Introduction: Global governance as ‘retreat of the state’?

This article is a critical examination of the claim that the emergence of private self-regulatory regimes in the transnational sphere signals a new trend of constitutionalization, located ‘outside the institutionalized political sector, in the “private” sectors of global society’ (Teubner Reference Teubner2012: 2) and outside the limits of nation-state based or intergovernmental control. This claim is a cornerstone of the project of a ‘transnational societal constitutionalism’ (see Teubner Reference Teubner and Teubner1997, 2010, 2011 and 2012) which aims at the reconceptualization of constitutional sociology and can be understood as a particularly far-reaching representation of constitutional pluralism.^{Footnote 1} It relates to the observation that the ongoing transnationalization of regulatory functions, such as rule-making, implementation and monitoring, is taking place within a global institutional architecture which lacks any centralized equivalent to the constitutionalizing state at the domestic level. The spontaneous and barely coordinated mushrooming of transnational private regulatory regimes is interpreted as a process of the self-constitutionalization of the transnational private sphere, decoupled from state-based constitutionalization, and adding up to a global, albeit fragmented constitutional order in its own right (Teubner Reference Teubner2011: 192). The merits of this view clearly lie in extending the notion of social control to the exercise of public authority by private organizations and institutions (Sciulli Reference Sciulli1992: 2001). While I do agree with Teubner that ‘transnational spheres are not constitution-free’, although ‘an equivalent of the constitutional subject of the national state is not so easily recognizable at the transnational level’, and that states have always ‘respected a certain autonomy of social sub-constitutions’ (2012: 6, 8–9), his claim of an ‘independent constitutionalization of autonomous social orders’ (2012: 17–18) does, however, over-emphasize the internal dynamics of transnational societal constitutionalization and tends to downplay the degree to which it is embedded in a political order of the international system in which state-based institutions have the ultimate authority to determine the scope of their own powers, to allocate constitutional functions to other bodies and to guarantee that acts of transnational private self-regulation ‘are justifiable in light of the constitutional norms recognized as playing a constitutive role for the establishment and exercise of public authority’ (Kumm et al. Reference Kumm, Lang, Tully and Wiener2014: 2). It sees heterarchy where there is hierarchy, and it does so because it mistakes lack of capability and willingness for lack of sovereignty.

Whether transnational private self-regulation takes on constitutional functions within or beyond the reach of an (inter-)state-based international order is not a purely academic question. It is of crucial importance for the extent to which the responsibility to procure the legitimacy of public authority exercised by private self-regulatory regimes can be shifted off to the private regulators themselves or still can be guaranteed by some kind of state involvement (Wiener et al. Reference Wiener, Lang, Tully, Poiares Maduro and Kumm2012: 3). To answer this question, I will proceed in two steps. First, a conceptual argument is developed which qualifies the idea of constitutional pluralism by identifying the emerging self-constitutionalization of private self-regulatory regimes in the transnational sphere as rule systems nested in a specific constitutional order of the international society, here described as ‘neo-Westphalian’. This concept differs from the idea of a global constitutional state as well as from a statist sovereigntist conception of global constitutionalism (Loughlin Reference Loughlin2014). Second, the implications of this ‘neo-Westphalian’ understanding for the responsibility to procure the legitimacy of the exercise of authority is discussed based on the claim that private transnational governance is not so private after all, but, in principle, taking place under the omnipresent shadow of public regulation which gives ultimate validity to private self-regulatory regimes (see Abbott and Snidal Reference Abbott, Snidal, Mattli and Woods2009; Börzel Reference Börzel2010). Here I will argue that, although the ultimate responsibility for providing democratic legitimacy continues to lie with the state/world of states, in practice the involvement of state actors and state-based international institutions does not guarantee a sufficient standard of legitimacy. Therefore, a substantial part of the burden of – initial – legitimation must be carried by those directly involved in private self-regulation by constituting and exercising public authority.

The governance perspective on international relations highlights the ‘explosion of transnational regulation outside the intergovernmental realm’ (Papadopoulos Reference Papadopoulos2013: 1) and the ensuing need to recalibrate the relationship between the state and non-state actors.^{Footnote 2} The concept of global governance is a persistent reminder of the fact that politics has long since ‘migrated away’ from the political institutions of the state and is now also happening elsewhere, for example, in highly diverse forms of transnational private self-regulation. During the past 20 years we have witnessed a growing transnationalization of global governance arrangements and, going along with that, a rise of ‘private authority’ in the sphere beyond the state (see Mathews Reference Mathews1997; Cutler, Haufler and Porter Reference Cutler, Haufler and Porter1999; Baumgart-Ochse et al. Reference Baumgart-Ochse2012: 1; see also Hall and Biersteker Reference Hall, Biersteker, Hall and Biersteker2002). When states share their regulatory authority with non-governmental organizations, business actors and international organizations, this diffusion of public authority entails a power shift as well as a role shift (Leibfried and Zürn Reference Leibfried and Zürn2005; Wolf Reference Wolf, Scherer and Palazzo2008). By the setting, implementation and monitoring of rules governing international affairs, such as exercising public authority through private law-making, non-state actors perform regulatory functions that were traditionally in the responsibility of the state or intergovernmental bodies.^{Footnote 3} The claim of a complex interplay among public and private forms of ordering, however loosely coupled, has been repeatedly stated, but more systematic research is needed to clarify how exactly they overlap or are interconnected (Knill and Lehmkuhl Reference Knill and Lehmkuhl2002; Wolf Reference Wolf, Scherer and Palazzo2008; Schuppert Reference Schuppert2011a) in order to gain a better understanding of the nature of the ‘neo-Westphalian’ political order.

The governance perspective also draws attention to the normative implications of the diffusion of public authority from the state to private actors (Börzel and Risse Reference Börzel, Risse, Grande and Pauly2005; Dingwerth Reference Dingwerth2007; Wolf Reference Wolf and Niesen2012a). While the pooling of public and private problem-solving resources may promise to increase effectiveness, the legitimacy of non-state actors supporting or – in the form of private self-regulation – even replacing regulation by the state or by intergovernmental institutions is viewed more critically because of the shift away from state-based institutional settings which, at best, can be held accountable by democratic procedures, to private governance arrangements which lack these instruments to hold them accountable (Graz and Nölke Reference Graz and Nölke2007). The legitimacy concerns raised against transnational private self-regulation are accentuated by the responsibility shift that is assumed to go along with the observed role shift and further specifies the normative ‘questions relating to the establishment and exercise of legitimate public authority’ (Kumm et al. Reference Kumm, Lang, Tully and Wiener2014: 3) at stake here. To what extent does the diffusion of authority in the sphere beyond the state affect the responsibility of the state/world of states to procure the legitimacy of private self-regulation? What makes public authority exercised by private actors in the sphere beyond the state legitimate authority? Are transnational private self-regulatory regimes justifiable in the light of the norms that are generated and implemented by way of private self-constitutionalization and that determine the manner in which public authority is to be exercised, or in the light of constitutional(izing) rules provided by a state-based political order of the international system? Seen through this normative lens, the ultimate question is about the direction in which the further development of the international political order should go if certain standards of legitimacy are to be met.

Like power shifts, shifts of responsibility have attracted the attention of IR scholars primarily from a state-centred perspective, as the vast and still growing literature on the responsibility to protect impressively confirms. There is a remarkable discrepancy between the broad debate on the responsibility shift from the state to intergovernmental, or supranational institutions of the international society of states to protect individuals from the human rights violations that are inflicted upon them by individual governments (see among many others: Bellamy Reference Bellamy2009; Ki-moon Reference Ki-moon2013), on the one hand, and the comparatively little research on the diffusion of responsibility that goes along with the shift from the public to the private sphere, on the other. The literature about ‘corporate social responsibility’ (CSR) is a major exception. It sheds light on the increasing number of transnational, cross-company and cross-sector self-regulatory initiatives, by which environmental, employment, human-rights, and anti-corruption norms and standards are being established and enforced by non-governmental actors (see among others Scherer and Palazzo Reference Scherer and Palazzo2008; Flohr et al. Reference Flohr, Rieth, Schwindenhammer and Wolf2010; Ougaard and Leander Reference Ougaard and Leander2010; Deitelhoff and Wolf Reference Deitelhoff, Wolf, Risse, Ropp and Sikkink2013). These initiatives provide a particularly clear illustration of the new interplay between public and private actors in the collaborative execution of public functions. The CSR debate extended the range of corporate responsibilities and raised the bar for defending corporate activities as legitimate (Koenig-Archibugi Reference Koenig-Archibugi2004; Börzel and Risse Reference Börzel, Risse, Grande and Pauly2005; Buchanan and Keohane Reference Buchanan and Keohane2006; Dingwerth Reference Dingwerth2007; Flohr et al. Reference Flohr, Rieth, Schwindenhammer and Wolf2010). While, 40 years ago, the activities of commercial enterprises could readily be justified by reference to Milton Friedman’s tenet that the sole social responsibility of businesses was to make a profit and stay within the law whilst doing so (Friedman Reference Friedman1970), now, as self-regulators rather than mere addressees of public regulation, businesses are expected to assume more extensive responsibilities (Haufler Reference Haufler2001; Ruggie Reference Ruggie2013).

But how far does this responsibility shift actually go? Does it also burden private self-regulators with the responsibility to procure the legitimacy of their contributions to global governance while at the same time unburden the state of this responsibility? Such a ‘retreat of the state’ (Strange Reference Strange1996) could result in undesirable legitimacy gaps. It is the cogency of this shift of the burden of justification onto private actors once they assume functions by which public authority is constituted or exercised, and the simultaneous unburdening of the ‘natural’ addressees of such requirements – namely states – which will be scrutinized in what follows here.

We shall begin by asking whether private self-regulation in the transnational sphere really is private self-regulation in the sense of ‘governance without government(s)’. For only if it is would non-governmental actors be left as sole bearers of responsibility in procuring the legitimacy of the public authority constituted and exercised by them (Section I). A typology will be drawn up of the forms in which the public sphere shadows private regulatory activity to demonstrate that purely private self-regulation, with no state involvement or support, does not exist in the constitutional frame provided by what will be identified as the ‘neo-Westphalian’ political order of the international system. What appears as a ‘retreat of the state’ will be reinterpreted as a new division of labour between public and private norm-setting and norm-implementation. This has implications for the attribution of the responsibility for meeting legitimacy demands on transnational private self-regulation. In order to clarify the substance and extent of this responsibility we shall consider what normative yardsticks, if any, can be called into service to assess these demands (Section II). An empirical case will then demonstrate the embedment of the private constitution and exercise of transnational public authority in a political order in which state-based institutions review constitutional functions fulfilled by private actors. The case of sports law has been selected for two reasons, which follow the article’s main line of argument. As an unlikely case, often cited as a model example of autonomous societal self-regulation, the lex sportiva renders particularly strong plausibility for the claimed non-existence of private self-regulation. Furthermore, and on closer inspection, the regulation of performance-enhancing substances can serve to illustrate the legitimacy procuring potential, but also the limits of the complex sharing of responsibilities between governments, international organizations and a transnational private self-regulatory regime. It shows how dispersed the loci are from which state-based institutions can fulfil their constitutional role in one thematic field alone, and reflects the fragmented setting of the ‘neo-Westphalian’ order, which is not a global constitutional state, but consists of a plurality of political and legal arenas and intergovernmental agreements. From these, a set of central constitutional norms can be derived with reference to which requirements for, and sources of, the legitimacy of the constitution and exercise of public authority by private actors in the transnational sphere can be assessed (Section III). Lastly, more general conclusions are derived from this empirical case regarding ways in which public and private actors can interact to increase the legitimacy of the exercise of public authority constituted by way of private self-regulation (Section IV).

I. How far does constitutionalization via private self-regulation actually go and where does this leave the state? A neo-Westphalian frame of reference

For a better understanding of the implications of the diffusion of authority for the locus of the responsibility to procure legitimacy for private self-regulation in the transnational sphere, we will take a closer look at the change in roles and the new division of labour between public and private norm-setting and norm-implementation. The main argument is that public authority exercised by private actors is always regulated self-regulation because all transnational private self-regulation takes place in the ubiquitous shadow of actual or potential (inter-) state-based regulation. (On this, see Schuppert Reference Schuppert2001; Abbott and Snidal Reference Abbott, Snidal, Mattli and Woods2009; Börzel Reference Börzel2010; Collin Reference Collin and Collin2011). It rests on a neo-Westphalian understanding of the political order of the international society which is fundamentally distinct from the notion of a (world-) societal self-constitutionalization as ‘global law without a state’ (Teubner Reference Teubner and Teubner1997) or ‘governance without government’ (Rosenau and Czempiel Reference Rosenau and Czempiel1992; Reinicke Reference Reinicke1998; Zürn Reference Zürn1998), separate from the world of states, in which ‘private regimes or rule-systems may operate without the state [and] neither emerge from nor rely on conventional forms of public lawmaking’ (Lehmkuhl Reference Lehmkuhl2008: 340, 342; see also Teubner 2012).

To answer the question where to locate the responsibility for providing legitimacy one needs to know the available candidates first. Here, different perspectives on global constitutionalization offer different answers. Seen from the extreme end of constitutional pluralism, self-regulatory normative orders generated by private self-constitutionalization emerge on an equal footing with institutions whose constitutions have been established through state authority while ‘each acknowledge the legitimacy of every other within its own sphere’ and ‘none asserts or acknowledges constitutional superiority over another’ (MacCormick Reference MacCormick1999: 104). The neo-Westphalian perspective differs from this assumption of private and state-based rule systems co-existing in non-hierarchical relationships.^{Footnote 4} It claims that the transnational rule systems that have been established and enforced by private actors are always embedded in a global political order which is still anchored in the inter-state system but transcending the political sphere of the world of states. The ultimate authority to allocate (and withdraw) constitutional functions to other bodies rests within the world of states rather than with those (i.e. private self-regulators) who constitutionalize ‘their own sphere’.

Although state-based in regard to identifying the bearers of ultimate constitutional authority, the neo-Westphalian frame of analysis is not static. It takes into account developments which have been described as ‘transformations of the state’ or as ‘de-nationalisation’ (Leibfried and Zürn Reference Leibfried and Zürn2005) and which have resulted in the pluralization of loci and levels at which statehood functions can be exercised unilaterally (by nation states) but also collectively (in inter-/supranational bodies). In contrast to the notion of a (global) constitutional state it implies a ‘thin’ concept of constitution, relying on the existence of constitutional(izing) rules which are not embodied in one single document but in a plurality of international agreements. Thus the neo-Westphalian perspective de-links constitutionalization from the idea of a (national or global) constitutional state as its centre and regards the institutions in which states exercise their ultimate constitutional authority as being widely scattered over different levels and sectors of governance. In this regard it is heterarchical.^{Footnote 5} This political order provides a ‘rudimentary’ constitutional framework for which only the world of states is in charge but which not only regulates inter-state relations via international self-regulation but also constitutes the only locus of competence in regard to the allocation of regulatory competencies to other – e.g. private – actors. In this regard it is hierarchical. Possession of this competence marks a fundamental distinction between the members of the world of states and the members of the world of transnational business or civil society. There may be different – in this case public and private – sites of constitutional authority, but there is also a clear hierarchy among them.^{Footnote 6}

The self-regulation projects of private norm-entrepreneurs have, without a doubt, resulted in a proliferation of loci of norm generation. But even if the establishment of private regulatory regimes in the transnational sphere goes beyond the scope of material regulations, in the sense of Hart’s (Reference Hart1961) primary rules or Kooiman’s (Reference Kooiman and Pierre2000) first-order norm-production, and also includes constitutional(izing) rules (‘rules for rule-making’), these loci lie within the shadow of a higher-order body of regulations/regulatory powers at the inter-state level and do not seriously call into question inter-state competence in the matter of the creation of the conditions for such private norm-production. Hence, what in the broadest sense of constitutional pluralism is alleged to develop ‘naturally’ (Teubner Reference Teubner and Teubner1997) in fact takes shape not in a worldwide pre-constitutional ‘state of nature’ but within the constitutional framework of the political order of the society of states within which the transnational production of norms by private actors is embedded. In this set-up, states are still actors – the only actors – with the ultimate competence to grant or deny authority and regulatory competence to non-state actors (on an essentially unilateral basis internally and multilaterally between each other). Within the ubiquitous shadow of possible (inter-) governmental control states can – at least in principle^{Footnote 7} – intervene at any moment to determine the conditions governing private norm-production.

This neo-Westphalian understanding, rather than distinguishing between ‘governance by, with and without governments’ (Zürn Reference Zürn1998: 169–71), highlights the way in which the one sphere is embedded in the other, how the two are related, and how they impact upon one another. In this respect, the transnational sphere is much more diverse than the intra-state sphere, where private legal entities may ‘lay down self-determined private rules within a framework delineated by state legislation’ (Mätzler Reference Mätzler2009: 144). The new ways in which public guarantor-responsibilities are discharged – ways that cannot really be comprehended on a narrow view of legal-style leverage – cover a broad spectrum that can range from restrictive pressure (compulsion, prohibition), through background support (acceptance, recognition) to encouragement (de-regulation, initiation). A variety of motives are reflected here, including ‘protection of endangered rights’ (through intervention), ‘protection of exercised rights’ (through recognition) and ‘reducing the burden on the state’ (through approval).

The public ‘shadow of hierarchy’ in which private self-regulation takes place may appear in a variety of roles and manifestations. A distinction can be made here between laissez-faire, reactive and proactive forms of public involvement in, and influence on, private self-regulation. These are set out in Table 1.

Table 1. Variants of state embedment of transnational private self-regulation^{Footnote 8}

This classification of the types of public–private interaction involved in transnational private self-regulation comprises both direct forms of involvement and less visible channels of state influence. In no case, however, do we have a situation where the state disappears completely from governance, or indeed where states/the world of states irreversibly disempower themselves. The following illustrations demonstrate that even in cases where the state, or the institutions of the world of states, keep their own contribution down to a minimum, in the form of tacit acceptance or deliberate inaction, they still cast a shadow of hierarchy – albeit an almost unrecognizably blanched one – over private self-regulation.

Responsible Care, the ‘world’s leading voluntary industry initiative’, may serve as one example of many to illustrate the state’s/the world of states’ laissez-faire role of non-prevention with regard to a transnational private self-regulatory effort. Responsible Care was originally initiated in 1985 by the Canadian Chemical Producers’ Association. Today, it is managed by the International Council of Chemical Associations at the global level. Its members comprise nearly 60 national chemical manufacturing associations who have committed themselves to promote the observance of environmental and work-related standards and who are responsible for monitoring implementation by the chemical companies in their countries.^{Footnote 9} The initiative has been and still is tolerated by governments as a private self-regulatory rule-system (Conzelmann and Wolf Reference Conzelmann, Wolf, Graz and Nölke2007). The same is true for the Equator Principles, the global private regulatory framework for environmental and social risk management for project finance (Wright Reference Wright, Biermann, Siebenhüner and Schreyögg2009; Flohr Reference Flohr2014). Its standards for due diligence to support responsible risk decision-making have been adopted by 79 financial institutions in 35 countries.^{Footnote 10} In both cases of voluntary private self-regulation, individual governments or intergovernmental institutions would have had the authority to step in at any time they decided to do so, but refrained from interference.

Examples for the – supportive or prohibitive – reactive role the public sector can play with regard to transnational private self-regulation are equally abundant. In the case of the official adoption of the environmental, social and economical standards set by the Forest Stewardship Council in order to promote the responsible management of forests (Gulbrandsen Reference Gulbrandsen2004; Dingwerth Reference Dingwerth2007 and Reference Dingwerth2008), the standards of this private transnational self-regulatory initiative have been backed up and explicitly recognized by those governments who subsequently adopted them as their own standards for public procurement. Even states’ or intergovernmental organizations’ reactive legal recognition of norms that have originally been set by private actors is a common practice. For instance, the banking sector’s Wolfsberg Principles to combat money-laundering was incorporated into national law (Wolfsberg Group 2002; Flohr Reference Flohr2014), and so were the food-security norms from the food industry’s codex alimentarius, originally a private initiative, into FAO and WHO agreements. Likewise, the so-called lex sportiva, commonly referred to as a prototype of transnational private self-regulation, has received national and international legal recognition and embedment.^{Footnote 11}

Corrective public interventions or even abrogations of private norm-setting initiatives can be exemplified by the Kimberley Process to stop the trade in conflict diamonds, in which states intervened by enforcing what had started as a private certification scheme in order to ensure that diamond purchases were not financing rebel movements (Grant and Taylor Reference Grant and Taylor2004; Haufler Reference Haufler, Potoski and Prakash2009a and Reference Haufler2009b; Grant Reference Grant, Lujala and Rustad2011; Jakobi Reference Jakobi, Jakobi and Wolf2013). A stronger intervention can be exemplified by the directive the EU set in force in 2009, specifying emission standards for new passenger vehicles after the European automobile industry’s implementation efforts fell short of their own voluntary self-commitments to reduce CO₂ emissions.^{Footnote 12}

Proactive forms of influence can vary from ‘soft’ invitations to collective private-self commitments, as in the case of the Global Compact initiated by former United Nations General Secretary Kofi Annan (Kell Reference Kell2003; Rasche and Kell Reference Rasche and Kell2010),^{Footnote 13} to delegating public functions of norm-setting and norm-enforcement to private actors, as in the case of the Chartered Companies which are also a reminder of the remarkably long tradition of transnational private self-regulation in the shadow of the state (Wolf Reference Wolf, Deitelhoff and Wolf2010).

II. The legitimacy demands of transnational private self-regulation

According to this reasoning we can no longer assume that – because states would have not any part at all in the process – the responsibility for procuring that private regulatory activities meet certain legitimacy demands has automatically passed on to the private actors who have assumed state functions. If there is an omnipresent possibility of backup guarantor or supervisory functions to be exercised, the ultimate responsibility of the state/world of states is at no point revoked, irrespective of how it is, or may be, exercised in practice, for example when governments take to deregulation and privatization.

Having said this, and assuming that this ultimate responsibility of the state/world of states also includes the responsibility to procure the legitimacy of transnational private self-regulation, the next question that needs to be addressed refers to the extent to which the public authority exercised by private actors in transnational self-regulations requires legitimation, and the possible ways in which legitimacy might be procured for it. The strongest rejection of the claim that transnational private self-regulation needs justification in terms of legitimate authority at all starts out from the modes of governance involved in transnational private self-regulation and from a fundamental division between public power and private freedom. Despite the misleading talk of a ‘lex mercatoria’, a ‘lex informatica’, or a ‘lex sportiva’, so it is argued, the modes of governance involved in transnational private self-regulation are located below the threshold of sovereign governance in the sense of exercising state-like legislative or executive power. Within the framework of voluntary commitments, there is no exercise of power in the narrow sense; there is simply use of the right to the collective exercise of individual autonomy. Private standard-setting is interpreted as ‘the free exercise of a fundamental right’ (Michael Reference Michael, Bauer, Huber and Sommermann2005: 435) to freedom of association as a constitutionally protected expression of civil liberty. Different from the generation and enforcement of collectively binding decisions based on the mandatory exercise of the sovereign authority of the state – the exercise of this freedom does not have to demonstrate its democratic legitimacy. The need for legitimation in the case of this classical exercise of state power is – it is claimed – fundamentally different from the need for (public) regulation in the case of ‘non-state self-regulation of public affairs’ (Collin Reference Collin and Collin2011: 8) in which there is a risk of private abuse potentially detrimental to the common good.

According to this view, even the kind of private norm-setting that one comes across within the framework of global governance would not, in principle, need to justify itself. As an instance of the exercise of fundamental rights, it would, in theory, qualify for protection and would only require regulation, if at all, in cases where the exercise of private freedoms accorded at (inter-)national level was damaging to the public good or violated basic laws. The problem of private self-regulation’s need for legitimation would dissolve into that of its need for regulation by the public sector. However, as Michael (2005: 444) has rightly pointed out, ‘the fundamental division between public power and private freedom … cannot be maintained in the case of private standard-setters who utilize a putative freedom to exercise power’. In the transnational sphere, in particular, the dividing lines between the public and the private exercise of power and control merge into one another in a multiplicity of hybrid manifestations of norm-setting and norm-enforcement. Within the framework of global governance, intergovernmental and private transnational rule systems are both often characterized by their voluntarism, informal nature, exclusiveness and self-authorization. In the case of transnational private self-regulation, these problems are magnified rather than minimized because, by making and implementing collectively binding decisions that infringe upon individual liberties, regulatory authority is exercised by actors who as a rule are not democratically accountable to those affected by these decisions. All these characteristics have the potential to breach fundamental rights or to fail to satisfy general-interest requirements, and all exercise of power requires constitutional limits (Poiares Maduro Reference Poiares Maduro, Avbelj and Komarek2012: 75). Transnational private self-regulatory mechanisms must therefore be justifiable in the light of the same constitutional norms recognized as playing a constitutive role for legitimate authority that apply to mandatory rule-making and rule-enforcement by the state. Are regulatory objectives of a socially oriented kind being successfully achieved with an appropriate balance of costs and benefits (criterion of public-interest effectiveness and efficiency)? Are matters of public concern being addressed in a sustained and credible manner (responsiveness criterion)? Do the addressees of regulation, and others affected by it, have ways of successfully influencing the regulatory process (self-determination criterion)? Is it possible successfully to control the exercise of regulatory authority and call its agents to account (accountability criterion)? (On this, see Flohr et al. Reference Flohr, Rieth, Schwindenhammer and Wolf2010: 203–8; Dingwerth Reference Dingwerth2007; Take Reference Take2012).

These questions effectively define the major normative preconditions for determining the nature and extent of public ‘shadowing’ required by transnational private governance contributions: the practical need for public intervention or non-intervention will depend on the extent to which such regulation, in its capacity as an exercise of freedom liable to abuse, has already imposed effective limits upon itself to prevent any damage to the public good or violation of fundamental rights.^{Footnote 14} Although only regulation by the public sector can link transnational private self-regulation with democratic mechanisms of political legitimation, this interrelatedness at the same time points to a joint (initial) private and (backup) public responsibility. Called upon in its capacity as guarantor, the public sector still holds the ultimate responsibility for procuring the legitimacy of authority exercised in global governance arrangements, including those based on transnational private self-regulation. This it can do, for example, by regulatory measures that incorporate private norm-setting into the constitutional set-up in such a way that the norm-setters, in exercising the freedoms granted to them, do not damage the public good or violate fundamental rights.

However, such expectations in the public sector’s actual capacity as guarantor may prove too high: public ‘shadowing’ by governments or intergovernmental organizations may themselves suffer from legitimacy deficits that their involvement would only pass on. This applies to the level of individual governments – which may not be democratic themselves – as well as to the supra- and inter-governmental level. Two of the examples already used to illustrate the forms and instruments of public influence in Table 1 may suffice to point this out: benchmarks set by the European Union in order to guide the self-regulatory activities of the European associations of the automotive industry with the effect of reducing CO₂ emissions suffer from the ‘democratic deficit’ of the EU itself (see among many others Majone Reference Majone1998; Scharpf 1999; Wolf Reference Wolf1999; Moravcsik Reference Moravcsik2002); likewise, the Global Compact initiative of the United Nations Secretary General has no mandate from governments (Ruggie Reference Ruggie2013: xxix). In each case, intergovernmental sources of legitimation would, in their turn, suffer from legitimatory deficits (Moravcsik Reference Moravcsik1994; Wolf 1999) which they would only pass on to the exercise of authority they are supposed to legitimate. The discrepancy between the options that theoretically exist for exerting (inter-)national regulatory influence on transnational private self-regulation and the actual implementation of these options is another reason why the contribution to the legitimation of transnational private self-regulation that may be expected from this kind of collective exercise of state power eventually comes up against its limits. Governments may be unable or unwilling to regulate private self-regulation in order to guarantee that the above-mentioned yardsticks of legitimacy are met.

Therefore, in cases where restrictive national or international public regulation of private self-regulation fails, or is not utilized, or has questionable normative credentials as a procurer of legitimation, external (public) and internal (private) sources of legitimacy need to be combined to ‘compensate for one another’s weaknesses, and play mutually reinforcing roles’ (Abbott and Snidal Reference Abbott, Snidal, Mattli and Woods2009: 46; Ruggie Reference Ruggie2013: 78). Against the background of these multiple accountability relationships within and outside the self-regulatory regime (see Black Reference Black2008: 157) the identification of the internal legitimatory potential of transnational private self-regulation once again becomes of interest as a complementary source of legitimation. Not being capable of democratic legitimation themselves,^{Footnote 15} functional equivalents could consist in the choice of particular, e.g. deliberative modes of governance whose non-mandatory character would entail a lesser requirement in terms of democratic legitimation. In addition to reducing the need for corrective external intervention by relying on ‘soft’ modes of governance, the establishment and exercise of authority by way of transnational private self-regulation could optimize their own resources for procuring (epistemic) legitimacy, based on technical or moral reputation, of the kind that are available to the economic or civil-society actors directly involved in self-regulation initiatives^{Footnote 16} and which, in conjunction with particular objects of self-regulation, may be capable of procuring some degree of internal legitimacy (Joerges and Neyer Reference Joerges and Neyer1997; Dingwerth Reference Dingwerth2007; Quack Reference Quack2010; Take Reference Take2012; Wolf Reference Wolf2012b;). Thus the participation of actors with practical knowledge can ensure expert legitimacy first and foremost for self-regulation initiatives involving material regulations – for example, technical standardization. By contrast, the involvement of actors who enjoy widely recognized moral authority as ‘commonweal norm-entrepreneurs’ could enhance the legitimacy, in particular, of private norm-setting initiatives whose functions included securing basic agreement on standards of appropriateness for public-interest activity.

III. The lex sportiva: Self-regulation by sports organizations in the WADC Code

Transnational self-regulation by sports organizations is characterized by the high degree of legislative, executive and judicative authority exercised by private regulators. The sports world’s so-called lex sportiva is, alongside the lex mercatoria, regarded as a model example of autonomous societal self-regulation. As an unlikely case, it is a particularly instructive example to illustrate the claimed non-existence of private self-regulation. Moreover, it exemplifies how dispersed and fragmented, in one thematic field alone, the places are from which state-based constitutionalization takes place in the neo-Westphalian setting. Contestation and justification of private self-regulation in terms of legitimate exercise of authority can best be studied by looking at the self-regulatory regime to prevent the use of performance-enhancing substances in light of the ongoing debate about the need for national anti-doping legislation and athletes’ criticism levelled at the self-regulatory mechanisms for violating basic human rights. Moreover, the dealing with these controversies offers insight into how public and private regulators can interact to procure legitimacy.

Long before the current global-governance discussion about the privatization of cross-border governance, the gradual evolution of private law in the world of sports was being cited as proof that a centralized state monopoly on power was not necessary in order to guarantee ‘public order, co-ordination and public goods’ (Burnheim as early as 1986: 221). Within the autonomous bodies of the sports world, state functions are fulfilled by non-state actors who have the power to lay down binding rules and to impose sanctions where these are violated. The non-state institutions in which this takes place enjoy their own authority whose claim to legitimacy is based on general acceptance by both the public and the actual addressees of the regulations, namely the athletes who make up the sports clubs and associations. An extensive privatization of the law, going well beyond a shift of the state’s monopoly on arbitration to the social sphere, therefore appears to have taken place in organized cross-border sports relations.

Contrary to this view, in what follows I will first go into the interaction between self-regulation and its public ‘shadowing’ in order to show that, even in this allegedly autonomous field, transnational private self-regulation takes on constitutional functions only within the reach of an (inter-)state-based international order which can give ultimate validity to it, or refuse to acknowledge its validity. A balance sheet will then be drawn up based on the four criteria mentioned above (public-interest effectiveness and efficiency, responsiveness, self-determination, accountability). Finally, the interaction between different political and legal arenas in order to deal with problems and deficiencies of the particular approach to self-regulation under consideration will be assessed against the background of two recent controversies about its alleged lack of effectiveness and violation of fundamental rights.

Indeed, virtually every sports association has its own national and international system of rules and arbitral bodies, at the head of which stands the Court of Arbitration for Sports (CAS) in Lausanne. The use of performance-enhancing substances has become one of the major regulatory problems facing sports law. In 1999, in an attempt to tackle this, the World Anti-Doping Agency (WADA) was set up, in the form of a foundation in Swiss law.^{Footnote 17} The World Anti-Doping Code (WADC) adopted at the second World Anti-Doping Conference in Copenhagen in 2003 came into force in 2004 and is an instructive example of private self-constitutionalization in action. Revisions of the code were adopted in 2007^{Footnote 18} (World Anti-Doping Agency 2009) and, most recently, on 15 November 2013 at the World Conference on Doping in Sport in Johannesburg.^{Footnote 19} This second full revision will be effective as of 1 January 2015 (World Anti-Doping Agency 2014).

The WADC represents a strong self-regulatory system by sports organizations to co-ordinate anti-doping measures worldwide. It is based on public authority exercised by non-state actors in non-state institutions. The particular self-regulation involved here in very great measure assumes legislative, executive and judicial state functions and replaces state regulation. With its anti-doping code, the system of control associated with this, and the international Court of Arbitration for Sports, WADA, sponsored by sports associations and the Olympic movement, is not only viewed as the ‘international police-force of high-level sports’;^{Footnote 20} it also gives the appearance of transcending all (sporting) sectors and acting as private lawmaker and judge. Its legitimatory requirements are further amplified by the fact that private self-regulation in sports seeks to achieve compliance primarily through the numerous sanctions listed in Articles 9 to 12 of the WADA Code (World Anti-Doping Agency 2009: 50–78) and only secondarily, for example, through ‘softer’ forms of governance such as exchange of best practices, or learning processes (World Anti-Doping Agency 2009: 13).

At the same time, the basically private and uniformized core part of this regime is supplemented by a fragmented, non-systematic public part which consists of a number of anti-doping laws at the national level and two intergovernmental conventions and one declaration at the international level. As early as 1989, in its Anti-Doping Convention the Council of Europe had already established ‘that public authorities and the voluntary sports organizations have complementary responsibilities to combat doping in sport’.^{Footnote 21} Several formulations in Article 4 leave no doubt as to the ‘whip in the window’, i.e. the threat of state regulation, which can count as a strong form of public ‘shadowing’ of the sports organizations’ self-regulatory efforts: ‘Parties reserve the right to adopt anti-doping regulations and to organize doping controls on their own initiative and on their own responsibility, provided that they are compatible with the relevant principles of this Convention’, or: ‘the Parties or, where appropriate, the relevant non-governmental organizations shall make it a criterion for the grant of public subsidies to sports organizations that they effectively apply anti-doping regulations’.^{Footnote 22} Article 7 entails a long list of measures to be taken by national and international sports organizations, among them ‘to introduce, on an effective scale, doping controls not only at, but also without advance warning at any appropriate time outside, competitions’.^{Footnote 23} In Article 8(2)(a) the state parties agree ‘to encourage their sports organizations to operate in a manner that promotes application of the provisions of this Convention within all the appropriate international sports organizations to which they are affiliated, including the refusal to ratify claims for world or regional records unless accompanied by an authenticated negative doping control report’.^{Footnote 24} Such ‘encouragements’ are admittedly weak instruments, but they are backed up by the threat of state regulation and the setting up of a monitoring group (Article 10 of the Anti-Doping Convention of the Council of Europe). To a significant degree the tightened rules for ineligibility by doubling the standard ban for serious doping offences from two years to four years in the revised Article 10.2 of the new World Anti-Doping Code which will come into force in 2015 (World Anti-Doping Agency 2014: 60–1) can be ascribed to the threat of mandatory state regulation.

The second intergovernmental convention was adopted at the 33rd session of the UNESCO General Conference in Paris in October 2005. UNESCO’s International Convention against Doping in Sport was the first-ever unified and globally binding international complex of anti-doping regulations. Largely based on the 2003 version of the WADC, it came into force in 2007.^{Footnote 25} On the one hand, in its preamble, the UNESCO Convention clearly states ‘that public authorities and the organizations responsible for sport have complementary responsibilities to prevent and combat doping in sport’. In Article 2 (‘Definitions’) it is also made clear that ‘in case of conflict the provisions of the Convention will prevail’. At the same time, in Article 4(1), states parties ‘commit themselves to the principles of the Code’ and, in various articles, to ‘facilitating the task of the World Anti-Doping Agency and anti-doping organizations operating in compliance with the Code’.^{Footnote 26} Moreover, the UNESCO Convention supplements the WADC with a commitment by the signatory states to institute suitable legal or other anti-doping measures within their own area of jurisdiction, for example against athlete support personnel (Article 9).

This indirect commitment on the part of the states, through the incorporation of private standards into international agreements, complements the shadow which state law casts over transnational private sports regulation with a remarkable ‘counter-shadow’ of private self-regulation over the state (Héritier and Lehmkuhl Reference Héritier and Lehmkuhl2008). This intermeshing of shadow and counter-shadow emerges particularly clearly in Article 22 of the WADA Code (‘Involvement of Governments’): ‘Each government’s commitment to the Code will be evidenced by its signing the Copenhagen Declaration on Anti-Doping in Sport of March 3, 2003, and by ratifying, accepting, approving or acceding to the UNESCO Convention’ (World Anti-Doping Agency 2009: 113). In the Copenhagen Declaration, ‘each participant recognises the role of the Code as the foundation in the world wide fight against doping in sport’ (Article 4.1) and ‘seeks to progressively adapt, where appropriate, their anti-doping policies and practices in sport to be in conformity with the provisions of the Code’ (Article 4.2).^{Footnote 27} The intention is thus that, via the UNESCO Convention and the Copenhagen Declaration, the signatory governments will indirectly pledge themselves to support the provisions of the privately instituted WADA Code and bring all other state-instituted anti-doping measures into line with it.

Siekmann (2012: 314) accurately describes this as a ‘regulatory hybrid’. However, not one of the sporting world’s international associations owes its existence or its scope for action to an international agreement in which the public sector devolves what were originally state competencies to the private sector or transfers state functions to it. There is therefore neither an officially decreed nor a negotiated instance of self-regulation here. In the absence of a mandate by the state or the community of states, the accepted benchmark when it comes to the exertion of influence by these is freedom of association, which generally enjoys constitutional protection. It imposes clear limits on state intervention and restricts the remaining isolated rights of intervention to the protection of basic freedoms and the rectification of any anti-competitive effects of private self-regulation.^{Footnote 28}

The legal status of the WADA as an entity in Swiss law, and the terms providing for half the WADA budget to be financed from the public purse, in themselves constitute a form of state backup of the regulatory initiatives in the anti-doping sphere.^{Footnote 29} Decisions of WADA’s sports court – the Court on Arbitration for Sport, CAS – are subject to the laws of the country in which the court is based – Switzerland. CAS awards may be challenged before the Swiss Federal Tribunal, for example, by invoking the so-called public policy exception of the Swiss Private International Law Act.^{Footnote 30} States may not sign up to the Code. Instead, the anti-doping regulations of the various sporting associations are generally incorporated into international agreements. The sporting associations’ codes were first recognized by the world of states in the Copenhagen Declaration on Anti-Doping in Sport, which was adopted in 2003 at the intergovernmental conference of the same name and was ratified by virtually every state in the world.^{Footnote 31} By means of this declaration of moral and political commitment, the ratifying states affirmed their commitment to the provisions of the WADA code and undertook both to support their national anti-doping organizations in implementing it and to gear their own anti-doping measures to it. Sporting organizations ‘that are not in compliance with the Code or applicable anti-doping rules adopted pursuant to the Code’ will lose all or part of their state funding. This represents not only a recognition, through juridification, of private standard-setting in (international) law but also, conversely, a self-imposed undertaking by states to adhere to regulation created by private actors.

The upshot of all this is a highly complex interplay of private and public co-regulation, with numerous interactions between, on the one hand, the rules generated, executed and monitored by self-regulation and, on the other, the standards set by national and international law. The ‘autonomy of sport’ plainly only operates within the framework accorded to it by (inter-)national law. However, it is not something conferred ex ante by national law, but the space in which it has been able to develop is one afforded by the state, and it is, in addition, underpinned and backed up by explicit ex post recognition in (inter-)national law. The role of the state must thus be described as ‘laissez faire’ and ‘reactive’ in terms of the types of state embedment distinguished for heuristic reasons in Table 1; it is complementing rather than supplanting. Nonetheless, it is a significant one: ‘National and international legal norms guarantee sport a regulatory power that allows it to set its own rules. Sport is not a state with its own decision-making powers or competence-competence. Its regulatory authority is not inherent; it is a competence derived from states or communities of states’ (Nolte Reference Nolte, Bumke and Röthel2012: 116).^{Footnote 32}

Having substantiated that the state/world of states is still in the game as the ultimate authority, it can also be concluded that the responsibility to procure the legitimacy of the public authority exercised by private self-regulatory regimes has not completely shifted to the sports organizations either. In fact, there is a constant need for some kind of state involvement in the lex sportiva in order to compensate for deficiencies in regard to legitimate authority. To make this clear, we will draw on the criteria which were introduced earlier as playing a constitutive role for the establishment and exercise of legitimate authority, i.e. public-interest effectiveness and efficiency, responsiveness, self-determination and accountability.

IV. Summary and generalizing conclusions

The case of the lex sportiva accentuates the need for new forms of checks and balances to assure that private self-constitutionalization meets certain legitimacy demands. The normative criteria applied here to sporting self-regulation demonstrate that although the latter has taken over many functions normally reserved to state-based political authority, and created extensive legitimatory mechanisms to justify these, it nonetheless remains dependent on a governmental/inter-governmental ‘shadowing’ that ensures a link back to the common good and the rule of law – a link only the state/states can provide. The anti-doping measures enshrined in the WADA code reflect complex interconnections between private self-regulation and the (inter-)governmental political structure in which that regulation is embedded. The exercise of private transnational authority operates in a space afforded by national and international law. The lex sportiva shows how the interaction between existing regulatory arenas who share the responsibility to procure the legitimacy of transnational private self-regulation can improve the legitimacy record of sporting self-regulation. In the realm of sports, this framework has produced hybrid links between private self-regulation and the public fabric in which this is embedded. The spheres of public and private regulation are closely intermeshed – in judicial or arbitration proceedings, for example where sport assumes primary responsibility for ‘prohibiting doping by means of its own system of rules, punishes violations via its own sports-based judicial procedures’ while the state ‘assumes a secondary guarantor function’ by providing a legal framework for this (Nolte Reference Nolte, Bumke and Röthel2012: 8). The functional division of labour between public regulation and private self-regulation in this area is based on the autonomy of sporting organizations, assured by international law and operating within the framework of international agreements. This does not exclude the possibility of norms relevant to sport also being incorporated into (inter-)national laws, or of higher-order inputs to governance also taking place via the self-regulation route. Thus, constitutional-style ‘rules for rule-making’ may very well also form part of transnational private self-regulation initiatives and codes of conducts within sport itself. However, such rules remain subject to public monitoring. A key function is thus retained by the state as the ‘ultimate accountable entity’ (Reissinger Reference Reissinger2010: 21 and 30).

The case of private self-regulation in sports confirms the state’s indispensable constitutionalizing function in guaranteeing a minimum of procedural requirements. Transnational private self-regulation invokes such guarantees on several levels here, among others in relation to support in the area of implementation or in regard to safeguarding the security of affected third parties. In this sense, public guarantee functions and private norm-setting and norm-enforcement can indeed be viewed as interdependent ‘two-way safety-nets’ (Röthel Reference Röthel2007: 763). The observations made here about the (inter-)governmental ‘shadowing’ of transnational private self-regulation corroborate the notion that (inter-)national influence is, in principle, both necessary and possible. When private actors assume public functions this does not unburden the state/world of states of their ultimate responsibility to mitigate the problems of legitimation associated with transnational private norm-setting and norm-enforcement by:

• • integrating the rules generated, executed and monitored by private actors into a national/international legal system that can exert a facilitating, corrective, or prohibitive influence on them in cases where fundamental rights may be violated or public-interest objectives may not be attained (‘external legitimation’)

• • at the same time helping to mobilize the sources of legitimation which private self-regulation has available to it in and of itself (‘internal legitimation’), which it does by –

• • guaranteeing space for the politicization, by society, of the deficiencies of private self-regulation and of its (inter-)national regulation.

The question of what actual combination of these core elements of regulated self-regulation is needed to deal with particular deficits of legitimacy can only be answered on a case-by-case basis. Some generalizing conclusions may, however, also be drawn from the sports world’s self-regulation. The state is still in the game and has a wide range of proactive, reactive and laissez-faire instruments available to it in discharging this responsibility, be it unilaterally or collectively. While, in their constitutionalizing function, the state and the international political order constituted by states, remain the ultimate reference points for procuring the legitimacy of global governance arrangements, a substantial part of the burden of – initial –legitimation can and must be carried by those directly involved in private self-regulation. The need for public regulation depends decisively on the extent to which private self-regulation, as a form of exercise of power open to abuse, takes initial regulatory action in regard to itself or, where appropriate, imposes limits on itself in order to satisfy the legitimatory requirements set out here. Vice versa, this initial regulatory action which can provide private self-regulation with a certain internal legitimacy remains equally indispensable when public regulation by the state/world of states fails because public actors are too weak, unwilling or their interferences suffer from legitimacy deficits themselves. External (public) and internal (private) sources of legitimacy need to be combined to compensate for one another’s weaknesses.

The way in which the lex sportiva is embedded not only in national but also in inter-governmental legal systems highlights the fact that, rather than there being one single place from which the shadow of state hierarchy can be cast across transnational private self-regulation, the levels at which the states/world of states can exert influence may be located in the international or the particular national sphere. How far they will actually fulfil the safety-net functions that fall to them, and under which conditions they will make practical use of the leverage theoretically available to them, is another story. The financial crisis has demonstrated how large this gap can be. Even in the world of sports, there remains a considerable gap between responsibility, on the one hand, and capacity and willingness for meaningful and effective involvement according to the guarantee functions listed above, on the other. Thus the case of the lex sportiva does not only illustrate the omnipresence, but also the limits of public ‘shadowing’ in regard to whether public guarantee functions are adequately exercised or not. Identifying an ‘ultimate accountable entity’ is one thing; putting its responsibility into practice is another. The political order of the international society as construed in neo-Westphalian terms provides a dispersed and fragmented constitutional-style legal framework with little reliable guarantees that states are capable or willing to enact their background role.

A basic consensus about the norms (or institutional ‘software’) that are generally recognized as central constitutional principles may still exist. However, the process of global constitutionalization that is currently going on within the neo-Westphalian setting does not yet reveal a realistic blueprint for a less fragmented, integrally connected and more reliable state-based ‘backup’ architecture (or institutional ‘hardware’) than the presently operating constitutional pluralism for monitoring the public authority exercised by way of transnational private self-regulation. In the meantime we have the postulate of private self-constitutionalization as an entity subject to regulation located in the midst of multiple sites of constitutional authority within a state-based patchwork of constitutionalizing rules that carries the ultimate responsibility to procure its legitimacy but suffers from a lack of instruments to make the state(s) live up to their constitutional responsibility in practice. Rather than being able to offer a ‘grand solution’ I join those who are still on the way to improve the ‘foundation on which grander ‘‘how to’’ proposals can be built’ (Black Reference Black2008: 137).