Hostname: page-component-74656f9884-g96k6 Total loading time: 0 Render date: 2026-11-09T15:05:10.012Z Has data issue: false hasContentIssue false
  • English
  • Français

IN THE TRAP OF A LEGAL METAPHOR: INTERNATIONAL SOFT LAW

Published online by Cambridge University Press:  03 August 2010

László Blutman
László Blutman
Affiliation:
Professor of International and European Law, University of Szeged, Hungary. Email: blutman@juris.u-szeged.hu.
Rights & Permissions [Opens in a new window]

Abstract

International soft law as a legal metaphor has become part of legal discourse. The author argues not only that the term itself is misleading and contradictory, but that the terminological problem indicates deeper doctrinal difficulties which soft law theories should overcome. Identifying two main streams of these theories, the article provides a short review of the fundamental arguments for the existence of soft law. It distinguishes between two lines of arguments, ie functional arguments and proximity arguments, evaluates their coherence and explanatory power, and finds that these arguments fail to lay a coherent and persuasive foundation for international soft law. The author concludes by pointing out that non-legal norms may have essential regulatory functions in international relations but the law and non-law distinction should be maintained, and separating soft law as a distinct category is unwarranted.

Information

Type
Article
Information
International & Comparative Law Quarterly , Volume 59 , Issue 3 , July 2010 , pp. 605 - 624
Copyright
Copyright © 2010 British Institute of International and Comparative Law

I. INTRODUCTION

In the early 1980s Prosper Weil mounted a serious attack against the conception of international soft law in two of his compelling articles,Footnote 1 triggering a number of reactions. Nonetheless, the name and concept of soft law has since come into much more general use. This, too, reveals an immovable fact: One can hardly fight against linguistic conventions as these do not necessarily obey the rules of semantics and logic, but are evolving in everyday discourse in an ‘organic’ way. The meaning of such a catchphrased, however, can erode over a short period of time and become undefinable. Relatively identifiable meanings thus could only be loosely linked together, tending to make up a web (a semantic ‘network of similarities’) and their semantic elements will end up losing their common features.Footnote 2

This is what has happened by and large to the expression of ‘soft law’ in international law since Lord McNair introduced the idea into legal discourse.Footnote 3 According to Aust there is no agreement about what ‘soft law’ means or indeed if it really exists.Footnote 4 In the Encyclopedia Thürer has come to the conclusion that in legal terms the expression has no clearly determinable content.Footnote 5 Other authors are facing practically the same difficulties. Abandoning the possibility of a uniform and at the same time meaningful definition, the concept seems to describe at least two separable phenomena. On the one hand, soft law can be treated as norms which do not appear in the form of a legal source recognized by international law,Footnote 6 but are of some legal relevance. On the other hand, soft law can be regarded by many as norms that appear in the form of a formal legal source recognized by international law, but are not enforceable owing to their generality, vague normative content or subjective nature.Footnote 7 I therefore will treat these two groups of soft law separately, identifying the above firstly mentioned norms as ‘soft law (I)’ and the latter as ‘soft law (II)’.Footnote 8 (This duplication, too, may imply the phenomenon of ‘double soft law’ if a norm falling within soft law (I) is too general and vague.Footnote 9)

In the following, I will be addressing the phenomenon of international soft law (I).Footnote 10 I will argue that soft law (I) is not in itself defined, and it cannot be defended in the context of international law. Therefore, as a vague, redundant and unfounded notion, its use may not be justified and can hardly withstand Ockham's razor.Footnote 11 First, I will consider the most fundamental arguments for the existence of international soft law (I).Footnote 12 (Hereinafter, soft law (I) is labelled simply as soft law, while the other type of soft law remains soft law (II).)

II. IN THE FRONTIER LAND OF INTERNATIONAL LAW

A. Norms and Instruments with the ‘Soft Law’ Label

According to the studies addressing soft law, the norms and instruments of soft law can be broken down into at least three groups. This may not be the case for every one of them as some only examine one category, while others highlight the other. But all in all, the following division and the scope of rules belonging to each category seems defensible.Footnote 13 In the first group the non-binding decisions of international organizations and bodies are included, eg non-obligatory decisions relating to extra-organizational relations, guidelines, programmes, declarations, directives, opinions, plans of action, programmes of action and recommendations (of them mainly resolutions of the UN General Assembly are standing in the front of analysis in international legal studies).Footnote 14 Bilateral or multilateral, and in a legal sense, non-obligatory agreements and declarations created together by States belong to the next category. These can take many forms: communiqués, declarations of intent, joint final declarations, final acts of summits, non-binding political, technical and other understandings, as well as recommendations adopted by international conferences, etc.Footnote 15 The third category comprises recommendations passed by non-governmental organizations (NGOs) which can nevertheless be of considerable impact in terms of international practice (eg the International Committee of the Red Cross).Footnote 16 While in the first case it is mainly about one international legal entity's act, in the second case we can talk about common statements, promises and expectations made by several legal entities by agreement. In the third case, normative measures put forward by non-governmental legal entities (NGOs) are placed at the centre of attention.Footnote 17

The rules of soft law do not appear in formal legal sources recognized by international law, and in legal terms they are not obligatory. According to traditional views, these lacking features are the indispensable, though not exclusive, elements of an international legal rule. In the views of the proponents of soft law just the absence of these features makes norms ‘soft’ as opposed to traditional rules of hard law.Footnote 18 Nevertheless, soft law can contain classic normative rules of conduct as well, and similarly to law, they can also be part of the complex mechanism regulating the activities of international legal entities and may have a great impact on the conduct and activities of international actors, including States.

Does all this justify the distinguished category and name of soft law? As could be seen there are authors who acknowledge the existence of a soft law (II), nonetheless they do not consider soft law (I) to be soft law, (Weil, Baxter). In fact, soft law poses serious doctrinal problems.

B. Why is it Soft, if it is Law?

Those using the name of soft law must in any event face a fundamental conceptual problem. A basic question is whether soft law is a type or variant of law. The name in itself implies that soft law is a type or variant of law with albeit a slightly decreased or inferior value. That soft law is to be regarded as part of traditional international law is not considered by all the advocates of soft law to be true.Footnote 19 However, in claiming this one would not need to insert the adjective ‘soft’ before the word ‘law’, since insofar as it were qualified as law, then it could be considered as it were, without any qualification: why would one want to have an adjoining adjective? This might only be warranted if soft law placed within the category of law were used to mean legal norms which are ‘less legal’ than the ‘hard’ legal norms. For example, in Guzman's view soft law describes legal norms that fall short of the classical definition of law.Footnote 20 Nevertheless, if this classical definition of law is rejected, because allegedly unable to explain the phenomenon of the so-called soft law (though used as a reference point to grasp soft law), what would be the new definition of law comprising soft law? As no such definition is provided, it is not clear how to tell soft law having a legal nature from other, non-legal norms on the one hand, and it is hard to explain why soft law is legal in nature but at the same time less legal than other, traditional legal rules, on the other.

If a new, possible and acceptable definition for law comprising soft law existed or were introduced one would not need a desperate distinction between soft and hard law because the extended concept of law denoted by this (apparently non-existent) definition would explain both phenomena under uniform criteria provided by this definition. Under an extended definition of international law hard and soft legal rules are bound to lose their ‘hard’ and ‘soft’ adjectives referring to a meaningless distinction from the perspective of uniform criteria provided by this definition. Legal rules would simply be legal ones, and one of them cannot be more or less legal than another one. So, soft law as a term in itself indicates that such a new, extended definition for law does not exist, or at least has not been introduced into legal discourse; that is why many authors lacking this new, extended definition have to use the term and concept of soft law itself to oppose and reject the classical definition and criteria of international law. However, provided that soft law is not, or is not considered to be, a variant of law, why should it be called law with or without an adjective? This is contradictio in terminis,Footnote 21 that can be overlooked in legal discourse, but which, at least in theory, is unfounded.

Soft law could be qualified as legal metaphor somehow were it not rather paradoxical and misleading to do so. Arend's wording is sharp and concise, (though the problem is more shaded and complex):

I find the use of the term quite troubling. … If a rule meets the criteria for law, then it should be called ‘law’. If, however, the rule is not binding—as soft law has been described to be—then it should not have law anywhere in its name. If the rule is not obligatory, but is habitually ‘complied with’, a ‘rule of the game’ seems to be a much more appropriate term. Given, in particular, the disdain with which international law is held by some international relations scholars, the concept of soft law seems to serve no purpose except to confuse matters and dilute the distinctive nature of legal rules.Footnote 22

One part of the problem is thus of terminological character. Is this worth finding fault? For example, Sztucki is rather lenient towards the term. Even though he finds it quite incorrect, he considers that in indicating or denoting complex phenomena such all-embracing and indefinite ‘capsule formulas’ may be used. In addition, he says that in international law there are many imprecise and incorrect expressions, and soft law can only increase this number at the very best.Footnote 23 The real problem, however, is that the expression is not only incorrect but it is also meaningless; the phenomenon behind it is nonetheless heterogeneous, and thus it can hardly be suitable for carrying out its denotative functions. Almost every commentator dealing with soft law stumbles against the fact that an exact and meaningful definition of soft law borders on the impossible. Mörth correctly points out that the expression can hardly be fruitful, if virtually every non-legal norm may be described with this expression.Footnote 24 Senden thinks soft law is an ‘umbrella concept’ that describes different types of documents and rules having disparate characteristics, carrying out dissimilar fuctions.Footnote 25 Thus, if someone used this expression, (s)he would have to be equipped with a user's guide, too, in order to insure conceptual correctness to a certain extent. So, we have a term that denotes something vague, by which every author means different things, or even worse, does not mean any defined phenomenon leaving the term in its undetailed generality in common legal discourse.

III. NEW PASSAGES BETWEEN LAW AND NON-LAW?

A. Binary Approach and Regulatory Force of Non-Legal Norms

Terminological problems point to deeper conceptual and doctrinal problems. So, in this sense, the termonological aspect is not superficial, but it also leads further since, with regard to soft law, the very existence of international law and the system of international legal sources are at stake. May the boundaries of law be maintained and could the distinguishing criteria be defended according to which the norms existing in international relations can be split into legal and non-legal norms?Footnote 26 The consistent application of traditional principles relating to the sources of international law would require sharp and formal separation of law and ‘non-law’ (binary approach) as I have demonstrated previously considering terminological issues. The core of this approach built upon some principles and criteria is a simple one, nevertheless a general question arises: could this be maintained as opposed to the soft law phenomenon?

The ‘either law or non-law’ approach is not new, moreover, it is somewhat dull and out of fashion nowadays. However, followers of soft law have not, as yet, opposed an overall, coherent and sweeping legal theory to it.Footnote 27 In positivist terms, there appears to be no great chance of developing such a theory.Footnote 28 What's more, they themselves share more or less certain basic positivist principles since their base of comparison is the recognised system of formal legal sources, and together with it is the axiom of the obligatory nature of law.Footnote 29 However, soft law theories share a misconception about the traditional approach relating to its assessment of non-legal norms (including soft law). They falsely—in either an explicit or an implicit way—attribute to the binary approach the claim that legally non-binding soft law would influence less the actions of a state than a binding legal norm.Footnote 30 Such a claim cannot be found among the necessary premises of the binary approach. The fact that soft law is not a legal norm does not mean that political, moral or other requirements would not be for some reason more compelling than a legal rule. As opposed to the legal consequences arising from violation of a legal norm, there can be non-legal restraints that urge States in a more effective way (even more than the legal norms) to comply with non-legal rules.Footnote 31 This mainly sociological aspect cannot thus be a determinant difference between soft law and legal norms, let alone one cannot make inferences from this about the relationship of the two types of norms.

Still, related to this, rules under soft law should not be regarded as a means of diminished or inferior value when it comes to controlling and influencing the conduct of those making their way on an international level. Guzman notes correctly that these norms are not necessarily second-best solutions failing to live up to the ideal (legal regulation) or just intermediate steps towards some ideal legislation.Footnote 32 Lack of legal form does not necessarily mean some kind of ‘diminished function’; these norms may well be as valuable as legal norms from a sociological point of view. For example, in OECD practice it has been realized that soft control and soft measures (dialogue, social exclusion etc.) are more effective and desirable than legal enforcement.Footnote 33 But, if a social norm has important regulatory role, why should it be coined as soft law?

It is doubtful that these rules would require a legal form to carry out their regulatory functions efficiently. Yet scholarly reflections on soft law usually commit the mistake of attributing at all costs some kind of legal nature to these norms in order to explain their regulatory functions. The answer to the efficiency or inefficiency of these norms, however, is not, most of the time, to be found in the coordinates of their relationship to formal legal rules, but in the detailed sociological analyses of international relations. Mörth asks on a correct note: Is it by all means necessary to talk about soft law instead of calling it ‘soft regulation’ or ‘soft control’?Footnote 34

B. ‘Grey zone’ Theories

The binary approach of ‘either law or non-law’ is not normally refuted, but rather, eluded. Objections, however, are not really convincing. Distilling the essentials of these objections we confront the following claims.Footnote 35 The binary concept simplifies processes in international relations, thus it is not able to handle conveniently new developments in practice. Cleaving desperately to the recognized system of legal sources, it is not capable of grasping and describing the complex reality, and, as a result, it will draw unfounded conclusions. Norms functioning or existing in international relations cannot be split into legal and non-legal norms; the expectations, constraints and obligations created by them have a gradual nature. Differences between them are not so much of a qualitative character than of a quantitative one: a norm may relate more or less to law, or can simply be non-law. In this respect, there is continuity between simple political or moral norms and traditional legislation.Footnote 36

Defending soft law conceptually, but maintaining its viability as well, requires breaking down the boundaries of the system of legal sources to a certain extent, or at least blurring these boundary lines.Footnote 37 Therefore those theories of soft law which more or less assume the existence of an autonomous or even separated international law within international normativity, but refute the binary approach, seem to rest on a specific, common premise. Between international law and non-law (non-legal norms containing obligations of a purely political, moral or technical nature) there exist one or more layers of norms with definable characteristics having a special relationship to the international legal sphere—a kind of ‘grey zone’ representing soft law or different types of soft law.Footnote 38

Soft law theorists try to establish that no line can be set up dividing legal rules from other international norms, because soft law norms may sometimes function very effectively and are usually regarded as authoritative by international actors, and that the formalism of the binary approach cannot grasp the specifity of those norms which play important role in regulating the activities of international actors without wearing the label of ‘international law’.Footnote 39 In Boyle's view, the norms of soft law may have a fine and versatile relationship to international treaties (as international rules of law) and norms of soft law cannot simply be viewed as non-law as it is suggested by the binary approach.Footnote 40 As these theorists are unwilling to elevate all norms having some (weak or strong) legal relevance to the stature of law, but at the same time will not relegate such norms simply to the status of non-law, it is necessary for them to presuppose at least three huge zones of international normativity: non-law, soft law as a grey zone, and law.Footnote 41

The metaphor of ‘grey zone’Footnote 42 or ‘interim zone’ can even be attractive. Despite this, the theory of soft law faces significant hardships. It is constrained to give reasons why these rules, as soft law, have to do with law when they are not obligatory in legal terms and when they have not been created within the framework of the recognized system of legal sources. Furthermore, it must have an answer the other way round, too, namely, to what makes them different from mere political and moral commitments that have not already been, or are not yet, of a soft law nature. Moreover, it has to be answered: What rules apply to concluding and terminating certain soft law norms (agreements)? How do they cease to exist as soft law norms? What are the conditions of their validity and applicability?

As long as they wish to find an answer to the boundaries and criteria of the existence of soft law standing in the grey zone, authors basically and usually refer to two general factors, to which complex arguments can be traced back. First, that soft law is observed and many times referred to by States and other international actors, and so, it in fact regulates international practice and functions like international legal rules do (functional arguments). Second, that soft law norms have special and close links to the sphere of law, which lift them out of the sea of other non-legal norms (proximity arguments).

IV. INTERNATIONAL SOFT LAW: ARGUMENTS AND OBJECTIONS

A. Functional Arguments

The essence of the functional argument is that soft law norms function like formal legal rules in the processes of their application by international actors.Footnote 43 The argument that norms falling under the category of soft law are observed by States and other international actors referring to them and regarding them as a base to shape their practice plays an essential role in soft law theories. When a non-obligatory norm is not observed by the States or other international actors, we can hardly speak about soft law norm in any sense. To analyse this argument or its various forms would be well beyond the reach of this article. Nevertheless, it is useful to emphasize in a general manner that those who use the functional argument in soft law discourse have no satisfactory answer to at least three basic questions.

1. They fail adequately to address why it follows that States shall observe a (non-legal) international norm, from the fact that many other international actors in fact and in many respects follow this norm in their practice. Observing voluntarily and generally a non-obligatory norm is sociological fact,Footnote 44 and is not in itself suitable for endowing this norm with (legal) normative force, only in special circumstances (see the criteria for the existence of international customary law). This is the case the other way around: non-compliance with an international legal norm does not generally deprive the norm of its normative force, only in special circumstances (eg the problem of desuetudo). Thus the fact that a norm determines or affects the behaviour of States and other actors will not change the legal or non-legal status of a norm and create legal or quasi-legal obligations; eg a political or technical norm will not become soft law, let alone get inside the territory of law.

This aspect has another side: What are the consequences when one or more States do not observe these norms? In trying to find an answer to this question one confronts the fact that violation of a soft law norm may ensue political consequences, but can hardly have any legal consequences; in other words, the observing of a soft law norm can hardly make the boundary between law and non-law traversable.

2. The supporters of the functional argument fail to give a satisfactory account of the distinction between simple, non-legal norms and soft law, and of how a non-legal norm can become a soft law norm in practice. Even the conditions of this transformation are obscure.Footnote 45 It is often said that more or less general compliance with or a strong commitment to this norm on the part of international actors may provide a condition for the soft law norm. It still leaves lots of questions open and unanswered. How to measure such compliance and commitments? How many States should follow a non-legal norm to become soft law? Do other international actors count or just States? How does one know when a non-legal norm has been elevated to the level of soft law; is there any authority to decide on this question? All the uncertainties otherwise surrounding the formation of customary law at the boundary of non-legal custom and international customary law are multiplied at the boundary of non-legal norms and soft law.Footnote 46

With this non-formalistic, functional background, in the absence of criteria, embarrassing answers have been offered. Peters and Pagotto concluded that the search for distinguishing parameters proves futile claims that the ‘I know it when I see it’ ad hoc recognition technique will resolve the problem of distinction.Footnote 47 Though in the context of soft law and hard law distinction, in MA Fitzmaurice's or Mörth's view each and every rule or an instrument must be judged on a case-by-case basis since no principles of distinction can be identified.Footnote 48 This leads to the third problem that the functional arguments cannot adequately address, namely the normative (legal) uncertainty.

3. For many authors the discourse on soft law is formulated in terms of formalism as being something bad and unable to describe real world realities against functionalism as being flexible and the only correct explanation for the practice in international relations. The conceptions of bottom-up lawmaking (eg Levit) or the inductive approach (eg Peters and Pagotto) supporting the concept of soft law are based on such functionalism. However, this strong antiformalistic tendency dissolving and deconstructing existing, positivist criteria in the name of adequate explanation of reality leaves functional theories without a firm conceptual basis on which they could be built upon making new distinctions. It must be borne in mind that law (and possibly international law) is to a great extent about formalism. Law can carry out its essential functions by the force of formalism which ensures that general standards prevail in specific cases over ad hoc, changing ideas, beliefs and interests. Formalism may bring (relative) normative certainty into international relations. By dispensing with formalist approach functional arguments cannot treat the problem of normative uncertainty to which they give rise.

States or other international actors entering an international commitment should know what kind of commitment they make, legal, quasi-legal or purely non-legal (political, moral). When committing to a norm, they should see in advance the consequences of the possible violation of this norm, the consequences of which may be different depending on the nature of the commitment. In a legal dispute an international court should know what norms are legal or what norms have legal consequences or what norms may provide a legitimate basis for deciding the dispute. If a state or an international court lacking distinguishing parameters cannot tell soft law norms from traditional legal norms how one could anticipate the conclusion of a legal dispute?

B. Proximity Arguments

For international legal theory and practice the proximity argument according to which the features of soft law norms are based upon the fact that they have some kind of special relationship to traditional legal rules may be more accessible and easier to accept. What is the core of this relationship? This is usually described with general remarks, such as rules falling within this category have or produce ‘legal effects’,Footnote 49 or are ‘of special legal relevance’Footnote 50, or which have certain ‘legal component’Footnote 51 or ‘legal value’Footnote 52, or lie ‘in close proximity to law’.Footnote 53 The question as to what is the exact content hiding behind these general expressions arises naturally.

Due to this proximity factor, as vague as it is, soft law can become, to a certain extent, associated with traditional international rules of law and along these lines it can be distinguished from simple political, moral or technical obligations existing in international space. In this respect, however, the main question is: what is the form and origin of such legal relevancy attributed to soft law rules? The legal studies devoted to international soft law topics are abundant in how they describe, most of the time citing examples, the nature of soft law norms' legal relevance, values or effects. It can generally be said that two basic patterns can be distinguished in the relationship between soft law and formal international law. One of these is when soft law presents itself as a precursor of emerging international legal norms, and the other is when soft law in a certain manner complements existing legal rules by reference or interpretation.Footnote 54

1. Soft law as a precursor and basis for legal norms

One starting point is when soft law provides the basis and material to the subsequent international law-creating processes, thus representing a step towards traditional law-making. In this sense it can become evidence of emerging international customary law with regard to either custom or opinio iuris,Footnote 55 or on the other hand it can constitute the material basis for the subsequently concluded international treaties. In this case, creation of the legal norm starts with non-binding international instruments falling outside the reach of traditional system of legal sources and will ultimately turn into a legal form later on. Several examples are found in current international law.Footnote 56

But, should a rule be furnished with legal status only because its content is used as the material basis for a legal rule coming into existence in a subsequent law-making process? If it were true, a paradox of parallelism would appear. On this view, the soft law character of a non-legal rule is predicated upon the creation of a legal rule with the same or similar content. Insofar as the criterion for a non-legal norm to become soft law is that it serves as starting material for a later legal rule, such nature of a non-legal norm could only be revealed as long as the legal rule based upon it comes into being. If no legal rule emerges the non-legal norm does not meet the condition of becoming soft law. However, when the new legal rule is established (eg an international treaty is concluded), then the soft law norm serving as its starting material more or less loses its significance, at least for the States that will become parties to the new international convention. In these circumstances, soft law can only come into being in a parallel manner, and can coexist only with its parallel ‘legal rule’.

If one accepts to make the condition less stringent by saying that non-law becomes soft law provided that it becomes part of the lawmaking process (thus it is no longer a condition that a legal norm should be created in the end), then another problem will appear for a different reason. It is quite doubtful when the lawmaking process begins, (especially when it is about the formation of an international legal custom). In this case, when will non-law become soft law (I)? Regardless of this difficulty, the significance of the soft law nature of a non-legal norm is nevertheless temporary. Insofar as a legislation comes into being, the norm based upon it will lose its importance (see above). If no international legal rule is created from the soft law norm made into part of the lawmaking process, this will mean that, concerning the content of the latter, no agreement between States has come to pass. This will take away the significance of the norm qualified as soft law, as it turns out that it does not constitute basis for the joint and congruent actions of the States. This soft law norm will hardly have greater theoretical or even practical significance than a simple political or moral commitment (non-law) which is in some cases observed by States and in other instances is not.

2. Soft law as a complement of legal norms

The second pattern is when norms classified under the category of soft law play an essential role during the application of existing legal rules, and will have gained legal relevance from this fact. Boyle ventures even further when he says that if soft law is linked to an international agreement, it loses its non-obligatory character, or at least goes under some legally relevant changes.Footnote 57 This special relationship would then provide a basis for forming a separate category of soft law between non-law and law. Concerning soft law and a priori existing international law, two basic types of cases arising from it can be distinguished: (1) a soft law norm is being transported within the trenches of the law by a traditional legal rule of reference; and (2) a soft law norm plays a role in the interpretation of a legal rule.Footnote 58

a) Soft law as the subject matter of the legal rules of reference

In international law a number of international agreements refer to rules of a non-legal nature, respect for which is made obligatory by such references. A clear and simple example is provided by eg article 18(1) (b) and (c) of the International Convention for the Suppression of Acts of Nuclear Terrorism where the State parties to the convention ensure that any nuclear material is held in accordance with relevant International Atomic Energy Agency safeguards; and have regard to physical protection recommendations and health and safety standards published by the IAEA.Footnote 59 Among well-known examples are some of the international conventions entered into by States in whole or in part for the protection of national, ethnic or religious minorities.Footnote 60

It remains a question whether the above IAEA recommendations and standardsFootnote 61 will become soft law rules merely because observation of their content has been made obligatory for the State parties by some international treaty rules of reference. If we make non-law an alternative condition of becoming a soft law norm by a rule of reference of an international convention as a rule to be observed, we will face a significant doctrinal difficulty. In this case, the non-legal norm has no sui generis legal effect beyond the fact that a legal rule makes the observation, respect and consideration of non-legal norm obligatory. However, from this constellation the existence of soft law as a third category between law and non-law cannot be derived. The parties to a treaty import by reference the originally non-legal content into the framework of the agreement, and by doing so they make it a legal obligation. Then why would this be a ‘soft’ obligation for the parties of the treaty? The originally non-legal content of a norm will appear in the recognised source by means of reference, and it will at the same time become legal obligation, in other words ‘hard law’ (traditional legal obligation) for the parties. For those not party to the treaty, it will only remain a simple non-legal norm since for them no legal effects would emerge from the fact that other States transfer the content of non-legal norms into a international treaty entered into by themselves (res inter alios acta)—without regard to the exceptional case that by doing so some universal legal custom would be created in this respect.

All this suggests that it cannot be the criterion of the formation of soft law as an interlocutory category the fact that an international convention refers to it explicitly thereby making it obligatory for the State parties. Still, the law and non-law dichotomy may be maintained in this respect.

b) Soft law as the basis of legal interpretation

There is a view according to which soft law gains its existence wholly or partly from the fact that it provides a basis for interpreting international legal rules.Footnote 62 The reason for this may be that the scope of the legal rule to be interpreted and the non-legal rule can overlap, and their subject matter is the same or very similar. But at the same time the legal rule to be interpreted may allude to the non-legal rule in some way, and, consequently, it can create access to the use of this latter during the interpretation.Footnote 63 By interpretation soft law is being connected to the normative content of the legal rule, and this may earn it a privileged rank among non-legal international norms in spite of the fact that it is not formally obligatory.Footnote 64 Such role of soft law norms is also evidenced by the legal practice.Footnote 65

Nevertheless, the question remains as to how the role in itself played by certain non-legal norms during interpretation substantiates the special status and soft law nature of these norms. There are certain doctrinal difficulties which are as follows: the theory does not make unambiguous the criteria that will separate soft law from non-law and there are more phenomena suitable for the criteria used for distinguishing soft law and (hard) law than the theory takes into account. This is the main direction the critique follows.

Although the tight relationship of soft law to traditional legal norms cannot be denied in this case either, and a number of authors like FastenrathFootnote 66 argue valiantly in favour of the special category of soft law, because of its role played in interpretation, the line of arguments are still intended only to explore the relationship between hard law and soft law as means of interpretation. It does not give us a satisfactory account of the criteria for distinguishing between non-law (the simple political, moral and other non legal obligations) and soft law, and thus leaves the other end of the concept of soft law wide open.

When does a non-legal norm fall into the interim soft law category? What are the criteria for this development? How many times do we have to rely on legal interpretation of a non-legal norm so that a tight relationship can be created with the sphere of law? In this respect, what is the decisive interpretation and who should carry out the interpretation? Can there be within a document (eg a resolution of the UN General Assembly) a rule which is soft law and another which is not? Can a soft law norm lose this character and be shoved back among the norms of non-law? What happens if, with regard to the interpretation, the practice is full of contradicitions? Has the precautionary principle, which is viewed as soft law,Footnote 67 lost this status after decisions by the WTO Appellate Body and the European Court of Human Rights which refused to interpret a treaty provision according to it?Footnote 68

For want of a suitable distinguishing criterion between non-law and soft law, anyone can classify any non-legal norm under soft law according to their wishes insofar as they can find any distant interpretative relationship with any existing law (and this is not among the hardest things to do). This also shows the uncertainty of soft law, and the incidental role played in legal interpretation in itself will not yield the necessary basis for establishing an undefined category between law and non-law.

At the same time, light should be shed upon the fact that extra-legal requirements supporting arguments in legal interpretation will not necessarily become soft law just because it plays a role in the interpretation of a legal rule. For instance, in the field of biomedical research there is an international legal requirement that says ‘any research must be scientifically justified, meet generally accepted criteria of scientific quality and be carried out in accordance with relevant professional obligations and standards under the supervision of an appropriately qualified researcher’.Footnote 69 This, however, does not, in any case, mean that the accepted requirements of scientific quality, the related professional rules or the system of researcher qualifications would become an international soft law norm only because this specific international legal norm refers to it, and serves as its base for interpretation.

Because non-legal norms constitute in several cases the external system of criteria for the performance of state obligations, they do not find themselves outside the zone of non-law and slip into a distinguished legal status of soft law. By analogy with domestic law, standards of the medical profession serving as the measure of the necessary level of medical service, and other medical protocols, will not become soft law in tort law merely because they provide standards to identify the extent of the obligation to exercise due diligence. Similarly, a scientific achievement can also be an important factor in determining the meaning and content of a legal norm, and thus, an essential tool of interpretation. But, can it be said that in the case of Kasiliki/Sedudu the scientific opinion, which, if read alongside with the legal rule to be interpreted in the case, creates normative expectations about the definition of ‘main river branch’ is considered to be within the scope of soft law?Footnote 70 Can there be included under the category of soft law the accepted tools or points of orientation of international legal interpretation like the preparatory documents of an international convention (des travaux préparatoires), the customs developed by States (that have not yet been transformed into legal customs) or the opinions of acknowledged international legal experts etc.? None of the above are presented by commentators in the soft law category,Footnote 71 although they conform to the same criterion (playing certain role in interpretation of legal rules) as other normative phenomena residing in the area of soft law for the same reason.

V. Conclusions

Norms under soft law are essentially (in a broad sense) social norms (political, moral, technical and other non-legal requirements) that can carry out an extremely important function in the coordination of actions of States or other international actors: they can raise expectations towards the arrangement of conduct of international participants; they can outline political, moral and technical (but not legal) requirements as well as mediate in the interactions of international participants. Of course, the regulatory function of these international norms is a complex issue of sociology, international relations theory and organization theories. The relationship of these social norms to international law is wide-ranging, multifaceted, nonetheless ever-changing and hard to classify. Several of these social norms have become the indispensable condition of the operation of the international legal system, especially in some branches of international law (eg protection of the environment, international trade and financial matters).Footnote 72

All the same, this group of international norms is not law and there is no valid and justifiable system of criteria which would take them out of the scope of other international social norms, which are even less linked to law. Consistent criteria that are effective in international practice which would establish an interim soft law category between law and non-law are nowhere to be seen. The two main lines of argument identified as functional and proximity arguments trying to prove the existence of international soft law in fact fail to lay a basis for this concept.

As to the terminological side of the issue, the proponents of international soft law can not even solve the conceptual self-contradictions this term implies. In common legal discourse by ‘soft law’ authors often mean different things or the phrase is used in its vague generality. ‘Soft law’ as a legal metaphor may have its own insidious charm but by semantic extension it has lost its identifiable conceptual content and become vague and sometimes even meaningless in particular situations. It has become a self-contradictory stock phrase, a phraseology of convenience in legal discourse. The content, the extension and distinguishing criteria of soft law continue to be undefined and just as vague, such that separating soft law as a distinct category within international normativity is unwarranted.

References

1 Weil, P, ‘Vers une normativité relative en droit international?’ (1982) 86 Revue général de droit international public 5Google Scholar, and P Weil, ‘Towards Relative Normativity in International Law?’ (1983) 77 AJIL 413.

2 As Wittgenstein pointed out, using the concept of ‘game’, features of a natural concept can only bear very loose ‘family resemblances’ (‘Familienähnlichkeiten’) at best, and are hardly suitable for the classic Aristotelian rules of concept formation and definition. L Wittgenstein, Philosophical Investigations (2nd edn, Blackwell, Oxford, 1967) 31–32.

3 Abi-Saab, G, ‘Cours Général de Droit International Public’ (1987) 207 Recueil des Cours de l'Académie de Droit international de la Haye (RdC) 9, 132Google Scholar; ÁM Bellido, ‘Soft law: mucho ruido y pocas nueces?’ (2004) 8 Revista Electrónica de Estudios Internacionales 1, 1, <http://www.reei.org/reei8/MazuelosBellido_reei8_.pdf> accessed 10 August 2009.

4 A Aust, Handbook of International Law (2nd edn, CUP, Cambridge, 2010) 11.

5 D Thürer, ‘Soft Law’ in R Bernhardt (ed), Encyclopedia of Public International Law Vol IV (ELSEVIER, Amsterdam, 2000) 452, 454. For this difficulty, see also JK Levit, ‘A Bottom–Up Approach to International Lawmaking: The Tale of Three Trade Finance Instruments’ (2005) 30 Yale J of Int'l L125, 127 and D Shelton, ‘Normative Hierarchy in International Law’ (2006) 100 AJIL 291, 319.

6 As it is set out in art 38(1) of the Statute of the International Court of Justice, but see also the question of unilateral legal acts of States.

7 Similarly eg MA Fitzmaurice, ‘International Protection of the Environment’ (2001) 293 RdC 9, 125; AT Guzman, ‘The Design of International Agreement’ (2005)16 EJIL 579, 583–584; LF Damrosch and others, International Law. Cases and Materials (4th edn, West Group, St Paul, 2001) 34.

8 Opinions in the studies advocating the concept of soft law are varied ast to which of these two types the term ‘soft law’ can be applied. Some expressis verbis acknowledge both forms, eg Thürer (n 5) 456 or K Nagy, Nemzetközi jog [International Law] (Püski, Budapest, 1999) 48–52. Some only accept the existence of soft law (II), eg Weil (n 1) 414 or RR Baxter, ‘International Law in “Her Infinite Variety”’ (1980) 29 ICLQ 549. There are authors who seem only to receive type (I) of soft law with approval, eg Aust (n 4) 11–12 or Péter Kovács, Nemzetközi jog [International Law] (Osiris, Budapest, 2006) 80 and 154.

9 E Roucounas, ‘Engagements parallèles et contradictoires’ 206 (1987) RdC 9, 176; similarly D Shelton, ‘Introduction: Law, Non–Law and the Problem of “Soft Law”’ in D Shelton (ed) Commitment and Compliance. The Role of Non–Binding Norms in the International Legal Systems (OUP, Oxford, 2000) 4.

10 These two types of soft law cannot be dealt with according to one set of criteria. Those having attempted such notwithstanding, either did not come to more general conclusions or were forced to separate the two types of soft law later on, eg Thürer (n 5) 455–456.

11 Law of parsimony, ie pluralitas non est ponenda sine necessitate (plurality should not be posited without necessity) attributed by many to one of the greatest English scholastics, William of Ockham, also applying to redundant terms and concepts in the process of description of a phenomenon.

12 In the law of the European Union, the founding treaties explicitly recognize non–binding norms (eg recommendations) within the system of Community or Union norms, see eg art 249(5) of the EC Treaty (Treaty of Rome, as amended). However, in the European Union the notion of these non–binding norms does not cover the concept of international soft law on account of differences in their creation and regulative functions. For a general and extensive analysis, see L Senden, Soft Law in European Community Law (Hart Publishing, Oxford, 2004).

13 For a good survey on the variants of soft law, see Damrosch et al (n 7) 142–158; see also MI del Toro Huerta, ‘El fenómeno del soft law y las nuevas perspectivas del derecho internacional’ (2006) VI Anuario Mexicano de Derecho Internacional 513, 533–537. <http://www.bibliojuridica.org/estrev/pdf/derint/cont/6/art/art12.pdf> accessed 10 August 2009.

14 Eg Roucounas (n 9) 174–193; A Cassesse, International Law (OUP, [specify (latest) edition] Oxford, 2001) 160–161. [This is the first edition. The 2nd (latest) edition of 2005 is not available for me.]

15 See eg H Hillgenberg, ‘A Fresh look at Soft Law’ (1999) 10 EJIL 499; F Münch and H-K Ress, ‘Non–Binding Agreements’ in R Bernhardt (ed), Encyclopaedia of Public International Law Vol III (ELSEVIER, Amsterdam, 1997) 606, 606–612. To make distinction between traditional international conventions or binding legal agreements and non–legal (political or technical) agreements sometimes poses serious difficulties, see eg Aegean Sea Continental Shelf (Greece/Turkey) [1978] ICJ Rep 3, 39–44, paras 94–107, and Dispute concerning Access to Information under Article 9 of OSPAR Convention (Ireland /United Kingdom) Permanent Court of Arbitration (Final Award of 2 July 2003) 30, para 90, <http://www.pca–cpa.org/> accessed 22 May 2009, see also P-M Dupuy, ‘L'unité de l'ordre juridique international’ (2002) 297 RdC 9, 134–135, and Fitzmaurice (n 7) 126–128. A separate problem is why States enter into non–legal (soft law) agreements instead of concluding formal treaties, but this issue is sociological in nature, see AE Boyle, ‘Some Reflections on the Relationship of Treaties and Soft Law’ (1999) 48 ICLQ 901, 902–904. So, there are explanations for the use of soft law by States founded upon, but to be separated from the explanations for the existence of soft law. I am only dealing with the latter problem in this article.

16 Bellido (n 3) 13–15 and generally see A-M Slaughter, ‘International Law and International Relations’, (2000) 285 RdC 9, 96–121. But this group of non–legal norms is not recognized as soft law by eg Thürer (n 5) 454.

17 However, the virtual boundaries of international soft law can be pushed further and, in theory, extended to all the normative phenomena, including unwritten norms, which play some role in international relations. So Francioni regards ‘elementary considerations of humanity’ used by the International Court of Justice in the Corfu Channel judgment (1948) as soft law which can be therefore drawn from unwritten sources, F Francioni, ‘International “Soft Law”: a contemporary assessment’ in V Lowe and M Fitzmaurice (eds), Fifty Years of the International Court of Justice: Essays in Honour of Sir Robert Jennings (CUP, Cambridge, 1996) 167, 169.

18 In the literature, these two features of traditional international legal norms are emphasized as important from the perspective of describing soft law in its relation to ‘hard’ law, see eg Bellido (n 3) 2; Boyle (n 15) 901–902; L Senden, ‘Soft Law, Self–Regulation and Co-Regulation in European Law: Where Do They Meet?’ (2005) 9.1 Electronic Journal of Comparative Law 1, 23. <http://www.ejcl.org/91/issue91.html>, accessed 22 May 2009; F Capotorti, ‘Cours general de droit international public’ (1994) 248 RdC 9, 116. However, it would be difficult to share Zemanek's position that binding nature of a norm only signifies that the obligation is enforceable if not performed, see K Zemanek, ‘The Legal Foundations of the International System’ (1997) 266 RdC 9, 141.

19 There are many authors who treat international soft law as part of international law. See eg Guzman (n 7) 583 or A D'Amato and K Engel (eds), International Environmental Law Anthology (Anderson Publishing Co, Cincinnati, 1996) 55. Following the lead of the prototype theory developed within cognitive psychology, Peters and Pagotto treat soft law as law, because ‘soft law is in the penumbra of concept of the law’, A Peters and I Pagotto, Soft Law as a New Mode of Governance: A Legal Perspective (2006) 1, 12, <http://www.eu–newgov.org/database/DELIV/D04D10_Democracy_and_New_Modes_of_Governance.pdf> accessed 10 August 2009. Even Koskenniemi seems to use the expression of soft law as ‘special law’, Fragmentation of International Law: Difficulties arising from the Diversification and Expansion of International Law. A/CN.4/L.682, 13 April 2006, para 490.

20 Guzman (n 7) 583–584.

21 Hillgenberg (n 15) 500; Bellido (n 3) 39.

22 AC Arend, Legal Rules and International Society (OUP, New York, 1999) 25.

23 J Sztucki, ‘Reflections on International “Soft Law”’ in J Ramberg and others (eds), Festskrift till Lars Hjerner. Studies in International Law (Norstedts, Stockholm, 1990) 549, 573, as quoted by Toro Huerta (n 13) 544.

24 U Mörth, Soft Law and New Modes of EU Governance—A Democratic Problem? (2005) 6. <http://www.mzes.uni–mannheim.de/projekte/typo3/site/index.php?id=476> accessed 10 August 2009.

25 Senden (n 18) 24., L Senden, ‘Soft law and its implications for institutional balance in the EC’ (2005) 1 Utrecht Law Review (No. 2) 79, 81, similarly M Mendelson, ‘Formation of Customary International Law’ (1998) 272 RdC 155, 360.

26 The boundaries of the international system of lawmaking are much more unstable anyway, compared to domestic law, and therefore the dividing line between law and non-law can be washed away and questioned more easily, see Dupuy (n 15) 135.

27 Since the concept of soft law itself presupposes the paradigm of a more or less traditionally featured, somewhat separated and autonomous international law, I will, therefore, stay within these limits, and will not refer to theories which, nurtured by legal realism or based upon it, dissolve the traditional international law in general international normativity, although followers of soft law have borrowed several arguments from them, see eg the New Haven school, Arend (n 22) 76–88; JHH Weiler and AL Paulus, ‘The Structure of Change in International Law or Is There a Hierarchy of Norms in International Law?’ (Symposium, Part 2) (1997) 8 EJIL 545, 550 and 559; RA Falk, ‘Casting the Spell: The New Haven School of International Law’ (Book Review) (1995) 104 Yale Law Journal 1991; Peters and Pagotto (n 19) 8.

28 Although, within a natural law framework, promising attempts could be made and a possible theory would not be inconceivable (basically supported by the concepts of promise, good faith and cooperation). In this setting do Witteveen and van Klink invoke Fuller's conceptual system of the principles of the inner morality of law, placing it into the framework of a variant of discourse theories and applying to international lawmaking. All this would lead to dismantling the existing system of international lawmaking and formal legal sources, W Witteveen and B van Klink, ‘Why Is Soft Law really Law? A Communicative Approach to Legislation’ (1999) RegelMaat (No. 3) 126–140. <chten.uvt.nl/bartvanklink/softlaw.pdf> accessed 3 July 2007. The adjective ‘soft’ may be losing its meaning in this new context since abandoning the presently recognized system of formal legal sources the mandatory character of soft law could be acknowledged.

29 As is referred to by Fitzmaurice (n 7) 126.

30 Guzman (n 7) 584; similarly Zemanek (n 18) 142.

31 A Halvorssen–J Hovi, ‘The Nature, Origin and Impact of Legally Binding Consequences: the Case of the Climate Regime’ (2006) 6 International Environmental Agreements: Politics, Law and Economics 2, 157, 169.

32 Guzman (n 7) 584.

33 M Marcussen, ‘OECD Governance through Soft Law’ in U Mörth (ed), Soft Law in Governance and Regulation: An Interdisciplinary Analysis (Edward Elgar, Cheltenham, 2004) 103, 112.

34 Mörth (n 24) 4.

35 Some basic objections to the binary approach are set out by eg Peters and Pagotto (n 19) 6–9.

36 The theory of ‘graduated normativity’ based upon these considerations ultimately may provide a theoretical framework for the existence of soft law. See U Fastenrath, ‘Relative Normativity in International Law’ (1993) 4 EJIL 305, 331–339 or Reisman's claims in D'Amato-Engel (n 19) 57–59. Nowadays, in the same or similar sense the term ‘relative normativity’ is also used, though this notion was originally applied and linked by Weil to the problem of soft law (II), Weil (n 1) 416–419, see also Fastenrath ibid 305–306, but see Shelton (n 5) 320.

37 See eg E Riedel, ‘Standards and Sources. Farewell to the Exclusivity of the Sources Triad in International Law?’ (1991) 2 EJIL 58, 58; Zemanek also claims that the obligatory and non–obligatory distinction does not hold too much water in practice, Zemanek (n 18) 141.

38 The theories of soft law which treat international soft law as part of international law (see above) do not need to use the ‘grey zone’ theory. They cannot furnish us with an acceptable definition for this extended concept of international law, and at the same time they are also bound to cope with the problem of drawing a line between non–law and law (this latter including soft law, as well) if they want to maintain an autonomous sphere for international law. So these theories, although inflating the concept of international law, paradoxically also have to apply binary approach, similarly to the traditional views. In the following I am principally dealing with the theories of soft law which do not consider soft law a part of international law.

39 Levit (n 5) 191.

40 Boyle (n 15) 913; see also Thürer (n 5) 453; Zemanek (n 18) 143; H Thierry, ‘L’évolution de droit international' (1990) 222 RdC 9, 75; JAP Ridruejo, ‘Le droit international à la veille du vingt et unième siècle: normes, faits et valuers’ (1998) 274 RdC 9, 61.

41 Levit (n 5) 191.

42 Weiler and Paulus (n 27) 554.

43 See eg Damrosch et al (n 7) 34; Peters and Pagotto (n 19) 7; Levit (n 5) 127.

44 In his excellent study, Ho analyses a specific example from the sociological perspective and explores the causes and motives for States to follow soft law norms, DE Ho, ‘Compliance and International Soft Law: why do countries implement the Basle Accord?’ (2002) 5 Journal of International Economic Law 647, 647–688; see also the studies addressing the same question in Shelton (n 9).

45 Levit seems to find a simple (and somewhat extreme) solution to this problem, terminating the non–legal sphere of international normativity. She claims that apart from treaties and customary international law, ‘other international rules and norms reside in the catch–all category of international soft law’ Levit (n 5) 127.

46 Mendelson has taken stock of the persistent difficulties in determining the existence of a international customary rule, Mendelson (n 25) 170–171.

47 Peters and Pagotto (n 19) 12.

48 Fitzmaurice (n 7) 131; Mörth (n 24) 8.

49 Senden (n 18) 23; Thürer (n 5) 454; C McCrudden, ‘The New Concept of Equality’ (2003) 4 ERA–Forum (No. 3) 9, 13; Thierry (n 40) 74.

50 Eg Thürer (n 5) 454.

51 Eg Guzman (n 7) 583.

52 Eg Ridruejo (n 40) 61 and Toro Huerta (n 13) 543.

53 Thürer (n 5) 454; similarly Toro Huerta (n 13) 539.

54 For similar conclusions, see Shelton (n 5) 321.

55 Here, it may suffice to refer to the famous part of the 1986 Nicaragua judgment, where the International Court of Justice assessed the legal value of UN General Assembly resolutions in the context of the formation of customary law, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States of America) (Merits) [1986] ICJ Rep 14, 99–100, paras 188–189, but see also Legality of the Threat or Use of Nuclear Weapons (Advisory Opinion) [1996] ICJ Rep 226, 254–255, paras 70–73.

56 It is well–known that the International Convention on the Elimination of All Forms of Racial Discrimination (adopted 21 December 1965, entered into force 4 January 1969) 660 UNTS 195 drew on the United Nations Declaration on the Elimination of All Forms of Racial Discrimination (General Assembly Resolution 1904 (XVIII) of 20 November 1963); see also, eg the Convention on Early Notification of a Nuclear Accident (adopted 26 September 1986, entered into force 27 October 1986) 1439 UNTS 275, and the Convention on Assistance in the Case of a Nuclear Accident or Radiological Emergency (adopted 26 September 1986, entered into force 26 February 1987) 1437 UNTS 133; their content is based on the following two IAEA gudelines: ‘Guidelines for mutual emergency assistance arrangements in connection with a nuclear accident or radiological emergency’ IAEA Doc INFCIRC/310 (1984), and ‘Guidelines on reportable events, integrated planning and information exchange in a transboundary release of radioactive materials’ IAEA Doc INFCIRC/321 (1985). See also O Schachter, International Law in Theory and Practice (Martinus Nijhoff, Dordrecht, 1991) 85; Boyle (n 15) 904 and Shelton (n 5) 321.

57 Boyle (n 15) 901.

58 It is worth noting that there are attempts to use the legal principle of good faith to push non–legal norms into legal territories. The basic assumption is that international actors' expectations relating to non–legal commitments, grounded in good faith and depending on the circumstances of adoption of these norms may increase their normative strength to the extent that these norms themselves acquire legal or quasi–legal status. See eg Fitzmaurice (n 7) 128–129; Bellido (n 3) 23–25; Schachter (n 56) 100; Roucounas (n 9) 172–173.

59 International Convention for the Suppression of Acts of Nuclear Terrorism (adopted 13 April 2005, entered into force 7 July 2007) (2005) 44 ILM 815. The references by legal rules have weaker forms which do not incorporate non–legal standards into treaties, but only turn these standards into a basis for treaty interpretation (n 63).

60 For instance art 20(1) of the Treaty between the Czech and Slovak Federative Republic and the Federal Republic of Germany on Good Neighbourly Relations and Friendly Cooperation of 27 February 1992, 1900 UNTS 69, declares that the contracting parties shall comply, at the minimum, with the political obligations set forth in CSCE documents, in particular the Document of the Copenhagen Meeting of the Conference on the Human Dimension of CSCE of 29 June 1990, as legally binding obligations. For such references to extra–legal standards in international conventions concluded by Hungary with its neighbouring countries, see Kovács (n 8) 154 and 327; K Nagy, ‘“Soft law” jellegű szabályok Magyarország kisebbségi rendelkezéseket tartalmazó kétoldalú szerződéseiben’ [‘Norms of “Soft Law” Nature in Hungary's Bilateral Conventions Containing Provisions on Minorities’] Acta Juridica et Politica Szegediensis (2000) Tomus LVIII Fasc 28. 399, 403.

61 The number of the possible examples is much higher, see eg Shelton (n 5) 321.

62 Zemanek argues that international courts and tribunals frequently take non–legal instruments into account, therefore the distinction between binding and non–binding instruments makes little sense in practice, Zemanek (n 18) 141.

63 There is a huge number of instances falling into this context where the preamble to an international convention refers to non–legal international instruments. On the other hand, there are weaker references incorporated into the provisions of international conventions alluding to non–legal norms which do not make these norms obligatory, but will open a path for their use during the interpretation of the provision. For such a weak reference, see eg art 13(2) and (3) of the Rotterdam Convention on the Prior Informed Consent Procedure for Certain Hazardous Chemicals and Pesticides in International Trade (adopted 10 September 1998, entered into force 24 February 2004) 2244 UNTS 337 which provide that in determining the environmental, health or other labelling requirements applicable to some classes of chemicals when exported relevant international standards shall be taken into account by the state parties. This is a weak reference, which can still establish the use of non-legal norms for interpretation. The strong and weak reference is not the same, although in both cases it is about a certain completion of the content of a treaty provision with the help of which it can be applied. When it comes to strong references, a certain non-legal norm is to be applied separately from or together with, but in both cases by the force of the provisions of the convention. In case of a weak reference pointing to a specific path of interpretation, however, an external factor (a non–legal rule) determines the scope of a treaty provision.

64 Fastenrath (n 36) 312–315; see also Riedel (n 37) 11; Capotorti (n 18) 117; Zemanek (n 18) 141.

65 For a starting point, see the statement of the International Court of Justice in Gabcikovo–Nagymaros Project (Hungary/Slovakia) (Judgment) [1997] ICJ Rep 7, 77–78, para 140, and also United States—Import Prohibition of Certain Shrimp and Shrimp Products (Appellate Body Report, AB–1998–4) WT/DS58/AB/R (adopted 6 November 1998) DSR 1998:VII, 2755, paras 130–131.

66 U Fastenrath, ‘Relative Normativity in International Law’ (1993) 4 EJIL 1 305–340.

67 See MW Janis and JE Noyes, Cases and Commentaries on International Law (2nd edn West Group, St Paul, 2001) 588–589.

68 EC Measures Concerning Meat and Meat Products (Hormones), (Appellate Body Report, AB-1997-4) WT/DS26/AB/R, WT/DS48/AB/R (adopted 13 February 1998) DSR 1998:I, 135, point VI; Balmer-Schafroth and others v Switzerland (Appl no 22110/93) Judgment of 26 August 1997, ECHR Rep 1997-IV, 25 EHRR 598, para 40.

69 See art 8 of the Additional Protocol to the Convention on Human Rights and Biomedicine, concerning Biomedical Research (adopted 25 January 2005, entered into force 1 September 2007) CETS 195. For a similar provision see art 4 of the Additional Protocol to the Convention on Human Rights and Biomedicine, on Transplantation of Organs and Tissues of Human Origin (adopted 24 January 2002, entered into force 1 May 2006) ETS 186.

70 Kasikili/Sedudu Island (Botswana/Namibia) (Judgment) [1999] ICJ Rep 1045, 1060 and 1064–1065, paras 20 and 30.

71 Shelton (n 5) 320.

72 Cassesse (n 14) 160; Fitzmaurice (n 7) 126; Aust (n 4) 11–12.