Hierarchy in International Law: A Sketch

Martti Koskenniemi¹

Full text available: PDF format ^*

Geometry is the archetype of modern mind. The grid is its ruling trope ... Taxonomy, classification, inventory, catalogue and statistics are paramount strategies of modern practice. Modern mastery is the power to divide, classify and allocate - in thought, in practice, in the practice of thought and in the thought of practice. Paradoxically, it is for this reason that ambivalence is the main affliction of modernity and the most worrying of its concerns.²

Legal reason is a hierarchical form of reason, establishing relationships of inferiority and superiority between units and levels of legal discourse.³ Sometimes law's hierarchical character is elaborated as an essential aspect thereof. Classical and modern naturalisms, for instance, often conceptualize the law in terms of systemic derivations that assume the existence of relationships of entailment between normative units and levels. St Thomas Aquinas posits a many-layered structure of behavioural norms deriving from other norms, existing at progressively higher hierarchical levels.⁴ The naturalism of an argument in this respect (i.e., in respect of superior norms), however, is not dependent on its being an explicit part of naturalist theory. The concept of `fundamental', used in human rights law, as well as the ideas of jus cogens or imperative norms and rules valid in an erga omnes way each presuppose relationships of normative hierarchy that implicate some form of moral naturalism.⁵

Law's hierarchical character is by no means, however, only a naturalist credo. It is shared equally by its two main contestants, formalism and the social concept of law. The best-known example of the former is, of course, Hans Kelsen's Pure Theory of Law. For Kelsen, what is specific to legal norms is that they enjoy a `validity' (in contrast to moral goodness or sociological effectiveness), which they receive by delegation from norms assumed to exist (or to be valid) at hierarchically higher levels. These latter norms, again, receive their validity in a similar way from norms at even higher levels ... and so on until we reach the basic norm whose validity can no longer be derived from normative delegation, but is a transcendental (or perhaps cultural) presupposition that must be made in order for what we know of the validity of other legal norms to be true.⁶

Social concepts of law employ less articulate conceptions of hierarchy.⁷ They build upon a priority of a sociological (often economic, but equally psychological or biological) base to a normative (legal, moral, institutional) superstructure. A basic level, often called `reality' (or the `will of the sovereign' or `basic social needs'), is installed in a hierarchically controlling position vis-à-vis other, ephemeral aspects, such as law. Frequently, the relationship is portrayed in an instrumental way. The determining base has recourse to the superstructural element in order to better fulfil its internal dynamism. If turning Hegelian hierarchies on their head was the Marxian battle-call for social revolution, the same turn and the same hierarchy is used by neoliberal politics to buttress the foundational position of the market for social reconstruction. The language of `levels', `reflexions' and `determination' betrays a rigorously hierarchical structure of thought behind the social conception.

Whatever difference there is between naturalism, formalism and realism, each portrays law in a hierarchical light. While the systemic aspects of hierarchy are highly elaborated in formalism, but less so in naturalism or realism, each understands the law as a working out, or a making express, of normative superior/inferior relations, conceptualized in terms of what is good (right) and what is bad (wrong), what is valid and what is not, or what works and what does not.⁸

* * *

In this paper I shall draw a rapid sketch of the role of hierarchy in international legal thought and practice. My argument is developed in three parts. The first part sets out a (non-exhaustive) typology of hierarchies employed by jurists when dealing with international law topics. The second part presents a (deconstructive) challenge to the typology and the modes of juristic discourse operating it by arguing that each superior/inferior relationship can always be reversed to produce its contrary by the use of impeccable legal argument. The third part looks beyond deconstruction and outlines a way of thinking about international law that respects the principle of reversibility but sees law as nonetheless a meaningful social practice.

One caveat is in order. The following presentation is a violent schematization. Within the confines of this paper, it is not possible to deal with the very large number of argumentative strategies used by lawyers to constantly challenge and reestablish threatened hierarchies, deal with hard or boundary cases in innovative ways and break the rules of the game so as to ensure its continued, subtle success. Also, it may be possible to outline many more modes, theories, or argumentative and hierarchical structures than I have been able to canvas. The scheme has no pretension of being a grand theory - not even `the' postmodern theory - of international law. I have simply tried to provide a scaffolding, as it were, or a summary, of certain critiques that have been presented and responses given in an attempt to `reconceive' or `reimagine' international law since neither the ideal of a stable and structured legal reason nor the prevalent intuitionist pragmatism seem able to reflect upon themselves without embarrassment.

I

Hierarchical relationships are employed in each of the three modes of juristic discourse, distinguished from the more concrete towards the more abstract as the modes of control, exegesis and philosophy.

The mode of control is directed at behaviour, seeking to describe it by reference to a binary normative code of `lawful'/'unlawful' in all its innumerable transformations (right/duty; conformity/violation; permitted/prohibited).⁹ The mode of control appears in at least three different contexts of juristic practice.

In dispute-settlement, parties present contrasting and normally incompatible theses about whether a particular behaviour should be seen as `permitted' or `prohibited'. It is the judge's function to fix the place of the contested behaviour within the relation. Apart from a substantive qualification, dispute-settlement always also involves the reaffirmation of the hierarchical position of the judge, or the third party vis-à-vis the litigants, or public power vis-à-vis political passion.

In diplomatic discourse as well, contestants seek to qualify behaviour in binary terms. However, unlike in dispute-settlement, no authoritative resolution need necessarily be forthcoming. On the contrary, diplomatic discourse often leaves the final determination suspended and instead moves for a procedural reconciliation. Whether through agreement or procedural displacement, the process involves a re-entrenchment of the authority of the disputants over the subject-matter, or sovereignty over law.

Legislative discourse seeks to control future behaviour by producing generic statements about the lawfulness (illegality) of particular types of behaviour. Ideal hierarchies (good/bad) inform legislators' moral beliefs and institutional objectives. They also provide the basis from which legislative hierarchies are inferred, which in turn appear as more or less successful reproductions of the former in terms of legal permissions or prohibitions, law over sovereignty.

The mode of control is directed towards the qualification of social behaviour. Two types of hierarchy are implicated. One is a hierarchy between forms of behaviour, the other is a hierarchy of controlling authorities. The two hierarchies are ambiguously linked. On the one hand, controlling authority is itself assumed to be controlled by requirements of substantive legality; on the other hand, substantive legality is derived from controlling authority. The relationship between the two has been a perpetual puzzle within the mode of control: does law defer to authority or vice versa?

Hierarchical thought is not exclusively directed at behaviour but equally at the law itself. Employing the mode of legal exegesis, lawyers grapple with problems of the content of the law and its systematization.¹⁰ The outcome of exegesis is not the qualification of particular behaviour but the abstract elucidation of what norms say and how they are linked into systemic wholes. In interpretation, exegesis seeks to establish the better of two contested meanings of legal norms as linguistic (for example, treaty text) or behavioural units (for instance, customary behaviour). Systematization aims to make explicit the origin and relationships of norms so as to answer questions about normative authority and to solve problems of normative conflict.¹¹ International law exegesis breaks down into two doctrines that have their own problematique.

Sources doctrine grapples with the origin and binding force of international norms. It deals with treaties, customs and general principles of law, attempting to link ambiguous or contested meanings to something outside those sources themselves, most frequently to ideas about party consent and international justice, so as to resolve their meaning.

Sovereignty doctrine grapples with the opposition between effectiveness and legitimacy. It deals with sovereign rights, competence of international and national bodies, jurisdiction, territory and immunity, and hopes to justify a contested allocation of power either by reference to sociological realities or normative ideas about them.

Sources and sovereignty constitute opposing openings into the legal substance. Although they seem initially incompatible (law/power), exegetic manoeuvres constantly reduce them into each other. From the perspective of sources doctrine, `sovereignty' has no independent normative sense but is just a pale shadow of results received within it: jurisdiction (for example) is what treaties, customs or general principles say it is. From the perspective of sovereignty doctrines, again, the meaning of legal sources is merely to describe whatever sovereignty, in particular relationships, means and what sovereigns have decided. This structure provides a perpetual movement for exegesis, while ensuring that its horizon is never lifted from what it is that can be seen from its two alternative openings: a uniform and stable `law' and a uniform and stable `sovereignty'.

Finally, the mode of philosophy poses directly the question of the nature and possibility as well as the limits of the knowledge of the law. The mode of philosophy is expressed normally in terms of various theories that are often borrowed from outside the legal discipline and seeks to grapple with the ontological and epistemological issues of legal control and exegesis.

Theories of the nature of law seek to describe the law either in terms of social processes or in terms of normative standards or, briefly, `facts' and `ideas'. They appear not only as grand theories but as implications of the methodological or historical accounts of international law's growth as a social practice and a scientific discipline. They employ distinctions such as positivism/naturalism; formalism/realism; Western/Third World approaches.

Theories of the limits of law seek to establish the relationship between law and neighbouring fields of discourse, such as politics, history, diplomacy, literature, and the corresponding academic factions of sociology, political theory, anthropology. The relevant hierarchies are constituted from arguments about whether law is prior to or derived from such disciplines.

Theories about nature and limits are also interconnected. If law indeed seems like a matter of behavioural regularities, then sociology becomes its foundational discipline; if critique is what we wish to do, then law can only exist as a political Utopia.

The three modes are not independent from each other but normally set themselves again in hierarchical relationships. For example, a practitioner or a diplomat might consider control as prior and exegesis and (a fortiori) philosophy as secondary and (if at all necessary) perhaps in an instrumental relationship to it. A university professor might believe exegesis to be the superior mode; control would then become its derivation and philosophy its auxiliary. A legal theorist might view both exegesis and control as inferences (of first and second degree) from resolutions attained within the mode of philosophy. And so on.

^* The free viewer (Acrobat Reader) for PDF file is available at the Adobe Systems.

1 Professor of International Law, Department of Public Law, 00014, University of Helsinki, Finland.

2 Z. Bauman, Modernity and Ambivalence (1991), 15.

3 A unit (or topos) of legal discourse is any distinguishable entity of legal meaning; a level is a group of topoi identifiable by a common property. To speak of levels instead of groups is to highlight the hierarchical character of the principles of identification. Thus, regular treaty provisions constitute topoi that are normally situated at a level different from the topoi of legislative purpose or principle underlying the treaty. This paper deals with the reversibility of the hierarchical relationships between such levels.

4 Cf. H. Tolonen, Nature and Justification. Modes of Derivation According to the Aristotelian Tradition in Natural Law (in Finnish, with English Summary, 1984).

5 Recourse to Latin is an interesting compliment that modern scepticism makes to its naturalist heritage.

6 Cf. H. Kelsen, Reine Rechtslehre. Einleitung in die rechtswissenschaftliche Problematik (1934), at 66-67 (§ 29).

7 On `social concepts of law', cf. my `The Normative Force of Habit: International Custom and Social Theory', 1 FYBIL (1990), at 77 et seq.

8 Hierarchy, in other words, means difference in a normative light. There is no such thing as a non-normative hierarchy.

9 For a taxonomy of some such relations, cf. W. Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning (1978).

10 For these two functions of legal `doctrine', cf., e.g., A. Aarnio, Denkweisen der Rechtwissenschaft (1979).

11 Systematization takes place normally by abstracting general principles from individual rules or deriving rules from general principles. These two strategies reverse the hierarchical relationships between two levels: either rules are normatively determining and principles only descriptive consequences thereof or vice versa. Cf. my `General Principles: Reflexions on Constructivist Thought in International Law', XVIII Oikeustiede-Jurisprudentia (1985) 117.

© 1990-2004 European Journal of International Law
All comments and suggestions should be sent to webmaster
This site is part of the Academy of European Law online, a joint partnership of the Jean Monnet Center at NYU School of Law and the Academy of European Law at the European University Institute.
This file was last modified: Tuesday, October 14, 2003 12:43PM