Case study research and critical IR: the case for the extended case methodology

Introduction

‘What is a case study and what is it good for?’ ¹ While asked and answered explicitly in positivist and neopositivist literature, this question is yet to receive substantial attention in non-positivist research in International Relations (IR). ² This article addresses this gap from the perspective of critical approaches in IR. It builds on the recent fertile season of methodological reflections in critical IR, strategically concerned with reclaiming discussions on methods and methodology from the exclusive domain of positivist scholars, and substantively focused on making explicit the grounds for producing knowledge and evaluating knowledge claims. ³ In the recent surge of interest towards methods and methodology, ⁴ much attention has been devoted to the scientific status of knowledge produced through non-positivist approaches. This development not only contests the tight link between positivism and methodologically sound research established in the classical literature on methods in IR, ⁵ but it also explores new possibilities for pairing methodologies and methods in ways that expand, rather than constrain, our research possibilities. ⁶ As Jackson argues, science is in fact ‘a pluralist endeavour’ that encompasses, following Weber, every ‘systematic empirical analysis that aims to produce knowledge rather than to produce innerworldly effects’. ⁷

Despite this deeper methodological awareness, critical IR scholarship has yet to articulate its own methodological standards for justifying case selection, the role of case studies in producing knowledge and the terms of comparison between cases. ⁸ While critical scholars have suggested that positivists and post-positivists may be ‘looking for different things’ in a case study, ⁹ critical IR practice has more often unreflectively adopted existing approaches than reflexively articulated its own approach to case study research. This is especially striking if we consider the attention given to other aspects of critical methods in IR, as well as the existence of other case-based approaches (such as ethnographies and genealogy) that could have provided a fertile ground for these discussions. ¹⁰ This article explicitly asks how we can conduct case study research in critical IR in a way that is methodologically explicit and transparent about its (non-positivist) research choices. To this end, it proposes the extended case methodology, inspired by the work of the sociologist Michael Burawoy, as a non-positivist route to case study research in critical IR.

More specifically, our article addresses scholars from various strands of critical IR who, following Jackson’s categorisation, recognise themselves in the ‘reflexivist’ tradition, grounded in the ontological-epistemological belief that we cannot know the world as an entity that is entirely separated from us, and rejecting the idea that we can only know what we can directly experience. ¹¹ Feminist, Marxist, Bourdieusian and post-colonial approaches are some of the traditions identified as a part of critical IR. ¹² Feminists are singled out as those who have made the biggest contribution to the development of reflexivity and positionality as a cornerstone of critical methodologies in IR. ¹³ These critical IR scholars, in their pursuit of holistic understanding and perspectival social knowledge, can benefit from a case study methodology that can help make knowledge claims about the social world in a way that systematically relates the observation of the local context to power relations and transnational power structures. While not being the only possible methodological choice for non-positivist case studies, and without wanting to play down the diversity of critical IR approaches, the extended case methodology contributes to IR methodology debates by responding to the need for non-positivist standards for conceptualising and carrying out case study research.

Before we proceed, a terminological clarification is necessary. Concurring with Hansen, ¹⁴ and Barkin and Sjoberg, we define methods as the techniques of data collection and analysis employed during research, and methodology as the overall research strategy of a project or a ‘bridge that links research questions, epistemology, and methods’. ¹⁵ Still agreeing with Barkin and Sjoberg, ¹⁶ as well as Jackson, methodology is here understood as informed by ontological-epistemological commitments, which shape how we conceive the relationship between the mind and the world, and the type of knowledge claims we make. ¹⁷ Once methods and methodology are redefined in these terms, Burawoy’s reference to the ‘extended case method’ appears as a misnomer from our perspective. ¹⁸ Insofar as it is concerned not with specific techniques for gathering and analysing data (i.e. methods), but with the strategy employed to make knowledge claims about a social phenomenon, we define our approach to case study research for critical IR as the extended case methodology. This terminological shift also signals the extent to which Burawoy’s original conceptualisation has been reworked to respond to the needs and questions of critical IR.

The main argument of this article is that the extended case methodology responds to the need for a case study methodology in critical IR. Characterised by case ‘construction’ rather than selection, and by the distinctive role of theory throughout the research process, the extended case methodology has a flexible approach to theory building, and it is concerned with contextually rich, complex accounts of the social world, while attributing an emancipatory function to the production of knowledge. The first section of the article highlights the problems faced by critical IR when adopting positivist approaches to case study research, and it draws on the contribution of critical IR scholars in outlining the main features of a critical approach to methodology in general, and case studies in particular. The second section argues that the extended case methodology constitutes one such approach. It thus briefly presents the methodology drawing on Burawoy’s work, before suggesting three ‘addenda’ that highlight the extended case methodology’s contribution to critical IR. The article concludes with an illustration of how the extended case methodology can shed light on marginalised aspects of transitional justice processes in post-war Bosnia and Herzegovina, thus substantiating our claim that the extended case methodology provides a viable and rigorous alternative to positivist approaches available to critical IR scholars engaged in case study research.

A missing link? Case study research and critical IR methodology

Case studies are a very popular methodology for social science researchers. Much of their appeal stems from the possibility to generate in-depth insights into a case and avoid the necessarily succinct representation of individual cases given by large-N studies. Predominant understandings of case study research are built on a positivist epistemology characterised by specific understandings of causality and causal mechanisms. Researchers are asked to justify what contribution to knowledge a single case study can actually make, and positivist approaches have developed specific answers to this question. We argue that these are not necessarily applicable to critical IR research, due to fundamental ontological and epistemological differences, which in turn translate into different research objectives and standards for evaluating knowledge claims.

Within the positivist domain, there are two dominant tendencies in case study research, ¹⁹ each reflecting different understandings of causality and logics of inference but sharing ontological commitments to a mind-world separation to which critical ‘reflexivist’ IR does not subscribe. The first, ‘traditional’ tendency inserts case studies, at least implicitly, within a comparative framework. Gerring illustrates this clearly when identifying three types of single case research: in each of them, the inference leading to generalisation, considered the aim of a case study, depends on either spatial or temporal covariation in/within a single unit, or on a combination of the two. ²⁰ Even if understood as merely supportive to multi-case research, ²¹ single cases could serve the purpose of a plausibility probe, of a building block for new theories, or provide descriptively rich accounts that could improve future uses of a case for comparative research. In this tradition, single case studies are thus nested into comparative research through the logic of representativeness driving case selection. An emphasis on the comparative function of case studies is indicative of a conception of causality deriving from Hume and Mill, which responds to the criteria of constant conjunction, the logic of control, directional relationships and disproving alternative explanations. Neopositivist research applies this framework in trying to falsify explanations. ²²

Recent methodological innovations in case study research have instead focused on causal mechanisms and process tracing methodologies. While not necessarily rejecting the first, authors following this second tendency put more emphasis on identifying the steps and relationships that, combined, constitute a specific causal process. In sum, causality responds to a ‘causes-of-effects’ approach, where the aim is to trace the originating causes of an observed outcome through the establishment of the specific processes and linkages producing it. ²³ The quest for causal mechanisms also drives case selection. For example, scholars might select cases because they are ‘least likely’ (and thus can potentially provide strong support if the theory is not falsified in that instance), ‘most likely’ (which can illustrate the mechanism at work, and identify spuriousness in causal relationships that were established statistically), or because they are deviant. ²⁴ Notwithstanding the arguably more complex view of the social world in this latter group of scholars, both these tendencies remain committed to the possibility of knowing something independently of our position vis-a-vis the object under study, combined with a concern for generalisation and explanations that can ‘travel’ from one context to another.

While valuable routes for the production of knowledge, these case study methodologies can be problematic for critical IR – its internal diversity notwithstanding – for three main reasons, pertaining, respectively, to fundamental ontological and epistemological differences, to the trade-off between context and generalisability, and to the difficulties of bridging epistemologies through methodology. First, the positivist quest to unveil causes, effects and mechanisms that are both objectively knowable and independent of the researcher’s own position and preferences clashes with the foundational assumptions of critical approaches, for which facts ‘are social and historical products, partly determined by the collective action of human beings in society’. ²⁵ Hence, the researcher’s own words and deeds contribute to the very emergence of social facts, giving critical theory its double role as ‘both an intellectual and a social act’, seeking to make sense of society ‘by taking a position outside of society while at the same time recognising that it is itself a product of society’. ²⁶ Feminist epistemology, which has significantly contributed to philosophical and methodological awareness in critical IR, ²⁷ forcefully made the point that knowledge is related to our position in the social world, and called for taking into consideration subaltern voices and perspectives. ²⁸ Critical researchers will thus be attentive to context and power structures, and adopt a relational and iterative approach to sense-making and knowledge production, which links theoretical insights to the observation of the social world. ²⁹

Second, in light of this attention to context, critical approaches consider the pursuit of external validity at the expense of contextual awareness either undesirable or altogether impossible. Critical researchers are suspicious of uniformity and not interested in elaborating designs that ‘control’ for specific factors to isolate causes and their impact on studied outcomes. Rather, they look at constitutive relations when showing how the things we know come to be, and put emphasis on producing layered accounts of human agency. ³⁰ While selecting cases by ‘controlling’ for certain factors and thus maximising generalisability is an appropriate move in positivist case study research, this is problematic for critical researchers: not only because it prevents a nuanced analysis of the context, but also because it assumes that the researcher’s point of view remains unchanged throughout the stages of the research process (from design to fieldwork, or from fieldwork in different sites to the analysis/interpretation of findings).

Finally, given these differences concerning not only epistemology and ontology but also the very purpose of theorising, attempts to bridge epistemological differences through methodological agreement are bound to be problematic. This temptation is especially strong with case studies, as their versatility makes them adaptable to different research purposes. Indeed, some scholars have noted that a case study can be equally appreciated by people holding different epistemological stances, although each might be looking for ‘different things’ in it. ³¹ Price and Reus-Smit even argue that the more ‘moderate’ critical interpretivists are able to engage in cross-epistemological dialogue with positivists on key methodological issues. ³² This attempt, however, is fraught with challenges. First, because the attempt to elaborate a critical constructivist way of making contingent truth claims and generalisations masks fundamentally different beliefs about the nature of knowledge and objectivity underpinning the positivist/interpretivist divide. On this point, we agree with critiques of attempts to ‘bridge’ epistemologies. ³³ Second, critical scholars are not necessarily interested in generalisation and recurrence: they would rather seek to destabilise them, and question who or what these generalisations are for. ³⁴ Ultimately, the fact that good case studies can be appreciated across different epistemologies does not eliminate the need for methodological approaches to case study research unmoored from positivist criteria.

In different ways from the extended case methodology proposed here, case-based research has been already put to fruitful non-positivist uses in IR, for instance, through genealogy and ethnography. The former, as the study of the conditions of possibility of an idea or discursive construction, has been used by IR scholars to investigate ideas such as sovereignty or finance. ³⁵ In genealogical research the ‘case’ under consideration is the social phenomenon itself, rather than an instance of it. Case studies, however, are interested in the observation of an instance (a ‘case’) for the purpose of making inferences about a phenomenon. Ethnography, understood as a methodology concerned with long-term engagement and observation of a field, can potentially be carried out through case-based research, and employ some of the existing case study designs as well as the extended case methodology. ³⁶ However, case study research is not necessarily ethnographic, insofar as it can rely on a variety of methods of data collection and analysis beyond participant observation. Moreover, some forms of ethnographic research, such as multi-sited ethnography, are not necessarily case-based in the conventional sense of the term. ³⁷ In sum, the undoubted contribution of ethnography to case study research in IR does not erase the need for a non-positivist case study methodology that is not inherently tied to participant observation.

Thus, in presenting the extended case methodology, this article attempts to make explicit the process of generating knowledge from a single case, contributing to a much-needed discussion on the development of a critical reflexivist approach to case studies. ³⁸ In doing this, the article draws on the contribution of debates that have addressed questions related to methodology in IR, to the scientific status of knowledge produced through non-causal, non-positivist approaches, ³⁹ and to methods recast as performative rather than representational, and as political rather than value neutral. ⁴⁰ For the purpose of this article, we draw on these debates to delineate some features of critical methodologies in IR, which we argue are reflected in the extended case methodology. A non-positivist, critical methodology should – first – acknowledge that the production of knowledge is not neutral, but a political (and emancipatory) endeavour. It will thus rely on methodological strategies and methods that are relational, dialogic and participatory. ⁴¹ Critical research rejects the separation between ‘what’ is studied and ‘who’ is carrying out the study, and places the researcher within the social relations surrounding the object under research, where positivist scholars would isolate themselves from it. The use of methods of data collection and analysis reflects these positions. Second, since critical theory is self-reflective, and knowledge is situated, ⁴² a critical methodology for case studies should pay attention to the evolving nature of power relations and social forces, and the way in which they shape societies while in turn being constituted by them. ⁴³ Methodologically, this entails a focus on complex, rather than linear, relations, and the need to think holistically rather than rigidly separate different levels of analysis. Third, because social structures are never stable nor present themselves uniformly, critical theory approaches usually adopt a flexible approach to theory-building. ⁴⁴ From a methodological perspective, this means that theory does not determine how we see the world, but drives our exploration of it and our research questions. It also means that theorising entails confronting one’s thinking with those living in the social world under scrutiny, without imposing epistemic superiority on them, but at the same time without uncritically reproducing their claims or assumptions. As the following section shows, the extended case methodology possesses these key features of a critical methodology, from its approach to generating knowledge to its use of methods and the role of theory in the research process.

The extended case methodology

Critical IR approaches are thus driven by questions that challenge established assumptions about the social world, and seek answers to such questions in ways that destabilise dominant narratives and foreground subaltern voices. They are also relational: not only should researchers see dialogue as the privileged point of entry into the research, but meaning emerging from such encounters should be related to the power relations and social forces at play. The extended case methodology constitutes a non-positivist avenue to case study research that responds to the needs of critical scholars. The first part of this section outlines Burawoy’s extended case method, which inspires the approach presented here. Because of its origins in a different field, the move towards an extended case methodology comes with three addenda to ensure that our approach responds to the specific challenges of conducting case study research in critical IR.

The extended case methodology at work: researching transitional justice in Bosnia and Herzegovina

What follows is a necessarily brief account of the benefits of using the extended case methodology, with reference to research conducted by one of the authors on the socioeconomic dimension of transitional justice in Bosnia and Herzegovina. Illustrating the methodological process behind a research project is always a difficult endeavour, especially because writing forces us to discuss in a linear fashion what is actually a messy and iterative process. With this difficulty in mind, this section highlights the workings of two key features of the extended case methodology: the process of case construction through extension, and the role of theory in the extended case methodology. We then outline the recovery of marginalised perspectives, the emphasis on contextually rich accounts, and the attention for complex relations as elements of the extended case methodology that make it especially suited to critical IR scholarship.

Before doing this, let us give a brief overview of the research project. This should be taken as a description of the final product aimed at providing context, whereas the paragraphs that follow discuss the methodological process through which the research developed. The research addressed the question of how post-war countries deal with the consequences of socioeconomic violence perpetrated during war, and what role does socioeconomic justice play in transitional justice processes. Tackled with reference to the Bosnian War, the question was situated within debates about whether (and how) transitional justice frameworks can incorporate socioeconomic concerns. The study entailed researching experiences of socioeconomic violence during war, the formulation of socioeconomic justice claims based on these experiences and analysing how international interventions shaped the possibility of putting forward these claims. The bulk of the empirical research consisted of in-depth interviews with local communities, but interviews with international officials, activists and a small amount of participant observation at activist meetings and protests were also conducted. The researcher also kept field notes of contextual observations emerged from interviews and informal interactions.

Positivist approaches to case studies require a specific formulation of the case under consideration at the beginning of the research: this is necessary to justify the selection of a specific case, and in turn to show how its analysis can contribute to broader knowledge on the topic. ⁷⁷ However, in cases where the aim of the research is to disrupt dominant narratives about a certain phenomenon (such as what constitutes ‘transitional justice’) and analyse marginalised voices and communities, the features of the case only become clearer as the research progresses. In this example, the research departs from an interest around the fate of workers’ communities through the Bosnian War, whose stories are largely untold by the literature, and the question of whether this dimension of the conflict bears any relation to transitional justice processes. As in Burawoy, the first step of the research consists in the researcher’s intervention onto the field, which they approach through preliminary theoretical reflections. ⁷⁸ In this instance, the idea driving the beginning of fieldwork, coming from the work of critical scholar Nancy Fraser, ⁷⁹ was that what happened to these workers’ communities could be characterised as ‘socioeconomic injustice’. In Fraser’s work, socioeconomic injustice is defined as rooted in the political economy, and including, for example, ‘exploitation (having the fruits of one’s labor appropriated for the benefit of others); economic marginalization (being confined to undesirable or poorly paid work or being denied access to income-generating labor altogether), and deprivation (being denied an adequate material standard of living)’. ⁸⁰ Following a preparatory phase, fieldwork was conducted through interviews and observation in two post-industrial Bosnian communities over a combined period of 9 months.

The construction of the case then occurs through the process of extension, as interviews and observation reveal patterns of wartime socioeconomic violence, and show that different forms of injustice (including interethnic violence) overlap and intersect in people’s lived experiences. In turn, these patterns lead to other questions, then explored through interviews, about how local understandings of justice in Bosnia were shaped by these experiences. It is through this incremental process, going back and forth between new discoveries and theoretical reflections, that the research evolves: local conceptions of justice now appear also structured by power relations, embodied for instance in the institutional features of post-war Bosnia and the political-economic reforms of the post-war period that perpetuate and deepen, rather than alleviate, socioeconomic injustice. ⁸¹ Hence, following Burawoy, in this second stage social situations become social processes through the reproduction of regimes of power. ⁸² In the following extension, from social process to the external field of forces, the extended case methodology allows us to see how these reforms are sustained by transnational processes of neoliberal restructuring, manifesting themselves also through post-war international interventions combining the goal of establishing democracy, peace and justice with marketisation and economic liberalisation. In sum, while case selection in positivist research is an initial step, in the extended case methodology case construction is a part of the whole research process, and occurs through extension. Such an approach is much closer to ‘abductive’ reasoning, which is iterative and incremental, ⁸³ than to the hypothetico-deductive or inductive models, and their various combinations, that usually feature in positivist research.

The flexible approach to theory is another distinctive feature of the extended case methodology as an alternative to conventional case study research. As mentioned above, Nancy Fraser’s work constitutes the theoretical point of departure for this project. As a political theorist building on the tradition of critical theory, Fraser has addressed questions related to democracy, justice, feminism and other pressing political issues of the post-Cold War era. ⁸⁴ Most relevantly for this research, she developed a multidimensional conception of justice (first including redistribution and recognition, then adding the third dimension of representation), ⁸⁵ and argued that apparent tensions between claims based on valuing difference (recognition) and equality (redistribution) are resolved when we take a transformative approach to redressing different types of injustice. ⁸⁶ Moreover, an important contribution that her work makes to the field of transitional justice is that justice itself, in its different dimensions, is defined on the basis of the originating injustice that it should remedy. While more conventional approaches focus on specific crimes or violations of International Humanitarian Law, here local experiences (and interpretations) of violence are placed at the centre of the analysis.

In the extended case methodology, the theory is then modified through its encounter with anomalies in the field: puzzling issues that lead us to rethink and further develop our framework. Here it is useful to remember that theory does not fix expectations nor does it translate into hypotheses: rather, the theory is a prism through which we approach and interpret reality, which guides, but does not determine, our research. ⁸⁷ In the context of doing fieldwork among local communities in Bosnia, anomalies emerge when actors and forces external to Fraser’s analysis come to define the process of dealing with wartime injustice. Relating her work to a post-war country thus requires adaptation. On one hand, injustice committed during war is characterised by a specific temporal element: the conflict in Bosnia ‘froze’ the effects of socioeconomic violence because the local population was prevented, between 1992 and 1995, from mobilising to address such injustice. On the other hand, the post-war international intervention invested time and resources onto transitional justice programmes that dealt with violations of International Humanitarian Law, neglecting the socioeconomic dimension of wartime violence. Moreover, it effectively removed socialism, and the redistributive policies it inspired, from public debate. ⁸⁸ Discussions of socioeconomic issues were only to be driven by the goal of facilitating Bosnia’s transition to free-market capitalism. The bracketing of socioeconomic issues and the long temporal trajectory of wartime violence become important elements in understanding the Bosnian case, as they all affected how people recount and make sense of wartime experiences, and how they develop justice claims.

The continuous dialogue between researcher and participants, and between fieldwork material and theoretical reflection leads to theory reconstruction as a part of both case construction and knowledge generation. This occurs, in this example, through the theorisation of Bosnia’s post-war and post-socialist conditions: respectively, the role of the war in complicating both the analysis of injustice suffered and the process of redressing it, and the ambivalent role played by the socialist past (a point of reference for people’s socioeconomic justice claims, and antithesis of the political and economic model envisaged by the international community). The theorisation of the second condition builds on Fraser’s conception of the ‘post-socialist age’ as an era marked by the absence of an alternative ‘emancipatory project’ in the wake of the fall of socialism, ⁸⁹ but revises it and contextualises it through the input of scholars of ‘actually existing’ post-socialism in Bosnia, and of dealing with the past. ⁹⁰ Through the research process, the initial theoretical framework is adapted to post-war countries, and becomes more responsive to the different ways in which justice claims develop – not only as a way of redressing experiences of injustice, but also in response to the conditions set by local and international politics. The difference between theory reconstruction and the role of theory in positivist research also lies in its reaction to anomalies: in the extended case method the theory is enriched through its contextualisation, whereas in positivist research uncovering specific conditions of the case under observation limits the scope and explanatory power of the theory.

To conclude, three additional features make the extended case method more suited to critical scholars interested in case study research. First, insofar as it is concerned with disrupting established assumptions about what post-war justice should be about, and as it puts in the foreground a neglected section of Bosnia’s society, this research retrieves marginalised communities, their stories and justice claims that have been historically left aside because they do not fit with conventional narratives of the Bosnian War and the ensuing transition. Second, as briefly mentioned above, the extended case method privileges contextuality over the generalisability of findings to a broader set of cases. The case of Bosnia is thus interesting not because it is representative or deviant, ⁹¹ but because it helps us understand socioeconomic justice in transitional societies in a contextually rich way, and more specifically through the complex relations between war experiences on the ground, political economy, international intervention and transnational processes of liberal democracy-building and neoliberalisation. Finally, the use of the extended case method predisposes the researcher towards attention to complex, rather than linear, relations. As the discussion of the extended case method and the example show, the thread linking local experiences and narratives, power relations and external interventions does not go in one direction. Local conceptions of justice, for instance, derive from wartime experiences of violence, but these are interpreted through the lens of post-war and post-socialist transformations. At the same time, justice claims shape the international intervention in some ways. It is also by labelling social justice claims as the legacy of socialist nostalgia and practicing its ‘reform agendas’ that the international community constitutes itself as a (neo)liberal actor. Once again, the extended case method differs from positivist approaches, as the focus here is placed on constitutive (rather than causal) relations. ⁹²

While not exhaustive, this section has provided an overview of how using the extended case method in practice differs from positivist approaches to case studies. This approach provides a methodological avenue to research the complex question of how societies deal with the consequences of wartime violence. It does so by focusing on the value of context-specific, in-depth research that challenges dominant narratives of post-war transitions and justice processes, allowing for a flexible approach to theory-building.

Conclusion

Discussions of methodology have traditionally contributed to setting the criteria for doing ‘social science’ that accept or reject certain modes of writing, and that put certain approaches in a privileged position over others. ⁹³ Recent IR debates on critical methodologies and methods have largely reflected a different spirit, rejecting the idea that there is a single route to knowledge. Save a common commitment to ‘systematic empirical analysis that aims to produce knowledge’, ⁹⁴ the standards by which knowledge claims about the social world are assessed and evaluated also depends on the researcher’s specific ontological and epistemological commitments. This much has undoubtedly emerged from these debates, aiming at breaking the methodological dominance of positivist approaches, and its attendant exclusionary effects. As a contribution to this scholarship, this article has presented the extended case methodology as a non-positivist route to systematically conducting empirical analysis, thus providing a viable alternative to conventional approaches to case study research. While rejecting the concern with representativeness and generalisation of positivist case study research, the extended case methodology represents an explicit attempt at detailing the process of generating knowledge, centred on a dynamic interaction with theory and working around extensions of ‘situated knowledge’ on the micro-level in connection to macro fields of forces. Insofar as it takes context and positionality seriously in its approach to theory-building, it sees and studies the world in complex rather than linear terms and it seeks to generate explicitly political and emancipatory knowledge, the extended case methodology is entirely consistent with the ontological and epistemological commitments of most critical IR.

Ultimately, the extended case methodology relies on a substantially different understanding of the ‘case’ compared to positivist research. In the latter, a case study is either an instance or a counter-instance of an existing theory, or conducive to the development of one through the in-depth study of causal processes. Within the extended case methodology, the case is a construction that emerges during the research process, through the constant relation of observation and theory, and international and global forces and processes by which it is shaped and on which it reacts back. It is this very relation that allows us to stretch the ‘situated knowledge’ ⁹⁵ gained from the case through the extensions distinctive of the extended case methodology. This article has provided an overview of this alternative methodological approach, and a brief example of how it could work through the study of socioeconomic justice in post-war Bosnia. This is, however, only meant to be a starting point for further discussion and practical application of the extended case methodology. Far from being prescriptive, this discussion of the extended case method is an invitation to further engage with case study methodologies from a critical perspective. This engagement might start from an exploration of how the extended case methodology might be adapted for case study research that is not fieldwork-intensive, but is based more on archival or document analysis or elite interviews, for instance.

In addition, as it opens an explicit discussion about the use of case studies in non-positivist IR, this article also points towards important questions for comparative research. In the extended case methodology, the final step of theory reconstruction not only affects our understanding of the case at hand, but as it transforms theory it also alters the conditions of existence of other instances of the phenomenon under study, thus transforming the terms on which case comparison is carried out. Burawoy himself has tried to address this challenge in some of his more recent work. ⁹⁶ Similarly, Jackson has argued for the need to develop a reflexivist approach to case comparison in IR, which has long been the domain of neopositivists. ⁹⁷ Thus, while contributing to the critical methods literature by introducing the extended case methodology as a non-positivist route to conducting case study research, this article also suggests that this methodology might facilitate reflection on the role and usefulness of comparative research in critical IR.