Two analogies lie at the core of Professor Waldron's article. The first is the claim that the standard analogy by which the state in international law is like the individual in domestic law is misleading; the state in international law is more like a government agency in domestic law. The second is that international law is (or is like) a species of public law and should be treated as such by domestic legal systems. I examine both claims, arguing (a) that even if we accept the first analogy it does not get us to the deeper levels of respect and commitment to international law that Waldron argues for, and (b) that the ‘floating normativity’ inherent in the second claim leads Waldron to overlook the specific organizational and structural conditions of international law. This leaves Waldron's position weakest where it should have most to offer: namely, in instances where our commitment to international law on one hand and the rule of law on the other seem to pull in opposite directions.
Like reaching for that extra piece of chocolate, we can’t resist resorting to analogies when it comes to thinking about international law. We know we shouldn’t, but we just can’t help ourselves. ‘Analogies to domestic law are impermissible’, Joseph Weiler once remarked, ‘though most of us are habitual sinners in this respect.’1 What's problematic about analogies is not so much that the act of being compared to something else heightens the risk of misunderstanding or devaluing the object being described. Rather it is that when pressed into the service of normative theory analogies have a propensity to short-cut argument. Such-and-such is like so-and-so; therefore such-and-such should be treated like so-and-so. There's nothing necessarily objectionable about the first step. It's the next step that needs to be treated with real caution. There may be plenty of very good reasons, even if such-and-such is like so-and-so, not to treat the two in the same way. It depends.
Jeremy Waldron's article, ‘Are Sovereigns Entitled to the Benefit of the International Rule of Law?’, is in a sense misnamed. The article is uncommonly rich and, while it does tackle near the end the question the title poses, at its core lie two analogies, one immediately apparent, the other less so. The first is the claim that the analogy by which the state in international law is like the individual in domestic law is misleading; the state in international law is more like a government agency in domestic law. The second – and ultimately more far-reaching – claim is that international law is (or is like) a species of public law and should be treated as such within domestic legal systems. I interrogate both these arguments in what follows.
Analogy 1: The State is Like an Agency
Waldron argues against the tendency to think in terms of an individual model when it comes to conceptualizing the role of the state in international law. The individual model holds that ‘[a]s individual humans are the subjects of domestic law, so nation-states are the individual subjects’ of international law (at 328). He cites Hobbes here, and indeed the tendency has deep roots. Carl Schmitt noted how personification was central to the conceptual construction of international law in the Baroque Age.2 ‘These states were conceived of as magni homines. In human fantasy, they actually were sovereign persons, because they were the representative sovereigns of human persons, of the agents of old and newly crowned heads, of kings and princes not precisely specified.’3 After the Peace of Westphalia, the practice of political relations was patterned in terms of such constructions. All significant authors, Schmitt argued, from Hobbes to Leibniz to Kant, have claimed that in international law states live as ‘moral persons’ in a state of nature, confronting one another as sovereign persons with equal legitimacy and equal rights.4
Against this model of personification, of individual sovereign agency, Waldron posits an analogy that has its origins in the more prosaic (though still, in another sense, baroque) world of the modern administrative state. Building on Abram Chayes's insight that if states are the ‘subjects’ of international law they are so as government bodies are the ‘subjects’ of public law,5 and out of Edward Rubin's analysis of the rule of law in the context of legislature–agency relations,6 Waldron identifies some of the contours of the international rule of law based on an agency model.
One of Waldron's objections to the individual model is that it inclines us (or at least the unwary) to see states as the subjects of international law, as opposed to agents who operate within an international legal order and who have important roles to play in that order, not least in the creation of new international law. We can avoid this error if we think of the state as operating in international law rather in the manner of an agency or similar (subordinate) governmental body. It is almost an axiom of public law that such agencies have no rights, but rather bear duties and responsibilities.7 Certainly, they have no freedom to operate outside the law or to uncover loopholes within the law that they might exploit.8 By extension, then, we might argue (a) that states do not have an inherent interest in freedom of action at the international level. An insistence on ‘the regularity and law-bound character of state action is undiminished’ (at 341) at this level, since the state ‘remains a governmental entity whose dangerousness continues to generate’ rule of law concerns (at 342).9
In addition, we might say (b) that the state as an agent of international law creation and enforcement comes under something like a positive duty to act in furtherance of the international rule of law. International law is to be internalized within the institutions of the state in a particular way. The fact that so often in international law states act as norm creators and enforcers serves even to heighten the obligation they have to take international law seriously. States should not take a purely instrumental view of international law, paying lip service to it if in their interest, ignoring it if not. Just as we want individuals to think about the (domestic) law they are charged with obeying (at 337–338), so too do we want governments to consider international law as an ‘independent vector’ in their decision-making (at 330). These requirements apply most directly to government lawyers, a point made more forcefully in the J’accuse! passage of an earlier article on a similar theme, where Bush Administration lawyers were excoriated for traducing international law in pursuit of the ‘war on terror’. Governments are bound ‘to show themselves devoted to the principle of legality in all of their dealings’, Waldron wrote, and ‘legal advice should be given in a spirit that embraces the importance of the international legal order and the obligatory character of its provisions’.10
I respect Waldron's objectives, deeply humane as they are. And he shares Walter Benjamin's conviction that ‘in the ruler, the supreme creature, the beast can re-emerge with unsuspected power’.11 The new analogy certainly seems preferable to the old.12 But I am uncomfortable with it. As with most analogies, it obscures almost as much as it illuminates. I worry that it may lead us to overlook what is distinctive about international law. It also has the tendency to short-cut normative arguments, including over fundamental questions like the sovereign equality of states.13 (Although Waldron might make the opposite point – that the old analogy in fact worked in just this way, and to the detriment of global justice.) Moreover the agency analogy, especially when combined with a strong Kantian conception of international law in which the interests of individuals rather than states take centre stage (at 325), has the potential to reduce the state to a juridical shell, and a subordinate one at that, in which its citizens just happen to have their abode. Not only would that position misrepresent the current state of development. International law still seems best understood as serving primarily the justice of order rather than constituting an order of justice. It is also not clear that it is normatively attractive. Recent events in Iraq and Afghanistan have reinforced the basic (Hobbesian) insight on which the international system rests, uncomfortable though it is for liberals of a certain stripe, that weak and failed states are ultimately more of a danger than rogue ones.14 In this world, security (hence order) is prior to justice.15
Waldron does not go down this route in his article, and in any case it is not the point that I wish to press. Let's assume that Waldron's agency analogy is the right one. I have no problem in accepting proposition (a). That governments (and hence states) are creatures of law and thus have no inherent and, as it were, extra-legal freedom of action sounds like Public Law 101 to me. Even Hobbes, who makes frequent appearances in Waldron's article as the representative theorist of unbridled sovereign authority, did not deny that the state was a thoroughly juridical (artificial) construct. I am less sure, though, about proposition (b). Not that it is a misstep, but rather that I am not quite sure that it serves to ground all the duties on the state that Waldron envisages. A complete disavowal of the idea of personification of the state, combined with strong Kantian talk of the rights and well-being of individuals being something like the telos of international law (at 325), decapitates (or at least bridles) Leviathan. But it leaves little conceptual space for talking about the interest of a state (or political community). But we talk about the state having, say, an interest in seeing a particular state of affairs realized at the international level. This talk is not nonsensical. And Waldron certainly doesn’t deny that instrumental behaviour – action to achieve particular ends – is at least in principle perfectly compatible with the rule of law.16
So let's say that the state can have an interest. Just like the agency in administrative law. Let's now return to Waldron's agency analogy. We can think of the dynamics of administrative law in terms of a game. The legislature, let's assume, sets the objectives or goals of the game. Agencies are players within the game so constructed. Courts (and other regulatory bodies) act as referees. Standing back, we can see that the level of commitment to the rule of law varies according to the actor's position in the game. We do not generally expect the agency to have precisely the same attitude to the law as, say, a judge. Certainly, we expect (and indeed require) an agency to operate within the framework of law. But, as a player, we also allow that considerations of policy and effectiveness are paramount in its decision-making.17 So we allow the agency some leeway when it comes to interpreting the law to suit its own interest (more correctly, in its interpretation of what the policy goals selected for it require). That's the point of legal doctrines like Chevron in US administrative law, which prevents judges from interfering with anything other than an unreasonable agency interpretation of the relevant statute.18 A similar point was made by the UK House of Lords about the application of the Human Rights Act (itself an example of the domestication of international law). Public officials, the Law Lords ruled, are not judges, and so are not required to work with law in the same way that judges are.19 As Lord Hoffmann pithily expressed it, public authorities cannot be expected to take decisions with ‘textbooks on human rights law at their elbows’.20
Leaving aside our scruples about arguing from analogy, even if we assume that the agency analogy is the right one, it is not clear that it gets Waldron all the way he wants to go. It certainly gets him some of the way. It might support, for instance, the relatively weak but not inconsequential proposition that states in international law, like agencies in domestic public law, have an obligation to work within the law. It might also ground a duty not to frustrate the purposes of the law.21 But the analogy by itself does not provide a basis for deeper levels of respect and commitment to international law. On the agency model, government agencies are players within the administrative law game, not referees. They therefore have a different relationship to law from referees, and so fall under different obligations. Hobbes compared laws to hedges, ‘set, not to stop Travellers, but to keep them in the way’.22 We might, say, then, that the obligation of the state at international law, if it resembles that of an agency in public law, is to refrain from bypassing the hedge of international law or from trying to hack a path through it. On the analogy, it is not its job to maintain it and keep it fit for purpose.
Analogy 2: International Law is Like Public Law
Let us turn now to the second analogy – if that indeed is what it is. Underpinning the agency analogy that we’ve just explored lies a conception of international law as public law. According to this approach, international law either is, or more likely should be treated like, (domestic) public law. The approach works best, I suspect, in relation to the application of international law in domestic legal contexts. But as this is an increasingly important part of international law it is not to be scorned.23 The theory is not clearly articulated in the article, but operates as something like a partially articulated premise of the main arguments within it. It also coheres with what Waldron has written elsewhere on the topic, particularly his article on ‘The Rule of International Law’ which argues that a duty rests on governmental actors everywhere to act in furtherance of international law,24 and his article on foreign law and ius gentium which articulates a vision of a common law of nations that connects with but is not subsumed by international law.25
An internationalist by persuasion, I have little truck with Scalia-esque26 (or for that matter Little Englander27) parochialism when it comes to looking abroad for normative inspiration. And the attempt to think seriously about the relationship between (domestic) public law and international law is one of the most pressing juridical challenges of our time. That said, I remain unconvinced about aspects of Waldron's argument on this point. It is particularly striking that the argument here is weakest on precisely the grounds where Waldron is typically at his best. One of the virtues of Waldron's legal theory is the consistent attention it pays not so much to the substantive aspects of law (to use his own terminology) but more to its formal and procedural aspects.28 He is usually very good on law's institutional arrangements and manifestations. But such aspects come but dimly lit here. What we are left with instead is the sort of theory that Waldron himself is (rightly) critical of in other contexts – a kind of floating normativity29 whose imprecations for us to be better are not always matched by the quality of their legal and institutional analysis.30
I don’t think Waldron and I differ on the public law essentials. Law is a technique of government that tends to come patterned with a set of (not always particularly well defined) values that are seen as inherent to its operation. Even Hobbes accepted this. For him, the exercise of sovereign authority (through civil law) is accompanied by an internal code (through natural law) by means of which the law is interpreted and applied.31 In its application, civil law is interpreted and applied through what we might call (although Hobbes didn’t) canons of legality – including, for instance, the injunction that the judge must always suppose that the legislator intends to act in accordance with Equity.32 Natural law wraps round and insinuates itself within positive law. Perhaps this is what Hobbes meant when he said that natural law and civil law ‘contain each other, and are of equall extent’.33 Positive law is the Sovereign’s, as are the courts, but the strictures of legality are not, at least not in any straightforward sense.34 We can express the same point sociologically (and historically). Lawyers, often in combination with other elite groups, when confronted with exercises of legal authority, tend to press for means through which to channel, structure, and control it.35 This is what helps drive the mechanism that Waldron describes as the rule of law and it is why, even in less than perfect political and social contexts, someone like E.P. Thompson might feel fit to describe it as an unqualified human good.36
Now, I don’t want to argue that the rule of law can only attach (say) to legislation. I agree with Waldron's criticism, expressed in another paper,37 of the crude positivism of the kind expressed by Justice Holmes in Taxicab, which denies the very possibility of ‘a transcendental body of law outside of any particular State’.38 My point is that Waldron makes precisely the opposite mistake that he identifies in Holmes. Holmes in Taxicab was all positivity and no normativity, all authority and no principle. Waldron here is almost all normativity and very little positivity. The usual Waldron hallmarks are missing. Where is the attention to form? Where the focus on procedures?
Let us bring the discussion down a couple of theoretical notches. I want to close by identifying two areas in which the floating normativity that characterizes Waldron's argument here appear most jarring when set alongside his work in public law theory. We can identify, first, the strange absence of legislation. Waldron is our great theorist of legislation. Over the years, he has persistently taken to task those for whom legislation is the ‘dirty little secret’ of legal scholarship.39 He has worked hard to make us see it as a ‘dignified mode of governance and a respectable source of law’,40 and has denied what others see as the antipathy between legislation and the rule of law.41 Waldron deserves credit for insisting against the crowd that we should take legislation seriously as the central act of modern law-making. It is no good just focusing on the ius of law, we must also pay attention to lex, and unpack the ius-like qualities of lex.
But when it comes to analysing the operation of the rule of law and international law in this article there is no mention of legislation. It is not as though legislation or something like it – analogies again! – doesn’t exist in the international sphere. The UN Security Council has been described as a ‘world legislator’,42 or at least as ‘having entered its legislative phase’.43 Waldron might have brought his unparalleled sensitivity towards the legislative context to bear on this claim, and might have contributed to the debate about whether (and if so how) the Security Council might be made subject to the international rule of law.44 Instead, there is silence.
This point goes further than simple disappointment that a scholar so invested in legislation as a legal form should make such an omission. It raises questions about the value of the theory itself. For recent actions of the Security Council have thrown up a particularly tricky dilemma for those – like Waldron, like most of us – for whom both international law and the rule of law are important. We might call this the ‘Kadi problem’, after the European Court of Justice case of that name.45 The problem arises as a result of a series of Security Council Resolutions that set up lists targeting individuals suspected of involvement with terrorism. These Resolutions require states to freeze the assets and deny employment, educational, and travel rights to those listed.46 Accompanying procedural protections are very minimal. The question raised when individuals targeted by these lists take their case to a national (or, as in the ECJ's case, supranational) court is this: what do you do (or what does the ‘international rule of law’ require) when the international law you are being asked to apply47 (indeed you are obliged to apply it) is less rule of law compliant than you would like or that your own domestic public law principles would dictate? A number of answers to this question have been essayed by various courts and commentators.48 My point, though, is not so much that Waldron has said nothing about this sort of rule of law problem, one that is likely to occur with greater frequency as international law grows in importance and penetration.49 It is rather that his theory does not even have the potential to say anything about it. Does the agency analogy mean that states are obliged as though by a superior legal authority to act the way the Security Council prescribes? Or does the argumentative and public law character of international law mean that states must internalize Security Council resolutions in a way that accords its own understanding of what the rule of law requires? There is no way of knowing. Either works; and neither.
A second silence within the article is perhaps even more glaring. Alongside his theory of legislation, Waldron's reputation in public law circles rests on his articulation of a rights-based case against judicial review.50 The limited role for the courts in his theory is a function of his defence of a theory of legislation as a theory of democratic self-government.51 But an implication of an approach to international law that sees it as akin to domestic public law, and that brings with it a duty on officials to further the international rule of law, must surely be an increase in adjudicative institutions involved in international law decision-making. Such institutions are needed to uphold the international rule of law by making sure other actors respect it. This theme is not discussed in the article, but it should be. For Waldron's argument seems not only to predict but to welcome a proliferation of international and transnational courts and tribunals (and panels, etc.) and an increased workload for existing institutions of this sort. It also seems to involve an increase in the role that domestic courts play in articulating and enforcing international law. Not just courts, of course. Legislatures and government officials have significant roles to play too. I am certainly not going to fall into the trap of establishing a binary opposition between legislatures as creatures of interest (‘bad’) and courts as forums of principle (‘good’).52 But to bridle Leviathan in the way that Waldron would like requires not just floating normative prescriptions, but institutions and pathways with which to enforce them. And domestic courts clearly have a significant role to play in this.53
Now, this is at one with global trends towards judicialization.54 But what does it do to the dignity of legislation? An increase in judicial review, particularly at the domestic level, must involve at least some transfer of decision-making authority away from legislatures to courts. Waldron argues against such transfers elsewhere. Law (and rights) are subject, he says, to disagreement all the way down. Legislatures are therefore the most suitable place to work through such disagreements. Judicial review does not do away with disagreement but simply repackages it in another form. To favour courts over legislatures is therefore tantamount to stealing decision-making authority away from the most legitimate institution, the legislature, and giving it to ‘a few black-robed celebrities’,55 with the effect of sapping a culture of active citizenship. If we believe what Waldron says about law and legislation elsewhere, then it is odd that the sort of transfers of competence that he denounces in domestic public law should be welcomed at the international level. Perhaps Waldron sees some significant difference between the two. But that is not the thrust of his article, which (rightly) seeks consonance and overlap between international law and public law. And the fact that the two are now often entwined cannot be denied. As James Crawford observes, ‘it is not too much to say that in many of these areas the role of international law is to reinforce, and on occasions to institute, the rule of law internally’.56 Why, then, should public law have a different structural operation where international law is present as compared to when it isn’t? If the dignity of legislation and the culture of active citizenship to which it relates are compromised by the operation of substantial judicial constraints at the domestic level, then why is the same not true when those same courts are applying international law?
I have made clear my reservations about some of the main lines of argument in the article, and have repeated the familiar objection about attaching too much weight to domestic analogies when thinking about international law. But these critical observations should not, I hope, obscure but rather serve as a testament to a rich article that invites us to think afresh about the rule of law in its various iterations with international law. It raises deep questions about structures of legality, and in particular about the way that we think about domestic public law structures in an era marked by the interpenetration of international and domestic law.