Читать книгу Claves del derecho de redes empresariales - AA.VV - Страница 9

CAPÍTULO 2

Network Constitutions: a Response to the Crisis?

GUNTHER TEUBNER

INTRODUCTION

In recent years, transnational corporations (TNC) were involved in a number of scandals that shocked the global public. Ecological catastrophes, like the Exxon Valdez, Shell in Nigeria, inhuman labour conditions, child labour, the repression of union members, the disastrous price policy during the Aids crisis in South Africa, the complicity of transnational corporations in corruption and human rights violations, and with special vigour, the recent corporate abuses in the banking crisis, drastically increased the public awareness of the negative effects brought about by the transnationalization of commercial activities. In parallel, these ramifications triggered a plethora of political initiatives aimed at regulating them through binding legal norms. However, both the strong resistance of transnational corporations against national and supranational regulations as well as the difficulties to achieve effective regulation via protracted international agreements led to the failure of many of these initiatives.²¹ Nonetheless, one result of this shortfall is particularly noteworthy. Instead of the aspired binding state regulations, a different species of transnational non-state legal regimes spread in huge numbers around the globe — the “voluntary” codes of conduct of transnational corporations.²²

Today, these codes exist in various forms, yet two basic variants predominate. On the one hand, the heavy public criticism, disseminated by the media globally, and the aggressive actions of protest movements and civil society non-governmental organizations (NGOs) force numerous transnational corporations to develop corporate codes “voluntarily”. They commit themselves to standards in the issue areas labour conditions, product quality, environmental policies, consumer protection and human rights and promise their implementation (short and imprecise: “private” codes). On the other hand, the state world establishes — through agreements under international law or through the norms of international organizations — codes of conduct for transnational corporations (again, short and imprecise: “public” codes).

How are the subjects of these private and public codes defined? It would be a misunderstanding to determine them as single formal organizations—TNCs, on the one side, and international organizations, on the other. Rather, a network revolution has taken place in both legal spaces. Extensive networks have developed between different organizations, which then facilitate the understanding of the entire configuration as the relationship between two different, mutually closed networks. On the one hand, private codes have already transcended the confines of individual companies. They have extended their validity to corporate groups and conglomerates that transcend national boundaries and encompass in some cases thousands of individual companies. Under pressure from the public and civil society organizations, their scope was even extended beyond the boundaries of corporate groups. With contractual regulations, powerful corporate groups are able to bind their suppliers and their distribution chains to their corporate codes and use the contractual mechanism also to introduce effective monitoring and sanctioning systems.²³ On the other hand, interlinks in the codes of the state world can be found. Here, manifold connections between the corporate codes of the ILO, the OECD, the United Nations Organizations, and the European Union have emerged.²⁴ Insofar the parlance of codes of corporations is misleading. This would suggest a world in which the markets are populated by isolated companies. Today’s universe of interconnected enterprises is rightly called a network society. And the adequate expression would be codes of corporate networks.

In many cases, “public” codes for corporate networks remain mere recommendations with no effects whatsoever. And the self-commitments in “private” codes are often only strategic attempts to pre-empt state regulation by the non-binding declaration of intent, or they are mere public relations strategies without involving any effective change of behaviour.²⁵ But there are some empirical studies which deserve particular attention. They demonstrate that in some cases the corporate codes brought about real change; hence that they improved labour conditions, increased environmental protection and pushed through human rights standards.²⁶ It is particularly noteworthy that these studies do not only document success stories; they also specify the social and legal conditions that have to be given, if the codes are meant to be successful.²⁷ Permanent NGOs monitoring or binding contracts with civil societal certification bodies are likely to be among the most important conditions for success.

What is special about the interplay between private and the public codes for corporate networks? My thesis is: As a reaction to recent crises not only tendencies of a private juridification but also of a constitutionalization beyond the state take place. While the debate on legal pluralism, private ordering, and transnational governance concentrate on the question whether new forms of law without the state are emerging in the globalization process, I would like to focus on the question whether in reaction to the crisis stateless societal processe are also developing genuinely constitutional norms. As I have extensively argued elsewhere, this point is based on a concept of constitution that is not limited to the nation state and implies that also non-state societal orders develop autonomous constitutions under particular historical circumstances.²⁸

I try to support this thesis with the following arguments, meant to highlight that the codes of corporate networks feature functions, structures and institutions of genuine constitutions, network constitutions:

1. To the extent that private and public codes juridify fundamental principles of a social order and establish rules for its self-restraint at the same time, they fulfil central constitutional functions for corporate networks.

2. As one among several reactions to the present crisis, a shift within these network constitutions takes place, a shift from constitutive to limitative rules.

3. With their characteristics of double reflexivity and binary meta-coding, both network codes develop genuine constitutional structures.

4. As constitutional institutions, the two codes do not form a hierarchy of public and private constitutional rules, but an ultracyclical linkage of qualitatively different networks of constitutional norms.

Thesis # 1: Constitutional Functions: Constitutive and Limitative Rules

Codes for corporate networks take part in two opposing waves of constitutionalization of the world markets. Advancing Karl Polanyi’s ideas about the transformation of modernity, one can even argue that transnational constitutionalism is part of a “double movement”.²⁹ Also in the trajectory of corporate constitutional law, the expansion of economisation is accompanied by counter movements, which reconstruct the “protective covering of cultural institutions”.

The first constitutional movement is identified by neo-materialist critics of a “new constitutionalism” as well as by ordoliberal advocates of a world economic constitution, naturally with diametrically opposing evaluations.³⁰ The Washington consensus of the last thirty years has pushed ahead politically with this first surge of constitutionalization of the world markets. It did not only trigger political regulation but fundamental principles of economic constitutionalism. They aimed at providing worldwide operating corporations with unlimited latitude for action, which encompassed ending governments holding shares in corporations, combating trade protectionism and freeing commercial enterprises from political regulations. In this vein, the International Monetary Fund and the World Bank have developed regime constitutions, whose guiding principle is to open up national capital markets. An integral part of these constitutionalization tendencies is the corporate governance of multinational corporate networks, whose principles encompass a high degree of corporate autonomy, the capital market-orientation of company law, and the establishing of shareholder value.³¹

This wave of “neo-liberal” constitutionalization is clearly characterized by its constitutive function, i.e. its focus on providing transnational corporations with a high degree of autonomy. It is fixated on the problem that the worldwide extension of corporate activities is hampered by the segmentary differentiation of the world into nation states. Constitutive rules of this kind serve for releasing the dynamic of commercial enterprises at the global level.

Thesis # 2: Externally Enforced Constitutional Self-Restraint

In the long run, however, it is not sustainable that a corporate constitutionalism restricts itself to its constitutive function in such an one-sidedly “neo-liberal” fashion. It is only a matter of time before the systemic energies released trigger disastrous consequences — alongside their indubitably productive effects. Now a fundamental readjustment of constitutional politics will be required to deal with the outburst of social conflicts. Globalized markets and corporations put — without being significantly hampered by nation state counter programs — a strain on society and the environment through the “negative effects of their own differentiation, specialization and high-performance orientation”. This is the moment when Polanyi’s “dual movement” makes its presence felt, which, as Streeck argues, identifies

“… the not just plural but inherently contradictory forces responsible for the specific dynamism of capitalist development, making it move, not linearly, but in its fits and spurts, and in cyclical waves of institutionalization and de-institutionalization”.³²

In such processes of “dynamic disequilibrium”, which alternate between liberation and limitation of systemic energies, the tipping point has now been reached. After a long constitutive phase, combating the risks of unrestrained liberalisation has now become indispensable. Limitative constitutional norms are now needed rather than constitutive ones.

This is the situation after dismantling nation state regulations at a transnational level. While global function-specific communication is no longer hindered by nation-state production regimes, the constitutive constitutional politics of the Washington consensus has overriden many of the limitations that nation-states placed on the dynamics of the function systems. Unburdened by nation-state restrictions, the systems are now placed to follow, globally, a programme of maximising their partial rationality. Despite they differ in their theory assumptions, sociological analyses in the tradition of Karl Marx, Max Weber, and Niklas Luhmann all agree on the consequences of this diagnosis. Whether the laws of motion of capital, or the rationalisation of spheres of social action, or the dynamics of functional differentiation — all identify the destructive energies created by the one-sided function-orientation of a social sector. Globalisation has an accelerating effect. The dismantling of national production regimes releases destructive dynamics in the global systems; destructive dynamics in which the one-sided rationality-maximisation of one social sector collides with other social dynamics.

Now, it is imperative to readjust constitutional policy. In the second wave of constitutionalization, instead of the constitutive, the limitative function of constitutional norms is in demand. As one among many reactions to the crisis, the codes of corporate networks partake in this second wave when they restrict corporate activities in the name of public responsibility. They try to overcome the primacy of shareholder value in favour of a stakeholder-orientation as well as to realize self-restraint in the areas of labour, product quality, environment and human rights.³³

This is the message of societal constitutionalism. Not only measures of public regulation, but also the “private” constitutions of corporate networks are faced with the task: How can a sufficiently large degree of external pressure be generated on the corporate networks to push them into self-limitations on their options?

Why self-limitation and not outside limitation, though? Does not experience show that self-limitation merely serves to set the fox to keep the geese and that excesses can only be prevented through outside influence? Equally, though, does not experience also show that attempts at trying to control internal processes through external interventions regularly end in failure? At this point societal constitutionalism does a difficult balancing act between external intervention and self-direction. A “hybrid constitutionalisation” is required in the sense that in addition to state power, external societal forces — that is, formal legal norms and “civil society” counter-power from other contexts (media, public discussion, spontaneous protest, intellectuals, protest movements, NGOs, trade unions) — exert such massive pressure on the expansionist corporatist networks so that they will be constrained to build up internal self-limitations that actually work.

However, workable limitations can take effect only within the system’s own logic, not outside it.

“Every function system determines its own identity … elaborating semantics of self-interpretation, reflexion, and autonomy. The mutual dependencies of the subsystems can no longer be normed in general. Indeed they can no longer be legitimised at all as a condition for order at the overall social level.”³⁴

The difficult task of co-ordinating the function of a social system and its environmental tasks at a sufficiently high level can be tackled only through system-internal reflexion, which can certainly be prompted from the outside but cannot be replaced. This is why there can be no external political definition of transnational corporate constitutions, but only indirect political impulses or constitutional irritations. The knowledge regarding which kind of self-limitation can be selected does not even exist as such. It cannot simply be accessed, but rather has to be generated internally first. Endogenous growth imperatives can be combated only with endogenous growth inhibitors. The knowledge required to do so cannot be built up by an external observer as centrally available experiential knowledge, but only out of the combined effect of external pressures and internal discovery processes.

No one knows how such a capillary constitutionalisation might work in practice. Ex-ante prognoses are by definition impossible. And, for that reason, there is no alternative but to experiment with constitutionalisation. The application of external pressure means that the self-steering of politics, or law, or other subsystems, creates such irritations of the corporate networks, that ultimately the external and internal programmes play out together along the desired course. And that cannot be planned for, but only experimented with. The desired course for corporate network constitutions is limitations of the endogenous tendencies towards self-destruction and environmental damage. This is the core of the constitutional problematic, this difficult handling of the corporate networks’ self-transformation.

Thesis # 3: Constitutional Structures: Double Reflexivity and Binary Meta-Coding

1. Structural Coupling of Reflexive Mechanisms

The codes would not establish constitutional structures, if they only introduced primary rules governing corporate activities in the fields of labour, environment and human rights. The critical threshold is reached, when the codes lay down secondary rules concerning the identification, interpretation, amendment, competences for the enactment and delegation of primary rules. Typically, codes of corporate networks show a three-tiered hierarchy, in which the interplay between primary and secondary rules is discernable indeed. The top level consists of the general principles of the corporate constitution; the middle level regulates enforcement and monitoring; while the lowest level includes concrete instructions for conduct.³⁵ At the top and middle level, a plethora of such secondary rules can be found. They come close to constitutional norms in the strict sense, since they produce as higher-ranking meta-norms a sort of reflexivity of intra-company law. But secondary norms as such do not constitute yet a constitution.

Only the peculiar double character of corporate codes, which I call the double reflexivity of legal norms and social structures, turns them into constitutional norms. If law plays a supportive role in the self-constituting of a social order beyond its function of conduct control, dispute settlement, regulation and frame setting, it creates constitutional law. A corporate constitution in the strict sense only emerges, if a structural coupling of a particular kind comes into being between the corporate organization and the law.³⁶ Transnational corporate constitutions link reflexive processes in the economic organization with reflexive legal processes; in other words, they link fundamental principles of the organization with secondary legal rules.³⁷

An autonomous, non-state, non-political, and hence genuinely societal constitutionalization occurs in the codes of transnational corporations, since they juridify reflexive social processes that concern the relationship of the company with its environments by linking them to on their part reflexive legal processes, i.e. standardizations of standardizations. Under this condition, it is reasonable to talk of elements of a genuine constitution within the corporate codes of transnational corporations. The codes show indeed typical elements of a constitution: regulations concerning the establishment and practice of organizational decision-making (procedural rules of the corporation) and the definition of the system boundaries (fundamental rights of individuals and institutions vis-à-vis the corporation).

The norms at the top level of corporate codes are especially geared towards these conditions. They regulate the fundamental decision-making processes of transnational corporations, which concern the relationship with their human and natural environments, especially the relationship with the employees whose fundamental rights are respected by the organization. The “guidelines” at the top level have constitutional character, since they are not only mere behavioural norms, like the rules at the lowest level. Rather they are explicitly higher-ranking norms, phrased as general principles and serving both as starting points for intra-corporate norm-generation and as yardsticks for the internal and external review of norms. This requires certain institutional arrangements, especially procedural roles, which are responsible for setting, modifying, interpreting and implementing the primary rules. It is therefore especially the middle level of control and implementation bodies that mediates between abstract principles and concrete corporate decisions. Thus, private codes do not only generate autonomous law as private ordering; at the same, they constitute their own constitutional foundations without being dependent on public codes — they generate literally constitutions without the state.

2. Binary Meta-coding of the Corporate Constitution

The endpoint of the constitutionalization of a corporation is reached, when a specific binary meta-coding develops. The meta-coding oscillates between the values “code-compatible”/ “code-adverse”, both with regard to the corporate constitution. A meta-coding exists in this case, because such a constitutional code subjects the already binary coding of intra-company legal norms to an additional examination, namely whether they conform to the requirements of corporate constitutional law. Here, the hierarchy between simple and constitutional law emerges, which is typical for all constitutions. The legal code (legal / illegal) is subordinated to the constitutional code (constitutional / unconstitutional). However, there is something peculiar to the constitutional meta-coding. It is not only hierarchically superior to the legal code but at the same time also to the economic code. It therefore subjects to reflection all economically binary coded operations of the corporation, whether they comply with the principles of the public responsibility of the corporation or not.

The constitutional meta-coding is therefore a hybrid. It serves as a fictional unity for two different constitutional reviews within the corporation. It is, on the one hand, placed hierarchically above the legal and, on the other hand, above the economic binary code. Therefore, it assumes a different meaning depending on whether it reviews the economic or the legal code. In economic contexts, it serves for reflecting the social responsibility of the company and seeks to identify strategies for environmentally friendly economic activities. In the context of corporate law, it introduces the distinction between simple and constitutional law and reviews simple legal acts for their compliance with the values and principles established in the corporate constitution. The meta-coding triggers the re-entry of fundamental principles of economic organization into the law as constitutional principle and vice versa the re-entry of law in the corporate organization.

Thesis # 4: Constitutional Institutions: Private and Public Codes in an Ultracycle

Even if in this way constitutional functions and structures can be identified, it remains still difficult to capture the institutional structure of corporate codes in more theoretical detail. Some authors describe them as the “new sovereignty” of transnational corporate networks and stress thereby their unrestrained self-regulation.³⁸ However, this does not do justice to their numerous normative dependencies on the environment; because the currently relevant corporate codes emerge from the conflicts between three groups of actors —, civil society groups, transnational corporations plus their delivery and distribution organizations and supranational institutions — whose mutual relations remain however unclear.

1. The Inversion of Nation State Hierarchies

Other authors try to model these relations as “governance triangles”.³⁹ This is, however, similarly inadequate for grasping the social embedding of the codes. It suggests mistakenly that a transnational equivalent emerges to the state-organized neo-corporatist triangle of the European welfare states. In comparison to the nation states, one is here however confronted with a totally different constellation in the relation between these three social forces. Also the model of “multi-level-governance” is hardly appropriate for capturing the peculiar interplay of the two transnational types of code.⁴⁰ In the nation state, corporate constitutions could certainly be conceived as a multi-level arrangement of constitutional norms, legal and judicial rules, on the one hand, and intra-organizational private ordering on the other. But its transfer to global corporate constitutions is mistaken. The different conditions of the transnational as well as the results of the first wave of societal constitutionalization, especially the high autonomy of transnational corporations, have fundamentally changed the relations between public and private collective actors compared to the corporate constitutions of the nation state. In the drastic words of an observer:

“Contract replaces law; networks of relationships replace a political community; interest replaces territory; the regulated becomes the regulator.”⁴¹

In the corporate constitutions of European nation states, as is well known, the linkage between public and private norms took place in hierarchical formations. The corporate constitution was based on a clear primacy of the state in the form of constitutional, statutory and judicial norms. The private ordering of corporations remained clearly subordinate to state law; it remained limited to those spaces of autonomy state law had left. This hierarchy of norms can be captured in the conceptual pair hard law / soft law.⁴² The state enacts hard law in company law, in the law of co-determination and in regulation law in the form of binding and sanction-reinforced norms. In contrast, intra-corporate norms are only a kind of soft law. As a manifestation of private autonomy they are not recognized as genuine legal norms, because their obligatory nature and enforcement depend on state recognition, and because they are subject to the review of state courts, whose results often repeal and change them.

In comparison to this traditional hierarchy, one can detect significant changes in the transnational codes which do not match the standard categories. In the interplay of the two corporate codes, a downright inversion of the hierarchy between state law and private ordering can be observed. A dramatic reversal takes place especially in the hard-law / softlaw quality of the public and private corporate codes: Now, it is the state norms that feature the quality of “soft law”, while the mere private ordering of transnational corporate networks emerges as new forms of “hard law”.

The norms under international public law, which, for instance, the UN enacted in the Codes of Conduct for Transnational Corporations, are not comparable with the binding norms passed for the corporate constitution by parliaments and constitutional courts of the nation states. Although it was initially planned in the 2003 “Draft Norms on the Responsibilities of Transnational Corporations” that a supranational regulatory body should directly regulate the conduct of transnational corporations with the help of sanction-reinforced norms binding under international law,⁴³ the massive resistance of influential nation states and of the corporate lobby marked a turning point. The finally passed version contained merely “soft law”: non-binding recommendations whose implementation cannot be enforced by legal sanctions.⁴⁴

On the other hand, intra-corporate network codes are merely non-state private ordering, but in fact they are the governing law of the land with a high degree of binding force and effective sanctions. Private law doctrine still contests vehemently their genuine legal character, since it insists upon the state deduction of normative validity claims and does not recognize private ordering as law. And it is only gradually that economically and sociologically inspired concepts of law emerge, which assign legal character to the normative orders of private transnational actors — under particular circumstances.⁴⁵ Intra-company and inter-company codes are directly binding for the actors involved, and they are equipped with effective sanctions, which are executed by newly created compliance departments.

In this way intra-corporate and inter-corporate organizational law seals itself off from the state law. In direct opposition to the usual norm-hierarchical relation between state and private law, public codes do not work as the constitutional basis for the authorization of the private codes. They produce their own validity from the linkage of primary and secondary norms in the realm of private ordering. They constitute a closed non-state system of legal validity, which is itself structured hierarchically. As already mentioned above, the top level encompasses the general principles of the corporate constitution, the middle level regulates enforcement and monitoring, the lowest level includes concrete rules of conduct. They thus generate their authorizing foundation themselves by their own constitutive rules. And intra-company rules, which regulate conduct according to the legal code, are reviewed themselves according to the constitutional code.

2. Hypercycle and Ultracycle

Inversion of hierarchy does still not go far enough. While there is a clear factual and normative primacy of the private over the public codes, the primacy is of no hierarchical nature. Rather more appropriate is the exclusion of the public by the private. State norms are not subordinated to private norms, rather they are banished from the interior of norm setting into the corporate environment. With this, the notion of a unitary legal space of state and private rules becomes obsolete. Instead, two independent legal spaces develop, an autonomous, privately ordered, coercive inner law of corporations and a state regulated ensemble of normative recommendations of conduct.

While it seems obvious that they form two mutually closed legal spaces it is not easy to determine what constitutes their closure. In any case, it is not the operative closure of social systems in the strict sense that separates them. Their closure is not based on the difference between their operations, as both code-orders are processed by the same type of operations — legal acts. Rather, it is a mutual structural closure that arises from two differences. One is the strict limitation of their space of validity: private codes claim validity for the members of the transnational corporations, public codes claim validity for the contracting states. The other is their different quality, as binding norm, on the one hand, and as mere normative recommendation, on the other. In terms of systems theory: The inner differentiation of the global legal system arises not through the emergence of a new kind of legal operations which would trigger an operative closure between the newly created subsystems. Rather the validity symbol is transferred in such a way that it creates boundaries between different legal orders. It brings about the structural closure by defining boundaries between different spaces of validity. Traditionally, the validity spaces are defined by territorial boundaries as in nation states, region or cities, in the transnational context they are of a issue-specific, functional or jurisdictional kind. Hence, one has to distinguish clearly between different forms of closure, operative and structural, which consequently also result in different ways how legal orders open up to each other. In this way, the private and public codes constitute two mutually closed legal orders, between which no transfer of validity takes place, but which influence each other in different ways.

As I said at the beginning, in both legal spaces extensive normative networks have developed between different organizations, which then allows to understand the entire configuration as the relation of two different, mutually closed normative networks. The interrelation of these two closed code-networks certainly does not match the traditional relationship between private and public corporate constitutional norms. Often, it is therefore attempted to conceive the relationship itself as a single large network or even a network of networks, as a metanetwork between state and private actors.⁴⁶ This is not necessarily mistaken, but relevant differences disappear. The relation can be captured in more detail in the difference between “hypercycle” and “ultracycle”.⁴⁷ A hypercycle emerges when communicative operations within a closed network form cycles that are interlinked in a circular way. In contrast, an ultracycle emerges when a cycle of mutual perturbations is developed between closed networks. Within private codes of corporate networks, interlinkages are of a hypercyclical nature; between the cyclical legal operations, which connect to each other within different formal organizations (i.e., TNCs, their suppliers, and their sales organizations), interorganizational direct connections are developed. The validity symbols of private ordering are directly transferred via intraorganizational law and interorganizational contracts. Within this network of private legal operations, the private norms have a direct binding effect on participants and in instances of norm infringements, sanctions are ordered. In this way, a closed scope of private ordering emerges through the hypercyclical linkage of TNCs and other commercial enterprises.

In an entirely different way, these mutually interlinked private codes are connected to public codes. For this kind of connection, unlike the model of the hypercycle, the ultracycle model is appropriate. Although public codes define certain politically desired obligations and establish the boundary between permitted and banned activities, they are only informal recommendations and mere appeals for certain conduct. They are also valid law, yet in a paradoxical form; they are law in force but without legal sanctions. This means that private codes, which present themselves as self-referentially closed validity circles, are not only entirely independent from public codes when it comes to their validity, but also that public codes cannot even connect normatively to private standardizations. They do not participate in the normative unity of the intra-corporate and intercorporate codes. Only from the outside, they can appeal, suggest, motivate, urge, or even compel, but they cannot command or suspend validity. They are only external irritations for the inner-validity cycle of private codes. The codes of the United Nations, the ILO, the OECD, and the European Union are mere constitutional impulses, which—certainly with great influence—international organizations send toward TNCs. Whether they indeed coagulate there into binding constitutional norms is not decided by the institutions of the state world but in the inner processes of private organizations.

3. Learning Pressures: Internal Self-limitations due to External Constraints

If under certain circumstances, interrelations between the network codes emerge, then an ultracycle arises, a perturbation cycle between the public and private codes. In the usual descriptions of how private and public soft-law regimes are interrelated, the fundamental difference is concealed which exists between the hypercycle of private codes and the ultracycle of public and private codes. This should, however, not tempt one to dismiss ultracycles as mere political window dressing, as mere white noise of the state world, unable to affect the intra-company codices at all.⁴⁸

What matters are learning pressures, i.e. internal changes induced by external constraints. Both elements have to be present in order to enable public and private codes to act in combination: an internal change of cognitive and normative structures and external pressure directed towards it. Otherwise ultracyclical linkages do not emerge, and public codes remain external impulses with no effects. Here, the above-mentioned special quality of mutual closure becomes apparent which makes possible a special quality of mutual opening. A transfer of validity between both is out of the question, instead learning pressures, that is other mechanisms of mutual opening, are developed.

At this point, one of the most significant changes in the legal structure becomes visible, which occurs in the transition to world society. Niklas Luhmann described this change in the following way:

“at the level of the consolidating world society, norms (in the form of values, regulations, purposes) do not anymore steer the pre-selection of the cognitive, rather vice versa the problem of adaptation through learning gains structural primacy and the structural conditions for the capability of all subsystems to learn have to be supported.”⁴⁹

This means that the code-orders of the two networks do not simply communicate via the medium of law. The validity of normative expectations is not transferred from one network code to another via legal operations. Instead, learning processes of corporate legal codes are triggered, often even compelled, by non-legal media — by expertise, by political and societal power, by normative persuasion as well as by monetary incentives and sanctions. In this context, cognitive primacy does not mean that corporate codes lose their legal-normative quality and only function as mere cognitive expectation. It is only the relations between the two normative networks that become de-normativized. While the codes themselves remain normative orders, the relations between them switch to cognitive mechanisms.

What does the first element of learning pressures — cognitive learning — consist of? The public codes only provide “templates”, behavioural models, principles, best practices, recommendations for the private codes. The ultracyclical linkage of both network codes triggers learning processes, which do not take place as validity transfer of rules within one legal order but run across the boundaries of mutually closed orders. Their particularity is that they do not amalgamate the involved orders into one unitary legal order with common legal operations, rather they are reconstructed via complex cognitive processes.⁵⁰ It is exactly this separation that makes possible a cognitive surplus value, which is generated when the sparks of perturbations jump across the boundaries between the involved codes. This may even lead to normative innovation. The ultracycle does not end their autonomy, rather it uses their autonomy to produce new norms, both of hard law in intra-company codes and of soft law in the codes of the state world.

What makes the learning effect special? Corporate groups can use the public codes to gauge what societal expectations they face, without having to follow them completely. In this way the public codes counterbalance the tunnel vision developed by the private codes and provoke their re-orientation towards a transnational public policy. Public codes provide constitutional stimuli for learning similar to the normative demands placed on corporate networks by protest movements and civil society organizations.

What does the second element — pressure — consist of? In this learning process, legal sanctions do not play a prominent role. Rather extra-legal mechanisms are responsible for the effect that corporate networks take public codes as compulsion for learning, and develop their own codes for their particular circumstances. In no way are these extra-legal mechanisms inferior to legal sanctions. First and foremost, it is inter-organizational power processes — unilateral pressure and political exchange — that force commercial enterprises to develop corporate codes. It cannot be stressed enough that this external pressure is an indispensable condition for corporate codes to exert an effect at all.⁵¹

According to the hitherto existing experience, nation states and the international organizations of the state world have generated the necessary power resources, yet only to a certain extent. The power pressures of a public network, of protest movements, NGOs, unions, non-profit organizations and public opinion have proven to be crucial. Economic sanctions often tip the scales. The sensitivity of consumers, of whose purchase behaviour corporations are dependent upon, and of certain groups of investors, who exert economic pressure on the commercial enterprises is decisive.⁵² It remains to be seen, whether the state world will take a leading role in exerting stronger external pressures on corporations after the financial crisis. In this context, the latest news rather feed scepticism. In any case, although they would change the balance between internal and external regulation, they would not make disappear the difference between hypercycle and ultracycle.

Behind the metaphor of “voluntary codes”, therefore, lies anything but voluntariness. Transnational corporate networks enact their codes neither on the basis of their understanding of common good requirements nor due to motives of corporate ethics. They comply only “voluntarily”, when massive learning pressures on them are exerted from the outside. The learning process does not proceed within the legal system from code to code via validity transfer of rules, but on a long-winding detour through other social systems and other media of communication. It is not sufficient to describe this as if legal sanctions are simply replaced by social sanctions. This would conceal the drastic effects such circuitous learning pressures have. In the described ultracyclical “translation processes”, system boundaries are in fact transcended; a perturbation cycle emerges between legal acts, pressures of political and societal power, cognitive operations of epistemic communities, normative persuasion and economic sanctions, which then goes back to legal acts in the other code. The original content of the public recommendations is dramatically changed, when they undergo a complicated translation process into different worlds of meaning. When the soft law of the public codes is “translated” into the language of expert knowledge, which develops models and organizes monitoring, this creates special effects. Different outcomes occur when it is translated into the inter-organizational power of political negotiations between international organizations, NGOs and transnational corporations, different again when it is reconstructed in the reputation mechanisms of the public or in monetary incentives and sanctions. Other changes occur when finally it is “re-translated” into the legal parlance of the hard law of intra-company and inter-company codes. These rather indirect connections between both network codes highlight that the auto-constitutionalization of the corporate world does come about neither due to intrinsic motives of voluntariness nor due to the sanctioning mechanisms of state law, but due to a circuitous translation process in which different learning pressures come to bear.