Methodologies of Rule of Law Research: Why Legal Philosophy Needs Empirical and Doctrinal Scholarship

Rule of law is a concept that is regularly debated by legal philosophers, often in connection to discussion of the concept of law. In this article, the focus is not on the substance of the conceptual claims, but on the methodologies employed by legal philosophers, investigating seminal articles on the rule of law by Joseph Raz and Jeremy Waldron. I argue that their philosophical argumentation often crucially depend on empirical or legal doctrinal arguments. However, these arguments remain underdeveloped. I explore how these arguments could be linked to approaches related to rule of law in different fields of legal scholarship and investigate how the methodologies of these fields may complement each other. Thus, the article aims to provide an argument for a specific form of triangulation of three kinds of approaches to the rule of law: philosophical, social-scientific and legal doctrinal. This method of triangulation is illustrated by a discussion of the World Justice Project’s Rule of Law Index.


Introduction
The rule of law is one of legal philosophy's favourites. It is a concept that is closely linked to the general questions of the concept of law, to the debate on descriptive versus normative accounts of law, and to other central concepts of legal philosophy, such as rules and authority. However, the rule of law is not simply a hobbyhorse for legal philosophers; it plays a central role in the practice of law as a normative idea, which legal systems around the world should uphold. What the rule of law consists of is therefore also of significant practical importance. The starting point for my argument here is that the importance of the rule of law idea also has methodological consequences: It requires attention to the way the rule of law functions and is perceived. Thus, this article raises the following observe these principles makes the law incapable of governing conduct, and conformity to these principles is a matter of degree. Raz adds a third principle: that the making of particular legal orders needs to be guided by general rules conforming to the first two principles. All of this is a conceptual argument based on the guidance function of law. However, the rest of the principles Raz enumerates no longer directly follow from the basic intuition. Principles four to seven all concern the place and functioning of the judiciary in a legal system: its independence, the importance of principles of natural justice, the power to review rule-of-law conformity and access to the courts. The eighth principle is rather specific, on limitation of the powers of crime-preventing agencies, which I will not discuss further. I am interested in the list of court-related principles. Where do these come from?
Can law only guide its subjects if there are independent courts, which are easily accessible? How do these principles follow from the conceptual claim about the character of law? Raz does give an argument for the independence of the judiciary linking it to guidance, but with a twist: (4) The independence of the judiciary must be guaranteed. It is of the essence of municipal legal systems that they institute judicial bodies charged, among other things, with the duty of applying the law to cases brought before them and whose judgments and conclusions as to the legal merits of those cases are final. Since just about any matter arising under any law can be subject to a conclusive court judgment, it is obvious that it is futile to base one's action on the law if when the matter comes to adjudication the courts will not apply the law and will act for some other reasons. 8 At first sight a guidance argument is made here: "just about any" legal question can be subject to a court judgment, so you need to know what these judgments are in order to know what the law asks of you. This argument, however, needs a few extra steps. Unlike Hart, who gives an extensive argument for the need for rules of adjudication in legal systems, Raz only posits that it is essential to have courts. A further argument is needed to show why general rules combined with particular orders are not enough. This could be done on the basis of Hart's work. 9 More crucially, the independence of the judiciary does not really follow from this reasoning. It is the duty of judges to apply the law in order to provide guidance, because people need to know what the law is, and things become complicated if judges do not apply the legal rules. But what does this have to do with independent decisions by judges? Why might it not be an administrative official, subordinate to the government, who applies the rules? The guidance function of law does not generate a conceptual argument that requires independence. It seems much more likely that Raz uses the character of existing legal systems that he knows about to come up with this principle.
Supporting arguments for principles five to seven are even less informative. For example, Raz states that the principles of natural justice include "open and fair hearing, absence of bias, and the like" and to be "obviously essential for the correct application of the law". 10 Are these principles essential to get at the facts of court cases or for interpretation of the laws? What else is essential other than a fair hearing and impartiality? It also matters whose guidance is meant: Is it the citizen who is involved in the case? That would be strange, because court cases are backward-looking: They consider whether the law has been applied correctly. If it is guidance for the broader citizenry, it seems that the principles of natural justice are not so important. To them, the outcome of the case matters.
Principles of natural justice are important for citizens as a guarantee of the fairness of procedures they may be involved in, but that is not a matter of guidance.
There are fine arguments for these principles of judicial independence, impartiality, etc., but they are not conceptual; they refer to actual legal systems and the experiences within these systems. The four principles concerning the judiciary make sense, once we place them in the context of legal doctrinal arguments about the structure and functioning of legal systems. Before I go into the nature of those arguments in section 4, I turn to Waldron's procedural rule of law.
In 'The Concept and the Rule of Law', and a number of related articles, 11 Jeremy Waldron takes issue with the standard division of rule-of-law theories into the categories of formal and substantive theories. 12 Formal theories require only form-related characteristics of legal rules such as clarity and prospectivity, while substantive theories also require certain content, such as non-discrimination or protection of individual rights. Waldron disagrees with this categorization because he believes a crucial strand of rule-of-law theorizing does not receive adequate recognition in either the formal or the substantive view. He argues that the prime value of the rule of law is procedural: It is the access to courts and the procedures within them. To some extent, the procedural rule of law is complementary to the formal aspects; the eight principles of Raz are partly formal and partly procedural. However, Waldron argues, there are situations in which demands of procedural rule of law are in tension with formal requirements. 13 Procedural rule of law demands that citizens have a voice in court procedures, but this entails that their arguments are listened to and may be used to interpret legal rules in novel ways, which makes law less predictable, therefore detracting from a central formal requirement.
What I am interested in here are Waldron's arguments for highlighting procedure in the rule of law.
His starting point is a comparison between a philosophical and a popular idea of the rule of law. He acknowledges that in legal philosophy the rule of law is primarily thought of as formal (e.g., by Raz and Fuller), but he claims that in popular opinion the rule of law is seen as procedural: When people say, for example, that the Rule of Law is threatened on the streets of Islamabad or in the cages at Guantanamo, it is the procedural elements they have in mind, much more than the traditional virtues of clarity, prospectivity, determinacy, and knowing where you stand. They are worried about the independence of the Pakistani courts and about the due process rights of detainees in the war on terror. 14 Waldron thus claims that in the eyes of ordinary people rather than theorists, the formal elements of the rule of law are subordinate to procedural elements, which include the independence of courts and procedural rights.
His reasoning for the importance of procedure is further supported by a conceptual argument that, like the rule of law, the concept of law too is more court-centred than most theorists acknowledge.
Of the five essential requirements for counting as a legal system, this is the first he discusses: "First and foremost, I do not think we should regard something as a legal system absent the existence and operation of the sort of institutions we call courts." 15 Waldron offers this as a suggestion, and does not provide further arguments other than, again, what people ordinarily think. "Most people, I think, would regard hearings and impartial proceedings -and the accompanying safeguards -as an essential rather than a contingent feature of the institutional arrangements we call legal systems." 16 Waldron has much to say about the further moral reasons to value the procedural element in law and the rule of law, but here I want to draw attention to his strategy of countering the conceptual accounts of legal philosophers by appealing to "how the term is ordinarily used" and what ordinary people "have in mind". 17 To summarize the point of this section, it is useful to give a brief characterization of how I understand conceptual argument as practiced by legal philosophers. Conceptual argument is the explanation or justification of the meaning of central concepts of law, such as rule of law, in terms of other concepts in a theoretical argumentation. Such argumentation often builds on basic intuitions, such as Raz's notion of the guidance function of law. It may also use the common-sense meaning of a concept as a building block in a conceptual account, but on its own this is not sufficient. 18 A good conceptual argument gives a coherent explanation or justification of the central concepts and thus goes beyond direct appeal to the ordinary understanding of a term.
Because Waldron so explicitly contrasts philosophers' arguments with what ordinary people think, I interpret Waldron's appeals to what people say and think as empirical claims: how people in fact, in political and popular debate, understand the rule of law. However, it is questionable whether we can rely on common sense and our own understandings of popular debate as an adequate source of such understanding. In the next two sections, I will argue that a turn to empirical social science and to doctrinal legal scholarship is helpful to support, and in some respects criticize or question, the nonconceptual arguments made in legal-theory accounts of the rule of law.

Making use of empirical social science
The starting point here is where I left the discussion of Waldron: the claim that people have in mind independence and due process when they think of the value of the rule of law. This is an empirical claim about the actual ideas ordinary people have about the rule of law, for which Waldron only advances support by way of media reports. Does the fact that people criticized the regime for Guantanamo Bay detainees in terms of the rule of law provide sufficient evidence that people believe that the procedural elements of the rule of law are more important than the formal ones? There are at least two reasons why this question should be answered in the negative. The first reason is that such a line of reasoning depends on the incidents that cause public outcry: These are extreme cases that are hotly debated, while the smaller problems of the rule of law and everyday experience with it are not discussed. The second reason is that it depends on a particular legal culture: Waldron's examples are American or common-law examples, and American legal culture is one of the most judicially oriented around the world. For instance, if one were to ask about rule-of-law problems in Bulgaria or Romania, chances are that general governmental corruption is seen as the most pressing rule-of-law problem. 19 If the idea is that a general concept of the rule of law needs to take account of popular views of the rule of law, it seems rather limited to look only at the Anglo-American setting.
To overcome these limitations and gather more robust evidence on what ordinary people think of the rule of law, it is helpful to consider empirical social scientific work. More particularly, because the argument depends on what people think of the rule of law, the scholarship on legal consciousness is a promising source for support of the claim that people regard procedural rule of law principles as the most important. Rather than opinion polling the general population's views on rule of law, legal consciousness research focuses on the everyday significance of law and the way this shapes and is shaped by what people believe and do. 20 In The Common Place of Law, Ewick and Silbey developed a typology of attitudes towards the law based on interviews. Distinguishing between positions 'before the law', 'with the law' and 'against the law', they show how people's attitudes towards law differ depending largely on the situations in which they find themselves. 21 Their research thus shows that different cultural attitudes to law are available: People may believe in law's autonomy and formal rules as a system that does not really concern their everyday life ('before the law'), they may see law as a game they can play in their own interest ('with the law') or they may see law as an oppressive force to be resisted ('against the law'). This research does not really support a general claim about the rule of law. Only the attributes of being 'before the law' have a relation to a rather formal idea of rule of law. Procedural fairness does not play a large role here, because people with this attitude do not see law as an opportunity for their own engagement. Neither the manipulation by the group 'with the law' nor the scepticism of being 'against the law' points to much faith in the link between shows a difference in emphasis at the very least. 28 The English rule of law emphasizes ordinary law administered by courts, as in the classic 19 th century formulation of the rule of law by Dicey. 29 The German 19 th century tradition of the Rechtsstaat was linked to struggles with the authoritarian state and explains the emphasis on formal legality, the idea that all state action must have a basis in legislation. The French idea of l'Etat de droit, emerging after the revolution, also needed to deal with the power of the legislature, but the French debate also referred to the Declaration of the Rights of Man, which might explain why in the French legal tradition both a strong focus on the legislature and on equal rights can be found. This is not so much a matter of understanding the historical shape of the rule of law itself, but of understanding the legal traditions that give us the resources to think about the rule of law in certain ways, and not in others. The historical dimension may create awareness of the links between political developments, ideas about law, and institutional structures, and the way these historical configurations influence rule-of-law thought. Referring to popular understandings of the rule of law, but also to empirical research about those understandings is a historically situated argumentative step.

Building on legal doctrinal scholarship
In section 2, I argued that both Raz and Waldron make use of claims about legal institutions, in particular courts and court procedures, with little argument on how the court-related and procedural elements follow from the idea of the rule of law. In this section, I will examine what legal doctrinal scholarship can add to the support of that claim. In addition, I will argue that legal doctrinal research also points out new directions for rule-of-law theory, particularly through its attention for the interplay between different legal orders. Before going into the specific arguments, however, it is 28 Tamanaha Legal doctrinal scholarship can be broadly defined as the systematic study of legal norms in the various sources of law that form the basis of particular legal systems. 30 Most legal doctrinal work is therefore the study of (a part of) a particular legal order, of English, French, European Union or international law. In the account I give here, I generalize over common law and civil law traditions. 31 Legal doctrinal scholarship is the study of positive law, but in that study it takes on board the academic writing about positive law too (and in that sense is self-referential). In many legal systems, the distinction between primary sources (legislation, treaties, court cases) and doctrinal writings is independence of courts supports the rule of law because it ensures that the power to determine the meaning of legal rules is shared by institutions that are organized as separate bodies. On this view, it is not so much separation of powers, but a sharing of powers by separate institutions that promotes the value of rule of law. The fact that courts, the judges of which are not directly accountable to other government institutions, need to interpret and apply the rules produced by lawmakers creates an interplay between legislatures and courts that curbs the power of each. This argument entails that the independence of courts has a slightly different relationship to the overarching rule-of-law ideal than suggested by Raz, who sees courts as necessary to provide practical guidance to legal subjects.
Court interpretation may be important for that reason, but it is more important to prevent legislatures from creating arbitrary rules. This may still fit the Razian argument, but it does so indirectly: It is the complex of legislative and judicial work that prevents arbitrary rules, which may then contribute to law's guidance function.
In this discussion of the principle of court independence, it is already apparent that doctrinal scholarship not only provides supportive arguments but also reinterprets the idea of the rule of law ought to relate to each other. Thus, the direction of legal doctrinal research generates a conceptual challenge to reconstruct the meaning of the rule of law to respond to these legal developments.

Triangulation: complementary methodologies and disciplines
Each of the three disciplines has a distinct perspective with specific substantive research interests and its own methodological approach. I have so far been optimistic about the synergies between these three perspectives, but one could be more sceptical: If the questions asked and the answers given are articulated in a different language, is it worthwhile to try to combine them?
My argument so far assumed that the hints at empirical and doctrinal knowledge from within legal philosophy are a first step toward collaboration between the disciplines to find support for these kinds of arguments, but it is possible to argue in the other direction. Maybe it is advisable to steer clear of empirical and doctrinal arguments altogether and limit the legal philosophical argument to conceptual and/or value-based reasoning only. Various legal philosophers characterize and justify their own approaches in contrast to sociology of law, arguing against the need to engage with sociolegal work. 49 Raz, for example, sees legal philosophy as engaging with universal and necessary characteristics of law, while sociology of law describes particulars of various legal systems. As Raz states, sociology of law "provides a wealth of information" on contingent features of existing legal systems, while legal philosophy needs to focus on "those few features that all legal systems necessarily possess." 50 However, this division of labour is questionable: How do we know what the necessary features are without knowledge of the contingent features? As I have argued in section 2, features that are presented as necessary for the rule of law, such as independent courts, may turn out to be contingent upon further examination. As Giudice argues, there is good reason for legal philosophers to expand their research beyond the necessary and universal characteristics of law, 49 Hans Kelsen is perhaps the most famous defender of the separation of the projects of the science of law and sociology, following from his strict separation of 'is' and 'ought ' which he sees as a need to move from conceptual analysis to "constructive conceptual explanation" in order to address the observed pluralism of existing conceptions of law. 51 As must be clear from the earlier discussion, I agree with Giudice that we need to move beyond legal philosophy as a self-contained project. An important reason for this point of view is that even clearly Continuing then with the exploration of the combination of perspectives, an idea from social science methodology may serve as a starting point. In social sciences, an often-used methodology is that of triangulation: using different methods to answer the same question in order to corroborate or complement results. 55 The assumption is that finding the same results with different methodological tools strengthens the support for those results, and makes the conclusions better warranted. Thus, if a researcher interested in the role of cultural differences in the courtroom finds that his interviews with judges show that they regard defendants from cultural minorities differently, and that his courtroom observations point out differences in the interactions between judges and defendants from minority groups compared to defendants from the dominant culture, and that analysis of verdicts shows differences in justification, such triangulation may be said to strengthen the conclusion that judges are sensitive to cultural differences between defendants. 56 In this example, the methods discussed are all forms of qualitative research, and they can all be used to answer the same question. Such corroboration is more problematic when the combination is between methods that are very different in orientation. Triangulating quantitative and qualitative research methods is less likely to lead to comparable results, not least because these methods are suited to answer sense to combine methods to overcome these limitations. Thus, in the example of cultural differences in the (criminal) courtroom, we may want to include differences in sentencing outcomes, an aspect that is best researched by analysing a large data set.
Once the complementary form of triangulation is accepted as a valid strategy, interdisciplinary triangulation can be seen as a natural next step. 57 Here too, the methods complement each other and may serve to uncover aspects that the other discipline cannot see or can only partly understand.
Making the argument for complementarity between legal philosophy, empirical socio-legal science and legal doctrinal scholarship requires a sense of the distinctive contributions of each and possible connections between them.
Legal philosophy is distinctive in its use of conceptual arguments about core ideas in law, many of which are normative, such as rules or rights, or have a normative aspect, such as authority or courts.
The focus on normative ideas also creates a need for normative arguments to support the substance of these ideas. Characteristic of the methods of legal philosophy is the construction of theoretical arguments based on various conceptual elements leading to a coherent account of basic ideas about law. In practice, the methods are primarily discursive: examining arguments and counter-arguments, building on existing theoretical work and aiming to improve on that work by presenting an alternative or modified account. Abstract theoretical arguments are confronted with concrete elements in various forms: common sense judgments or everyday language use, hypothetical or real examples. 58 The aims of legal philosophy may differ: 59 They may be descriptive, aiming to make sense of the meaning and use of legal (normative) concepts, or normative, aiming to justify the use of values of the rule of law are taken as a measurement criterion or transformed into something that is part of people's actual beliefs and practices. In this article, I focus primarily on the latter, the research on cultural understandings of the rule of law, while there is also a good deal of research on the rule of law from a more quantitative angle. This would lead to including other disciplines such as economics and political science as the empirical perspective. 63 The reason for my focus on this form of socio-legal studies lies in the arguments that are needed to complement the views of Waldron and Raz: Their argumentation leads in this direction.
The distinctiveness of legal scholarship is the most difficult to grasp, but here the focus is on the normative content of the law as an institutionalized coherent practice. Legal doctrine focuses on the meaning of legal texts in a particular institutional setting, of a legal system or particular legal regime, often in comparison to other systems or regimes, and studies these texts in that context. Because of its contextual and institutional orientation, it can provide detailed descriptions that clarify relations between norms, procedures, roles and arguments. Because of its focus on coherence, it pays specific attention to the relation between the particulars of the text or case and the larger whole of the system. 64 Legal doctrinal scholarship does not question the value of the rule of law but is interested in how it works in specific legal contexts. 65 Considering the similarities and differences between the three disciplines, legal philosophy and doctrinal legal scholarship seem linked more closely than either of them is to empirical socio-legal science. The difference between the first two can be characterized as focus on the general versus the particular: where legal philosophy tends towards general arguments, doctrinal scholarship focuses on the way law is applied concretely in particular contexts, and on comparison of these particular instances. There are also significant similarities, which can be summarized as a hermeneutic methodology: linking descriptive and normative aims and arguments and using interpretive and argumentative reasoning. Empirical socio-legal science, by contrast, focuses on description and explanation, staying away from normative claims, and uses a data-based empirical method.
One way to view complementarity of these three disciplinary perspectives is to advocate a division of labour in a strict order: First, legal philosophy provides a careful argument for a particular conception of the rule of law; second, legal doctrinal scholarship shows how the rule of law conception is given a particular shape and applied in legal institutions and third, empirical science uses this conception to find out whether actual rule-of-law practices deserve to be labelled "rule of law," and whether what goes on in the rule-of-law institutions has meaning in people's everyday lives. In various combinations, such work is done. 66 However, more interesting combinations go beyond complementarity and critically reflect upon the other disciplines. Ideally, the criticized discipline takes on board the criticism and tries to reorient itself to incorporate it. Thus, the most demanding form of triangulation of perspectives is to allow for corrections of one's own perspective based on the criticism voiced by the others. This requires a sensitivity to each other's projects, acknowledging the value of the other perspective. It entails thinking about the way in which one's criticism may be received by the discipline under scrutiny, being careful to avoid complete dismissal and posing questions in such a way that it opens up new directions for the other discipline. This is not to say that fundamental critique is always unwarranted; there is an argument to be made, for instance, for that my classification of the three disciplines is wrong: Why distinguish these three? Other theorists advance an interdisciplinary methodology for legal theorists who draw the lines in different places.

An attempt to do this for rule-of-law research is Paul Gowder's book, The Rule of Law in the Real
World, in which he makes a combination of conceptual/normative legal philosophy with empirical work. He puts law and philosophy in the same camp, as the providers of "normative/conceptual tools". 68 However, he also lists law as one of the disciplines providing knowledge of "real-world institutions", 69 which makes it seem as if the discipline of law has a double function in Gowder's account: both the source of normative principles and the provider of knowledge of real institutions.
This points to the difficulty of distinguishing the normative and the descriptive in the legal field. 70 However, if our interest is in the justification of conceptions of the rule of law, it is important to keep track of the kind of argument advanced -conceptual, empirical, or institutional -because the support needed for such arguments differs. Another interesting argument, about the concept of law rather than the rule of law, is given by Michael Giudice, in Understanding the Nature of Law. Giudice argues for the need to combine conceptual legal philosophy with empirical social science and moral theory, leaving out legal doctrinal work. In his framework, it seems that legal doctrine does no more than provide the problems to which the combination of the other three disciplines gives answers. As will be apparent from my earlier account, I think this misses the important contribution of legal doctrinal scholarship as providing contextual arguments about the institutional side of the rule of law. 71 An objection to my idea of triangulating distinct perspectives could be that it stops short of what good scholarship should do, namely to integrate perspectives. Rather than argue that there are three different perspectives that need to speak to each other, one could argue that insights from each of these perspectives need to be synthesized into a disciplinary perspective of its own. 72 Dagan and Kreitzer make this argument in their defence of legal theory as a middle position between what they see as the dominant other approaches to law: law as craft, law and policy and sociohistorical analysis of law. Philosophy of law could well be added as a fourth corner to which their idea of legal theory relates: "One manifestation of the significance of law and craft for legal theory is legal theory's tendency to be less abstract than the philosophical, economic, or other theories with which it interacts." 73 What is crucial in their account is that legal theory is a "synthesis" necessary to deal with law as a complex and multi-faceted phenomenon. 74 The triangulation idea advanced here is more modest: Rather than claim that we need a synthetic discipline of legal theory, I think it is ambitious enough to ask of the separate disciplines working on the rule of law to take seriously the implications of the other disciplines for their own work. Thus, legal philosophers need to take on board the developments in legal doctrinal scholarship and sociolegal studies that are a reason to question certain preconceived ideas, like the notion that the state legal system is the central case to account for. 75 This may be reason to investigate how a universalizing approach of the rule of law can respond to pluralism arguments. How such triangulation might work is best explained by way of an example. independent adjudication." 78 However, when looking more closely, it becomes clear that Botero and Ponce have a more expansive, and substantive, view of the principles necessary to achieve rule-oflaw quality. Of the nine factors they distinguish, the fourth factor is fundamental rights, and in some of the sub-factors, other rights-based indicators appear, such as a free press and free elections.
Because this substantive component is not really supported by arguments other than an appeal to "a common theme" and to the United Nations' definition of the rule of law, 79 the legal-philosophical approach may serve as a valuable source of argument to complement the theoretical framework.
Especially Waldron's normative arguments for a procedural rule of law as enhancing human dignity and agency help to join rights to formal procedure. Because court procedures are crucial to participation of individuals in legal matters, allowing them to express their own views and confirming their status as human agents, they have both formal and substantive value. However, this does not generate an argument for full-blown fundamental-rights protection as part of the rule of law, which would need further reasoning based on dignity and equal participation. If individuals need equal opportunities to participate in procedures in order to recognize their dignity, they need the freedom and capacities to do so. Without meaningful access to justice or knowledge about procedural rights, a procedural rule of law will not be realized. Although this widens the scope of the rule of law beyond Raz's elements, Waldron's theory does restrict the scope of rights to those with procedural relevance. With regard to a second issue -the operationalization of the factors -a noticeable feature of the Index is that it is set up to be outcome-oriented, looking at how the rule of law is achieved in practice. 80 Thus, the questions asked concern experiences with government procedures and the justice system rather than the existence of such institutional elements. However, from a legal-doctrinal perspective this outcome-orientation seems somewhat overstated. Although some questions are about the outcomes of legal procedures, many of the questions concern the 78 Botero and Ponce, 'Measuring the rule of law', p. 5. 79 Botero and Ponce, 'Measuring the rule of law', p. 5. 80 Botero and Ponce, 'Measuring the rule of law', p. 16: 'The second fact is that, for the most part, our focus is on rule of law outcomes; as opposed to the institutional means, or inputs-including the legal and regulatory frameworks-to attain them.' experience with the institutions themselves: e.g., awareness of procedures, possibilities to participate and experiences of impartial treatment or corruption. From a doctrinal perspective, these are questions about the functioning of the institutions and procedures that involve the process rather than simply the outcomes. Although the focus shifts from a focus on outcomes to a more even attention to institutions and the way they function, doctrinal scholarship thus actually supports the content of the Index. In such a narrow application of triangulation, I tentatively conclude that it is possible to achieve a more coherent and better-supported account of the Index project than the one provided by the makers of the Index by themselves: restricting its meaning to a substantively coloured procedural rule of law and acknowledging the importance of legal institutions in this.
While a narrow form of triangulation is mainly useful to fill gaps in the argumentation for the Index, a broader form of triangulation is potentially more critical. To construct it, I return to the points made about pluralism earlier, which may be elaborated to apply to the Index. Such a broader form of triangulation does not take the framework of the Index for granted, but also questions underlying aims and assumptions of such an assessment of the rule of law in the world. The most important methodological change from a narrow form of triangulation is to make use of other socio-legal research than the empirical data of the Index, most importantly including theoretical implications of legal sociology and anthropology. And by asking what rule-of-law aspects are not included, doctrinal scholarship may be used for criticism by applying its insights on internationalization of rule of law.
Finally, the discussion in legal philosophy about the normative core of the rule of law can be used to challenge the choice of the factors of the Index and the aggregation of the scores.
This combined use of the three perspectives may be labelled a multi-faceted legal pluralism challenge: pluralism of legal orders, pluralism of legal cultures and contexts and pluralism of rule-oflaw values. I do not mean to suggest that a pluralism challenge is an inevitable direction for broad triangulation -Gowder, for instance, criticizes the Index for being multidimensional rather than onedimensional 81 -but it is a challenge that is fuelled by important insights from all three perspectives.
The pluralism of legal orders -in the sense of the interactions between international and national law -and its importance for the rule of law is a point made repeatedly by doctrinal scholars, as shown above in section 4. 82 In the Index, there is no reference to the international dimension of the rule of law (other than the use of the UN definition). A question worth asking is whether a country conforms to international norms, particularly, when it comes to rights protection, in the form of human rights treaties. From a doctrinal perspective, the internationalization of the rule of law cannot be ignored when fundamental rights are concerned: There is clear influence of international developments in national legal practices and national courts contribute to the development of these rights transnationally. 83 Socio-legal studies bring forward the importance of the pluralism of legal cultures, which also has implications for the reach of the rule of law and the ways to protect it. 84 The way in which the Index touches upon legal-cultural pluralism is interesting. It embraces pluralism in the ninth factor, informal justice, where it is acknowledged that the rule of law may be enhanced by effective dispute settlement by non-state actors, in traditional, religious or community-based forms.
Although the Index thus incorporates the insight that access to justice may be provided in ways other than through formal courts, in the final aggregate score of the Index this factor is not counted, because the Index compilers say the mechanisms are too varied and measuring their effectiveness is too complex. 85 This choice is significant: Not only do the limitations of the Index become clear -its methodology cannot cope with an aspect that is seen as necessary theoretically -it also points to a pluralism of underlying values, which also appears in a legal-philosophical perspective. If we admit that a full picture of rule-of-law realization needs to include non-state orders, the Index approach needs to be combined with other types of research, particularly the ethnographic, contextual method of legal anthropology. Thus, the socio-legal pluralism point also challenges the quantitative approach of the Index. From a different angle, the legal philosophical perspective may also do this: by developing the point of rule of law value pluralism.
A crucial debate in legal philosophy about the rule of law concerns the relationship between the ideal of the rule of law and other values and purposes. Both Raz and Waldron distinguish rule of law from these, but acknowledge the need to justify the rule of law in terms of further purposes, such as serving freedom through guidance (Raz) or protecting human dignity through procedural standing (Waldron). Both Raz and Waldron acknowledge that the rule of law may be justified by reference to different values. As discussed in section 2, Waldron even argues that the different strands of the ruleof-law ideal are in tension with each other: Whereas the formal rule-of-law aspects demand predictability to provide stability, the procedural aspects lead to openness by recognizing the arguments of individual agents, in order to acknowledge their agency. If this argument of internalvalue pluralism is accepted, the Index needs to be questioned on a more fundamental level. What are the values that the rule-of-law conception of the Index furthers? What is the underlying justification? This question is not explicitly addressed, but looking at the various factors, one sees links between social order and freedom, open government and participation, access to justice and non-discrimination. 86 Thus, the range of different values served by rule of law according to Botero and Ponce is even broader, and the set of values seen as integral to rule of law more varied, than those advanced by Raz or Waldron. This plurality leads to problems when it comes to measuring the rule of law. As Raz argues, value pluralism implies incommensurability, 87 which means that aggregating the scores of different factors, protecting different values, into one number is not possible. Aggregation leads to intriguing results; for example, Singapore scores 0.80 while Canada 86 Botero and Ponce, 'Measuring the rule of law', p. 10, 12 and 14 respectively. 87 Joseph Raz, The Morality of Freedom (Oxford: Clarendon 1986), pp. 321 ff. scores 0.81, but the factors on which they score well are very different. 88 Canada also scores well on the Freedom House index of 'Freedom in the world' (99/100) while Singapore does not make the list of free countries, only scoring 52/100. 89 If the rule of law is supposed to serve individual freedom, this discrepancy is problematic. The values of social order and individual freedom may not be as clearly linked as the Index suggests. I would conclude that reporting on the Rule of Law Index is more useful without aggregating scores, that it needs to be contextualized and read in combination with other sources.
Of course, this sketch of a critique of the Index on the basis of three perspectives is only a start, and it is clear that it could be done differently. 90 What I hope to have shown is that the combination of three perspectives strengthens the basis for criticism and that the three perspectives can be used to reinforce and supplement each other's arguments, when looking at a specific rule of law issue.
Moreover, although I have been critical of the projects of Raz and Waldron as separate legal philosophy, it turns out that their views make crucial contributions to the assessment of a concrete phenomenon such as the Index.

Conclusion: Asking the right questions
Thus, the bottom line of interdisciplinary cooperation may not be seeing other disciplines as the source of information that can be put to use in one's own perspective, but of the other discipline providing a different framing of the issues that challenges your own discipline to change its questions. Central in this regard are the methodological limitations of each discipline: Knowing what can and cannot fruitfully be asked should be central to the conversation. Having such a dialogue, however, does entail being open to interpreting the views from the other discipline charitably, and making a genuine effort to consider in what way these can be incorporated in one's own framework.
That means making sure the claims from the other discipline are well understood, but also that they can be turned into viable arguments in one's own context. 91 For legal philosophers, this implies that they stand to benefit from interdisciplinary cooperation if they go beyond using common-sense knowledge of the social context of law and beyond their own partial knowledge of legal doctrines and institutions. Legal philosophy needs to be open to these forms of interdisciplinarity if it aims to be serious about all of its arguments, including the empirical and doctrinal parts of these. Moreover, legal philosophy will be more relevant for the study of law more broadly if it is willing to take a more interdisciplinary course. 92 In combination with other perspectives, the philosophical study of the rule of law makes important contributions to our understanding of this central legal value. 91 One way of describing this process is James Boyd White's idea of translation between disciplines as requiring a double responsibility: staying true to the original claims, but also making the best of the receiving discipline, 'Establishing Relations between Law and Other Forms of Thought and Language'. 92 Compare Cotterrell, Sociological Jurisprudence, pp. 54-55.