Gralf-Peter Calliess & Moritz Renner, Bremen* FROM SOFT LAW TO HARD CODE: THE JURIDIFICATION OF GLOBAL GOVERNANCE**

Transcription

1 Gralf-Peter Calliess & Moritz Renner, Bremen* FROM SOFT LAW TO HARD CODE: THE JURIDIFICATION OF GLOBAL GOVERNANCE** I. Introduction In the context of globalisation legal doctrine remains relatively speechless. 1 This is surprising as globalisation processes directly influence law in various ways. On the one hand, it seems that in the ambit of global governance - understood as the ensemble of all forms of regulation that are oriented towards social values and have cross-border effects 2 - there is a trend towards non-legal forms of regulation. It is evidenced e.g. by the rise of Alternative Dispute Resolution (ADR) 3 and the birth of new trust-based regulation models in e-commerce contexts 4. On the other hand, however, we can witness the evolution of legal and semi-legal forms of regulation beyond the state which are often referred to as private 5 or hybrid regimes 6. While both these developments are readily conceptualised by predominantly economics-based theories such as governance * Gralf-Peter Calliess is Professor of Law at the University of Bremen and leads a project on New Forms of Legal Certainty in Globalized Exchange Processes at the Collaborative Research Center Transformations of the State ( Moritz Renner is a Ph.D. candidate at the University of Bremen and works as a research fellow in the said project. ** This paper is based on a presentation given at the workshop Law after Luhmann: Critical Reflections on Niklas Luhmann's Contribution to Legal Doctrine and Theory in Oñati (Spain), July 5-6, The authors would like to thank Oren Perez and Peer Zumbansen for organizing the conference and all the participants for their helpful comments. 1 Twining, W, Globalisation and Legal Theory (London, Butterworths 2000), at Zürn, M, Global Governance, in Schuppert, GF (ed), Governance-Forschung (Baden-Baden, Nomos 2005) 121, at For the case of Online Dispute Resolution (ODR) see Calliess, G, Online Dispute Resolution: Consumer Redress in a Global Market Place (2006), 7 German Law Journal No See Calliess, G, Transnational Consumer Law: Co-Regulation of B2C-E-Commerce, in O Dilling, M Herberg and G Winter (eds), Responsible Business: Self-governance in transnational economic transactions (Oxford, Hart Publishing [forthcoming]). 5 Bernstein, L, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry (1992), 21 Journal of Legal Studies Electronic copy available at:

2 without government 7 or economic governance 8, legal theory so far failed to grasp the intricate relationship between law and social norms in the context of global governance regimes. In the first part of this paper (II), we will therefore propose a functional analysis of the role of law that allows us to effectively contrast legal with non-legal forms of regulation. To this end we will draw upon concepts from the economics-based theory of Law and Social Norms which explains how both law and social norms contribute to stabilising the behaviour of economic actors. We will argue, however, that this actor-based approach fails to come up with a sufficiently selective definition of the function of law. We will therefore complement it with a communication-based approach building upon Luhmann s theory of social systems. On this basis we will introduce a distinction between the performance and the function of legal systems that allows for a more accurate analysis of the function of law in its relation to society as a whole. In the second part of our paper, the analytical framework we have developed in the first chapter will enable us to conceptualise the emergence of global governance regimes in terms of legal theory (III). To this end we will, after laying out the basic preconditions for the evolution of legal systems, briefly sketch three examples of global governance regimes. We will then show how these regimes enter into a performance competition to both domestic and international law by providing dispute resolution and regulatory services. We will argue that governance regimes develop into legal systems at the point 6 7 Teubner, G, Hybrid Laws: Constitutionalizing Private Governance Networks, in R Kagan, M. Krygier and K Winston (eds.), Legality and Community (Lanham, Rowman and Littlefield 2002). Rosenau, JN, and Czempiel, EO (eds), Governance without Government: Order and Change in World Politics (Cambridge, Cambridge University Press 1992). 2 Electronic copy available at:

3 where they assume the function that law fulfils towards society as a whole: the stabilisation of normative expectations. In our conclusion, we can then shed some light on the relation of law and social norms in the context of global governance (IV). If our hypotheses that (1) legal and non-legal governance mechanisms compete in terms of dispute resolution and behavioural control, that (2) the social function of law lies in the stabilisation of normative expectations and that (3) global governance regimes that take on this function can develop into legal systems hold true, then we will be able to point out both the differences and the interrelatedness of legal and non-legal forms of regulation, of law and social norms, on the global level. II. Law and Social Norms: A Functional Approach The question what is law is persistent in legal theory. 9 Although the question obviously also persists on the global level we have no intent to reopen this debate. Our goal is much less ambitious and at the same time much more pragmatic. Instead of contrasting e.g. positivist with legal-pluralist conceptions of law we want to approach the problem from a different angle. Borrowing from both Law and Economics and Systems Theory we want to outline elements for a functional analysis of different regulatory mechanisms that allow us to better describe global governance regimes in their oscillation between law and social norms. 8 9 Dixit, AK, Economic Governance, in Durlauf, S, and Blume, L (eds), The New Palgrave Dictionary of Economics (London, Palgrave Macmillan, 2nd ed. 2007). Hart, HLA, The Concept of Law (Oxford, Clarendon Press 1961), at 1. 3

4 1. Law and Social Norms One of the most notable functional comparisons of law and social norms has been elaborated in the early 1990ies in Ellickson s analysis of informal dispute resolution between neighbours. 10 Building specifically on his ideas, Law and Social Norms as school of thought developed a Law and Economics -based perspective on the social function of norms. It tries to explain regularities in collective behavioural patterns in terms of economic analysis. In this context, the function of both social and legal norms is seen in inducing co-operative behaviour, i.e. any behaviour that contributes to the resolution of collective-action problems. 11 Drawing on concepts from New Institutional Economics and Game Theory, Eric Posner, for example, uses the model of a cooperation game to explain conformity in social behaviour. 12 Posner tries to reconstruct aspects of social reality as a repeat prisoner s dilemma, i.e. put very simply - a situation where cooperative behaviour will be rewarded with future payoffs by keeping the game going while non-cooperative behaviour ( cheating or free-riding ) will be rewarded with short-term benefits but at the same time will end the game and thus foreclose the possibility of resolving the collective-action issue at stake Ellickson, R, Order Without Law: How Neighbors Settle Their Disputes (Cambridge, Mass., Harvard University Press 1991). Posner, E, Law and Social Norms (Cambridge, Harvard University Press 2000), at 11 ff. Posner, E, Symbols, Signals, and Social Norms in Politics and the Law (1998), 27 Journal of Legal Studies 765, at

5 Taking into account an initial situation of incomplete information, Posner introduces a signalling model 13 : Those players of the game that are willing to act cooperatively, he argues, have an interest in signalling this attitude towards other players by engaging in symbolic behaviour. The signalling is successful as long as the signal has the right cost structure to distinguish good cooperators from bad cooperators, the most striking example being the activity of gift-giving. Possible signals, i.e. social norms ascribing a meaning to a certain conduct, are introduced by public or private norm entrepreneurs i.e. the state as well as individuals or corporations. Thereby, norms are generated in a fundamentally heterarchical process: many people--including politicians, academics, novelists, journalists, and other cultural players--compete to be a successful norm entrepreneur (...). They propose signals; cooperators use the signal that seems most likely to create a separating equilibrium. There will be conflict and confusion, but pooling around a single signal may emerge, with benefits redounding to the norm entrepreneur(s) who proposed it. 14 This rather trivial model of two-party social interaction can be further developed by introducing informal (reputation-based) and formalized (organisational) third-party control mechanisms, 15 but all of these models share a common feature: They define as norms any type of symbolical behaviour that induces cooperation. This leads Law and Social Norms -influenced authors to emphasize the interrelatedness or even interchangeability of legal and social norms. Not only do social norms form the implicit basis of many commercial contracts: This model is a deliberate attempt to overcome the assumption made by earlier authors such as Ellickson that individuals comply with social norms just because they are internalised. Cf. Posner, E, Law and Social Norms: The Case of Tax Compliance (2000), 86 Virginia Law Review 1781, at 1818 f. Posner, n 12, at 774. Ellickson, n 10, at 131; Cf. Calliess, G, Grenzüberschreitende Verbraucherverträge (Tübingen, Mohr Siebeck 2006), at

6 in many contexts, transactors approach the task of reaching an agreement with an idea of how similar transactions are usually structured; they have in their minds an implicit form contract made up of clauses such as prices that are commonly negotiated, boilerplate provisions, and legal default rules. A party may be wary of suggesting too many deviations from the implicit form contract since these might be interpreted as a signal that he is a less reliable or more contentious trading partner than the average market participant. 16 Often parties might also rely on social sanctions rather than the state legal system when it comes to the enforcement of their mutual commitments. 17 To give an example, a party might terminate an ongoing business relationship and, thus, invoke damage to the reputation of the other party. 18 The role that law can play in this context is very limited: its symbolic mechanisms will sometimes enhance co-operation and sometimes interfere with it. 19 And as the chances of winning a breach of contract suit are pretty much random 20, there are good reasons to stick with non-legal enforcement mechanisms. This leaves us with a mixed picture of the theory s usefulness with a view to global governance regimes. On the one hand it aptly explains how the regulation of behaviour can work in the absence of state law. On the other hand, it does not distinguish between the function of law and non-legal norms. For its proponents, on the one hand legal rules can Bernstein, LE, Social Norms and Default Rules Analysis (1993), 3 Southern California Interdisciplinary Law Journal 59 at 71 f. Charny, D, Nonlegal Sanctions in Commercial Relationships (1990), 104 Harvard Law Review 391; Posner, E, A Theory of Contract Law Under Conditions of Radical Judicial Error (2000), 94 Northwestern University Law Review 749; Bernstein, LE, Private Commercial Law in the Cotton Industry: Creating Cooperation through Rules, Norms, and Institutions (2001), 99 Michigan Law Review 1724, at 1786:... the stability of this and other cooperative-based commercial systems may also be due, in whole or in part, to the fact that social norms of honor, particularly when reinforced through group activity, and a basic human desire to think of one s self as trustworthy are more powerful motivators of transactional behavior than economic models of behavior typically assume. Dietz, T and Nieswandt, H, The Meaning of Cognitive and Normative Expectations for the Emergence of Global Research and Development Cooperation (2007), TranState Working Paper No. 49, available at at 19. Posner, n 11, at 148. Posner, n 11, at

7 be and often are considered to be social norms 21 and on the other hand governmental provision of legal institutions is not strictly necessary for achieving reasonably good outcomes from markets Norms and the Legal System The reason why Law and Social Norms fails to uphold the analytical distinction of law and non-legal forms of regulation is that it reduces law to its usefulness for the coordination of economic actors. 23 From this point of view, law is just a means, but not the only means, of securing commitment between commercial actors 24. Law then is simply defined by the public status of the regulator involved in the performance of the coordinative function. In the context of global governance, however, for the absence of a world state the public or private status of regulators often remains unclear. Instead, States, industry, and civil society compete, intermingle, and work together in their regulatory efforts. Thus, the question arises how the analytical distinction between law and non-law can be upheld where transnational regulators have a hybrid status. We suggest to shift the focus from law s function for social actors to its function as a social system of communication. On the basis of systems theory, we will assume that law is a system of communications which are marked by the binary code legal/illegal, 25 and that this code is handled on the level of the second-order observations of the legal sys Eisenberg, MA, Corporate Law and Social Norms (1999), 99 Columbia Law Review 1253 at Dixit, A, Lawlessness and Economics (Princeton, Princeton University Press 2004), at 4. For a critical account see e.g. Hetcher, SA, Comments on Eric Posner s Law and Social Norms: Cyberian Signals, 36 University of Richmond Law Review 327 at 365 f. Hadfield, GK, Privatizing Commercial Law (2001), 24 Regulation 40 at 43. Luhmann, N, Law as a Social System (Hawkins ed; Oxford, Oxford University Press 2004), at 93. 7

8 tem. 26 We can then analyse the interaction of this communicative system with different social sub-systems as well as its interaction with society as a whole, i.e. the totality of all social communications. In order to do so we shall, following Luhmann, distinguish between the performance and the function of law in modern societies. 27 The function of law is given by reference to society as a whole, whereas performance describes law s relation to other functional subsystems of society like the economy, politics, etc. 28 a) Performance The possible performances of law towards other social systems are mainly constituted by behavioural control and dispute resolution. At this point there is a considerable overlap of a systems theory approach and the conceptions of Law and Social Norms we have discussed above. When analysing law as providing dispute resolution services and a certain level of behavioural control e.g. to the economic system, a plethora of functional equivalents to law comes into mind. The performances of law, that means, are not necessarily fulfilled by reference to the legal code. Instead, the behaviour of economic actors might as well (or even more efficiently) be regulated by social norms and ADR provides a good example for the resolution of social conflicts outside the legal system. b) Function The specific societal function of law, therefore, must be found on a different level. And in fact, when we turn to its relation to society as a whole, law fulfils only one function : the Luhmann, n 25, at 70. Luhmann, n 25, at

9 stabilization of normative expectations, i.e. expectations that are upheld even in case of disappointment. 29 This function is what, on the one hand, distinguishes law from every other functional subsystem and, on the other hand, determines the way in which legal communications are processed. Because [i]f law has the function of stabilizing normative expectations in the face of an unorganized growth of normative expectations (...), this can be achieved only by a selection of those expectations that are worth protecting. 30 We shall thus have a brief look at the mechanisms which provide for this selection. From a constructivist perspective, the process in which law decides which norms to protect is necessarily a recursive one: Whether a norm is a legal one or not can only be ascertained through observation of the recursive network that produces legal norms; that is, through an observation of the context of production which becomes a differentiated system through its operations 31 The counterfactual stability of normative expectations is thus guaranteed by the operative closure 32 of law in a legal system, i.e. the network of legal communications perpetually referencing to other legal communications. 33 This implies a new understanding of the relationship between structure and operation (process) which can be subsumed under the concept of autopoiesis: 34 the legal system produces and at the same time is a product of legal communications. This, in turn, presupposes the legal system s capability Luhmann, N, Die Gesellschaft der Gesellschaft (Frankfurt, Suhrkamp 1998), 757 f. Luhmann, n 25, at 148. Luhmann, n. 25, at 152. Ibid. Luhmann, n. 25, at 106 ff. Ibid. Luhmann, n. 25, at 81. 9

10 of second-order observation because the autopoietic legal system [i]n order to specify its operations as legal ones, (...) has to ascertain what it has done so far 35. At this point it establishes its very own logics of remembering and forgetting providing that those norms that are referred to in an institutionalised linkage of episodes 36 are condensed and confirmed, while others just fall into oblivion. 37 In common law systems this institutionalised memory is guaranteed by the reliance on judicial precedents and the doctrine of stare decisis. The doctrine acts as the invisible hand of the system by making sure that adjudication orients itself along the lines of a few leading cases that act as points of reference for later decisions. 38 What is law and what is not is then defined by the legal system itself, all law is valid by decision only. 39 At the same time, the selective reference to earlier judgements evokes the impression that each decision is but the product of earlier decisions. 40 The social function of law is thus guaranteed through the self-referentiality of legal communications. By observing its own operations the legal system selects those expectations that can be normatively upheld while at the same time making sure that it is normatively expected to expect normatively 41. This insight provides us with the basic elements of a terminological framework for analysing the respective role of law and social norms in the context of global governance Luhmann, n. 25, at 90. Teubner, G, Global Bukowina: Legal Pluralism in the World Society in G Teubner (ed), Global Law Without a State (Dartmouth, Aldershot 1997), 3 at 16. Calliess, G, Reflexive Transnational Law. The Privatisation of Civil Law and the Civilisation of Private Law (2002), 23 Zeitschrift für Rechtssoziologie 185 at 196. Shapiro, M, Toward A Theory of Stare Decisis (1972), 1 Journal of Legal Studies 125, at 133. Calliess, n 37 at 195. Calliess, n 37 at 196. Luhmann, n. 25, at 157 f. 10

11 The performances of law, we can conclude, might be equally fulfilled outside the legal system, which explains the rise of non-legal forms of regulation and dispute resolution on the global scale. The function of law, however, defined as the stabilisation of normative expectations, can only be fulfilled within the self-referential structures of a legal system. This will help us analyse whether and to what extent global governance regimes might develop into legal systems. III. Global Governance: An Evolutionary Perspective After having identified the function of stabilising normative expectation as the crossing line between legal and non-legal forms of regulation, we can lay out the conditions under which global governance regimes might cross this line. This evolutionary process can be described against the background of a competition in terms of performance and function between different forms of regulation. 1. Crossing the Line: From Soft Law to Hard Code As we have seen, the function of law is fulfilled where a legal system establishes selective mechanisms for the temporal stabilisation of normative expectations by observing its own operations. These mechanisms of remembering and forgetting, however, are only triggered where normative expectations are contested, where there are conflicts to be decided. Because only then there is occasion for a communication in terms of legal/illegal: 11

12 The departure point for the evolution of law is the initially barely marked distinction between uncontested and contested cases of disappointment. Only if conflicts can be verbalized (...) can a second-order observation arise, because only then is one obliged to decide who is in a legal position and who is in an illegal position. 42 The first enabling condition for the evolution of a legal system can thus be found in the verbalisation of conflicts. We can define this verbalisation as the communication of a social conflict in terms of legal/illegal and vis-à-vis a third party. Any governance mechanism can fulfil this condition where it provides for a third-party dispute-resolution procedure that is legalised in the sense that it effectively suppresses ad hoc and ad hominem arguments. 43 The second enabling condition for the evolution of a legal system we find by looking at the mechanisms of remembering and forgetting we have described above. This kind of second-order observation can only arise where there are points of reference for the interlinkage and mutual referencing of legal communications. These points of reference are most often constituted by precedents or, less likely, a doctrinal elaboration of legal principles. 44 In any case, these points of reference can only be reflected in legal communications if they are accessible for later communications, i.e. if they are made public. Therefore, the second condition is fulfilled where legal communications are published, most likely in the form of judicial decisions, but also in the form of textualized norms Luhmann, n. 25, at 246. Cf. Luhmann, n 25, 248 ff. This mode of operating may well be regarded as the constructivist surrogate for the generation of secondary rules as laid out in Hart, n 9, at 77 ff. 12

13 A global governance regime, we can thus conclude, might develop into a legal system at least under the conditions that (a) it provides for an impartial dispute resolution procedure in the above-mentioned sense and that (b) past decisions are published Performance vs. Function At which point this line is crossed, however, is also determined by the interdependence of law s performances and its function. As we have seen, the decision of contentious cases is crucial to the development of a governance regime into a legal system. Any such mechanism, however, will only be able to attract cases if it is regarded by potential claimants as performing satisfactorily in terms of dispute resolution and possibly behavioural control. Performance with regard to conflict-resolution can be measured against a number of different factors: accessibility (standing), speed (rules of procedure), affordability, and - most importantly attractive remedies. 46 Yet while enhancing its performance these factors might at the same time impair the regime s function as a legal system. There is a perpetual conflict between the capability to attract conflicts and the capability to resolve these conflicts in a way that contributes to the stabilisation of normative expectations with regard to society as a whole. From the perspective of legal discourse, this conflict is described in the categories of fairness vs. legal certainty. It is well illustrated by a brief look at legal history that shows how and with which consequences dispute resolution is performed inside and outside the legal system Calliess, n 37 at 43. Calliess, Billigkeit und effektiver Rechtsschutz (2005), 26 Zeitschrift für Rechtssoziologie 35, at 41 ff. 13

14 At various stages of their evolution, European legal orders have been in need of development aid from outside the law. 47 In classical Roman jurisprudence, aequitas as a maxim in adjudication helped overcome the formalism of the ius civile by providing for principles of conflict-resolution that were explicitly not based in positive law. 48 In praetorian judicature this led to a parallel development of the traditional ius civile and an emerging ius honorarium as jurist-made law. 49 And, many centuries later yet in a similar way, Equity as a supplementary body of rules began developing in the 15 th century alongside the English common law. It was mainly the lack of attractive remedies in common law (providing for damages only) as well as the need for more flexible forms of actions that led to the rise of this supplementary system of adjudication in other words: it was its ability to attract and verbalise conflicts. 50 But after having been administered by the Court of Chancery until the end of the 19 th century, Equity was re-integrated into the common law by the Judicature Acts just as the rapid development of aequitas in Roman law seems to have come to a halt around the 3 rd century: Each of them tended, and all such systems tend, to exactly the same state in which the old common law was when Equity first interfered with it. A time always comes at which the moral principles originally adopted have been carried out to all their legitimate consequences, and then the system founded on them becomes as rigid, as unexpansive, and as liable to fall behind moral progress as the sternest code of rules avowedly legal Calliess, Billigkeit und effektiver Rechtsschutz (2005), 26 Zeitschrift für Rechtssoziologie 35, at 46. Wieacker F, Römische Rechtsgeschichte (JG Wolf ed, München, Beck 2006), vol 2, at 89 f. Waldstein, W and Rainer, M, Römische Rechtsgeschichte (München, Beck 10 th ed 2003), 121 and 136. See e.g. Hanburj, H and Martin J, Modern Equity (J Martin ed., London, Sweet & Maxwell 14 th ed. 1994), at 44 f. Maine, HS, Ancient Law (1861), chapter 3, available at 14

15 The tension between (single case) aequitas/equity as deciding each case on its own merits and (systemic) justice as treating like cases alike, it seems, has in both instances been gradually resolved in favour of an all-encompassing legalisation of formerly nonlegal adjudication principles. Having started off as alternatives to the legal system in terms of performance, it seems that both aequitas and Equity became less and less capable of outperforming the legal system as they also took on its stabilising function. While it is obvious that these insights cannot easily be translated to today s regulatory mechanisms it demonstrates very well which factors fairness and case-orientation on the one hand, legal certainty on the other hand determine the oscillation of conflictresolution between legal and non-legal forms of governance. 3. Global Governance: Three examples In order to examine whether we can describe similar mechanisms in the context of global governance, we shall now very briefly present three examples for regulatory regimes on the global level. a) lex mercatoria In contemporary jurisprudence the most-discussed example of a global regulatory regime is definitely that of lex mercatoria or the New Law Merchant. Yet even its existence is subject to a heated discussion. 52 While some authors try to talk this body of norms into existence and even speak of a creeping codification 53, others see nothing Teubner, G, Global Bukowina: Legal Pluralism in the World Society in G Teubner (ed), Global Law Without a State (Dartmouth, Aldershot 1997), 3 at 7: war of faith. Berger, KP, The Creeping Codification of the Lex Mercatoria (Leiden, Kluwer Law International 1999). 15

16 more than a phantom conjured up by a few speculative Sorbonne professors. 54 Substantially, lex mercatoria claims to rely on the rules and usages of the international trademanship, thereby, alluding to the medieval law merchant. 55 Hard evidence for the reality of an emerging New Law Merchant, however, is only to be found in the references made to it in arbitral awards. 56 In fact, lex mercatoria is best categorised as a certain procedural setting in the context of ADR mechanisms and as the adjudication of conflicts according to general fairness-oriented principles. b) ICANN UDRP Our second example relates to the context of internet regulation, where the arbitration of disputes is provided for by the Uniform Domain Name Dispute Resolution Policy (UDRP) of the Internet Corporation for Assigned Names and Numbers (ICANN). The ICANN is a private non-profit organisation, responsible inter alia for the management of the domain name system, i.e. the global addressing system of the internet. In 1999, it adopted the UDRP as a quick, efficient, and cost-effective on-line dispute resolution procedure for domain name disputes. 57 Its primary goal is to deal with the practice of cybersquatting, i.e. the pre-emptive registration of trademarks by third parties as domain names. All registration service providers accredited with ICANN as registrars of the generic top-level Cf Teubner n 52, at 12 ff for an in-depth account of the discussion. Yet it is already debatable whether the medieval law merchant has ever been more than a set of procedural rules, a privileged daughter of the common law rather than a body of material norms. See Lerch, K, Vom Kerbholz zur Konzernbilanz (2004), 5 Rechtsgeschichte 107, criticizing inter alia the further-going approach by Berman, HJ, Law and Revolution (Cambridge, Harvard University Press 2004); similarly Cordes, A, The search for a medieval Lex mercatoria (2003), Oxford University Comparative Law Forum 5 (ouclf.iuscomp.org). For the different meanings of the term reflexive in this context see Calliess, G, Reflexive Transnational Law. The Privatisation of Civil Law and the Civilisation of Private Law (2002), 23 Zeitschrift für Rechtssoziologie 185, at 193 f.. For details see Calliess, n 56, at 202 ff. 16

17 domains.com,.org etc. have incorporated by reference the UDRP in the registration agreements with their customers, the individual domain name holders. c) CSR Our third example is constituted by a governance mechanism that is much discussed in both corporate and (public) international law: Corporate Social Responsibility (CSR). CSR refers to voluntarily adopted codes of conduct laying down the social and environmental policies of transnational corporations (TNCs). Increasingly, TNCs are also cooperating to adopt common standards for CSR. These collective efforts are reflected in initiatives like the OECD Guidelines for Multinational Enterprises 58, the UN-sponsored Global Compact 59 and, most recently, the UN Draft Norms on the Responsibilities of Transnational Corporations and other Business Enterprises with regard to Human Rights 60. CSR basically works as governance by self-commitment : it is not about reciprocal obligations but about the unilateral setting of standards in the fields of human rights, anti-corruption, labour conditions etc. 4. Performance Competition With a view to their performance, global governance regimes compete to domestic legal systems and international law in various regards. They are predominantly used as a way of resolving disputes outside domestic legal systems such as in the case of ADR mechanisms, but they can also step in the place of international law by regulating the behaviour of transnational actors the way CSR does Available at Available at 17

18 The performance of lex mercatoria, for example, is characterised by its embeddedness into ADR mechanisms. The reference to lex mercatoria in arbitral proceedings provides for the fairness-oriented and case-specific resolution of conflicts between commercial actors, thus providing dispute resolution services to the economic system. In this regard, it stands in a self-proclaimed competition to domestic legal systems. This performance competition, however, takes place on the level of procedures rather than with regard to substantive norms. The basic idea of international commercial arbitration is that the parties to a cross-border contract agree to submit their disputes for decision to a privately elected tribunal. An arbitration clause in their contract allows parties to effectively opt out of domestic legal systems. 61 The resulting advantages of private arbitration as compared to domestic legal systems are manifold. First, arbitration is cheaper, especially in high value disputes. Second, it is also quicker as there are no stages of appeal. Third, arbitration can provide for a perceivably neutral venue in the settlement of international disputes, whereas under traditional conflict of laws, the forum would usually be one of the parties country of domicile. Fourth, the arbitrators can be chosen according to the parties preferences. Last but not least, arbitral proceedings are non-public. In all of these aspects, lex mercatoria-based ADR mechanisms effectively outperform domestic legal systems. The performance of ICANN UDRP can be described in very similar terms. What UDRP basically provides for is an adversarial court-like procedure before an accredited arbitral Available at e7e8141b38cc1256d63002c55e8?Opendocument International Chamber of Commerce, Standard arbitration clause, available at All disputes arising out of or in connection with the present contract shall be finally settled under the Rules of Arbitration of the Inter- 18

19 panel, while the applicable law is restricted to a single substantive norm: the UDRP definition of cybersquatting. The performance of ICANN UDRP, thus, is very much comparable to that of other ADR mechanisms in that it provides dispute-resolution services to the business community. The most notable difference as compared to other ADR mechanisms is to be seen in the fact that with regard to its scope ICANN UDRP has a de facto monopoly: due to the centralised structure of the domain name system the transfer and deletion of domain names simply cannot be performed by any competitor. It is considerably more difficult to discern the specific performance with regard to our third example. As CSR guidelines are not directly legally binding, they are not intended be used in the adjudication of conflicts. 62 They rather serve as a means for building up social pressure: Even if transnational corporations and other business enterprises are not legally bound by the [UN Draft] Norms, there will be considerable pressure to comply with the Norms, since individual companies will be exposed to world public opinion - their customers in the respected forum of the UN. 63 The performance of CSR thus does not lie in providing dispute resolution to businesses or consumers but in a benefit to the global economic system: the regulation of (corporate) behaviour which levels the playing field for economic actors in terms of production standards. The example illustrates very well how corporations can act as private norm national Chamber of Commerce by one or more arbitrators appointed in accordance with the said Rules.. However, they may become legally binding indirectly in various forms: see the contributions to Olaf Dilling, Martin Herberg & Gerd Winter (eds.), Responsible Business: Self-governance in transnational economic transactions, Oxford: Hart Publishing (forthcoming 2007) Hillemanns, C, UN Norms on the Responsibilities of Transnational Corporations and Other Business Enterprises with regard to Human Rights (2003), 4 German Law Journal 1065; similarly Weissbrodt D and Kruger M, Norms on the Responsibilites of Transnational Corporations and Other Business Enterprises with regard to Human Rights, 97 American Journal of International Law

20 entrepreneurs and use CSR standards for signalling their reliability towards other economic actors, i.e. their consumers. Again, this is a performance that cannot be achieved by domestic legal systems as their reach is territorially limited. Nor can this be achieved by (public) international law as TNCs commonly are not regarded as having international legal personality. 64 From a performance perspective, however, it is impossible to characterise governance regimes as being based on either law or social norms because the performances we have outlined above dispute resolution as well as behavioural control can be achieved by both legal and non-legal means. 5. Functional Competition The interesting question, then, is whether out of this mere performance competition could also evolve a functional competition with domestic legal systems and international law. Both ICANN UDRP and arbitral practice referring to lex mercatoria are conceived of as Alternative Dispute Resolution, as a way of consciously settling conflicts outside the law. And CSR characterises itself as mere soft law, thus also locating itself outside the boundaries of legal discourse. Yet, as our historical examples show, the border lines of legal discourse are all but impermeable. Whether they are crossed or not depends, as we have seen, on the two basic preconditions that conflicts are verbalised in terms of legal communications vis-à-vis a third party and that these communications are published to serve as a starting point for second-order observation in other legal communications. 64 Brownlie, I, Principles of Public International Law (Oxford, Oxford University Press 6 th ed. 2003), at 65 20

21 With regard to the verbalisation of conflicts, transnational legal systems thus must provide for a procedure of third-party conflict-resolution. 65 Such a procedure is easy to make out with regard to transnational commercial arbitration as well as the ICANN UDRP. They both institute arbitral panels assuming the role of the alter ego, the generalised other, in the settlement of a dispute. 66 With regard to (self-)regulation by CSR standards, however, there is so far no court or other forum that provides for thirdparty dispute resolution. But this might rapidly change if e.g. domestic courts were to rely on collective codes of conduct when concretising (domestic or international) law standards regarding corporate behaviour. Already today, it can be observed that CSR is more and more becoming an issue to legal discourse. This development is mainly due to a rising tide of human rights litigation in the United States. Since the seminal district court decision in Filártiga v. Peña-Irala 67, U.S. courts are increasingly accepting civil damages actions for human rights violations under the 1789 Alien Tort Claims Act (ATCA). Also, recent US decisions, relying on language in the well-known Supreme Court ruling in Sosa v. Alvarez-Machain 68, have held that corporations are legally capable of violating customary international law and may be subject to jurisdiction under the ATCA. 69 Against this background, it does not seem too far-fetched to assume that future decisions of domestic courts might refer to CSR standards when concretising human f. Calliess, G, Billigkeit und effektiver Rechtsschutz (2005), 26 Zeitschrift für Rechtssoziologie 35, at 53. Calliess, n 56, at F.2d 876 (2d Cir. 1980). 124 S. Ct (2004). For a critical account of the jurisdictional bases of these decisions see Shaw, C, Uncertain Justice: Liability of Multinationals under the Alien Tort Claims Act (2002), 54 Stanford Law Review 1359; see as well for a recent example The Presbyterian Church of Sudan, et al., v. Talisman Energy, Inc. (June 13, 2005), summary available under 21

22 rights standards in customary international law 70 or general clauses in domestic private law 71, even if - as for now - no domestic court has made explicit reference to CSR codes such as the UN Draft Norms. In addition, new dispute settlement procedures might be created in the context of CSR. The OECD Guidelines for Multinational Enterprises, for instance, are supported by a unique implementation mechanism: through their specific instances facility, National Contact Points (NCP), government agencies responsible for the guidelines, offer to help parties resolve disputes. Since 2000 some 130 specific instances have been brought to the NCPs' attention. Most of them concerned the employment and industrial relations chapter of the guidelines and were related to business operations in developing countries. 72 The mediation and conciliation of these cases constitutes another instance of performance competition between law and ADR mechanisms. With regard to the second functional prerequisite for the emergence of a legal system we have to examine whether global governance regimes are capable of selecting the normative expectations that can be normatively expected. It is only through an institutionalised logic of remembering and forgetting that this can be achieved. Once a dispute resolution mechanism can observe, and refer to, its own decisions it guarantees legal certainty in its most basic form: by treating like cases alike. This goal is achieved by le At least to the extent that e.g. the UN Draft Norms on the Responsibilities of Transnational Corporations and other Business Enterprises with regard to Human Rights can be considered a reflection of already existing customary international law. As it is already practiced for international treaties, see e.g. AG Tauberbischofsheim, Neue Juristische Wochenschrift Rechtsprechungsreport 1992, 1098, holding that like the fundamental rights guaranteed by the Basic law, the rights embodied in the European Convention on Human Rights affect private law relations by virtue of the general clauses by the German private law. OECD, The Contribution of the OECD Guidelines for Multinational Enterprises to Managing Globalisation, Paper prepared for distribution at the meeting of G8 Labour and Employment Ministers, 6-8 May 2007, Dresden, Germany, 2007, available at 22

23 gal reasoning which describes the stabilisation of legal norms as a problem of consistency: 73 Every decision has to expressly integrate itself into the continuum of other decisions. It can be observed that arbitral tribunals, as soon as they publish their decisions, engage in this kind of discourse. This is most obvious with regard to ICANN UDRP tribunals: even though the UDRP Rules do not provide for the binding nature of precedent, there hardly is a web-published 74 panel decisions that does not refer to earlier cases employing the well known common law techniques of analogical reasoning. As a random example we might take the following reasoning of a UDRP Panel in the mcdonaldslovesjesus.com 75 case: The Panel notes that in the principal case relied upon by Respondent in support of its "noncommercial" use argument, Bridgestone Firestone, Inc. et al. v. Myers, No. D (WIPO July 6, 2000), the domain name registrant was a former employee of Bridgestone-Firestone and developed a website under the domain name <bridgestone-firestone.net> in order to offer constructive criticism of his former employer. The instant case is distinguishable in that, as noted above, Respondent is not engaged in any criticism of Complainant. 76 But the evolution of second-order observation mechanisms is even more important with regard to lex mercatoria which impossible reality is measurable [only] by the number of references made to it 77. The very existence of lex mercatoria principles, this means, is validated only to the extent that they are referenced in a choice of law clause by the parties to a commercial contract and subsequently adjudicated by an independent tribunal Luhmann, n. 25, at 257 f. Available at McDonald's Corporation v. The Holy See, Claim Number: FA , available at Emphasis added. Calliess, n 56, at

24 Adjudication can in this context be regarded as performative speech acts, 78 that literally talk into existence 79 lex mercatoria. These speech acts, however, gain their performative quality only by the linkage of episodes 80 that especially the doctrine of precedent provides for. But so far, there is only an inchoate practice of precedent and stare decisis in commercial arbitration 81 which, together with the lack of any institutionalised court hierarchy, 82 at least slows down the development of an autonomous legal system of lex mercatoria. We can thus conclude that with regard to the ICANN UDRP and, to a lesser extent, lex mercatoria the enabling conditions for the evolution of a legal system are met. Arbitral tribunals thus enter into a functional competition with domestic courts. Other private or hybrid governance regimes such as CSR standards still lack even those basic features and have so far failed to initiate the communicative process of generating legal norms. A comparison of the three regimes we have described is given below, with different shades of grey signifying the respective degree of legalisation : Cf. Dunn, PH, How Judges Overrule: Speech Act Theory and the Doctrine of Stare Decisis (2003), 113 Yale Law Journal, 493 at 498. Calliess, n 56, at 207. It is dubitable, however, whether this linkage can be described as a chain of narratives as suggested by Dworkin, R, Law as Interpretation (1982), 9 Critical Inquiry 179 or, more recently, Yoshin, K, What s Past Is Prologue: Precedent in Literature and Law (1994), 104 Yale Law Journal 471; for a pointed criticism of this kind of approach see Fish, S, Working on the Chain Gang: Interpretation in Law and Literature (1982), 9 Critical Inquiry 201. Teubner, n 36, at 18. The lack of institutional hierarchy might possibly be replaced by different forms of reputational hierarchy, though, ibid. 24

25 Performance Function Behavioural control / dispute resolution Verbalisation of conflicts Second-order observation mechanisms CSR Yes No No lex mercatoria Yes Yes Partial ICANN UDRP Yes Yes Yes IV. Conclusion: Evolutionary Anomaly or Survival of the Hybrid? The evolution of legal systems in the ambit of global governance can thus be described as a permanent reconfiguration of the conflict between performance and function. Where legal systems fail to perform satisfactorily and escape into legal formalism, alternative ways of dispute resolution will take their place just as the proponents of Law and Social Norms suggest. While many ADR solutions do not fulfil the function of guaranteeing legal certainty and thus stabilising normative expectations, they might develop towards formalised legal systems. The latter phenomenon is well illustrated by the examples of lex mercatoria and ICANN UDRP. Looking at this oscillation of global governance mechanism between law and social norms, the preceding paragraphs have already hinted at a possible answer to the more general question whether the central role of law in European societies should be regarded as an evolutionary anomaly as Niklas Luhmann predicted or whether law in the ambit of global governance might survive in new hybrid forms as Gunther Teubner suggests. Based on our brief tour d horizon of various forms of private and hybrid ordering on the global scale, there are basically two concluding statements that can be made 25

26 on this matter. On the one hand, it is obvious that state legal systems are challenged by non-legal forms of regulation in an increasingly tough competition, where they are oftentimes outperformed by ADR and soft law mechanisms. CSR provides a good example for how (corporate) behaviour may be regulated by social norms, its violations being sanctioned by public opinion only. On the other hand, a systems theory-informed approach to law and social norms also shows how transnational regulatory mechanisms can be legalised through the establishment of an institutional and procedural setting that allows for the autopoietic generation of legal communications. This trend is evidenced by the contemporary developments in lex mercatoria as well as under the ICANN UDRP. 26