International legal scholarship and the making of a ‘scientific self’

1. Introduction

Pluralising International Legal Scholarship starts off with a diagnosis. A ‘wave of change’ is going through international legal scholarship, the introduction claims, with non-doctrinal research methods gaining ground and causing a ‘paradigm shift’ within the field.¹ In the same breath it also identifies a gap in the literature: nothing much is said about the use of these non-doctrinal methods as such, nor is there a structured attempt to compare and contrast them with doctrinal approaches to international law. So the aim of the volume, as Rossana Deplano explains in the introduction, is to put forward ‘concrete applications of non-doctrinal research methods … with a view to assessing their interplay with the doctrinal-only type of inquiry’.² In other words, the brief to the contributors was to write about a particular method in relation to a concrete example: to show by doing what it can and cannot do compared to doctrinal approaches to international law.

Assessing the chapters in concert, Deplano identifies three themes running through the book: first, ‘a concern to generate more scientific results’, by which the pertinent contributors mean the use of a particular set of methodologies that purport to increase transparency and provide what may be called better data on international law.³ The second theme has to do with the scope of inquiry, which in almost all the chapters extends beyond formal legal texts. Third, and finally, Deplano argues that all chapters demonstrate ‘the inherently partial and incomplete nature of non-doctrinal studies, which makes them necessarily complementary to, rather than substitutive of, doctrinal analysis’.⁴

By claiming the complementarity of doctrinal and non-doctrinal methods, Deplano explicitly identifies those who need to be persuaded of the added value of the latter as the target audience of the book.⁵ This is repeated in the conclusion, where one of her main arguments is that the volume shows that the gap between non-doctrinal and doctrinal work is smaller than is usually presumed, as the latter feeds into the kind of non-doctrinal questions that are being asked.⁶ Specifically, the editors measure the added value of these methods in terms of the extent to which they succeed at producing a better ‘science’ of international law.⁷ To explain this benchmark, they devote a section in the introduction to ‘the relationship between theory and method’, a relationship they define in terms of ‘identifying … variables’ and ‘[t]esting the hypothesis’.⁸ Theory, in their understanding, ‘“is simply a reasonable and precise answer to the research question” aimed at enabling the researcher to extract an observable implication’, with researchers engaged in ‘framing a research design which is as objective as possible’.⁹

Yet, the editors’ invitation to move beyond doctrinal research creates a much wider space of contestation than their framing of the volume in the introduction and conclusion suggests; one in which highly contrasting ideas are put forward of what these non-doctrinal methods are and should be after. For example, whereas in Chapter 6 Giulia Gentile and Luigi Lonardo (the co-editors of the volume) claim that ‘the promise of quantitative empirical research is that it helps getting closer to “truth”’,¹⁰ in Chapter 3 Josef Ostřanský declares (legal) anthropology to be ‘little concerned with the notions of replication, causality, validity and representativeness … our epistemological position is that there is never a complete impartial image in social sciences’.¹¹ And whereas in some chapters quantification equals greater precision equals greater objectivity,¹² Huaxia Lai claims in Chapter 8 that ‘pattern identification is usually achieved at the cost of forsaking the fine-grained understanding of the data situated in its particular context’.¹³ The volume is marked by these strongly opposing views on how best to understand international law. Chapters 1 and 6 both strongly draw on quantitative analysis as a ‘more scientific’¹⁴ method, in contrast to Chapters 2 and 3, which are concerned not so much with improving the scientific character of international law as they are with what they call the ‘lived reality’¹⁵ of international legal practice. Chapters 4 and 5 operate somewhere in-between; Chapters 7 and 8 present (fundamental) critiques of Chapters 1 and 6.

So, what struck me most about the volume was not so much the positioning of the individual chapters vis-à-vis doctrinal work and what they contribute to that particular debate, but rather the dissonance between what the contributors argue is worthwhile scholarship versus the way the book is situated. By singling out doctrinal readers as their target audience and, what is more, by using positivism to persuade it of the added value of non-doctrinal work, the editors impose a particular ‘scientific ideal’¹⁶ on the volume, in so doing failing to fully take note of how deep the disagreement between the contributors runs. What I mean to say is that the volume ‘pluralizes’ much more fundamentally than its editors allow for or acknowledge.¹⁷ Though they claim that it merely aims to show ‘the role different non-doctrinal methods serve in developing international legal theory’,¹⁸ the book holds profoundly conflicting views on why these non-doctrinal methods are worthwhile efforts to begin with – what, exactly, constitutes valuable scholarship. Several of the contributors could well be envisioned as rejecting the claim put forward in the introduction that ‘an observable implication’ can be ‘extract[ed]’, or that quantitative analysis is ‘[m]ore objective in nature’ than anthropological research, with one of them explicitly denying this kind of objectivity to be either possible or the aim of his research.¹⁹ To some the choice between different methods comes down to a choice between more or less objectivity – and therefore also in the editors’ views better or worse science – but to others, this is simply not their main concern. In that sense the contributors themselves put paid to Deplano’s claim in the introduction that ‘[n]on-doctrinal methods … are … research methods not associated with any particular scholarly position’.²⁰ Though she refers here to different legal theoretical views, when read as a collection of contrasting ‘scientific ideals’ the volume happens to show precisely this. The differences between the chapters pertain to what the writers think non-doctrinal international legal scholarship is for:²¹ ‘what is worth looking at’, ‘how it should be looked at’ and what is the appropriate role of the scholar when doing so.²² Finally, then, the volume captures a profound and irreconcilable disagreement about how international law may best be known, and what kind of demands that way of knowing imposes on the international legal scholar.

On the following pages I first briefly summarize each chapter, with a particular focus on how the contributors write about the aims of their respective methodology and, if applicable, the part played by the researcher. In so doing my aim is to provide not just an overview of the volume, but also to highlight the effect of a sequential reading of the chapters. As ‘users’, we do not tend to read an edited volume sequentially, rather selecting those chapters we think suitable to our purpose or that pique our interest. Reviewing the volume, however, I read the chapters in the order in which they are placed in the book, and doing so reveals something which would otherwise be missed. The sequence in which the chapters are placed, and the very different ways in which the writers describe their work, means the disagreement between them slowly but steadily builds, as each chapter can be read as a response to the previous one. As a result, the methodological – or, as I shall explain in greater detail below, normative – struggle referred to above is gradually exposed. The last of the substantive chapters in particular, by Huaxia Lai, raises the stakes by explicitly questioning the supposed supremacy of empirical (quantitative) methods over doctrinal ones. Whereas the other chapters purportedly argue merely difference, Lai explicitly refers to an ‘empirical agenda’ and claims that a ‘careful and close reading of [international law’s] texts’ is needed ‘for understanding how international law works’.²³ In so doing I understood her to be advocating not just doctrinal work, as she claims she is doing, but also to be sympathizing with the qualitative approaches put forward in, e.g., Chapters 2 and 3 of the volume. As such, it puts the question of the hierarchy between the different methods it proposes front and centre.

What follows this overview is an attempt to account for the effect of that sequential reading; to explain the dissonance between the contributions and the way the editors gauge them. I do so by placing the volume alongside Lorraine Daston and Peter Galison’s 2007 Objectivity, a history of what they refer to as the ‘scientific ideal’ of objectivity, uncovered by means of images in scientific atlases.²⁴ Daston and Galison do not just trace the origins of objectivity as a scientific ideal and describe what came before and what followed it, but they also explain what kind of scientific subject follows from the particular ideals they write about. Reading this edited volume against their book allows for an argument in three steps: (i) that Pluralising is an assembly of scientific ideals rather than a description of methods, (ii) that a scientific ideal presupposes an ideal ‘scientific self’,²⁵ which in this particular understanding of objectivity is an absent scientific subject, and (iii) that the ideal and its subject materialize and manifest themselves in specific practices. To first show the contesting views between the contributors I closely consider the use of the word ‘objectivity’ as it appears in the chapters of those who claim it – Chapters 1 and 6 – to then juxtapose it with the scientific ideal put forward in Chapter 3.²⁶ This focus on the specific language use of the contributors means I understand that use as an academic practice, evidencing a scientific ideal as well a set of epistemological demands placed on the legal scholar.²⁷ In other words, if a scientific ideal and the accompanying ideal international legal scholar are understood as manifesting themselves in language, they are made and understood in a particular way in these texts – and my aim here is to show how and where the differences between these scientific ideals happen exactly. What I mean to counter in doing so is the idea that the volume merely presents an overview of non-doctrinal methods; rather, underlying it is a normative struggle that belies the positivist benchmark the editors put forward in the introduction and conclusion. The article subsequently elaborates on this last point, and interrogates the way in which the editors close the volume in their conclusion. This, finally, brings me to suggest that the volume sets irreconcilable demands on the scholar attempting to know international law better than we do now.

2. The chapters

Following Rossana Deplano’s introduction, Chapter 1 attempts to uncover how the PCIJ and ICJ assessed/assesses the existence and substance of general principles as a source of international law.²⁸ At the beginning of her chapter, Marija Đorđeska puts forward that we do not really know what general principles are, or how their precise content is ascertained.²⁹ In order to answer this question she turns to the ICJ and its predecessor, mining all of their advisory opinions and rulings to see how their references to general principles may be understood, and accordingly classifying and categorizing their use.³⁰ As a result, she says, it becomes possible to ‘provide a more accurate and objective image of what general principles are’, in so doing ‘present[ing] the methodology that led to an objective definition of general principles.’³¹ ‘Th[is] more objective (or “scientific”) approach to the research of general principles’, she concludes, ‘inspired more “scientific” results.’³²

The second chapter, by Sarina Landefeld, is radically different in tone. She starts by emphasizing the importance of acknowledging one’s theoretical framework as a researcher, claiming that doing so is less common in doctrinal approaches to international law than it is in International Relations.³³ In terms of the method it puts forward, the chapter proposes discourse analysis as a non-doctrinal qualitative method.³⁴ It briefly deals with the turn to history in international law, concluding that ‘constructivists … recognize that history does not have an external objective existence to be discovered, but is created by its writers’.³⁵ It is the least specific of all the chapters in the volume, in the sense that the method it deals with is not illustrated by means of a concrete question.³⁶ Rather, it positions constructivism at a theoretical level in relation to doctrinal methods, arguing that ‘[a constructivist] understanding of international law … evades generalizations and predictions … It is less interested in patterns of lawmaking and more in the idiosyncrasies … of particular legal norms or concepts’.³⁷ The chapter is a counterweight to the previous one in the way it speaks about the character of its findings: whereas the first chapter aims to come to ‘more scientific results’, the second starts from the presumption that the international legal world is constructed, and wonders how it came to look the way it does.³⁸ ‘[T]he constructivist approach’, Landefeld writes, ‘promises a more complex and nuanced view of the dynamics in international law’.³⁹

Something similar happens in Chapter 3. Drawing on an ongoing research project, Josef Ostřanský argues how the use of ethnographic research methods can improve understanding of how international investment law affects modes of governance in specific states.⁴⁰ The chapter aims to show what an anthropology of international law can do: most importantly, Ostřanský writes, it ‘may get closer to the lived reality of the actors influenced by [international investment law]. The core idea of ethnography is to be sensitive to everything around you’.⁴¹ This sensitivity means the researcher considers her- or himself very much part of the research; not just in terms of designing it, but also in terms of bringing about its results. ‘[S]ocial interaction is at the core of ethnographic research’, Ostřanský says; ‘[t]he typical positivist dividing line between the researcher … and the field … does not find its place in ethnography … the relationship between the researcher and the interviewed … becomes part of the data, not something external to it.’⁴² This also means that objectivity is not the aim here, for what results from this kind of research is, in Ostřanský‘s words, ‘a necessarily partial picture’.⁴³ Contrary to the method put forward in Chapter 1, however, this partiality – both in terms of the results as well as that of the researcher – ‘is not anything to worry about, as our epistemological position is that there is never a complete impartial image in social sciences’.⁴⁴

Considering the way they talk about their own methodologies, Chapters 4 and 5 occupy a middle position in-between the far ends of Chapters 1 and 3. In Chapter 4, Alice Margaria uses textual analysis of judgments as well as interviews with participants to uncover the European Court of Human Rights’ approach to European diversity in family life, as well as the experience of those involved in a particular case.⁴⁵ The chapter builds on Ostřanský’s one in the sense that Margaria bases her work in ethnographic approaches, yet she claims a slightly stronger connection to doctrinal work here: not only because she studies the texts of judgments but also because she explicitly argues ‘how the combination of doctrinal and non-doctrinal methods … can produce a (more) realistic and in-depth understanding of the approach adopted by the European Court of Human Rights … in a specific field’.⁴⁶ Such an approach, the conclusion to the chapter argues, ‘makes it possible to address questions … that could not be produced using one method’; specifically, it ‘leads to a more “grounded” vision of the law’.⁴⁷ Though from the perspective of quantitative approaches this might be problematic as it comes ‘at the expense of representativeness’, qualitative work is concerned rather with the added value of what Margaria refers to as ‘in-depth research’⁴⁸ by its bottom-up approach.⁴⁹ Similarly, in Chapter 5 Elizabeth Faulkner argues that archival research into the prohibition on child trafficking in international law, ‘illuminate[s] the practice of states at the grassroots level’ in a way doctrinal methods could never do.⁵⁰ The latter ‘would have demonstrated the significant shift in the language’ used in the pertinent treaties, ‘but would fail to demonstrate the impact that this shift in language had in practice’.⁵¹ The relation between doctrinal and non-doctrinal work here is rather more oppositional than in Margaria’s chapter: Faulkner explicitly argues that doctrinal work presents an image of the evolution of international law on trafficking which is contradicted by archival, non-doctrinal analyses.⁵² Nonetheless echoing the language used in Chapter 4, the method Faulkner outlines in hers ‘affords a richer analysis of the development of international law’.⁵³ In other words, in applying a different method we get closer to the practice of international law, resulting in a ‘richer understanding’ of it.

Chapter 6 echoes Chapter 1 in its advocacy for quantitative empirical methods. Written by the two co-editors of the volume its opening lines state that ‘[a] method is a way to achieve a result, and it is neutral to that result. If the method is sound, it should not influence the outcome’. The next couple of phrases nuance this claim, including with regard to the discipline of international law, ‘where results are never objective but always relative to the observation method’.⁵⁴ But ‘a suspicion of subjectivity’, they add, should not lead to a rejection of ‘the value of methodologies’⁵⁵ as such. Focusing on appeals at the European Court of Justice against decisions made by the General Court, in this chapter Giulia Gentile and Luigi Lonardo ‘aim … to offer an objective, quantitative basis for measuring … how the judgments correlate to (i) the formation of the court … and (ii) the length of appeal proceedings’.⁵⁶ Dealing specifically with judicial review of EU sanctions, the chapter concludes, inter alia, that the composition of the court affects both the outcome of an appeal as well as the length of the procedure.⁵⁷

These and similar kinds of conclusions, however, are heavily contextualized in Chapters 7 (by Gabriel Lentner) and 8 (by Huaxia Lai) of the volume. Chapter 6 itself goes some way towards doing this;⁵⁸ but the two chapters following it do so much more forcefully. First, the chapter by Gabriel Lentner constitutes an internal critique of the Empirical Legal Studies movement, putting forth as a reminder the rules of inference and adding a critical note on the current review process of international law journals, which does not check empirical work to the extent that it should.⁵⁹ The chapter explicitly deals with quantitative empirical research, and as such may be read as an engagement with Chapters 1 and 6 of this volume in particular. Rather poignantly, Lentner adds in his conclusion that ‘quantitative studies … certainly are not necessarily more “scientific” than other approaches that interrogate the law and its institutions outside of specific outcomes and the legal practice itself’.⁶⁰ Titled ‘The perils of quantitative research in international law’, his chapter is a firm critique of the current use of quantitative methods in legal scholarship: it calls for better empirical work and for a more appropriate (and possibly limited) use of these methods. Chapter 8, by Huaxia Lai, picks up on this last point and addresses it in an even more critical way, targeting what she perceives to be the politics of the methodological struggle. At several points Lai explicitly refers to an ‘empirical agenda’⁶¹ calling for ‘[b]igger data’ which, she argues, ‘is usually achieved at the cost of forsaking the fine-grained understanding of the data situated in its particular context’.⁶² Agreeing with Lentner that there is a particular time and place for quantitative work, she however moves beyond his critique to question the promises of this approach as such. Much is lost in the move from ‘subject matter expertise’ to ‘computational algorithms’,⁶³ she argues: ‘Automated content analysis is no substitute for careful and close reading of the texts, the nuanced differences of which are more relevant for understanding how international law works’.⁶⁴

By means of these last few words Huaxia Lai raises the stakes of the volume by placing the question of how best to understand international law front and centre. It is perhaps done somewhat involuntarily, as the last part of her chapter does contain an attempt to reconcile different methodologies: there, she specifically states that it ‘should not be taken as waging a campaign against empirical scholarship of international law’ and that ‘[c]lassical doctrinal analysis can be refined by borrowing from the empirical research tradition a more rigorous methodological standard’.⁶⁵ As it is, however, her plea for a ‘careful and close reading of texts’ as well as the repeated references to an ‘empirical agenda’ fully expose a second scission running through the book: next to the question of the relation between non-doctrinal and doctrinal methods, a second one is how these non-doctrinal methods are thought to relate to each other.

The next section zooms in on this latter question. I first elaborate on Daston and Galison’s book Objectivity which, as I explained in the introduction, lays the groundwork for three arguments in relation to this volume: (i) that Pluralising is an assembly of scientific ideals rather than a description of methods, (ii) that a scientific ideal presupposes an ideal ‘scientific self’,⁶⁶ which in this particular understanding of objectivity is an absent scientific subject, and (iii) that the ideal and its subject materialize and manifest themselves in specific practices. Specifically, I take a similar approach to Daston and Galison’s in the sense that my concern is how and in what sense objectivity is ‘made’ and what specific form it takes on in the texts of the contributors who claim it.⁶⁷ This means that I consider objectivity, to borrow Kenneth Gergen’s words, to be a ‘textual achievement’,⁶⁸ which entails a view of scholarship as academic writing constructing knowledge in a particular discipline.⁶⁹ Following Daston and Galison’s claim that ‘[s]cientific objectivity resolves into the gestures, techniques, habits, and temperament ingrained by training and daily repetition’ – in short, manifests itself in practices – I similarly adopt a ‘bottom up’ approach here by looking at the textual practices of the contributors to Pluralising.⁷⁰ I do so by closely considering the chapters of those who claim objectivity, that is the writers of Chapters 1 and 6, to then contrast it with the scientific ideal manifested in the language used by the writer of Chapter 3 of the volume. Whereas the first two explicitly claim or seek as much objectivity as possible, the latter contradicts this idea most radically. These practices reveal contradictory scientific ideals and, therefore, contradictory ideals of what makes a good scholar. In turn, this allows for a reflection in Section 4 on how these contradictions are subsequently handled by the editors.

3. The making of objectivity

‘[O]bjectivity has a history’, Daston and Galison proclaim on the opening pages of their book; it ‘has not always defined science’.⁷¹ The way they argue their point is by means of scientific practices, specifically, scientific atlases: collections of images of birds, bodies and flowers intended to instruct the user in a particular discipline. They argue that these images do not just represent something, but also tell us what they are thought to represent: the ‘average’ human bone disease, or the ‘particular’ shape of a milk drop.⁷² The history of science, Daston and Galison show, is marked by changes in these preferred ways of seeing, different ways of depicting reality – and disciplinary notions about what is possible and desirable with and through these images, change over time.⁷³ Different ideas of what constitutes good science both preceded, coincided with and followed the emergence of scientific objectivity in the mid-nineteenth century,⁷⁴ and as such ‘the history of objectivity is only a subset, albeit an extremely important one, of the much longer and larger history of epistemology’.⁷⁵ Therefore, to say that to do science means to be (as) objective (as possible) to the exclusion of other epistemological positions is ‘imprecise’, they claim, ‘both historically and conceptually’.⁷⁶ For example, eighteenth century scientific image-making required the scientist to get as close as possible to the ‘essence’ of a flower (an approach Daston and Galison refer to as ‘truth to nature’).⁷⁷ The scientist was constantly and deliberately interfering in what the viewer got to see. What is more, this interference was regarded as necessary to approach reality: rather than attempt to let it speak for itself as much as possible, nature was perceived to be able to do no such thing. Viewers needed help in understanding what a flower ‘truly’ or ‘really’ was, in essence.⁷⁸ With the emergence of objectivity in the mid-nineteenth century the scientist was removed from the scene, and was required to simply serve as a channel through which objects could be registered.⁷⁹ Next to being historically imprecise, the equation of science with objectivity is also conceptually so, Daston and Galison argue, as objectivity does not always mean the same thing. It translates differently in different situations: ‘The criterion may be emotional detachment in one case … recourse to quantification in still another; belief in a bedrock reality independent of human observers in yet another.’ But at its core, they say,

[t]o be objective is to aspire to knowledge that bears no trace of the knower … Objectivity is blind sight, seeing without inference, interpretation, or intelligence … objectivity is the suppression of some aspect of the self, the countering of subjectivity … If objectivity was summoned into existence to negate subjectivity, then the emergence of objectivity must tally with the emergence of a certain kind of wilful self, one perceived as endangering scientific knowledge.⁸⁰

In other words, objectivity is defined in relation to what it is not – subjectivity.⁸¹ The ideal of objectivity as understood here, Daston and Galison show, demands an absent scientific subject: it requires of the scholar as little interference as possible with the knowledge s/he is ‘discovering’.⁸² The pursuit of objectivity presumes a scientist who is taking herself out of the equation; eliminating to the extent that she can, in Daston and Galison’s words, ‘any trace of the knower’. What their account of objectivity therefore also shows is that scientific ideals are inextricably linked to the construction of a ‘scientific self’:⁸³ an idea of what constitutes good science also sets a specific set of demands on the scientist, and in case of this specific understanding of objectivity, it roughly means that the researching subject should be there as little as possible.⁸⁴ Notions of what makes for a proper way of seeing force the one doing the seeing into a particular posture: these practices ‘represent the knower’.⁸⁵

Coming back to the volume under review here, if objectivity is understood as seeking to suppress the subject, any method that envisions a more active role for the researcher is in direct contradiction to this scientific ideal.⁸⁶ This is precisely what happens in Pluralising. Approaches that do not aim for objective knowledge but instead, as some do in the volume, attempt to approach the ‘lived reality’ of the actors under study fundamentally undermine those assumptions of what constitutes good science. If objectivity and subjectivity are understood as the far ends of a spectrum the researcher traverses in her work, methods that put the scholar front and centre in the construction of knowledge operate in a different universe altogether. From the viewpoint of the demand of this specific understanding of objectivity this is, quite simply, bad science; more poignantly, it turns the scholar in question into a bad scientist. This is exactly why Daston and Galison refer to scientific ideals as a matter of ethics rather than competence.⁸⁷ ‘A [scientific] self must be practiced’, they write, and ‘[m]uch of epistemology seems to be parasitic upon religious impulses to discipline and sacrifice’.⁸⁸ Scientific ideals are first and foremost ‘epistemic virtues’, which

… earn their right to be called [so] by molding the self … As long as knowledge posits a knower, and the knower is seen as a potential help or hindrance to the acquisition of knowledge, the self of the knower will be at epistemological issue. The self, in turn, can be modified only with ethical warrant.⁸⁹

Given the ideal of objectivity, what this allows for is for the charge of subjectivity to become an ethical charge, implicating the scholar in question. ‘The surest sign that the values of objectivity deserve to be called such’, Daston and Galison say, ‘is that violations ignite indignation among those who profess them’.⁹⁰

A closer look at the chapters of the volume that claim objectivity shows how this works exactly. The word ‘objective’ (or objectively, objectivity) in the academic sense of the word (as opposed to its meaning as purpose or goal) is used in Chapters 1, 2, 5, and 6.⁹¹ Only Chapters 1 and 6 use it in the affirmative; that is, Chapters 2 and 5 argue against it being possible at all. For example, in Chapter 2 Sandrina Landefeld states that ‘constructivists themselves recognize that history does not have an external objective existence to be discovered, but is created by its writers’; in Chapter 5 Elizabeth Faulkner writes that ‘[t]here are numerous issues with archival formation and acquisition, problematizing their use and objectivity’.⁹² Here, I first turn to the objectivity claimed in Chapters 1 and 6 before turning to Chapter 3, which proposes a radically different scientific ideal.

Beginning with Chapter 1, recall how Marija Đorđeska attempts to come to a more objective definition of general principles, stating decades of scholarly disagreement on the question and instead looking at its use by the PCIJ and the ICJ to determine its meaning. As a result of her approach, she claims it becomes possible to ‘provide a more accurate and objective image of what general principles are’, ⁹³ a claim which is also presented in the title of her chapter: ‘General principles of law recognized by civilized nations: method, inductive-empirical analysis and (more) “scientific” results.’ Đorđeska’s pursuit of objectivity is a response to a perceived lack of consensus among scholars about what general principles are, how they should be defined and what is the proper methodology for ascertaining them.⁹⁴ The way she aims to achieve it is described in Section 1.2 titled ‘Scientific’ approach, which lists six steps, starting with a search for the word ‘principle’ in the Court’s jurisprudence (‘Step 1’) and ending with an account of some of the more difficult methodological choices that were made as part of the research (‘Step 6: Addressing the Challenges’).⁹⁵ What results is a total of 156 general principles identified in the PCIJ and ICJ jurisprudence, a definition of general principles as used by the two Courts, a classification of the different uses of the word as well as their legal basis (domestic, international and judicial) and a division into subtopics.⁹⁶

Returning to the work of Daston and Galison outlined above, the pursuit of objectivity as a remedy against scholarly dissent and the attendant construction of a scientific self takes on a particular form here. If ways of seeing are indeed also understood as ‘representing the knower’, the knower in this chapter – Marija Đorđeska – is mostly marked by her absence. Contrary to the writers of Chapters 6 and 3, which I will turn to below, Đorđeska does not explicitly address the role of the researcher in doing non-doctrinal work; rather, she hardly comes into view at all. Other than in an author’s note preceding the first footnote, no personal pronouns or possessives are used;⁹⁷ she refers to her own forthcoming book once in the third person.⁹⁸ The availability and use of impersonal forms differs across cultures, languages as well as age,⁹⁹ to name a few, but it is also associated with what Zohar Livnat refers to as ‘objective reporting’. Work in this style, he adds, is presented as

completely independent of the identity, personality or specific circumstances of the researcher carrying it out … a deliberate effort is made in scientific discourse to diminish the researcher’s presence in the text, resulting in an “objective” style of writing that ostensibly enables the facts to “speak for themselves”.¹⁰⁰

Viewed from this lens, most striking by far are the visual representations of the ‘findings’¹⁰¹ in Đorđeska’s chapter: it contains pie charts (e.g., representing the underpinnings of different types of principles) and cubes (‘classif[ying] all general principles under the same umbrella’), and refers to a ‘worksheet’ listing the 156 general principles identified in this way; a ‘Digest’; and mathematical formulas.¹⁰² Similarly, the section headers somewhat track the order of a science publication: (1) Introduction, (2) ‘Scientific’ Approach, (3) Inductive-Empirical Analysis, (4) ‘Scientific’ Results, and (5) Conclusion.¹⁰³ The ‘scientific’ approach taken by Đorđeska aims, in her own words, ‘to provide objectively verifiable data’; ‘a more accurate and objective image’ of general principles by drawing on ‘empirical research and inductive analysis’.¹⁰⁴

A similar hierarchy is established in Chapter 6. As in Chapter 1, the chapter findings are presented by means of graphs, pie charts and correlations.¹⁰⁵ Following three reflective sections on the relation between doctrinal and non-doctrinal studies and the ‘promises and perils’ of the latter, its headings similarly track those of a science publication: the methodology is followed by the research findings, which is in turn followed by a discussion and a conclusion. ‘[T]he promise of quantitative empirical research’, according to Guilia Gentile and Luigi Lonardo, is twofold: it both ‘helps getting closer to “truth”’ and it ‘promises a degree of accuracy that doctrinal methodologies might not have’.¹⁰⁶ The truth and accuracy pursued in this chapter pertain to the ECJ’s treatment of appeals against decisions made by the GC on EU sanctions: ‘The aim of our study is to offer an objective, quantitative basis for measuring several aspects of the functioning of EU sanctions’ appeals’, they write, ‘namely how the judgments correlate to (i) the formation of the court of the appeals which have led to the reversal of the GC’s decision and (ii) the length of appeal proceedings’. ¹⁰⁷

Contrary to Chapter 1, however, the writers of this chapter are far more present in their text than Đorđeska is in hers, for example by their use of personal pronouns and possessives – though they use the latter only in relation to their methodology. Section 6.5 of their chapter is titled ‘Our methodology’; the only other uses of ‘our’ or ‘we’ outside of this section are in relation to the word methodology itself, e.g., when they write that their chapter ‘takes the move from a schematisation of the methodology of previous scholarship … to the description of our methodology’.¹⁰⁸ In other words, the writers are present in designing the research but otherwise disappear from view. Rather than the researchers, others operate as the acting subject: the chapter itself (‘[t]he chapter focuses’), the research (‘this study contributes’), the findings (‘these findings … offer’) or the figures included in the chapter (‘Figure 6.3 indicates’).¹⁰⁹ The relation between the researcher and the method features most prominently in the opening salvo of their chapter, the first line of which reads that:

[a] method is a way to achieve a result, and it is neutral to that result … this is not true for all disciplines … A researcher might dismiss these disciplines in which the result depends on the methodology as sloppy, non-rigorous and/or unscientific … Interestingly, the same seems to hold also for more “scientific” domain … where it is the very presence of an observation that constitutes or modifies the phenomenon to be observed. Yet, the value of methodologies should not be a priori dismissed due to a suspicion of subjectivity.¹¹⁰

They elaborate on this a few pages further down, where they state the following:

At its most fundamental level, the promise of quantitative empirical research is that it helps getting closer to “truth”. Since all empirical data is, by definition, gathered through observation, experience and experimentation, it is, in principle, collected without bias. Arguably, the strength of scientific research depends on the ability to gather and analyse empirical data in the most unbiased and controlled fashion possible … Naturally, if taken at face value, the promise of ‘unbiased scientists’ is little more than chimerical. Since scientists – who are human and prone to error – can be influenced by prior beliefs and experiences, “science” lies in the fact that empirical data is often gathered by multiple people who independently replicate experiments … Science – that is, the aggregate result of individual researches – corrects the mistakes and bias which individuals, consciously or unconsciously, may commit.¹¹¹

In this view, more or less truth, greater and lesser objectivity, and better or worse science are matters of degree. Though full truth and objectivity may be out of reach, the writers say, it is in fact possible, necessary as well as worthwhile to attempt to get as close as one can. Moreover, the specific word choice of the contributors to describe the problem of subjectivity calls to mind Daston and Galison’s point that to do epistemology is a matter of virtue rather than competence; that the pursuit of objectivity comes with the ‘duty to discipline [oneself]’.¹¹² Gentile and Lonardo refer to a ‘suspicion of subjectivity’; to the necessity of collecting data ‘in the most unbiased and controlled fashion possible’ and how scientists, as ‘humans’, are ‘prone to error’. This is an explicit call to untangle to the greatest extent possible the scientist and the human being and to mitigate for the ‘errors and bias’ at play in any individual scientific effort. Not to try and do so exposes the individual researcher to the charge of subjectivity, that is to the charge of being a bad scientist in the moral sense of the word.

Given the above, it’s all the more interesting that in Chapter 3, it’s precisely the subject that takes centre stage. The chapter is full of personal pronouns: ‘I intend to make two claims’, its opening line reads; Ostřanský refers to ‘a collaborative multidisciplinary research project, of which I have been part’, further on in the chapter he states how ‘I already mentioned that …’. As do the writers of Chapter 6, he explicitly addresses the positionality of the researcher, but in a way diametrically opposed to Gentile and Lonardo’s views. The dominant ‘epistemic virtue’ is situatedness rather than objectivity: ‘Acknowledgement of the researcher’s positionality’, he writes, ‘implies that the relationship between the researcher and the interviewed is recognized and becomes part of the data, not something external to it’. Explicitly addressing the pursuit of objectivity as engaged in by others, he states how in anthropology ‘the positivist distinction between subject and object of knowledge is not presumed’.¹¹³ So in the kind of anthropology as engaged in by Ostřanský the ‘duty to discipline oneself’, to revert back to Daston and Galison, takes on a different form: rather than separate the scientist from the human, the researcher is highly aware of her or his own ‘positionality’,¹¹⁴ in Ostřanský’s words, and rather than repress it s/he makes awareness of it part of the research itself. ‘This reflexivity’, he adds, ‘uses the process of empirical enquiry to destabilize the positionality, frames of references and conceptual categories of the researcher herself’.¹¹⁵ Rather than the scholar being a hindrance to knowledge, s/he is very much part of its construction.¹¹⁶

In rejecting the positivist position and in being so explicit about the different aims of anthropological research, Ostřanský exposes himself to a possible criticism of being biased and of being far too present in his own research endeavours. That’s precisely the point, he would probably add, and as such his chapter is a direct engagement with and rebuttal of the positions held most prominently in Chapters 1 and 6. In the last, concluding section of this article I will elaborate on these opposing views about what constitutes good scholarship the volume contains – and how the editors deal with it.

4. International legal scholarship and the making of a ‘scientific self’

Different ideas of what is needed to really understand international law – or as Huaxia Lai says in this volume, what exactly is ‘relevant for understanding how international law works’ – bring Lorraine Daston and Peter Galison to write about epistemology ‘as the repository of multiple virtues and visions of the good, not all simultaneously tenable (or at least not simultaneously maximizable), each originally the product of distinct historical circumstances’.¹¹⁷ Read as a collection of scientific ideals, that is normative understandings of how international law may best be known, the volume under review is precisely this: a ‘repository of multiple virtues and versions of the good’. Some contributors claim that our knowledge of international law is improved by means of the values of generalizability, objectivity, precision, transparency and neutrality, whereas others hold a completely different understanding of what makes for good scholarship, be it ‘sensitiv[ity] to context’ (Landefeld), ‘situatedness’ (Ostřanský), engaging with the lived reality of those affected by the laws one studies (Margaria) or a bottom-up approach to the ‘actual practice’ of international law (Faulkner).¹¹⁸

Given the overarching aim of the volume, the editors’ question whether all these different methods actually improve our understanding of international law makes sense. Presuming moreover a somewhat sceptical doctrinal audience, the desire to persuade it of the added value of non-doctrinal work automatically follows. The question is, however, how this added value is subsequently measured; what is the yardstick for saying that something is a proper methodological undertaking or had better be left aside. By phrasing this question in terms of better (and therefore presumably also worse) science the ‘scientific selves’ of the editors emerge alongside that of their contributors. Though I previously provided some examples from the Introduction, their views come forward most strongly in the conclusion to the volume. It begins by repeating the volume’s initial question whether the non-doctrinal methods it contains also bring about a better ‘science’ of international law. Specifically, here Rossana Deplano asks whether the volume has indeed proven this point; whether ‘the empirically grounded type of inquiry is “more scientific” – hence, more precise – than the principled analysis of international law characterizing doctrinal research?’.¹¹⁹ What follows is an overview of the chapters, considered in terms of their scientific character and their added value to our understanding of international law compared to doctrinal work. The main argument put forward here is that each method contributes to international legal knowledge in its own way, even though they may differ in the degree to which they can be called scientific. ‘Scientific’, then, in this concluding chapter refers to greater precision, generalizability, objectivity, transparency and neutrality.¹²⁰ Yet it remains unclear who was asking after the scientific character of these non-doctrinal methods in the first place; more specifically, asking in this sense of the word. The editor’s engagement here with the supposed views of her imagined audience, as well as possibly her own, automatically puts the volume’s non-positivist methods on the defensive – and it shows in the way Deplano deals with them in her summary. For example, Alice Margaria’s chapter, which draws on ethnographic methods, is placed alongside the chapters on discourse analysis and archival research when Deplano writes how ‘[e]thnography poses similar problems about the generalizability of the findings’.¹²¹ Writing about Ostřanský’s chapter, Deplano puts forward that some of his methodological choices ‘at face value … [raise] concerns about the “scientific” character of ethnographic research’, and that his approach to his research question ‘raises questions’ of a similar kind.¹²² By invoking these concerns she engages in a dialogue¹²³ with her audience, something which is extended by her immediate response, for example when she counters that ‘because of its unbound and context-specific character, ethnography situates itself in an ideal position to prove or disprove axiomatic assumptions in the scholarly literature. The latter interpretation appears more convincing, since the methodology itself recognizes its own limits’.¹²⁴ In this view, ethnography may highlight something about international law overlooked by doctrinal work, despite its not being properly scientific. And with regard to Chapter 6 Deplano points out that ‘[q]uantitative analysis is not neutral either’,¹²⁵ adding that though the researchers themselves influence the research process in numerous ways, ‘[t]his … does not affect the objective and transparent character of the inquiry. The authors themselves acknowledge [the] issue [and] address and mitigate [it] in their analysis’.¹²⁶

It remains unclear who exactly is worried about mitigating bias other than the editors themselves; and to whom some of the purportedly less scientific methods ‘pose problems’ or raise either ‘concerns’ or ‘questions’. Stated somewhat crudely, they certainly do not to the contributors whose methods are interrogated in these terms. Put differently, though Deplano acknowledges different methods may be after different things, her benchmark for measuring their scientific character is a positivist one. ‘Concerns about the “scientific” character of ethnographic research’ may be raised by those pursuing objectivity, but the kind of pluralization engaged in here creates a space in which not just different – as per the editors’ claims – but fundamentally opposing scientific ideals are put forward. The volume, in that sense, ‘pluralizes international legal scholarship’ much more than it itself acknowledges.

Concluding, ours is one among many disciplines engaged in a ‘debate between positivism and its alternatives’:¹²⁷ international law is not the only discipline in which ‘epistemic anxiety’ produces the kind of volume as the one under review here.¹²⁸ This kind of anxiety emerges at disciplinary turning points, Daston and Galison say,¹²⁹ and in that sense it may be unsurprising to have Huaxia Lai claim ‘an identity crisis of international law’¹³⁰ in Chapter 8 of the volume. I’m not sure whether these struggles can indeed be best identified as a crisis or whether they constitute a permanent feature of international legal scholarship, but what Daston and Galison beautifully show by means of scientific atlases is that the way we look changes what we think to see, as our notions of ‘epistemic vice or virtue’ affect how we define and understand the ‘working objects’ of our scholarship.¹³¹ This problem of defining international law recurs in the conclusion of the volume, when Rossana Deplano concludes on the relation between doctrinal and non-doctrinal methods by stating that ‘[i]n the absence of a shared definition of international law, it would be counterproductive to ignore the inherently partial nature of any research method’.¹³² Interestingly given the purpose of the book, the editors themselves do not explain what they mean by ‘doctrinal’, except perhaps when the understanding offered in Chapter 6 by Giulia Gentile and Luigi Lonardo is understood as such. There, they refer to doctrinal work as based on ‘the letter of the law or … the reasoning of the Court decisions’.¹³³ As Deplano suggests in her conclusion, this is indeed a stark contrast with, for example, Sarina Landefeld’s understanding of international law as ‘regulat[ing] a specific interpretation of the social and material world which is reproduced in its terms’.¹³⁴ Yet a ‘shared definition’ isn’t just absent between doctrinal and non-doctrinal scholars; the same goes for the contributors to the volume. What is more, to say that the methods produced by these different understandings of international law – be they doctrinal or non-doctrinal – are ‘inherently partial’ does not quite cover the extent and depth of the disparities between them.¹³⁵ A chapter by Josef Ostřanský on general principles would render a completely different contribution to the current volume than Marija Đorđeska’s; conversely, a chapter by Marija Đorđeska on international investment law would already look completely different from the one written by Josef Ostřanský’s, let alone any differences between what they would ‘find’. The epistemic disagreements contained in Pluralising goes beyond them having different definitions of international law to include how they look, why they think they should do it in this particular way, and what that ‘way of seeing’¹³⁶ requires of their ‘scientific selves’.¹³⁷ In sum, the volume contains a struggle for the dominant epistemological virtue and the appropriate role of the international legal scholar as scholar. Any attempt to contain its contributions on equal terms is bound to fail, precisely because the normative premises under which the writers operate are mutually exclusive.