Causal Explanation in Brazilian International Relations

Explicação Causal nas Relações Internacionais Brasileiras

Abstract

How do Brazilian IR scholars tailor causal explanations in their research? Searching for causes has become paramount to the scientific endeavour in the social sciences, and in recent years the lexicon of causality has come under scrutiny, with many scholars revising the standard Humean and covering law approaches to cause. Far from being consensual, current philosophical and meta-theoretical debates in IR reveal a rather kaleidoscopical picture of causation. However, Brazilian IR scholars seem to remain unfamiliar about what it means to advance causal claims. In this paper, I have a twofold goal: to advance the causal talk by rehearsing recent meta-theoretical debates on causality; and to examine the ways Brazilian IR scholars in the subfield of Brazilian foreign policy analysis (BFPA) tailor causal explanations. To achieve the latter, I perform a cartography of the articles published in five Brazilian top-tier journals in the discipline – Brazilian Political Science Review, Dados, Contexto Internacional, Revista Brasileira de Ciências Sociais and Revista Brasileira de Política Internacional – from 2018 through 2022. I conclude that BFPA research is founded primarily in non-Humean understandings of causality, but Brazilian scholars might benefit from further awareness on the various meanings of causal explanation.

Keywords
causality; explanation in International Relations; meta-theory; International Relations theory; Brazilian foreign policy analysis

Resumo

Como os acadêmicos brasileiros de RI adaptam as explicações causais em suas pesquisas? A busca por causas se tornou fundamental para o esforço científico nas ciências sociais e, nos últimos anos, o léxico da causalidade passou a ser examinado, com muitos acadêmicos revisando as abordagens padrão humeanas e da lei de cobertura da causa. Longe de serem consensuais, os debates filosóficos e metateóricos atuais em RI revelam um quadro bastante caleidoscópico da causalidade. No entanto, os acadêmicos brasileiros de RI parecem não estar familiarizados com o que significa apresentar alegações causais. Neste artigo, tenho um objetivo duplo: avançar o discurso causal ensaiando debates metateóricos recentes sobre causalidade; e examinar as maneiras pelas quais os acadêmicos brasileiros de RI no subcampo da análise da política externa brasileira adaptam as explicações causais. Para alcançar o último objetivo, realizo uma cartografia dos artigos publicados em cinco revistas brasileiras de primeira linha na disciplina - Brazilian Political Science Review, Dados, Contexto Internacional, Revista Brasileira de Ciências Sociais e Revista Brasileira de Política Internacional - de 2018 a 2022. Concluo que a pesquisa da ABPF está fundamentada principalmente em entendimentos não humeanos de causalidade, mas os pesquisadores brasileiros podem se beneficiar de uma maior conscientização sobre os vários significados da explicação causal.

Palavras-chave
causalidade; explicação em Relações Internacionais; metateoria; teoria das Relações Internacionais; análise da política externa brasileira

Introduction

“If we cannot find reasonable evidence about the causes of social events, we have little chance of fully understanding who we are, where we have been, or where we are going”. – Harold Kincaid (1996: 5)

In recent decades, causal inquiry has drawn attention of philosophers and social scientists alike interested in how causal explanation is tailored in scientific work. In philosophy, a variety of theories of causation and causal explanation have emerged, reinvigorating debates about the meanings of cause (Lenine and Machado 2023). Causal explanation has also been in the spotlight in the natural sciences for a long time, but philosophical disputes reveal multifaceted understandings of the explanatory enterprises in these disciplines. Some fields in physics do resort to causal arguments, although do not make them central to tailoring explanations (Ladyman et al. 2007; Woodward 2003); other quarters, like quantum mechanics, are rather sceptical of causality due to the time symmetry of certain quantum phenomena (French 2006; Ladyman et al. 2007). In biology, causal arguments are primarily mobilised in evolutionary theory (Mayr 2008), as well as in molecular biology (Woodward 2010). Similarly, in the discipline of IR different understandings of causal explanation coexist, each of which premised on specific ontological, epistemological, and methodological commitments (Kurki 2008).¹ Explaining international phenomena in causal terms encompasses a variety of practices that intertwine each of those commitments in defining what a causal explanation is and should be.

Despite the centrality of causation to IR, scholars often disagree about what they conceive of cause and causal analysis. Most frequently, causation has been regarded as a feature of neopositivist forms of research, where the quest for causes is manifest in the statistical orientation of quantitative research (Bueno de Mesquita and Fowler 2021). Alternative conceptions of causation have reacted to the regularity-based, quantitative-oriented approach of positivist works, proposing instead other lexicons to pin down causal relations. Some define causality as the probability of a cause generating an effect (Gerring 2005); whereas others advocate, based on within-case and comparative analysis, the search for necessary and sufficient conditions (Goertz and Mahoney 2012); and still others make distinctions between the properties that constitute the world and causes (Wendt 1998); or prefer to speak of two separate domains of inquiry, namely explanation (where causality resides) and understanding (focused primarily on interpreting the meanings of social phenomena) (Hollis and Smith 1990). All these approaches attempt to trace, with different priorities, the ontological, epistemological, and methodological underpinnings of causation in order to understand how we explain international phenomena, and what kinds of questions we answer (Grynaviski 2013; Suganami 1996; Wendt 1998).

Giving the preceding, my goal in this paper is twofold: for one, by rehearsing debates on causality in the discipline, I aim to unleash a reflection on causal explanation in Brazilian IR that is still under-appreciated by the national literature on theory and meta-theory of IR. Secondly, I conduct an exploratory cartography of causal explanation in the field of Brazilian foreign policy analysis (henceforth, BFPA)², aiming to characterise causal approaches in the works published in five top-tier journals in Brazil, according to Qualis Capes 2017-2020: Brazilian Political Science Review, Contexto Internacional, Dados, Revista Brasileira de Ciências Sociais and Revista Brasileira de Política Internacional.³ This second goal is meant to build the connection between the meta-theoretical debates that inform research on international phenomena with the explanatory practices of BFPA. As a field that is primarily driven by the interest in generating local interpretations about Brazil’s relations in the international arena and the logics underpinning decision-making processes of foreign policy, BFPA is the perfect locus to assess the distinguishing characteristics of causal explanations in Brazilian IR. Evidently, it provides a partial picture in the face of the diversity of research fields and subfields in the discipline, yet one that aids in re-interpreting the conditions under which causal explanations are advanced in the literature.

At this point, a caveat is in order. Causal explanation is just one type among different forms of explanatory discourse (Mantzavinos 2016). Historically, causal explanation emerged as a response to the deductive-nomological model of explanation, whereby a set of initial conditions and covering laws operates as an explanans for a given explanandum. Unificationist, mathematical, statistical, and equilibrium explanations are just a few examples of the explanatory pluralism in science, denoting how explanation is a rather multifaceted enterprise (Rice and Rohwer 2021). In this paper, however, I focus on the particular case of causal explanation and how it has been conceptualised in IR.

The paper is divided into three sections. In the first one, I introduce debates on causation in the discipline, focusing on mainstream approaches to cause-as-regularities, causal narratives, the critical realist critique, and interpretative causation. In the second section, I discuss the methodological approach used to categorise the research output in the aforementioned journals vis-à-vis categories of causal explanation derived from the aforementioned meta-theoretical discussions. I also present the results of the cartography of causal explanation, relating them to similar works on theoretical and methodological preferences of Brazilian IR researchers. Then, I proceed in the third section to discuss these results in terms of what causality means to Brazilian IR, especially to BFPA.

The quest for causes: contending conceptions of causation in contemporary IR

The philosophy of science has quarrelled about what cause is since time immemorial, but much of the contemporary debate is centred on David Hume’s conception of causality and Carl Hempel’s covering law model of explanation. Indeed, these two frameworks are mentioned in virtually all discussions of causality not only as a starting point for rehearsing central ideas in the philosophy of explanation, but also and more importantly as a critique of causal inquiry (Kurki and Suganami 2012).⁴

According to Hume’s definition, causality would be composed of two dimensions: the regularity dimension of a cause, which may be discovered by observing the real world, for ‘causes and effects are discoverable, not by reason but by experience’ (Hume 1952: 459); and the counterfactual dimension, which assumes that if event A had not happened, B would not have happened. Both stances of Hume’s idea of cause and causation have sparked debates in the philosophy of science, which, in turn, have spilled over to other scientific domains, including the social sciences, where it is known as the constant conjunction model of causal explanation.

In the beginning of the 20th century, Hempel and Oppenheim (1948) provided a formal analysis of explanation: to explain something is to subsume a situation under a law. The explanandum – what gets to be explained – is accounted for by the force of a law that provides generality and, more importantly, captures the connection between A and B (Hempel and Oppenheim 1948; see also Barker and Kitcher 2014; Dowding 2016). The explanans – what actually explains – is given by a covering law. Hempel and Oppenheim’s model assumes that the explanans consists of statements of antecedent conditions and general laws that, via logical deduction, build the connection with the explanandum, or the description of the empirical phenomenon (Hempel and Oppenheim 1948: 137; see also Hempel 1965). In their view, in order to explain a certain event, one has to specify: the initial conditions, the underlying covering law, and the event that logically (i.e. deductively) unfolds from this setting.

Both models of explanation exerted profound influence in the nascent social sciences of the 20th century. The neo-Humean model accords to patterns of empirical regularities a pride of place in the social sciences (Beach and Pedersen 2019; Jackson 2017; Kurki 2008), and by subsuming causality to the constant conjunction of events, it eschews dwelling about the metaphysics of causes and related concepts, such as mechanisms, powers, forces, tendencies, capacities, just to name a few (Bhaskar 2008; Cartwright 1999, 2007). In a similar vein, the covering law model has become an important metric for generating knowledge in the social sciences, and the pursuit of general laws of social and international phenomena has been considered the ultimate goal of scientific inquiry by some quarters (e.g.Cuzán 2015, on laws in political science).⁵

It is in the context of the embracing of and reacting against the aforementioned models of causal explanation that the recent turn to philosophy of science in IR has gained thrust. Substantive engagements with the various versions of scientific realism (Kurki 2008; Patomäki and Wight 2000; Wendt 1987, 1999), as well as with pragmatism and interpretivism (Beach and Pedersen 2019; Bennett and Checkel 2015; Yanow and Schwartz-Shea 2015) have resulted in the revaluation of the central tenets of the neo-Humean and Hempelian models of causation. They also respond to the causal framework advanced by Gary King, Robert Keohane, and Sidney Verba (King, Keohane and Verba 2021), which has provoked a series of reactions since the publication of the first edition of their book in 1994. Underlying this process of reassessment is the goal of pluralising the views about causation by placing the ontological, epistemological, and methodological commitments entailed in causal inquiry under scrutiny (Kurki 2008). The resulting picture is far more diverse than the allegedly universal logic of causal inference advocated by King, Keohane and Verba (2021). In what follows, I briefly present some of the central conceptions of causality that emerged from these engagements and that now coexist with (and dispute) neo-Humean and Hempelian models of causal explanation.

The regularity model

The neo-Humean regularity approach to causes is dominant in IR, having acquired the status of the mainstream view of causal explanation (Kurki 2006; Jackson 2014, 2017). In its contemporary version, Hume’s regularities and constant conjunction translate into correlation, i.e. covariation between two variables (Goertz and Mahoney 2012; King, Keohane and Verba 2021). As a consequence, causality is inextricably tied to finding correlations between variables. It is informed by the empiricist legacy of searching for observable patterns of regularities, which are understood as the firm evidential basis for scientific knowledge.

King, Keohane and Verba (2021) epitomise this conception of causality as correlation and covariation, and their work Designing Social Inquiry (DSI, henceforth) has influenced much of the current debate not only about political methodology, but also, and more importantly, about how we conceive of causality. By and large,

[...] the modern debate was set off by King et al. (1994[2021]) and their claim that there is one logic of inference’, and ‘the underlying implication of the logic of inference is that only quantitative evidence can (a) pin down causation and (b) test hypotheses drawn from theory. (Dowding 2016: 162)

Inference is central to KKV’s argument: we make scientific advances by using facts that we know to speak about what we do not know. The authors conceive of two different types of inference: descriptive inference, which ‘is the process of understanding an unobserved phenomenon on the basis of a set of observations’ (King, Keohane and Verba 2021: 54); and causal inference, which is related to causal explanation, summarised in the following formula: ‘the causal effect is the difference between the systematic component of observations made when the explanatory variable takes one value and the systematic component of comparable observations when the explanatory variable takes on another value’ (King, Keohane and Verba 2021: 80). As one can easily anticipate, KKV’s concept of causality derives from statistics (indeed, it is a derivation of the Rubin-Holland model of causal inference), and unsurprisingly the language they use to describe causal relations reflects the very language of dependent and independent variables, covariation, and correlation. Ultimately, the causal logic described in DSI pervades much of the quantitative analyses in IR, especially in the American context (Lebow 2022).

Since the publication of DSI, political scientists have contested KKV’s understanding of social inquiry, and the more problematic claim of one single logic of inference (Gerring 2017; Hawkesworth 2015). It is not my goal to rehearse these critiques, but I would like to mention two consequences of this debate that affect discussions on causality. The first concerns the move to other forms of causal analysis which rely on necessary and sufficient conditions (Goertz and Mahoney 2012). To pin down causation, according to this view, implies looking for the conditions that cause a certain effect, either under a comparative perspective, or a within-case analysis. Advocates of necessary and sufficient conditions do not rely on the statistical apparatus of regularity-oriented scholars to construct their causal analyses, preferring instead to map these conditions. This has had methodological consequences for methods such as Qualitative Comparative Analysis (Ragin and Amoroso 2019), as well as to comparative-historical analysis (Thelen and Mahoney 2015).

The second consequence results from the engagements with mechanistic explanations and the philosophies of science underlying mechanistic thinking (Hedström and Ylikoski 2010). The search for causal mechanisms has become a potent response to the arguments of quantitative scholars, and it has generated a new stream of research with important ontological, epistemological, and methodological reflections (Beach and Pedersen 2019; Bennett and Checkel 2015). Process tracing⁶ exemplifies the turn to causal mechanisms, but the polysemy of the latter has resulted in various interpretations of the former, as well as in the interaction with different – and not always reconcilable – philosophies of science, such as scientific realism in its various flavours. Nonetheless, both types of thinking (i.e. conditions and mechanisms) characterise part of the critique of regularity-based approaches and provide alternative ways of conceptualising causation in IR.

Causal narratives

Drawing on the historiographic tradition of IR, some scholars have argued that the intellectual practice of the discipline has its foundations in the construction of narratives (Kurki and Suganami 2012; Lebow 2021; Suganami 1996, 2008). These narratives are explanatory to the extent that their goal is to impart an understanding about international phenomena, answering questions about events in the international arena. In particular, causal narratives convey explanations in virtue of tailoring the connections between events.

A causal narrative is a story that explains by connecting the two ends of the occurrence of an event: the beginning, where the event has not yet occurred; and the end, where the event has already unfolded. But this is no ordinary story: Suganami (2008: 334) enumerates three ingredients of causal narratives, namely chance coincidences, mechanistic processes, and human acts. Explanation, thus, emerges as the result of filling in the gap between the two ends with knowledge that answers the questions of our interest, i.e. ‘we need to show how, through what intelligible connection, the beginning (x) and the end (y) were linked; and to show this is to give a narrative account that has a beginning, middle, and end structure’ (Kurki and Suganami 2012: 404, emphasis in original).

Advocates of causal narratives argue that to explain a set of phenomena is not simply a matter of identifying correlational evidence, as if it were a self-evident response to a ‘why-question.’ Rather, causal narratives do their job by explaining why and how things happened in that particular case (Jackson 2017: 706; see also Suganami 1996). Evidence by itself is just that: evidence. It does not move from answering the essence of the questions that drive us to study international phenomena. More is needed, and the causal narrative supplies this need by ‘[diving] into the specifics of a particular case’ (Jackson 2017: 706), demonstrating how a sequence of events unfolds, mapping the landscape of alternative causal claims, and probing which causal explanation does better in imparting an understanding about those events. To a large extent, to offer a causal explanation is to render the knowledge of a cause more intelligible or familiar (Kurki and Suganami 2012; Suganami 2008; Jackson 2017).

The challenge faced by causal narratives consists in offering a causal account that does not solely rely on postulating a sequence of events. Some authors doubt that a descriptive narrative is capable of pinning down causation (Beach and Pedersen 2019), others question the ambiguity entailed in the ways historical narratives are constructed (Dowding and Miller 2019), and still others advance alternative methodological strategies of historical process tracing to tailor causal explanations (Bennett and Checkel 2015). Despite these challenges, Kurki and Suganami (2012) and Suganami (2008) advocate that causal narratives constitute a deeply rooted element of the disciplinary development of IR, a tradition traceable back to the early works of E. H. Carr and Hans Morgenthau (or even back to Thucydides in Ancient Greece). According to them, the explanatory purchase of a given causal narrative depends on how the academic community of IR evaluates the claims it advances, in a process of internal debate among scholars to determine which narratives are considered successful in explaining international phenomena, according to criteria that account for what counts as good explanation or not, such as ‘empirical testability and corroboration, coherence and intelligibility, a capacity to incorporate more detailed explanations, when required, without undermining themselves’ (Kurki and Suganami 2012: 415). Furthermore, they emphasise the narrative dimension of virtually all other forms of explanation: even works that rely on the neo-Humean model must, at some point, construct a narrative that makes sense of the causal links postulated as the result of statistical tests.7

The critical realist critique of causation

Dating back to the 1980s and 1990s, IR scholars looked for philosophical realism as a meta-theoretical source to reflect upon causality (Dessler 1989, 1991; Wendt 1987). Philosophical realism departs from the premise that the world we inhabit exists independently of our own existence, i.e. reality exists independently of what we think about it (Godfrey-Smith 2021). Insofar this premise states an idea that may sound self-evident – why would the existence of the real world depend on what we think about it? – what exactly the terms ‘world’, ‘reality’, and ‘independently’ mean has been theorised in a variety of ways, ultimately giving rise to different strains of philosophical realism. Scientific realism, in particular, has gained thrust in the discipline thanks to Wendt’s (1999) articulation through constructivism. But other scholars derived new insights from Roy Bhaskar’s (2008) critical realism, offering an alternative perspective to Wendt’s more positivist-based version of philosophical realism (Patomäki and Wight 2000; Kurki 2008; Wight 2006).

Bhaskar’s⁸ original conception of causation invoked the notion of causal powers and tendencies to rework the meaning of causes in open systems. Open systems are different from the controlled environment of laboratory experiments, where scientists control for the myriad interferences that may affect the phenomenon of their interest in what could be defined as a closed system: instead, the social world (and, to a large extent, the natural world) is a complex system where a variety of circumstances, factors, and causes interact to generate phenomena (Bhaskar 2008). In this sense, nomothetic models of causality, by being premised on the operations of closed systems and the requirements of necessity, produce causal laws that fail to incorporate the complexity of the real world, and due to this fact may not be realisable once the closed system is brought to the open landscape of the social world. Bhaskar, then, postulates the concept of powers to disentangle the role of causes in open systems:

To say that a thing has the power to do ø is to say that it will do ø, in the appropriate circumstances, in virtue of its nature (e.g. its intrinsic structure or genetic constitution). [...] To ascribe a power is to make a statement about possibilities which may not be actualized and which are possessed by the thing whether or not they are known by men. (Bhaskar 2008: 231)

The articulation of Bhaskar’s philosophy in IR has generated important insights to debates on causation. According to Kurki (2007: 364), the philosophical tenets of critical realism in respect to causation are: 1. Ontological priority to causes, which exist as real forces in the real world; 2. Causes are often unobservable, rendering empiricist methods inadequate to unravel causation; 3. Causes operate in a complex network of interactions, never in isolation or in a X-causes-Y fashion; and 4. Causes assume many forms, material and immaterial ones (e.g. ideas, norms, reasons, social constructions and structures) (see also Kurki 2008, and Patomäki and Wight 2000). As causes are frequently unobservable, critical realists in IR emphasise the search for the causal structures that lie beneath the surface of observable patterns, adopting an agnostic position in respect to the ways one uses them to generate understanding about the causal complexity of the world. The language of causal powers,⁹ thus, underscores precisely the open nature of natural and social systems, where causes are always embedded in a larger, pre-existing causal context (Kurki 2007; Patomäki 2017). In this sense, critical realists, in opposition to regularity theorists, prioritise the ontological dimension of causation, a feature that the latter eschews interrogating. Furthermore, critical realism is grounded on the theses of epistemological relativism, which stresses that knowledge is socially constructed, hence contextual and fallible; and judgemental rationalism, which establishes that knowledge claims are subject to various interpretations and explanations that need to be compared in order to assess their merits and truth-value (Patomäki 2019: 192).

Interpretative causation

In the more recent turn to interpretivism, scholars have directed their attention to the peculiar characteristics of hermeneutic explanations, and in this process began to reconsider the meanings that causation assume in their research practices (Guzzini 2017; Jackson 2015, 2017; Lebow 2015, 2022; Norman 2021; Yanow and Schwartz-Shea 2015). Traditionally, interpretivism has been associated with the constitution-cause divide, which is premised on the distinction between what constitutes the properties of real world structures, on the one hand; and the relationships of cause and effect that they are capable of generating (Wendt 1998; Ylikoski 2017). In this divide, interpretivism would naturally side with constitutive explanations, for the research focus is centred around social structures, cognitive models, cultures, identities, norms, systems of meaning, and local contexts. However, recent meta-theoretical interventions have produced novel understandings about the role of causation in interpretative research by either re-signifying the meaning of constitution and its relation to cause, and how they are articulated in singular causal analysis (Guzzini 2017; Lebow 2015; Jackson 2015); or by eschewing conflating causation and constitution, as well as reconceptualising causal explanation under a different ontological basis (Norman 2021).

Singular causal analyses belong to a Weberian tradition of configurational causation, whereby ‘a causal configuration [...] is identified and demonstrated by showing how the configuration in question interacts with a range of possible factors, where possible factors emerge from a historically grounded study of the situation itself’ (Jackson 2015: 279; see also Jackson 2014). Causality, thus, results from the operation of a multiplicity of causal mechanisms historically unique and situated in a specific context. In constructing the analysis, researchers consider a variety of factors related to the patterns of signification resulting from the cultural resources permeating a given setting (Jackson 2015, 2017). These patterns constitute the causal mechanisms that, once set in motion, have the capacity to generate outcomes in the real world. But possessing the capacity or disposition to generate an outcome is no warrant that it will do so, because the mechanism is immersed in a wider web of events that interact with it, leading to its triggering or not (Guzzini 2017). Consequently, an interpretivist mechanism is not a deterministic material device that always produces the same result as in regularity-based approaches to causation: rather, many paths may ensue depending on how a specific mechanism interacts with other mechanisms. Therefore:

Thinking causality in terms of dispositions and process avoids pure contingency and yet allows for a causal analysis which is not constant conjunction or efficient causality. Such causal mechanisms are part of, but not reducible to a theoretically informed explanation of a wider process that remains open. (Guzzini 2017: 752)

This raises the question of complexity and how a causal analysis deals with a world that is an open system of various interacting factors. As reality is intrinsically complex, with intertwined layers of social processes, Lebow (2015: 65-66) postulates inefficient causation as a concept and strategy to identify and interpret causal webs, embracing their complexity in a narrative that captures the systems of meanings, cognitive frames of reference, processes, and mechanisms, all of which generate particular, idiosyncratic outcomes. Inefficient causation emphasises that the world is pervaded with multiple causes and causal chains, which interact and operate across analytical levels and social domains to produce the outcomes we observe. Consequently, one has to draw a causal map through an iterative process, for ‘tacking back and forth between putative causes and their effects might lead us back, step by step, to deeper levels of possible causation’ (Lebow 2015: 66). More fundamentally, inefficient causation departs from a complex world where sources of causes cannot be subsumed in a straightforward fashion to a set of factors which operate in isolation: rather, the complex web of phenomena and actors generate multiple combinations of causes and causal chains.

In light of the aforementioned understandings of causal explanation, Norman offers an alternative framework to conceive of singular causation drawing on Woodward’s (2003) counterfactual interventionist theory of explanation. Norman (2021: 945) contends that ‘[s]ingular causal explanations capture relations between events (broadly defined) and constitutive explanations capture relations between the characteristics of some social whole and some aspect or part of that whole,’ but both of them are cemented on a counterfactual basis where explaining is always conceived of in terms not only of the relationship between explanans and explanandum, but also in respect to alternative, counterfactual worlds. To explain an event means to provide an understanding of why things were the way they were instead of some other possible way. By resorting to a well-established model of causal explanation in the philosophy of science, Norman eschews the need for reconceptualising causation in terms that fit into interpretative research. As he sees no intrinsic ontological difference between causal claims in other forms of research and causal claims in interpretative research, postulating an alternative ontology of causality does not necessarily solve the challenges faced by interpretivists in their inquiries. Instead, he advocates for reframing the terms of the constitutive-causal divide, for ‘constitutive explanations help make sense of how events are interpreted and acted upon by socially situated agents and why causal processes are triggered and unfold in particular ways’ (Norman 2021: 947).

All the preceding debates advance specific conceptions of causation, some of which dive into ontological reflections to reconsider their epistemological and methodological implications, whereas others eschew investing so much effort in metaphysical discussions. Nonetheless, the fact that scholars engage with different philosophies of explanation to inform their research demonstrates how contentious the issue is in the discipline, with some quarters arguing that there is no substantive difference between the allegedly competing conceptions of causation (e.g., Gerring 2005), whilst others prefer to embrace the multiple meanings of cause co-existing in IR research (e.g., Humphreys 2017).

Causality in Brazilian IR: examining explanations of Brazilian foreign policy analysis

One of the most prominent fields of Brazilian IR is that of foreign policy analysis. Since the advent of the first undergraduate and graduate programmes in the national academia, IR scholars have devoted large efforts to map the history of BFPA and understand the rationales behind foreign policy decisions and decision-making processes (Casarões 2018; Reis da Silva 2023; Salomón and Pinheiro 2013). The burgeoning literature of BFPA is, therefore, an interesting starting point to examine the kinds of explanations advanced by researchers, and how they attend to ideas of causality.

To be sure, the field of foreign policy analysis is particularly sensitive to beliefs about causality due to the practical uses of foreign policy recommendations and analyses for policy-making and decision-making processes (Chernoff 2007). Having an understanding of cause and effect, broadly understood, allows policy makers to grasp the consequences of today’s decisions on the future. How causality is, thus, perceived depends not only on theoretical frameworks, but also on the meta-theoretical principles that inform these theories. These principles are reflected in theory choice, methodological approaches, and scholars’ own conceptions of causal explanation, which are all interconnected in the course of an analysis of foreign policy.

In recent years, bibliometric research has drawn a clearer picture of some theoretical and methodological preferences of different fields in Brazilian IR. Medeiros et al.’s (2016) study of 35 South American journals (out of which 16 are Brazilian) is still one of the most extensive attempts to characterise the discipline in the region. Their findings demonstrate that mainstream theories are dominant in the literature, especially liberalism, realism, constructivism (which, together, account for more than half of theoretical choices in Brazilian journals), while methodological information is remarkably absent in the vast majority of their sample. Further replications of this type of research have yielded similar results (Novelli 2022), with the persistent failure of Brazilian IR scholars to render their methods and methodological discussions transparent in their works, which may be explained by limited awareness to methodology in the curricula of undergraduate and graduate programmes (cf. Albuquerque, Mesquita and Brito 2022), and to the historical-descriptive tradition that is still prevalent in the national discipline (cf. Carvalho, Gabriel and Lopes 2021). More importantly to the present discussion, in the specific field of foreign policy analysis, Medeiros et al. (2016) find similar trends that mirror the general image of Brazilian IR: liberalism, realism, and constructivism are the dominant theories, followed by qualitative and quantitative methodologies.

This scenario aids to understand the place of causal explanation in Brazilian IR and, more specifically, in BFPA. In order to further evaluate the meta-theoretical principles which underlie the theoretical and methodological choices of Brazilian scholars associated with this field, I surveyed all papers on BFPA published in the Brazilian Political Science Review (BPSR), Contexto Internacional (Contexto), Dados, Revista Brasileira de Ciências Sociais (RBCS) and Revista Brasileira de Política Internacional (RBPI) from 2018 through 2022, aiming to assess whether causality is a concern; and, should that be the case, what aspects of causality resonate with the preceding meta-theoretical debates. As one might anticipate, this corpus is limited to articles published in a small number of journals, hence the findings are not generalisable beyond these constraints.10

As this is, to the best of my knowledge, the first assessment of causal explanation in Brazilian IR,¹¹ I framed the debate in terms of four categories of causality that sum up the central tenets of the previously discussed conceptions of causal explanation: regularity causality, which emphasises the regularity approach of the Humean and covering law models (King, Keohane and Verba 2021); causal narratives, which refers to the construction of causal narratives that connect different events into a coherent story to impart an understanding (Suganami 1996, 2008); hermeneutic causation, which draws on interpretative causation and its different approaches to incorporate patterns of meanings into the form of singular causal analysis (Jackson 2015; Lebow 2015, 2022; Norman 2021); and causal powers, which invokes (explicitly or implicitly) philosophical realist ideas of tendencies, capacities, and powers (Bhaskar 2008; Kurki 2008; Patomäki 2019). These categories respond to the philosophical desiderata of the causal debate in IR, for the literature is still hesitant in its engagement with philosophical understandings about causation and, more importantly, explanation. For instance, the categories mobilised by Mantzavinos (2016) and Rice and Rohwer (2021) – such as covering law, equilibrium, statistical, causal, mechanistic, unificationist, and mathematical explanation – do not directly resonate with current discussions of explanation and causality in IR. Therefore, the categories hereby proposed are aligned with current meta-theoretical debates in the discipline. Evidently, they are not meant to be exhaustive, but rather to aid in delineating the explanatory routes taken by Brazilian scholars when the causal talk is invoked. Furthermore, they are not rival nor exclusive, for different criteria of causal explanation may coexist in the same work (see, for example, Chernoff 2014).

From the outset, it is worth noting that Brazilian scholars rarely make direct mentions to causal explanation, which, thus, requires different strategies than those used in typical bibliometric studies to identify the type of explanation a given work aims to advance. In order to make the present assessment possible, I proceeded by initially screening the papers according to whether the author(s) explicitly presented the goal of tailoring an explanation. In RBPI (N = 29), 17 papers made explicit reference to explanation, often using the verb ‘to explain’ in various conjugations; in Contexto (N = 12), that number was 8; in BPSR (N = 8), that was 7 articles; in Dados (N = 2) and in RBCS (N = 2) only 1 article in each of them aimed to explain. After I screened the papers, I proceeded to identify the type of explanation by fully reading each work.¹² In this process, I analysed the full set of papers, irrespectively of their stated goal. This was a necessary step for not all authors stated their research goals in terms that are commonly associated with explanation. Furthermore, this procedure provided insights about what authors aimed to do in papers where explaining is not the primary focus. Papers were coded based on the aforementioned categories, plus the extra descriptive-understanding category, which comprises ‘non-explanatory’ research.¹³ I was also open to inductive insights that provided further information about how explanation is conveyed in each work. Methodological choices were also identified as a supplementary category that allowed to assess the connections between certain categories of explanation and methods. The results are displayed in tables 1 and 2.

The results in tables 1 and 2 require further scrutiny. In Contexto and RBPI, which are journals disciplinarily located in IR, not only BFPA works are founded on causal approaches that resonate with interpretative causation, but also with causal narratives. To be sure, the findings demonstrate that causal narratives tend to mobilise the ontological arsenal of interpretivism, which extrapolates from the description of how events unfold in a causal sequence. This is an important distinction in respect to other types of narratives that do not aim to advance causal claims. Furthermore, these papers were more likely to resort to qualitative methodologies to pin down causation.

The picture is rather different in BPSR, Dados and RBCS: these journals are disciplinarily located in political science (the case of BPSR) and the social sciences (Dados and RBCS), and the frequency of IR papers they publish is strikingly smaller than in Contexto and RBPI. Furthermore, in their disciplinary locations, quantitative methodologies are more commonly used than in IR journals. This fact responds for the 50.00% frequency of regularity-based types of explanation of BFPA. As one can easily note, this is not the standard approach in Contexto and RBPI, where not only regularity causality accounts for 25.00% and 10.34% of articles, respectively, but also quantitative models are scarcely used – a finding that resonates with Casarões’ assessment (Casarões 2018: 419).

Comparing the results of these two sets of journals (one that is disciplinarily located in IR and another which is not) reveals interesting patterns of explanatory enterprises. Firstly, BFPA reflects the general trend of a nonquantitative Brazilian IR (Medeiros et al. 2016; Novelli 2022), defined by methodological practices that are not always specified in a transparent fashion (Albuquerque, Mesquita and Brito 2022; Medeiros et al. 2016). This has important implications for how explanations are tailored, and, in the case of causal explanation, it affects the articulation of causal arguments, for ontology, epistemology, and methodology are intrinsically connected in the explanatory enterprise (Kurki 2008). Unsurprisingly, causal narratives, and hermeneutic causation are dominant in BFPA, which has long been a field characterised by a descriptive and narrative tradition (Carvalho, Gabriel and Lopes 2021). However, it is important to note that this shift has produced interesting engagements with ontological and epistemological debates on how ideas and identities produce foreign policy outcomes. As for the set of non-IR journals, and despite the samples’ smaller size, the intersection with other disciplinary practices may have affected the frequency of regularity-based explanations. This finding deserves further scrutiny in subsequent iterations of this analysis and may provide illuminating understandings of the place of regularity-based approaches to explanation in Brazilian IR (I shall discuss this further in the next section). A number of analyses of BFPA relied on the language of ideas, identities, and discourses – sometimes in direct connection with constructivist theories – to advance causal arguments. Although not explicitly, the construction of causal narratives in these papers often resembled the concept of causal powers/forces, with ideas and reasons being capable of generating specific outcomes in foreign policy. In these papers, ideas were central to complement the focus on material factors in the tailoring of causal analyses. For example, Silva (2020) analysed the effects of chancellors’ perceptions and images in the bilateral relationship between Argentina and Brazil; Kotyashko, Ferreira-Pereira and Vieira (2018) examined the role of self-images and identities as drivers of norms contestation (specifically, Responsibility to Protect) by Brazil and Russia; and Barros and Gonçalves (2021) demonstrated how changes in foreign policy actors’ ideas generate outcomes in South American integration. Methodologically, authors affiliated with this approach were more likely to resort to multiple techniques to map ideas, namely interviews with foreign policy-makers and discourse analysis. Other methodologies involved content analysis, QCA, process tracing, and other forms of textual, archival, and document analysis.

The case of causal powers is particularly interesting. The lexicon of ideational resources and mechanisms in causal explanations suggests the notion of powers, broadly understood. To be sure, the articles themselves did not engage with critical realist perspectives or other philosophical approaches that mobilise the language of powers, tendencies, or capacities, but the use of ideas and mechanisms as elements of the explanatory enterprise opens venues for further meta-theoretical engagements with scientific realism and other philosophies of science.

What can be said about the descriptive-understanding category? Papers in this category were commonly structured around descriptive narratives, which are not causal in the sense that causal evidence is presented and mobilised throughout the analytical enterprise, nor any sort of mechanism or causal pattern is delineated as part of the argument (Beach and Pedersen 2019). Although their intent may be grounded in a view of explanation-as-description (Dowding 2016: chapter 3), works that rest on descriptive narratives do not postulate nor probe causal arguments or mechanisms. This by no means diminishes their value in terms of academic inquiry, for they can be the point of departure for more causally oriented research. As for the understanding side of the category, authors who subscribed to this view not only mobilised hermeneutic frameworks in their papers, but also explicitly stated their research goals in terms of ‘understanding’ instead of ‘explaining’.

Toward new understandings (and awareness) of causality in Brazilian IR

The results presented in the previous section, though restricted to the field of BFPA and constrained by the short time frame of analysis and small number of journals assessed, are illuminating about the current state of meta-theoretical and philosophical debates in Brazilian IR. As pointed out, in recent years scholars have produced interesting assessments of the research output in the discipline, many of which converge to the persistent unawareness of methodological issues (Casarões 2018; Lopes, Nicolini and Carvalho 2022). Therefore, a discussion of causality in Brazilian IR – and in BFPA, in particular – cannot ignore the sobering results of bibliometric works, for they involve not only ontological and epistemological concerns, but also methodological ones. However, as the scope of this paper was restricted to BFPA, in the remainder of this section I shall focus on how the Brazilian scholarship may foster awareness on meta-theoretical debates about causation, and how we can advance national research on foreign policy analysis by incorporating specific versions of causality that fit more neatly with the current practice of IR research in the national academy.

To a large extent, the BFPA literature analysed follows the central tenet of the explanatory enterprise advanced by Suganami in the 1990s: ‘To state the cause of an event is to explain its occurrence. To explain the occurrence of an event is to render its occurrence more intelligible than before’ (Suganami 1996: 139). Most works in this analysis engage with causal explanation not through a direct and explicitly stated concern with causality, but rather by mobilising an epistemological and methodological apparatus capable of pinning down causation. To be sure, ontological commitments regarding causal claims are rarely stated in a transparent fashion: they rather underpin the broader ontological framework informing the research – for example, when authors postulate that ideas, motives, and material resources possess the power or force to generate outcomes in foreign policy; or when they invoke causal mechanisms whose operations generate certain outcomes. This is not in itself a major concern, for it is not necessary for a researcher to indulge in meta-theoretical incursions prior to presenting the structure and findings of her research. What is necessary, though, is a background awareness about the existence of different forms of tailoring causal claims in order to avoid treating causality as a neopositivist enterprise.

The articles in the corpus mostly resonate with conceptions of causation advanced by critical realists and interpretivists. Moreover, the research practice of Brazilian scholars in BFPA seems to be primarily oriented towards narratives, either causal or descriptive. Most works rely on qualitative data of various forms, such as discourses, official documents, texts, messages, interviews, and ethnography, just to name a few. The neo-Humean, regularity-oriented approach to causality does not cope with the idiosyncrasies entailed in these research designs. Consequently, other conceptions of causal explanation are required in order to make sense of a praxis that is first and foremost centred on the role of ideas, motives, social constructions and change in generating foreign policy outcomes. If ideational resources are so important to BFPA, then taking them seriously requires the language of causal powers, the tailoring of causal narratives, and the use of causal mechanisms.

This meta-theoretical shift necessitates the engagement with philosophies of science that conceive of causality in terms that expand the ontological apparatus beyond material (and observable) resources, whilst embracing the metaphysics of causes in ways that are capable of being mobilised for practical goals in research and policymaking. In this sense, scientific realist and interpretative accounts provide two venues for exploring new understandings about causation with more compatible ontologies (with the epistemological and methodological implications that follow them) with the current praxis of BFPA research. In so doing, it may be possible to shed light on the frequently emphasised ‘methodological obscurity’ of Brazilian IR (Albuquerque, Mesquita and Brito 2022; see also Medeiros et al. 2016): by specifying causality in alternative ontological grounds, one can make epistemological and methodological choices that are consistent with one’s research goals without having to rely on a strict formula for research design specification. More importantly, it invites researchers to think of their research as a holistic construction, where ontology, epistemology, and methodology are intertwined in a productive fashion (Kurki 2008).

Critical realist perspectives on causation resonate with much of the current state of causal inquiry in Brazilian IR. As a philosophy of science, it offers meta-theoretical insights to discuss the ontological implications of explanations that depend on powers and mechanisms to render a set of phenomena intelligible (Patomäki and Wight 2000; Wight 2006, 2007). More importantly, critical realism embraces the causal potentialities of ideas, norms, social constructions, which figure in most debates of the BFPA literature (Casarões 2018; Lopes, Nicolini and Carvalho 2022). Furthermore, its epistemological and methodological pluralism is an asset (see Kurki 2008; Patomäki 2019) if one considers the variety of research strategies used by Brazilian scholars to explain international phenomena. However, making the move to critical realism – or to scientific realism, in general – requires re-thinking how Brazilian researchers engage with philosophical discussions in the course of their academic trajectory. As Albuquerque, Mesquita and Brito’s (2022) analysis of the syllabi of graduate courses of IR in Brazil demonstrates, debates about causal explanation are frequently held in methodology courses, where classical international handbooks are frequently used to teach the next generation of students. Introducing scientific realism and alternative philosophies of science, such as interpretivism, face the challenge of making room in syllabi that tend to reproduce canonical understandings of science, epistemology, and methodology.

Interpretative approaches to causation also figure as alternatives that can clarify the relationships between constitution and cause, and can specify methodological choices in the course of the research endeavour. The focus on contextualising local realities and subjects in BFPA requires engaging with the constitutive side of explanation in ways that clarify the link between what constitutes reality and the properties of these constituents; and the causal processes that generate certain outcomes as a result of the powers entailed in the properties of the constituents of the world. Lebow’s (2015) inefficient causation, Jackson’s (2017) Weberian adequate causation, Guzzini’s (2017) interpretative process tracing centred on mechanisms-as-capacities, and Suganami’s (1996, 2008) causal narratives articulate both sides of constitution and causation in ways that incorporate the complex web of phenomena, whilst imparting understanding about how this complexity operates. But researchers might also be interested in making a clear distinction between both types of explanation entailed in the constitution-causation divide, in which case Norman’s (2021) proposal of interpretative process tracing may offer a more productive methodological device to work out the distinguishing features of constitution and causation, namely by working out ‘how changes in intersubjective worlds play into our causal explanations of well-specified social and political outcomes,’ which, in turn, ‘enables a sharper differentiation of dispositions of particular social systems from causal analyses that map relations between events’ (Norman 2021: 951). What is interesting about interpretative process tracing – both Guzzini’s and Norman’s versions – is that it provides a framework that combines ontological, epistemological, and methodological commitments that articulate the current praxis of BFPA research with the methodological desideratum of research transparency.

At this point, a few caveats regarding regularity-based frameworks of causation are in order. Advocating for philosophies of science that provide alternative conceptions of causal explanation by no means entail a complete rejection of neo-Humean frameworks to analyse international phenomena, especially in terms of its orientation toward quantitative methodologies. Brazilian IR is still in its infancy in terms of quantitative analyses: few works rely on statistical data and models to advance causal claims. The discipline might benefit from neo-Humean and even Hempelian frameworks, for they rely on extensive generation and use of data on important international phenomena in various domains of IR inquiry, and ignoring them obfuscates a set of methodological possibilities concerning causation. Furthermore, the nomothetic orientation entailed in quantitative methodologies satisfies the desideratum of producing knowledge for the management of social affairs (Suganami 2008: 347, fn. 41). In the case of BFPA, interdisciplinary engagements have led to interesting findings based on the tenets of regularity-based causality. However, quantitative analyses are still scarce in this field: for instance, Casarões (2018: 419) maps few attempts to orchestrate analyses based on quantitative methods (and, as a consequence, on neopositivist forms of causal explanation). I concur with him that a quantitative approach might provide new insights for BFPA; likewise, regularity theories and the covering law model may generate invaluable causal knowledge on foreign policy processes and outcomes.

This does not imply that Brazilian IR must change its self-image to mirror the international output of prestigious journals and schools (especially, American IR), but a revaluation of the locus of quantitative research based on neo-Humean and Hempelian approaches to causality is a necessary step in raising awareness about the different modes of tailoring causal explanation. It is undeniable that these frameworks of causality contribute to producing knowledge of international phenomena: the question is how one uses them as the understanding of cause/causation. As IR is now a global discipline, with different practices emanating from local forms of knowledge production (Acharya 2021), a pluralist approach to causality is ever more urgent in order to serve the different research goals of the various practitioners of IR in the Global South. In so doing, we may respond to the Global IR desiderata of avoiding ‘cultural exceptionalism’ (especially Western exceptionalism) and discovering ‘new patterns, theories, and methods from world histories’ (Acharya 2016: 5, fn. 2), which can pluralise our understanding of causes and causality (see Acharya 2023: 275). The bibliometric analysis showed how different concepts, theories and methods are creatively mobilised in BFPA to tailor causal explanations, which is essentially the central tenet of the Global IR movement (Acharya 2021; Blaney and Tickner 2013; Tickner 2013). As the discussions above have demonstrated, the ontological, epistemological, and methodological commitments of our research have consequences not only for the academic goal of producing explanations, but also for political decisions about how to use them in concrete, real-world settings.

Conclusion

Causal analysis has been central to IR inquiry, but how one understands what ‘causal’ means is subject to different interpretations. Its centrality resides in the fact that causes are used by various audiences to produce certain desired outcomes. As Humphreys (2017: 662) argues: ‘[causal claims] are used by politicians, policymakers, academics, and ordinary citizens to satisfy curiosity, to resolve disputes, to justify actions (including, for example, the use of force), to counter criticism, to articulate and defend viewpoints, to allocate resources, and more.’

Reflecting upon the meanings of causation is paramount to the conduct of inquiry in IR. Throughout this paper, I rehearsed some of the most central accounts of causation in the discipline, and assessed their penetration in Brazilian IR, more specifically, in BFPA works. Differently from traditional bibliometric research, where theoretical and methodological preferences are easily identified by their mentioning in the text, the type of explanation advanced in a given work is less explicit, not only because explanation is treated as a self-evident problem, but also because delving into philosophical and meta-theoretical details is rarely a research concern, especially in more empirically oriented studies. The main thrust of my argument, hence, consisted in filling in this gap by showing how causal inquiry is implied in Brazilian research of international phenomena, even when scholars are not aware of the explanatory enterprise they are engaging with.

The findings of this research demonstrate the predominance of hermeneutic causation and causal narratives in BFPA articles in IR journals, and a more nuanced picture in journals that are not disciplinarily located in IR, where regularity-based forms of causal explanation are present in a significant number of works. Causality, in BFPA – and, to a large extent, in Brazilian IR –, is primarily advanced through alternative paths that do not replicate the neo-Humean, neopositivist forms of causal explanation. To be sure, there is room for deeper reflections about how these alternative paths and the philosophies of science they involve might generate interesting insights about our practices as IR researchers. This is paramount to overcome the descriptive nature of Brazilian IR scholarship, which still responds for a substantial part of our research.

The picture of BFPA drawn above is just one among the many other pictures of the kaleidoscope of Brazilian IR. Other fields most certainly mobilise the conceptions of causation discussed in the paper, as well as other conceptions that could not be covered in my presentation of the meta-theoretical debates in IR. Nonetheless, the maxim of reflecting upon one’s ontological, epistemological, and methodological commitments is valid to any other field of academic inquiry in the discipline. What we need to interrogate, even if it does not come to the light in the course of describing the underlying assumptions of our works, is how explanation is produced in our research, what its properties are, how it imparts an understanding about international phenomena, and, finally, what constitutes good explanation. The answers we give to those questions are not merely a matter of meta-theoretical curiosity but do influence how we conceive of world politics and how we can act upon it. Following the epigraph in the opening of this paper, making sense of causation is a necessary step in understanding where we are and have been, but, most importantly, how we shape our future in the international scene.

Notes

• 1
  For a long time, meta-theoretical discussions about explanation and causality in IR have been guided by Thomas Kuhn’s understanding of paradigms and scientific revolutions, and Karl Popper’s falsifiability thesis. They still inform the self-image of the discipline, but other engagements with the philosophy of science (e.g. scientific realism) have been more productive in discussions about causal explanation.
• 2
  In this paper, BFPA refers to the output of articles on Brazilian foreign policy analysis published in Brazilian journals over time.
• 3
  Apart from Contexto Internacional and RBPI, the other journals publish on a variety of issues and themes that are specific to other disciplines, such as political science, sociology, and anthropology. This variety allows to assess how the intermingling with other disciplinary practices affects the explanatory enterprises in BFPA.
• 4
  Current debates about explanation in the philosophy of science acknowledge the plurality of understandings of cause/causation (Cartwright 2007; Mantzavinos 2016; Rice and Rohwer 2021). Mapping all these approaches is a far greater goal than the one I advance in this paper, but it is worth noting that the philosophy of explanation admits of variegated explanatory enterprises, some of which have already been incorporated in IR meta-theory (e.g. Bhaskar, Cartwright, Woodward).
• 5
  In IR, this hunt for laws is epitomised by the Democratic Peace debate.
• 6
  Even KKV concede that constructing causal explanations requires specifying, at some point, how processes unfold between cause and effect (King, Keohane and Verba 2021). Parsons (2015: 518) notes that this move distances not only KKV, but also other Humean-oriented researchers, from the rigidness of Hume’s constant conjunction – indeed, much of the current practice of IR research that is quantitatively oriented admits of process tracing analyses and mechanistic explanations (see also Gerring 2010).
• 7
  Game theorists also introduce their models and results via narratives, not only as part of enunciating the setting that the game is supposed to represent, but also to render results more intelligible to audiences that are not familiar with the mathematical details of the model. From another perspective, as Suganami (2008) points out, Bates et al.’s (1998) analytic narratives exemplify the use of narratives to impart understandings whilst using rational choice theory-based approaches to analyse social phenomena.
• 8
  A complete discussion of Bhaskar’s philosophy is beyond the scope of this article. For a comprehensive introduction to critical realism, see Collier (1994).
• 9
  For a discussion of causal powers in the philosophy of science, see Cartwright (1999, 2007).
• 10
  A more robust analysis would require extending the time frame and including more journals, and other types of publications, such as books, book chapters, dissertations and theses. This is beyond the goal of this article, but the results hereby presented should work as a starting point for further meta-theoretical research.
• 11
  Lenine et al. (2023)’s analysis of the uses of process tracing in Brazilian IR also discusses issues of causality from a methodological standpoint
• 12
  Chernoff (2014) uses a similar approach to assessing the progress of explanation in three subfields of international security. First, he proceeds by rehearsing philosophical debates about explanation, and then selects exemplary IR works in nuclear proliferation, democratic peace, and balance of power debates according to a series of procedures that aim to identify the more relevant studies in each subfield, ending up with a sample of ten to twelve works in each of them. Despite the small number of ‘cases,’, Chernoff was able to draw a picture of the explanatory enterprise in these distinct areas of international security. Therefore, quantitative limitations should cause no trouble in this sort of meta-theoretical investigation.
• 13
  Although explanation may also take the form of descriptions (Dowding 2016: 36), and considering that explanation and understanding are intertwined (Chernoff 2007, 2014), the distinction hereby presented is informative about how Brazilian IR scholars conceive of their research goals and practices.

References

• Acharya, A. 2016. ‘Advancing Global IR: Challenges, Contentions, and Contributions’. International Studies Review 18 (1): 4-15.
• Acharya, A. 2021. ‘Global International Relations’. In T Dunne, M Kurki and S Smith (eds.), International Relations Theories: Discipline and Diversity (5th edition). Oxford: Oxford University Press, pp. 304-321.
• Acharya, A. 2023. ‘Before the Nation-State: Civilizations, World Orders, and the Origins of Global International Relations’. Chinese Journal of International Politics 16: 263-288.
• Albuquerque, R B de, R Mesquita and R V L Brito. 2022. ‘Obscuridade metodológica: um mapeamento da formação em métodos na pós-graduação em Relações Internacionais e áreas afins no Brasil’. Revista Brasileira de Ciência Política 39: 1-25.
• Barker, G and P Kitcher. 2014. Philosophy of Science: A New Introduction. Oxford: Oxford University Press.
• Barros, P S and J de S B Gonçalves. 2021. ‘Crisis in South American regionalism and Brazilian protagonism in Unasur, the Lima Group and Prosur’. Revista Brasileira de Política Internacional 64 (2): 2009.
• Bates, R H, A Greif, M Levi, J-L Rosenthal and B R Weingast. 1998. Analytic Narratives. Princeton: Princeton University Press.
• Beach, D and R B Pedersen. 2019. Process-Tracing Methods: Foundations and Guidelines. Ann Arbor: University of Michigan Press, 2nd edition.
• Bennett, A and J T Checkel. 2015. ‘Process tracing: From philosophical roots to best practices’. In A Bennett and J T Checkel (eds.), Process Tracing: From Metaphor to Analytic Tool. Cambridge: Cambridge University Press, pp. 3-37.
• Bhaskar, R. 2008. A realist theory of science. London: Verso.
• Blaney, D L and A B Tickner. 2013. ‘Introduction: claiming the international beyond IR’. In A B Tickner and D L Blaney (eds.), Claiming the International. London: Routledge, pp. 1-24.
• Bueno de Mesquita, E and A Fowler. 2021. Thinking Clearly with Data: A Guide to Quantitative Reasoning and Analysis. Princeton: Princeton University Press.
• Casarões, G. 2018. ‘The evolution of foreign policy studies: Four perspectives’. In B Ames (ed.), The Routledge Handbook of Brazilian Politics. London: Routledge, pp. 406-429.
• Cartwright, N. 1999. The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge University Press.
• Cartwright, N. 2007. Hunting Causes and Using Them: Approaches in Philosophy and Economics. Cambridge: Cambridge University Press.
• Carvalho, T, J P N Gabriel and D B Lopes. 2021. ‘‘Mind the Gap’: Assessing Differences between Brazilian and Mainstream IR Journals in Methodological Approaches’. Contexto Internacional 43 (3): 461-488.
• Chernoff, F. 2007. Theory and Metatheory in International Relations: Concepts and Contending Accounts. New York: Palgrave Macmillan.
• Chernoff, F. 2014. Explanation and Progress in Security Studies: Bridging Theoretical Divides in International Relations. Stanford: Stanford University Press.
• Collier, A. 1994. Critical Realism: An Introduction to Roy Bhaskar’s Philosophy. London: Verso.
• Cuzán, A G. 2015. ‘Five Laws of Politics’. PS: Political Science & Politics 48 (3): 415-419.
• Dessler, D. 1989. ‘What’s at stake in the agent-structure debate?’ International Organization 43 (3): 441-473.
• Dessler, D. 1991. ‘Beyond Correlations: Toward a Causal Theory of War’. International Studies Quarterly 35 (3): 337-355.
• Dowding, K. 2016. The Philosophy and Methods of Political Science. London: Palgrave.
• Dowding, K and C Miller. 2019. ‘On prediction in political science’. European Journal of Political Research 58 (3): 1001-1018.
• French, S. 2006. ‘Structure as a Weapon of the Realist’. Proceedings of the Aristotelian Society 106(2): 167-185.
• Gerring, J. 2005. ‘Causation: A unified framework for the social sciences’. Journal of Theoretical Politics 17 (2): 163-198.
• Gerring, J. 2010. ‘Causal Mechanisms; Yes, But...’ Comparative Political Studies 43 (11): 1499-1526.
• Gerring, J. 2017. Qualitative Methods. Annual Review of Political Science 20: 15-36.
• Godfrey-Smith, P. 2021. Theory and Reality: An Introduction to the Philosophy of Science. Princeton: Princeton University Press.
• Goertz, G and J Mahoney. 2012. A Tale of Two Cultures: Qualitative and Quantitative Research in the Social Sciences. Princeton: Princeton University Press.
• Grynaviski, E. 2013. ‘Contrasts, counterfactuals, and causes’. European Journal of International Relations 19 (4): 828-846.
• Guzzini, S. 2017. ‘Power and cause’. Journal of International Relations and Development 20: 737-759.
• Hawkesworth, M. 2015. ‘Contending Conceptions of Science and Politics: Methodology and the Constitution of the Political’. In D Yanow and P Schwartz-Shea (eds), Interpretation and Method: Empirical Research Methods in the Interpretive Turn (2nd edition). Londres: Routledge, pp. 27-49.
• Hedström, P and P Ylikoski. 2010. ‘Causal Mechanisms in the Social Sciences’. Annual Review of Sociology 36: 49-67.
• Hempel, C G and P Oppenheim. 1948. ‘Studies in the logic of explanation’. Philosophy of Science 15 (2): 135-175.
• Hempel, C G. 1965. Aspects of Scientific Explanation and Other Essays in the Philosophy of Science. New York: Free Press.
• Hollis, M. And S Smith. 1990. Explaining and Understanding International Relations. Oxford: Clarendon Press.
• Hume, D. 1952. ‘An Enquiry Concerning Human Understanding’. In R M Hutchins (ed.), Great Books of the Western World. Chicago: University of Chicago Press, pp. 451-509.
• Humphreys, A R C. 2017. ‘Introduction: problems of causation in world politics’. Journal of International Relations and Development 20: 659-666.
• Jackson, P T. 2014. ‘Broadening Causation Without Diluting Causation’. Qualitative & Multi-Method Research 12: 3-6.
• Jackson, P T. 2015. ‘Making Sense of Making Sense: Configurational Analysis and the Double Hermeneutic’. In D Yanow and P Schwartz-Shea (eds.), Interpretation and Method: Empirical Research Methods and the Interpretive Turn. Abingdon: Routledge, 2nd edition, pp. 267-283.
• Jackson, P T. 2016. The Conduct of Inquiry in International Relations: Philosophy of science and its implications for the study or world politics (2nd edition). London: Routledge.
• Jackson, P T. 2017. ‘Causal claims and causal explanation in international studies’. Journal of International Relations and Development 20: 689-716.
• Kincaid, H. 1996. Philosophical Foundations of the Social Sciences: Analyzing Controversies in Social Research. Cambridge: Cambridge University Press.
• King, G, R O Keohane and S Verba. 2021. Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton: Princeton University Press, New Edition.
• Kotyashko, A, L C Ferreira-Pereira, A V G Vieira. 2018. ‘Normative resistance to responsibility to protect in times of emerging multipolarity: the cases of Brazil and Russia’. Revista Brasileira de Política Internacional 61 (1): e001.
• Kurki, M. 2006. ‘Causes of a divided disciplines: rethinking th concept of cause in International Relations Theory’. Review of International Studies 32: 189-216.
• Kurki, M. 2007. ‘Critical Realism and Causal Analysis in International Relations’. Millennium: Journal of International Studies 35 (2): 361-378.
• Kurki, M. 2008. Causation in International Relations: Reclaiming Causal Analysis. Cambridge: Cambridge University Press.
• Kurki, M and H Suganami. 2012. ‘Towards the politics of causal explanation: a reply to the critics of causal inquiries’. International Theory 4 (3): 400-429.
• Ladyman, J, D Ross, D Spurrett and J Collier. 2007. Every Thing Must Go: Metaphysics Naturalized. Oxford: Oxford University Press.
• Lebow, R N. 2015. Constructing Cause in International Relations. Cambridge: Cambridge University Press.
• Lebow, R N. 2021. ‘Classical realism. In T Dunne, M Kurki and S Smith (eds.), International Relations Theories: Discipline and Diversity. Oxford: Oxford University Press, 5th edition, pp. 33-50.
• Lebow, R N. 2022. The Quest for Knowledge in International Relations: How do we know? Cambridge: Cambridge University Press.
• Lenine, E and R Machado. 2023. ‘Das causas na Ciência Política: uma intervenção filosófica necessária’. Cuadernos de Filosofía Latinoamericana 44(129): 1-34.
• Lenine, E, E Grizenti, A Bia, and B Cardoso. 2023. ‘Process tracing na Ciência Política e nas Relações Internacionais brasileiras: uma análise bibliométrica (2012-2023)’. Carta Internacional 18(3): e1365.
• Lopes, D B, J P Nicolini and T Carvalho. 2022. ‘200 Years of International Relations in Brazil: Issues, Theories, and Methods’. Oxford Research Encyclopedia of International Studies. DOI: 10.1093/acrefore/9780190846626.013.744
• Mantzavinos, C. 2016. Explanatory Pluralism. Cambridge: Cambridge University Press.
• Mayr, E. 2008. Isto é biologia: a ciência do mundo vivo. São Paulo: Companhia das Letras.
• Medeiros, M de A, I Barnabé, R Albuquerque and R Lima, Rafael. 2016. ‘What does the field of international relations look like in South America?’ Revista Brasileira de Política Internacional 59 (1): 1-31.
• Norman, L. 2021. ‘Rethinking causal explanation in interpretive international studies’. European Journal of International Studies 27 (3): 936-959.
• Novelli, D H. 2022. ‘A identidade do campo das Relações Internacionais no Brasil: uma análise a partir da produção científica em seus principais periódicos na década de 2010’. Carta Internacional 17 (1): e1168.
• Parsons, C. 2015. ‘Before eclecticism: competing alternatives in constructivist research’. International Theory 7 (3): 501-538.
• Patomäki, H. 2017. ‘Praxis, politics and the future: a dialectical critical realist account of world-historical causation’. Journal of International Relations and Development 20: 805-825.
• Patomäki, H. 2019. ‘The Promises of Critical Realism in the 2020s and beyond’. Teoria Polityki, 3: 189-200.
• Patomäki, H and C Wight. 2000. ‘After Post-Positivism? The Promises of Critical Realism’”. International Studies Quarterly 44 (2): 213-237.
• Ragin, C E and L M Amoroso. 2019. Constructing Social Research: The Unity and Diversity of Method (3rd edition). Thousand Oaks: SAGE.
• Reis da Silva, A L. 2023. ‘A produção sobre política externa em Programas de Pós-Graduação no Brasil (2000-2019)’. Carta Internacional 18 (1): e1307.
• Rice, C and R Rohwer. 2021. ‘How to Reconcile a Unified Account of Explanation with Explanatory Diversity’. Foundations of Science 26: 1025-1047.
• Salomón, M and L Pinheiro. 2013. ‘Análise de Política Externa e Política Externa Brasileira: trajetória, desafios e possibilidades de um campo de estudos’. Revista Brasileira de Política Internacional 56 (1): 40-59.
• Silva, A C. 2020. ‘Diverse images, reverse strategies: Brazilian foreign ministers’ perceptions and the Brazil-Argentina rapprochement (1974–1985)’. Revista Brasileira de Política Internacional 63 (1): e006.
• Suganami, H. 1996. On the Causes of War. Oxford: Clarendon Press.
• Suganami, H. 2008. ‘Narrative Explanation and International Relations: Back to Basics’. Millennium: Journal of International Studies 37 (2): 327-356.
• Thelen, K and J Mahoney. 2015. ‘Comparative-historical analysis in contemporary political science’. In J Mahoney and K Thelen (eds.), Advances in Comparative-Historical Analysis. Cambridge: Cambridge University Press, pp. 3-36.
• Tickner, A B. 2013. ‘Core, periphery and (neo)imperialist International Relations’. European Journal of International Relations 19 (3): 627-646.
• Wendt A. 1987. ‘The agent-structure problem in international relations theory’. International Organization 41 (3): 335-370.
• Wendt, A. 1998. ‘On constitution and causation in International Relations’. Review of International Studies 24 (5): 101-118.
• Wendt, A. Social Theory of International Politics. Cambridge: Cambridge University Press.
• Wight, C. 2006. Agents, Structures and International Relations: Politics as Ontology. Cambridge: Cambridge University Press.
• Wight, C. 2007. ‘A Manifesto for Scientific Realism in IR: Assuming the Can-Opener Won’t Work!’ Millennium: Journal of International Studies 35 (2): 379-398.
• Woodward, J. 2003. Making Things Happen: A theory of causal explanation. Oxford: Oxford University Press.
• Woodward, J. 2010. ‘Causation in biology: stability, specificity, and the choice of levels of explanation’. Biology and Philosophy 25 (3): 287-318.
• Yanow, D and P Schwartz-Shea. 2015. ‘Wherefore “Interpretive”: An Introduction’. In D Yanow and P Schwartz-Shea (eds.), Interpretation and Method: Empirical Research Methods and the Interpretive Turn. Abingdon: Routledge, pp. xiii-xxxi.
• Ylikoski, P. 2017. ‘Micro, Macro, and Mechanisms’. In H Kincaid (ed.), The Oxford Handbook of Philosophy of Social Science. Oxford: Oxford University Press, pp. 21-45.

Publication Dates

• Publication in this collection
  19 May 2025
• Date of issue
  Jan/Abr 2025

History

• Received
  11 Aug 2022
• Accepted
  17 Mar 2024