2. Four perspectives on international law

What are these different perspectives? Initially, it is useful to distinguish four, which reflect four different questions we can ask about international law.Footnote ⁵

First, the doctrinal perspective focuses on questions of legal validity and justification. It asks: What are the existing legal norms and how do they apply in particular contexts? Do human rights norms apply extraterritorially, for example? Does customary international law require states to limit their greenhouse gas emissions, in order to prevent damage to other states? Does Crimea have a right to secede from Ukraine? Do states have the right to use self-defence against attacks by non-state actors and, if so, under what circumstances? Can states use trade measures to promote climate protection? More generally, what is the legal status of UN General Assembly resolutions? What is the result of an invalid treaty reservation? Do jus cogens norms depend on state consent?

Second, what I will call the normative perspective asks: What should be the legal rule, as a matter of policy or morality? For each doctrinal question about what the law is, there is a parallel question about what the law should be. Should human rights apply extraterritorially, for example, or Crimea have the right to secede from Ukraine, or states have the right to use force in self-defence against attacks by non-state actors? What should be the effect of General Assembly resolutions or invalid treaty reservations? When we evaluate existing legal rules as moral or immoral, wise or foolish, adequate or inadequate, or when we advocate the creation of a new legal rule, we are adopting the normative perspective.

Third, the explanatory perspective asks causal questions about how international law develops and operates, and what its consequences are. Why do new legal rules emerge – rules about diplomatic immunity, for example, or protection of the ozone layer, or free trade, or self-defence, or secession? Why is the rule of law stronger in some societies than others? Does international law simply reflect power relationships or does it serve to constrain power? To what extent does it affect behaviour and under what circumstances? These questions about the causes and effects of international law reflect the explanatory perspective.

Finally, the instrumental perspective asks how international law might be used to achieve desired ends, such as the promotion of human rights, environmental protection, or free trade. It sees law as a tool, and focuses on how we can use that tool most effectively. In tackling a problem such as global warming, should we proceed at the global or regional level, through hard or soft law, using what kinds of incentives or sanctions, with what exit options, and so forth?

These four perspectives are, of course, interrelated in various ways. The instrumental perspective, for example, raises normative questions about our desired ends, doctrinal questions about the existing law, and explanatory questions about how to move from here to there.

Nevertheless, the four perspectives are conceptually distinct. The doctrinal perspective is that of insiders – judges in deciding cases and lawyers in advising clients – and focuses on issues of interpretation, application, justification, and validity. In contrast, the explanatory perspective views law from the outside, as an object to be explained, and is the perspective of social science and history: social science to investigate general causal mechanisms, and history to explain the origin and effects of particular legal rules and institutions. The normative perspective focuses on what the law ought to be, and is the perspective of moral and political philosophy. And the instrumental perspective is concerned with problem solving, and is the perspective of legislators and policy analysts.

Distinguishing between the four perspectives is important for two related reasons. First, it helps us avoid the trap of focusing on a single perspective, thinking we are perceiving the entire elephant. Second, failure to distinguish the four perspectives can lead people to talk past one another. They can appear to disagree, when they are in fact addressing different questions. Consider, for example, the dispute about whether judicial decisions are based on legal or non-legal reasons. A judge's political views, biases, or mood on a particular day may be the cause of his or her decision, and hence may explain the result. But they are not a reason or justification for the decision; they do not make the decision valid. The difference here is not a difference between two conflicting views, but between the explanatory perspective, which focuses on the process of discovery, and the doctrinal perspective, which focuses on the issue of justification.Footnote ⁶

A similar confusion underlies criticisms of the application of international relations theory to international law. States’ economic and political interests may help explain various features of the legal landscape: what legal norms have emerged, why they take the form they do, why they have been more effective in some issue areas than others, and so forth. But international relations theory cannot tell us whether a particular international rule is justified or is a valid rule of inter-national law. It can help answer explanatory questions, but not normative or doctrinal questions.

3. Precursor of realism: classical legal thought

Stewart Macauley once observed, ‘It is easier to describe what those who came to be known as realists were against rather than what they were for.’Footnote ⁷ So, to understand legal realism, it is helpful to understand its antagonist, ‘classical legal thought’, which dominated American jurisprudence in the late nineteenth and early twentieth centuries.Footnote ⁸

Classicism had many elements. Although it is perhaps best known for its conservative appellate opinions protecting business interests against labour unions and government regulation,Footnote ⁹ two features of classical legal thought are particularly important for our purposes. First, classical legal thought (at least as portrayed by the realists) saw law as a self-contained, autonomous discipline, distinct from politics and morality, focused exclusively on doctrinal issues. Second, it believed that doctrinal questions could be answered purely through the method of logic.

According to classical legal thought, law is a science, aimed at expounding ‘universal principles of justice and moral order, . . . as prevalent, unchanging, and authoritative as the law of gravity’.Footnote ¹⁰ Classical legal theorists believed that doctrinal questions can be answered through a distinctive process of legal reasoning that is abstract and categorical. Although classical legal theorists, in reality, had a strong ideological bias in favour of laissez-faire capitalism, they claimed that law is a logical system that is ‘neutral, natural, and apolitical’.Footnote ¹¹

Because classicalism conceptualized law as an autonomous system of logic, it regarded ‘social science data with Olympian indifference’.Footnote ¹² For a classical legal theorist, it would have been as nonsensical to ask normative questions about whether a legal rule is good or bad, or explanatory questions about how a rule developed, as it would have been to ask these questions about rules of logic. Policy considerations, morality, ideology, the personal sympathies and assumptions of the judge – none of these factors matter in the development of the law, since judges do not make the law, they simply find it.Footnote ¹³ Judges have ‘no more discretion to invent a legal rule on instrumentalist grounds or policy preferences than a chemist ha[s] to dictate the outcome of an experiment’.Footnote ¹⁴ Legal science develops in the same way as logic and mathematics, through an autonomous process of abstract reasoning.

4. American legal realismFootnote ¹⁵

American legal realists disagreed about many things, but they agreed in rejecting the two basic tenets of classical legal thought: formalism and autonomy. First, they believed that doctrinal questions cannot be answered solely through the method of logic, a view that Jerome Frank irreverently called ‘legal fundamentalism’Footnote ¹⁶ or ‘rule-fetichism’ (sic).Footnote ¹⁷ Instead, legal decision-making necessarily involves extra-legal factors. Second, legal realists viewed the law in instrumental terms, as a means to an end. As such, they were interested not only in doctrinal questions about the existing law, but also normative questions about the appropriate ends of the law, explanatory questions about how the law operates, and pragmatic questions about how it can be used to achieve these ends. Finally, legal realists’ opposition to classical legal thought had a political as well as a jurisprudential motivation. Legal realists wished to advance their progressive politics by debunking the formalistic, apolitical justifications offered by classical legal jurists for their conservative decisions.

In rejecting classical legal thought, legal realism was part of a broader movement that included prominent judges such as Oliver Wendell Holmes, Jr., Louis Brandeis, and Benjamin Cardozo; the sociological school of Roscoe Pound;Footnote ¹⁸ legal historians such as Charles Beard;Footnote ¹⁹ and, later, the legal process school, the New Haven School, and the various ‘law and’ approaches prominent in American legal scholarship today.Footnote ²⁰ Holmes is undoubtedly the leading progenitor of legal realism, although he did not share its progressive politics. Both legal realism's anti-formalism and its interest in the normative, explanatory, and instrumental perspectives are already prominent in his work.Footnote ²¹ Indeed, one could say that, after Holmes, the rest is commentary.

American legal realists picked up where Holmes left off. Holmes planted the seeds of legal realism, but took a primarily doctrinal approach in his own judicial opinions and believed that legal rules determine most cases.Footnote ²² In their quest to study how the law really works, the legal realists displayed a deeper skepticism of legal reasoning, and thus represented a sharper departure from classical legal thought.

As Morton Horwitz explains, legal realists

In its most extreme form, rule skepticism denies the importance of legal reasoning – indeed, even its possibility. If legal doctrines are completely indeterminate, then they cannot be used either to decide new cases or to explain old ones. We must rely on morality and social policy in making decisions, and the social sciences in explaining legal outcomes – that is, we must rely on the normative and explanatory perspectives, not the doctrinal perspective. As Felix Cohen wrote,

Cohen's first question – How do judges actually decide cases? – led realists to empiricism. The goal of legal realists was to be able to predict how judges decide cases, through an understanding of the regularities that govern judicial decision-making.Footnote ²⁵ In contrast to formalists, who believe that judges make decisions based primarily on legal reasons, legal realists believed that ‘judges are (primarily) responsive to nonlegal reasons’,Footnote ²⁶ such as perceptions of fairness, the parties’ relative power, perhaps even what the judge ate for breakfast.Footnote ²⁷ This emphasis on the importance of non-legal factors in judicial decision-making is what Leiter calls the ‘Core Claim’ of legal realism.Footnote ²⁸

Cohen's second question – How ought judges to decide cases? – led realists towards the normative and instrumental perspectives, and the use of law to pursue progressive politics. If the existing legal rules do not determine the outcome of a case, then we need to focus on what the outcome ought to be, and how we can effectively achieve it.

To my mind, legal realism's answers to both of Cohen's questions are unexceptionable. What is problematic is Cohen's separate claim that these are the only two questions one can meaningfully ask about the law and that all else is nonsense – and the related claim that empirical regularities of judicial decision-making are the law, and that we can determine what the law says empirically. As usual, Holmes expressed this claim most pithily: ‘The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law.’Footnote ²⁹

As a philosophy of law, the prediction theory has many flaws. Most fundamentally, it denudes law of its normativity: the law's character as a standard, rather than simply a reflection, of behaviour. But if we do not equate empirical prediction with legal doctrine – if we understand prediction as addressing the explanatory rather than doctrinal perspective – then the legal realists’ interest in prediction is much more defensible. Explaining how judges decide cases – and, more generally, how the legal system actually works – is extremely important, not only for Holmes's ‘bad man’, who cares only about consequences, but also for the legal practitioner and the legal reformer, who wish to be able to influence the legal system effectively.

Moreover, as Shaffer notes, it is not clear the extent to which the ‘prediction theory of law’ was intended seriously by realists as a theory of law, or was put forward primarily for rhetorical effect.Footnote ³⁰ Most legal realists, although skeptical of legal doctrine, were not entirely dismissive of it. Doctrinal issues were not what interested them most; they were interested instead in cases where the rules run out and that must therefore be decided on the basis of extra-legal factors. But they accepted that, in many if not most cases, the law determines – or at least should determine – the result.Footnote ³¹

5. A legal realist approach to international environmental law

American legal realists focused on judicial decision-making, which plays a much smaller role in international than in domestic law. Nevertheless, as Shaffer's very interesting introductory essay to this issue demonstrates, legal realism's anti-formalist, pragmatic perspective is highly relevant to the study of international law, as many international law scholars have recognized for some time.

In thinking about the implications of legal realism for international law generally, and international environmental law in particular, the starting point is to view international law instrumentally. International environmental law is a tool to address problems such as climate change, stratospheric ozone depletion, loss of biodiversity, and so forth. It is created by humans in order to serve human ends.Footnote ³²

Second, a legal realist approach to international law is empirical in orientation. Because international environmental law is only one among a number of tools one might use to address an environmental problem, it is important to understand how it works. What are its distinctive features? What are its strengths and weaknesses as compared to other approaches? Is it likely to be effective in addressing a particular problem? What might make it more effective? Are non-legal approaches possibly preferable? These questions do not have doctrinal answers. They require us to consider explanatory issues.

Third, to understand how international environmental law works, we need to focus on the processes by which it is developed, implemented, and enforced – the actors involved, the wellsprings of their behaviour, and the ways they interact. What is the role of private and sub-national actors in developing international environmental norms? How do norms change over time? What is the range of options for promoting compliance with them?

To illustrate the differences between a legal realist and a non-legal realist approach to international environmental law, consider the problem of climate change. A doctrinally-orientated scholar might focus on whether states have an obligation under international law to reduce emissions – for example, on the basis of the duty to prevent significant transboundary harm or the precautionary principle, or as a matter of human rights law. He or she might try to bring a case before the International Court of Justice, or have a professional society undertake a study on the principles of international law applicable to climate change.Footnote ³³

In contrast, a legal realist would ask: Does the duty to prevent transboundary pollution or the precautionary principle have much actual influence on state behaviour? Do states even use these norms as standards of conduct in evaluating each other's actions? Would an advisory opinion by the International Court of Justice change anything on the ground?

Legal realism – at least as I understand it – does not advance any theory to answer these questions, since they are empirical in nature. For this reason, legal realism is, in principle, compatible with the entire spectrum of explanatory theories, including rational choice, liberalism, constructivism, even international relations realism (contra Shaffer). The question for a legal realist is: Which theory (or combination of theories, since they are not necessarily incompatible) best describes the world? Personally, I agree with Shaffer that international relations realism does not do well empirically, but there is nothing in the concept of legal realism that necessarily excludes international relations realism. Whether international law is epiphenomenal or plays a causal role is a question of fact.

What legal realism does exclude are attempts to narrow the subject matter of international law to traditional doctrine. Much international environmental law scholarship, of course, is concerned with doctrine. The two leading treatises in the field are almost entirely doctrinal in their content.Footnote ³⁴ Although I do little doctrinal scholarship myself, doctrinal analysis is both interesting and significant, and I accept that doctrinal scholars are studying an important part of the elephant. But only a part. Even if we are concerned, like doctrinal scholars, with describing the operative norms or standards of behaviour relevant to international environmental problems, we need to look well beyond the traditional sources of international law. We need to consider private and public-private standard-setting activities, which are increasingly prominent in the environmental field. We need to consider non-binding instruments, such as the UN General Assembly resolution banning driftnet fishing or the decisions of conferences of the parties. We need to consider the activities of sub-national actors, such as provinces and cities. And we need to consider the sui generis approaches to compliance that international environmental regimes have developed, which aim not to determine state responsibility, but to make international environmental law more effective in the future.Footnote ³⁵

Which of these various phenomena are ‘legal’ in nature? The question would be important if we were litigating a case in court (although, even here, doctrine is often not decisive, as the legal realists demonstrated). But it is much less important when, as is usually the situation, we are outside the courtroom, in the messy, diffuse world of international environmental politics.

Moreover, even when we enter the world of doctrine, we need to understand its limits. The duty to prevent transboundary pollution, for example, has been called the ‘cornerstone’ of international environmental lawFootnote ³⁶ and has been recognized by the International Court of Justice as part of the corpus of international law relating to the environment.Footnote ³⁷ But transboundary pollution remains common and few states assert violations of customary international law in response. The observation by Oscar Schachter more than two decades ago still remains true today: ‘To say that a state has no right to injure the environment of another seems quixotic in the face of the great variety of transborder environmental harms that occur every day’.Footnote ³⁸

6. Is international legal realism new?

Shaffer claims that international legal realism is new. Although I am in broad sympathy with Shaffer's exposition of the New Legal Realism, I question its novelty. Shaffer's six core claims – that how international law obtains meaning, operates, and changes are empirical and pragmatic questions (Claims 1 and 2); that international law should be viewed in processual and transnational terms (Claims 3 and 4), that international law matters (albeit conditionally, depending upon the context) (Claim 5), and that it cannot be reduced to either logic or politics (Claim 6) – these propositions would have all been familiar to such seminal thinkers of the last several decades as Abram Chayes, Louis Henkin, Oscar Schachter, and Michael Reisman.

Consider, for example, Chayes’ classic study of the Cuban Missile Crisis, written forty years ago.Footnote ³⁹ First, it is strongly anti-formalist. He says of the question – ‘Was the quarantine of Cuba legal?’: ‘A question put in that form is bound to elicit over-generalized and useless answers. The object of a first-year law school education is to teach students not to ask such questions.’Footnote ⁴⁰

Second, Chayes’ approach is highly pragmatic. As Chayes says, international lawyers ‘must avoid the temptation to deal with very difficult political and moral issues as though they could be resolved by rather simple and very general legal imperatives’.Footnote ⁴¹ Instead, international lawyers need to ‘focus on the relation between law and the social and political facts out of which it grows’.Footnote ⁴²

Third, Chayes’ approach is empirical. It attempts to answer questions about the role of international law in international politics through an extended empirical study of a particular incident, the Cuban Missile Crisis.

Finally, Chayes’ approach focuses on the legal process rather than on legal rules. Chayes spends comparatively little time analysing the relevant international law doctrines. Instead, he provides an extended description and analysis of the interactions among the various actors involved, and how they used legal doctrines and institutions in fashioning a response to the crisis.

Similarly, Professor Michael Reisman, throughout his work, has emphasized that to describe the legal system accurately,

Here we have, in a nutshell, the key features of legal realism, both old and new: skepticism about the formal rules, a comprehensive view about what counts as law and the relevant actors, a focus on the legal process, and, most importantly, an emphasis on determining what's really going on.Footnote ⁴⁴

Shaffer recognizes that his approach has many precursors, but claims that there are two new features of the new international legal realism: its empiricism and its focus on transnational processes.Footnote ⁴⁵ These are, of course, both important strands in recent international law scholarship, but neither is unique to legal realism. Just as the empirical study of domestic law owed much more to law and society scholars such as Willard Hurst rather than to the legal realists, who did little empirical work themselves, the proliferation of empirical studies of international law have been undertaken largely by political scientists rather than international lawyers and are part of the larger trend towards the social scientific study of international law. Similarly, the focus on transnational legal processes is neither new (Philip Jessup lectured on transnational law in the 1950s and Henry Steiner and Detlev Vagts first published their casebook on Transnational Legal Problems in 1976) nor unique to legal realists. Given the growing importance of transnational legal processes, it is appropriate that legal realism focus more on transnational issues – just as it is appropriate that it focus more on emerging problems such as cyber warfare. But there is a difference between saying that legal realism should pay more attention to transnationalism and saying that transnationalism has changed legal realism in any fundamental way.

7. What legal realism contributes to the study of international law

Whether or not international legal realism is new, it is still provides a valuable perspective on international law. One value is its emphasis on empirical work – an emphasis not unique to legal realism, but one of its key features. In his essay, ‘Science Is Not Your Enemy’, Steven Pinker chastises the humanities for their (occasional) inattention to empirical evidence, and observes, ‘a more scientific mindset would realize that an empirical proposition demands empirical verification’.Footnote ⁴⁶ The same can be said of international law philosophers. Consider, for example, Hans Kelsen. Although he characterizes his approach as a ‘pure theory of law’, it is brimming with unsupported empirical assertions. For example:

• • ‘It is remarkable that of the two sanctions, reward and punishment, the latter plays a much more important role in social reality than the former.’Footnote ⁴⁷

• • ‘Legal orders usually contain rules according to which one of the two norms [that command mutually inconsistent behavior] is invalid or may be invalidated.’Footnote ⁴⁸

• • ‘The development of law from primitive beginnings to its present stage in the modern state displays . . . a tendency that is common to all legal orders. It is the tendency gradually and increasingly to prohibit the use of physical force from man to man. . . .’Footnote ⁴⁹

These propositions may be true, but they are not necessarily so. Their truth is a matter of fact, so empirical study is crucial.

Legal realism is also a salutary reminder that international law is not a ‘brooding omnipresence in the sky’, in Holmes's marvelous phrase – a position that (almost?) no one actually espouses, but that, I think, still lurks in the legal and popular unconscious (and leads social scientists to be skeptical of legal scholarship).

It serves as a useful counterpoint to a new variety of formalism, which continues to resist the social scientific study of law. Jan Klabber's critique of the use of game theory to understand treaty law provides a good illustration.Footnote ⁵⁰ Klabbers argues that ‘applying game theory to the law of treaties is . . . a politically retrograde step, perpetuating the idea that international law serves the (narrowly defined) interests of states and states alone’.Footnote ⁵¹ But whether the rules of treaty law serve the interests of states and states alone is an empirical question, which cannot be answered through conceptual analysis. Moreover, if the rules of treaty law do reflect state interests, this is important to know, not only for sovereigntists, who think international law should serve state interests, but also for those, like Klabbers, who view this approach as retrograde.

Similarly, legal realism has a huge amount to offer for sources theory, which continues to be dominated by the ‘official story’ reflected in the ICJ Statute, namely, that there are three sources of international law (treaties, custom, and general principles), and that judicial opinions and scholarly writing are only evidence, not sources, of international law. That this approach does not reflect reality was recognized by no less a figure than Robert Jennings, the former president of the ICJ, who once observed,

But this critique notwithstanding, sources theory largely continues along traditional lines, rather than trying to determine how states and other international actors actually identify international legal rules – the sources and tests of validity that they, in fact, accept and use.Footnote ⁵³

8. What legal realism may leave out

Realism has an intuitive appeal. We all want to be realistic, don't we? But the appeal of realism depends in part on what it is contrasted with. If the opposite of realism is delusion, then realism looks good. We want to base our actions on the world as it is, not the world as we’d like it to be. But if we contrast realism with idealism, then the choice is not so clear. On the one hand, there is a virtue in being realistic about what one can accomplish; that is the essence of pragmatism. On the other hand, there is a danger that pragmatism can devolve into an apology for the status quo.Footnote ⁵⁴

I think this is at the root of much of the disquiet with legal realism: namely, that it has the effect of ‘disenchanting’ the law, to use Weber's famous phrase. Legal realism denies ‘law's unity, ahistorical essence, or transcendental meaning’Footnote ⁵⁵ – its utopian, universalizing potential. Instead, legal realism focuses on law's ‘instrumentality and functionality, its secularism, its pragmatism, its historicity, its dubious moral grounds, its fragmentary nature, and its lack of transcendental meaning or essence’.Footnote ⁵⁶

Like others who wish to re-enchant the law, critics of legal realism don't see law ‘“merely” as a social instrument, a political compromise, a dominating structure, a historical contingency, or a cultural artifact’. Instead, they retain a vision of law as committed to ‘transcendental wisdom, truth, and values’.Footnote ⁵⁷ On this view, international lawyers have a duty, as lawyers, to promote the rule of international law, rather than simply effectuate whatever states wish.

Legal realists may think such a vision is misguided, even delusional. And they may be right. But, as with religion, the question remains, Is the enchanted view of law a good or a bad delusion, a helpful or an unhelpful one? If we take the pragmatic, instrumental strain in legal realism seriously, that is the ultimate test.

American legal realists of the 1930s tended to be optimists. They believed that law can be a ‘robust, expansive, problem-solving enterprise’, and that ‘by breaking the shackles of formalism, they would liberate it, unleash the power of the law to solve problems’.Footnote ⁵⁸ Although Shaffer is silent in his introductory essay to this issue about whether the new international legal realism has a political orientation, his other scholarship (as well as his example of the role empirical work plays in uncovering bias) suggest that he conceives of it as having the same progressive politics as the original legal realists.Footnote ⁵⁹ Indeed, I wonder whether part of his agenda is to rescue legal realism from the conservative slant of much of its progeny, such as law and economics scholarship, rational choice theory and, within international legal scholarship, the equation of realism with a focus on the practice of the most powerful states.

But what if the power of law to effectuate change depends, not so much on unmasking its realities, but on a belief in its ‘majesty’ – on our ‘reverence or veneration’ for the rule of law?Footnote ⁶⁰ What if ‘the majesty of the law is not an obsolete irrationality but the prime feature of the phenomenon of law’?Footnote ⁶¹ Then, as Harvey Mansfield argues, the legal realists, by seeking to disenchant the law, would be the ones who are out of touch with how law really operates.Footnote ⁶²

In any event, we need not go as far as Mansfield to feel, like Weber, that something is lost in the disenchantment of the world and its replacement by the ‘iron cage’ of rationality. One casualty is the internal, doctrinal point of view, which seeks to determine the content of international law, based on legal rules of validity (what Kelsen called the Grundnorm and Hart characterized as the rule of recognition). The fear of critics is that the New Legal Realism, by de-emphasizing the internal point of view and the concept of legal validity, deprives international law of the very features that make it a distinctive enterprise, and that allows it to serve as a check on power.Footnote ⁶³ That is why Klabbers, in response to calls that international lawyers adopt the tools of social science to better understand how international law works, urges international lawyers to ‘jealously guard the relative autonomy of their discipline’ and to ‘be ready to defend its values and its modesty, its purity, if you will’.Footnote ⁶⁴

Ironically, this move to a new formalism by European scholars such as Koskenniemi and Klabbers, although diametrically opposed to the New Legal Realism in many respects, shares its progressive politics. The difference hinges on the question: Is law's formality a progressive force, by serving to curb the (mis)use of law as an instrument of power, or is it a conservative force, because it turns a blind eye to questions of social justice. If the former, the task is to defend law's formality; if the latter, to debunk it.