomethodological perspective, the practice of trusting is a precarious accomplishment that depends “on a feel for correct timing, skill, observance of proper form, and on frames that organize what is foregrounded and what is bracketed” (Eyal, 2019: 60). Just like calculating anticipation, trusting is a specific way of relating to the future based on particular bodies of knowledge and practical skills. 14
In the broader trust literature, one of the most powerful ways in which trusting has been theorized as a practice is through an extension of Pierre Bourdieu’s critical sociology. Morten Frederiksen (2014) suggests viewing trust as a practice of shared anticipation shaped by the actors’ history and positioning in a social context. 15 Habitus, the embodied past of individual and collective experience, functions as a generative principle that gives rise to trusting by providing a sense of familiarity with certain situations and the schemata by which judgments about trustworthiness are made. Furthermore, the manner in which individuals trust is shaped by the social situation in which they find themselves. Their practical sense affords an intuitive understanding of their situation and the actions and perceptions that are appropriate to their position within it. Trusting occurs when multiple actors in a situation align their respective practical sense, creating a shared anticipation of what lies ahead. What Pouliot (2010: 35) observes concerning practices in general, namely that they unfold at the “intersection of embodied dispositions and structured positions,” thus also holds specifically for practices of trusting. “A propensity to trust in a given situation [is] generated between habitus and familiarity on one side and the nature of the situation on the other” (Frederiksen, 2014: 175). To fully incorporate this line of thinking into IR, future research could explore how the ability to trust and be trusted, as well as the agency it confers, is shaped by the habitus and field position of international actors.
One advantage of this approach is that it can shed more light on the emotional aspect of trusting practices. As Pouliot (2010: 40) notes in his outline of a Bourdieusian account of international security, trust is not propositional knowledge but rather an “inarticulate feeling” that expresses embodied past experiences. To remain true to the relational and processual commitments of practice theory, this perspective requires that emotions not be reduced to mental or physiological states and that their sociocultural embeddedness be fully considered (Bially Mattern, 2011; Koschut, 2014; Rösch, 2021). In his article on emotional labor in international organizations, Deepak Nair (2020: 581; drawing on Hochschild, 1983) reflects on how international bureaucrats manage their emotions to inspire trust in their national diplomatic counterparts. Building on these insights, future research could develop the complementary perspective of how nurturing a sense of trust within oneself can also be part of the emotional labor involved in cultivating diplomatic relations. This would contribute to a more comprehensive understanding of trusting as a complex technique for relating to others.
Theoretical purity or eclecticism?
While all of the practice approaches discussed in this article share an appreciation for the relational and processual nature of social life, as well as for situated agency, they differ in some of their premises and in their theoretical ambition. Therefore, it would be unhelpful to think of them as proposing one single practice theory of trust. Instead, I identified three theoretical options for thinking about trust in practice-theoretical terms: understanding trust as a product of practices, as a background and precondition for practices, or as itself a practice of trusting. How are these options related to each other, and how can they help us study trust in international politics?
It is important to note that all three options are available within a strong practice ontology and a more eclectic approach. Some practice theorists consider practices to be the “‘smallest unit’ of social analysis” (Reckwitz, 2002: 249) and attribute basic ontological qualities to them (Schatzki, 2016). Theirs is a perspective that, as Sally Haslanger (2018) puts it, places practices first. All three theoretical options can be used—either alone or in combination—to develop a theory of trusting based on a strong practice ontology. In a world of practices “all the way down,” trusting can be thought of as a practice preceded and followed by other practices as social life unfolds. Within a strong practice ontology, selecting one of the three options simply means choosing a research focus. For our purposes, in a given research project, are we most interested in the practices of trust building that precede trusting; the practices that follow from and are enabled by trusting; or trusting itself? 16 Practice theory also provides resources for thinking about how these practices are connected. For instance, the concept of anchoring practices (Neumann and Sending, 2011; Swidler, 2001) could help us rethink the relationship between trust and cooperation. While IR scholars traditionally think of trust as a necessary precondition for most forms of cooperation, a practice approach could redefine the relationship as one in which practices of trusting anchor practices of cooperation. Such a perspective could advance the theory of security communities along the lines first suggested by Pouliot (2010). On his account, diplomacy (which could be further theorized as a practice of trusting) appears to actors as a self-evident way to resolve conflict and hence grounds other practices of cooperation within a security community.
However, not all practice scholars adhere to a strong practice ontology. Many operate with ontologically less constraining notions of practices. For example, Emanuel Adler and Pouliot (2011b: 28) treat practices as a “conceptual focal point” around which theories with different ontological commitments can converse. 17 Within an eclectic approach to theory building that integrates elements from different theoretical traditions (Cornut, 2015a; Sil and Katzenstein, 2010), each of the three theoretical options can be combined with other arguments. For example, Walker (2022) offers one of the thinnest practice accounts surveyed in this article. She provides a model of how trust building can be conceived as a set of practices that lead to a mental state of trust. Even approaches that consider some forms of trust to be practices do not necessarily have to abandon the concept of mental states entirely. As Davies (2022) and Edwards (2019) suggest, one could think of practices of trusting as coexisting with the mental state of trust. According to them, trusting and trust are different phenomena that enable cooperation in similar ways and can substitute for each other, at least to some extent. Möllering argues that observing the process of trusting within socio-material contexts provides a complementary perspective to studying trust as a mental state of suspension. For him, trusting and trust are facets of the same phenomenon. Thus, these approaches leave room for practices of trusting and mental states of trust within a single theoretical framework. Similarly, trust can be theorized as a background to practices either within a strong practice ontology, as ethnomethodologists do, or alongside mental states such as anxiety and fear, as some scholars of ontological security propose.
In summary, a practice perspective on trust can be incorporated into a comprehensive theory of practice with its own strict ontology, which views practices as the building blocks of the social world. However, a practice perspective can also inform less ontologically purist theoretical projects that use practices as an analytical concept alongside other ideas, such as mental states, social norms, or even rational choice. The three theoretical options explored in this article do not offer a complete theory of trust, but rather, they offer building blocks that researchers with different ontological commitments can use in projects of varying scope, from mid-range theorizing to grand theory.
Conclusion
In this analytical essay, I have demonstrated how a practice perspective can shed light on complex trust dynamics in ways that established IR literature, with its mentalist orientation, cannot. Given the vibrant but thus far largely disconnected debates in IR about trust and practices, the time seems ripe for a concerted effort to advance research on trusting as a distinct agenda. As I have argued, this agenda can gain coherence from a shared commitment to focusing on relations, processes, and agency. Systematizing and expanding on a number of recent texts that embrace these commitments, I have outlined three ways to theorize trust in practice-theoretical terms. First, trust can be understood as a product of practices; second, trust can be explored as a background and precondition for practices; third, trust can be conceived as itself a practice of trusting. These options are not mutually exclusive but can be combined with each other and with other theoretical arguments from practice theory and other fields. Depending on their theoretical commitments, some colleagues will seek to develop a theory based on a strong practice ontology, while others will prefer an eclectic approach that adds practice-theoretical elements to existing mentalist frameworks.
In conclusion, I would like to briefly highlight two issues that fall outside the scope of this analytical essay but deserve further attention as the debate on practices and trust progresses. First, a more systematic examination of how trust dynamics in diplomacy compare to trust dynamics in other practical contexts would be worthwhile. The nascent literature on trusting surveyed in this article, as well as practice theory in IR more generally, leans heavily on diplomatic practices. Beginning with a set of practices that is easily recognizable due to its accredited agents and established rituals is certainly a fruitful research strategy. However, to gain a more complete picture, one should supplement it with studies on a wider range of practices. This would enable us to ask questions such as: Why is it easier to build trust in some practical contexts than in others? Do some practices require more or different forms of trust than others? Do diplomats trust differently than other practitioner communities? Among the literature surveyed in this article, Pouliot’s (2010) early work on the NATO–Russia Council, as well as the articles on intelligence cooperation by Lagerwaard and de Goede (2023) and Tuinier et al. (2023), offer hints about trust dynamics in contexts primarily populated by military personnel and intelligence officers. Studying trust dynamics in practices such as warfare (building, for example, on Bode, 2023; Wilcox, 2015), intervention (Autesserre, 2014; Pingeot, 2018), global governance (Kortendiek, 2024; Sondarjee, 2024), and activism (Anderl and Salehi, 2025; Holthaus, 2023) promises to be similarly instructive.
Second, IR practice theorists should consider the implications of the arguments presented in this article for the opposite concepts of trust, namely mistrust and distrust. In doing so, they can take their cue from the mentalist literature on trust. Some recent work in this field seeks to move beyond the simplistic understanding that trust is the absence of mistrust or distrust, and vice versa. Instead, they treat these concepts as distinct mental states that can coexist (Saunders et al., 2014; Six and Latusek, 2023). The implications for IR of this innovation in mentalist trust research are currently being explored by Cervasio et al. (2025). For practice theorists, an attractive theoretical option could be to think of mistrusting or distrusting as a distinct process rather than the absence of trusting. Starting from this premise, they could explore what a focus on relations, processes, and agency reveals about mistrust and distrust. Furthermore, they could explore the theoretical implications of treating mistrust and distrust as products of practices, as preconditions of practices, or as practices themselves. Expanding the research agenda to include contexts beyond diplomacy and incorporate the opposite concepts of trust would provide a clearer understanding of international politics by identifying the various dynamics of trust, as well as of mistrust and distrust, that characterize its diverse practices.
Footnotes
Acknowledgements
A draft of this article was presented on the roundtable “Trust in international practices” at the 2025 Annual Conference of the International Studies Association in Chicago. I am grateful to the participants and audience members for the inspiring discussion. I would also like to thank Kristina Lepold, Max Lesch, the editors of Cooperation and Conflict, and the anonymous reviewers for their helpful comments on earlier drafts of the article, and Ciaran Cronin for editing the final version. Frederik Hermle provided excellent research assistance.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
1
There are also interesting connections between trust research and the literature on friendship in international politics that warrant further exploration. See, for example, Haugevik (2018), Koschut and Oelsner (2014), and 
.
2
For an overview of the two literatures, see Haukkala et al. (2018), Rathbun (2018), Ruzicka and Keating (2015), Wille (2025b), and, respectively, Adler and Pouliot (2011a), Bueger and Gadinger (2018), and .
3
6
For two influential perspectives in IR, see Adler and Pouliot (2011a) and .
7
Relationalism and a process perspective are often subsumed under the label “relationalism,” even though, from an analytic point of view, they are different commitments. I treat them separately here because each has specific implications for the study of trust.
8
For further accounts of trust as a process from literatures other than IR, see, for example, H. Jackson (2014) and .
9
Some contributions to these literatures use more than one of the theoretical possibilities, or at least do not clearly distinguish between them. This is also why, in the following, different arguments from the same authors will sometimes be discussed in the context of more than one theoretical option.
10
By framing “performances” of trust as less than real, Davies opts for a moderate constructivist position.
11
According to Hopf, this renders trust, which he understands as a reflexive attitude, superfluous. However, a practice-theoretical account of trust as suspension would be consistent with his conception of habit as the subjective elimination of uncertainty. See Versloot (2022) and .
12
This can also be seen in Luhmann and Giddens, who treat trust among actors and trust in social systems as distinct but closely related phenomena.
13
Of course, a theory that operates with two types of trust—one a mental state and the other a shared practice—cannot entirely avoid the abovementioned methodological challenges arising from the unobservability of mental states.
14
15
Frederiksen’s Bourdieusian theory of trust is unique in that it accords equal importance to diachronic and synchronic elements. Other approaches mostly focus on the diachronic element of trust as a habit. See, for example, Ma (2023) and .
16
It should be noted that, even within a strong practice ontology, more than one conception of trust can be combined. For instance, address in detail the relationship between vertical trust in the international system and horizontal trust among states and other international actors. For them, trusting in the continuity of social life and trusting other actors are distinct phenomena, even though both can be explained in practice-theoretical terms.
17
For a critique, see Kustermans (2016) and .