A SELF-REGULATION PARADOX: Notes towards the Social Logic of Regulation Willem J. Witteveen Readers are reminded that this work is protected by copyright. While they are free to use the ideas expressed in it, they may not copy, distribute or publish the work or part of it, in any form, printed, electronic or otherwise, except for reasonable quoting, clearly indicating the source. Readers are permitted to make copies, electronically or printed, for personal and classroom use. Abstract In the context of the regulatory problems of the Dutch welfare state, the idea of legally conditioned self-regulation seemed an attractive alternative for other strategies of regulation. It was adopted as official government policy in the early 1990s.Yet, it has not been applied consistently and seems not to have affected the style of government regulation in its appointed domains. This article seeks an explanation for this development in the mismatch between legislative strategy, undertaken at the national level, and normative expectations in society generally about what regulation is to be and to achieve. Following Sch?n and Rein, two cultural models or frames are posited: the production model of legislation versus the situational model of legislation. Self-regulation puts both models under strain. In order to work, the strategy of legally conditioned self-regulation requires a different social logic circumventing the paradoxical nature of the injunction to self-regulate. This autonomy paradox is a variant of the spontaneity paradox described by Watzlawick et al. (1967). The command to ‘be autonomous’ requires an institutional structure in which metacommunication is possible. It is argued that a constructive social logic for a strategy of legally conditioned self-regulation should make use of Selznick’s theory of responsive law. 1. The emergence of legally conditioned self-regulation The idea of self-regulation as an alternative or supplement to government regulation has a long history in the Netherlands, with arguably some of its roots already in ideas and practices of autonomy and self-rule developed during the Dutch Republic. It became official government policy in 1981, when the first Lubbers cabinet declared its interest in deregulation. Following then dominant Anglo-American trends set by Margaret Thatcher and Milton Friedman, deregulation was initially conceived of as the elimination of unnecessary rules preventing free markets from functioning properly. This impulse was very quickly deflected, however, into a call for not fewer but better rules. Deregulation, in the Dutch context, was understood in qualitative rather than quantitative terms. And the way to achieve this improvement of the rules, making them more manageable in social settings, was understood in terms of self-regulation: What better way to have effective rules than by having them drafted by well-organized groups of actors concerning their own domain? Education, health care and labour relations were often mentioned as the domains that could generate, through the organizations already established there, the rules that would actually be needed to let practices of education, health care and labour relations perform better. Rather than centralized regulation by distant bureaucracies, self-regulation promised dialogically formulated and locally effective regulation by actors with a stake in the work processes and their outcomes. The idea has been controversial from the start, however. While its proponents mentioned a number of advantages of self-regulation - such as the mobilization of local knowledge, a greater tendency towards living by the rules, an increased sense of responsibility with the actors involved and less need for government supervision over application of the rules - its opponents pointed to serious disadvantages. Self-regulation would reflect differences of power and organization between the actors participating in it; it might lead to evading general responsibilities (or shifting burdens onto excluded groups); self-regulation would be selectively enforced and it could even be a cover operation for illegal practices; self-regulation might also lead to a more bureaucratic organization of domains; standards developed through self-regulation might differ too greatly from area to area, hurting the ideal of equality before the law. A stalemate ensued, especially since the proponents of self-regulation in turn could point to the well-known disadvantages of centralized government regulation, or rather overregulation: a surplus of inflexible and impracticable rules threatening to make life too complicated and leading to suboptimal processes and outputs. It was at this juncture that an almost Hegelian synthesis between conflicting ideas of regulation and self-regulation arose: legally conditioned self-regulation. Under this legislative strategy, the legislature initially states a framework of conditions (both substantive and procedural), within which citizens, companies and social organizations are free to develop their own rules, after which a government agency inspects the results of norm formation and norm application to ensure that the legal conditions are met. When the strategy does not work and, as a result, the actors involved do not succeed in creating rules through cooperation, rules which are then also applied through social pressure internal to the domain rather than through state controls, the fall-back option is to take over the regulative process and again supply centrally made rules. Of course, after this procedure the government agency is better informed about the problems besetting self-regulation and so presumably also better able to make rules that are both in conformity with centrally posited objectives and informed by local conditions. In the Ministry of Justice white paper Zicht op wetgeving (Legislation in Perspective) of 1991, this legislative strategy was officially recognized. As one of the authors of this white paper makes clear, the strategy of legally conditioned self-regulation is especially attractive when the government aims to regulate professional conduct: when there is not too much difference between the interests of the individual and the group on the one hand and the public interest on the other, and when effective compliance with government regulation is difficult or impossible to achieve. 2. Legislative strategy and its discontents In view of the enthusiasm with which this strategy of legally conditioned self-regulation was received and the wide area of its potential application, it is rather surprising to see that since 1991 the strategy has not been applied all that frequently and seems not to have affected the whole style of government regulation in its appointed domains. Legally conditioned self-regulation has indeed been prominent in the fields of privacy law and primary education (in regard to articulating the objectives of teaching, the so-called kerndoelen), but these are areas where at the same time other legislative strategies have proliferated simultaneously. On the whole, it must be concluded that the areas that in theory lend themselves to self-regulation have in recent years also been visited by streams of orthodox government regulation, often emanating from the European Union. As a result, those participants in processes of self-regulation who really aim at taking the margins set by the central legislature as serious conditions for self-rule are confronted with intrusions into their domain of free deliberation by uncoordinated regulations in which the government denies them this deliberative freedom. Of course, that may well be the work of another government agency, pursuing its own legislative strategy, but the effect of such uncoordinated interventions is that the potential space for deliberations about self-rule is turned into an illusion. Imagine, in a shift of perspective, that you are a responsible actor in one of the domains supposedly engaging in processes of legally conditioned self-regulation. You may, for instance, imagine yourself to be the principal of a primary school. From this position, you would not see a coherent legislative strategy at all. On the one hand, there would be initiatives taken by the Ministry of Education to organize a broad debate about the desirability of certain forms of regulation; it is even prepared to listen to suggestions about the rules that should actually be removed so that schools can find their own solutions for pressing social and educational problems. Let us suppose that some of these consultations indeed result in a degree of deliberative freedom for your school to experiment with the rules and that after some time, upon positive evaluation, a whole set of schools comes to adopt the new line of regulation. (This scenario has not yet occurred in practice, but it is part of the rhetoric of educational policy.) Even when everything goes according to such a scenario of legally conditioned self-regulation, as the principal of your school you will still be confronted with new directives almost weekly on all kinds of other aspects of school life and these directives, originating perhaps also from other departments or from local government, do not fit into a consultative strategy; they are presented as legislative commands. Very likely, in your position of responsible actor in the domain of primary education, you will not experience a marked increase in deliberative autonomy at all, and you will not rationally suppose that you yourself are a very meaningful or powerful actor in a process of self-regulation. Indeed, the real actors in processes of self-regulation, from your vantage point, will be organizations such as the trade unions, and you will, as an individual principal, have limited access to and knowledge of their negotiations with the Ministry. The outcome of the process of legally conditioned self-regulation will most likely be as binding on you and as external to you as the old-style central rules would be emanating weekly from government bureaucracy. Seen from below, from what Griffiths calls the vantage point of the work floor of social life, the rules made through legally conditioned self-regulation will not be distinguishable from rules promulgated in any other way; at most they will be perceived as the results of a somewhat different technology of control, as another instrumentality in a top-down approach rather than as an enabling strategy of autonomy working its way up from the bottom. It is important to remember, at this point in the argument, that Sally Falk Moore’s famous idea of semi-autonomous social fields applies at the level of practices of regulation and that it means that our imaginary principal is always already finding herself situated in such a semi-autonomous social field the core characteristic of which is that it can generate some of its own rules in defiance to the rule-making and rule-enforcing capacity coming from outside the field but at the same time is never completely autonomous and will have to obey at least partially the outside rules or transform them into workable arrangements. The theory of legally conditioned self-regulation supposes too much harmonious and coordinated interaction by positing a strategy that all actors collaboratively engage in and not allowing for the clash of different strategies and power differentials within the social fields. (In game-theoretical terminology, the strategy tacitly assumes a positive-sum game and neglects the conditions under which zero-sum games, leading to prisoner dilemmas, could be avoided.) 3. Two cultural frames Ending our thought experiment for the moment, we will now focus on the match or mismatch between legislative strategy, undertaken at the national level, and expectations of a normative kind about law and legislation prevalent in the legal and political culture generally. What is regulation supposed to be and to achieve? This question directs us to the presence of cultural frames, defined by Sch?n and Rein as ‘underlying structures of belief, perception, and appreciation on which people and institutions draw in order to give meaning, sense, and normative direction to their thinking and action’. Frames, on this understanding, are cultural models helping people to organize their experiences and to give direction to their responses. They function in the plural. Often there are competing frames, selecting different aspects of a situation as salient or interpreting events in different terminologies and at the same time providing attitudes and sets of responses geared to the frames’ organization of knowledge. When each competing frame gives different directions concerning courses of action in social settings, we can also distinguish their differing social logic: their projections of preferable scenarios to be followed in practice. It is my impression that Dutch political and legal culture contains at least two such frames concerning regulation and that both these frames have explanatory power and normative saliency, while remaining contradictory models. Choosing between the two frames is difficult, and likely to be avoided, with people not choosing an applicable frame so much as sliding into it as the situation seems to demand. The first cultural frame is the production model of legislation. It sees legislation as a continuous process of production. Rules are made and changed, according to an iron logic of democratic self-governance. Regulations are the momentarily binding outcomes of democratic procedures that call them into being. As public deliberation continues and new problems are responded to in a legal way, the rules will inevitably change and again be momentarily binding outcomes of democratic decision-making. This frame is pragmatist in orientation and offers a procedural interpretation of ideals of both representative and deliberative democracy and the rule of law. The second cultural frame is the situational model of legislation. It regards legislation as a permanent background noise in modern society, or, more positively formulated, as a normative setting that is present in all areas of social life. Legislation is one of the necessary conditions for an orderly society and, therefore, the laws should be a stabilizing background factor, coming into the focus of attention only when conflicts cannot be solved in another way. While it may sound old-fashioned and certainly less dynamic than the production model, the situational model is firmly entrenched in legal culture, expressing normative orientations referring to the principles of legality (such as the ideal of legal certainty itself) and to political freedom. Not democratic decision-making is central to the frame, however, but reasonable governance and judicial conflict resolution. The best rules are those sustained in social life, manifested in practices that work; the situational model is opposed to the bureaucratic tendency inherent in the production model. How does self-regulation fit in with these two conflicting cultural frames about legislation and regulation? It would seem that self-regulation puts both models under strain. The production model is based on a distinction between official regulation and social practices; the democratic arena is the locus of ongoing legislative discourse, while society is the place where rules perform (or fail to perform) their intended functions. Self-regulation blurs these boundaries and extends the reach of the production model to the full domain of the regulated. This extension weakens the cognitive organization of experience the frame makes possible (it includes too many disparate phenomena) and it raises questions the frame does not provide easy answers to: what, for instance, is the official status of private legislators? While the strategy of legally conditioned self-regulation promises to reduce ambiguities somewhat, it does not give the same easy orientation as the elementary cultural model did. The situational model, at first sight, seems made for self-regulation since it refers regulation to social practices and local contexts; as long as self-regulation succeeds in creating background certainties, it performs the way the model would expect. But there is a problem here as well, not of extension but of implosion. By allowing the making of rules into the frame, rather than focusing on the situatedness of rules already generated, the situational model suffers a loss in stability. Its logic comes under strain: how can rules be both givens, ready to be interpreted or manipulated by participants in social life, and new creations, being imposed upon the situation? When a strategy of legally conditioned self-regulation is introduced, the situational model will after a while see both the imposed conditions and the locally produced rules as part of the same scenic complex of new background certainties and it will bring into view a new distinction: between the context of production and a new context of use, since in the logic of this frame there can be no rules which are not situated in some way. The model has to duplicate itself and, in so doing, loses much of its simplicity and its facility for cognitive organization. Putting the two cultural frames for regulation under strain, the strategy of legally conditioned self-regulation does not fit in well with the social logic of either model. It needs a social logic of its own. Can this perhaps be discovered by investigating the nature of the rhetorical claims made under the strategy on participants in social systems who are enjoined to engage in self-regulation? 4. ‘Be autonomous!’ Let us, therefore, imagine a sender and a recipient of the message of legally conditioned self-regulation. The sender (the legislature) issues a command to obey rules of one’s own making but to remain within the ambit of the law. But the command imposing this injunction to be an autonomous lawgiver must be obeyed faithfully, obediently. For the recipient this becomes a paradoxical rhetorical claim. How can one obey the rule and at the same time be a law unto oneself? The attitude that is being commanded is the opposite of the attitude that is needed to comply with the command. We thus have a ‘contradiction following correct deduction from consistent premises’, that is to say, a paradox. Watzlawick and his co-authors (1967) distinguish three types of paradoxes - the logical paradoxes, paradoxical definitions and pragmatic paradoxes; it is the third type of paradox that seems to be at issue here. The prototypical example is the utterance ‘Be spontaneous!’ This is ‘an injunction demanding specific behavior which by its very nature can only be spontaneous’. The sender typically places the receiver of the message in an untenable position, for to comply with the injunction the receiver would have to be spontaneous within a frame of compliance, of non-spontaneity. The self-regulation paradox shows the same stifling social logic, demanding autonomy within a frame of obedience, of non-autonomy. ‘Be autonomous!’ Watzlawick and his co-authors develop their diagnosis of paradoxical injunctions following the interactional pattern of the spontaneity paradox out of a therapeutic concern with the very real pathological effects the paradox can generate; it is at the basis of the generation of schizophrenic symptoms. Further research leads them to the hypothesis of ‘double bind’, a pathological condition in which a person finds herself punished for correct perceptions. A double-bind interaction consists in three ingredients: - Between two or more interacting parties there is an intense relationship, such as exhibited by family life and loyalty to a creed or ideology, but also including ‘contexts influenced by social norms or tradition’. - In such a context, a message is given which is structured in a way ‘that (a) it asserts something, (b) it asserts something about its own assertion and (c) these two assertions are mutually exclusive; thus, if the message is an injunction it must be disobeyed to be obeyed’. - ‘The recipient of the message is prevented from stepping outside the frame set by the message, either by metacommunicating (commenting) about it or by withdrawing; this makes the logically meaningless message into a pragmatic reality.’ When we apply these psychological insights to the self-regulation paradox and take into account the less intense relationships between official actors on the side of the government and actors in social fields, we can diminish the concern that a double bind generating something akin to organizational schizophrenia will be a frequent phenomenon, but still the comparison will show a number of factors that may indeed lead to pathologies at the level of organizational interaction, and thus to a faulty social logic of processes of regulation. For instance, think of situations where: - the stakes between interacting agents are high, as a result of mutual dependencies in reaching their differently defined objectives; - a legislative strategy implicitly metacommunicates a discrepant attitude, thus creating confusion (‘this legislative message is an injunction that must be disobeyed to be obeyed’); - the legislative strategy is unidirectional, not a response to an offer for a regulation strategy coming from the regulated field; - the legal command is backed by force; - the legal command cannot be escaped, negotiated or evaded (thus becoming a pragmatic reality). Especially when these factors are cumulatively present, a legislative strategy of legally conditioned self-regulation may easily backfire, leading to pro forma responses, and not achieve the desired effect of bounded regulative autonomy. But there is a more important lesson to be learned from the ‘pragmatics of human communication’. It concerns the conditions for success rather than the factors responsible for pathological interactional patterns. For the message of legally conditioned self-regulation to be taken seriously as an injunction to rule making and to evade the traps of its paradoxical structure, it is necessary to create an institutional structure in which metacommunication can occur, so that receivers can communicate back to senders. Consequently, both sides to the interaction can adapt their mutual expectations in the light of the legitimate point of view of the other. It will also be helpful if the official announcement of legally conditioned self-regulation is not issued as a command but as the outcome of previous negotiations (as a quasi-contract perhaps), so that it is no longer a unidirectional but a multidirectional strategy. It concerns matters for which the use of force is not resorted to, instead relying on the organizing power of established social norms. And it will even be helpful to openly acknowledge the options of escape, negotiation and evasion in order to accept, if needs be, some of them as opportunities for a constructive interpretation of the purposes the parties to the self-regulation process already multilaterally agreed on. 5. The social logic of responsive regulation By turning the faulty social logic of the autonomy paradox employed by self-regulation strategies on its head, the contours of a different, constructive social logic of responsive regulation become visible. This refers us to Philip Selznick’s remarkable synthesis of sociology, ethics and law in The Moral Commonwealth. Responsiveness is an institutional virtue: it requires an adequate and purposive institutional reaction to the problems and challenges it is facing. There are three dimensions to responsive legal institutions: outreach, empowerment and competence. Outreach requires legal institutions with a structure of decision-making treating all relevant interests ‘as potential claims to intrinsic worth and therefore as objects of moral concern’. This requires a translation of the values of the master ideal of legality (such as openness and transparency) into workable and fair procedures. Empowerment, in line with the foregoing requirement, means an acknowledgment of the limits of law. ‘A responsive legal order is not set over society. . . . The vitality of a social order comes from below, that is, from the necessities of cooperation in everyday life.’ Selznick here refers to the socio-legal tradition of legal pluralism, which ‘posits the moral worth of institutions close to the people, that is, based on shared experience, reflecting shared sentiments, sustained by practical needs’. Selznick of course does not deny the need for higher-level regulation, but demands an attitude of responsiveness on the part of the regulators, striving after ‘maximum feasible self-regulation’. Outreach and empowerment together stand for forms of regulation in which not the rules and their application are the central features but a sustained collective attempt to solve problems. ‘Responsive regulation is more problem-centered than rule-centered, more persuasive than coercive.’ This does not exclude the use of force on the side of the overarching community, but posits it as a last resort. Rules are needed, compliance mechanisms as well. ‘Responsive regulation is not the same as deregulation.’ Responsive regulation distinguishes itself, however, by the democratic quality of the legal processes leading to such ultimate rules and in the permanent willingness to consider new solutions to recurring problems. Mobilization is needed, in the light of the articulated collective purposes, values and norms of the community. ‘Responsive law is less interested in rule-compliance than in mobilizing energies for the achievement of public purposes.’ When the dimension of competence is added to this rather idealistic picture of responsive legal institutions, it becomes clear that the institutions of the law must prove their legitimacy as the community’s regulators by their active response to socially perceived problems. ‘The law must be an effective instrument for dealing with change and meeting social needs.’ Selznick on this count again stresses the limits of law by opposing institutional specialization and bureaucracy as ineffective kinds of institutional response. All government agencies should ideally be conceived of ‘as a single institution made up of complementary and cooperative (as well as antagonistic) parts’. This scenario of responsive regulation, outlined here in rather summary fashion, proposes a different social logic of regulation. It does not depend on a central institution issuing an injunction to engage in self-regulation. On the contrary, responsive regulation starts out by acknowledging the regulative power of the social order, allowing actors in this social order to articulate their concerns in governmental decisional structures. It does not depend on one-way communication. On the contrary, it envisages an interactional process of articulation of common purposes, referring to the values and norms held in the community. It does not depend on previous decisions about the legal conditions the centre formulates within which self-regulation must subsequently take place. On the contrary, it requires outreach, empowerment and competence on the side of the officials in charge of institutional functioning, in order to stimulate collective efforts at the coordinated articulation of both the problem and the desirable response (and then only partially in terms of regulation needed). It avoids the self-regulation paradox by asking, analysing and listening rather than by authoritatively stating an objective, reserving the right to do so at a later stage of the interactions. Deliberating autonomous actors engaging in such a responsive institutional framework do not easily get entangled in the pathology of the double bind. What the ideal of responsive regulation especially makes clear is that the social logic of regulation requires constant sensitivity to those qualities of institutions and practices through which a moral commonwealth shapes and reshapes itself. There must be reciprocal relations between the authorities and the public, between the governors and the governed, between those engaging in self-governance and those stimulating them to do so. As a result, it would seem that practices of responsive regulation are a prerequisite for practices of legally conditioned self-regulation. Indeed, when the social logic of regulation follows the model of responsiveness, it will be hard to distinguish the two. 6. Endnote But can the history of 25 years of failing to establish legally conditioned self-regulation on a grand scale in the Netherlands really end on this optimistic note? When there are actually two incompatible cultural frames (the production model and the situational model), would this not suggest that all possibilities for a change of direction and for more effective legislative strategies must reckon with this fact? Can a different kind of social logic concerning regulation ever take hold? Will the lofty ideal of responsive regulation not inevitably be distorted to fit one of the two dominant frames? It is hard to answer these objections with any degree of certainty. After all, there are only weak signs that official actors are turning towards an ideal of responsive regulation and responsible self-regulation. (I do find, however, openings for this idea in the recent Ministry of Justice white paper on regulation, entitled De bruikbare rechtsorde (A Practical Legal System; 2003). It is, moreover, to be expected that the various governmental agencies will not automatically adopt the line held by the Ministry of Justice. If the implementation of regulation is a persistent problem, so is the implementation of legislative policy. If we compare the social logic implied in the two cultural frames with the social logic proposed under a conception of responsive law, however, we will find interesting connections. Like the production model of legislation, the model of responsive regulation assumes democratic self-governance, albeit not by supporting the idea of political primacy for government and parliament. But it is similarly pragmatist in orientation and geared to procedural values of deliberative democracy and the rule of law, promising - seen through the lens of the production frame - a different and superior, pragmatic kind of instrumentality based on multilateral exchanges and on a widened knowledge base and the inclusion of more stakeholders. Like the situational model of legislation, responsive regulation awards primacy to the social order, to practices people engage in and local knowledge utilized therein, and it shares Ehrlich’s preference for normative legal pluralism. But for the situationalists it has the message that even very effective semi-autonomous social fields are connected to a supporting environment and that the people acting within these systems are primarily citizens of a wider community with important shared values (and a lot of controversy over their meaning). When we see this, we situate the social logic of responsive regulation in between the social logic espoused by the two cultural frames of the production model and the situational model. Perhaps, then, the idea of responsive regulation can be a bridge between them. Sch?n and Rein developed their notion of cultural frames in order to be able to help actors caught up in their implicit conceptual schemes reflect on how their perceptions obstructed the solutions of intractable policy dilemmas. These authors subsequently envisaged methods for bringing together ‘framed’ actors in conversational settings in which they would engage in ‘frame-reflective policy discourse’. The evidence of various of such reflective conversations suggested that under certain conditions it would indeed be possible to escape from restricting frames and construct new ones. Perhaps this would also be feasible in legislative conversations involving participants in the legislative process, designed in such a way as to stimulate them to look into their own schemes of conceptualization. ‘Participants in such a conversation must be able to put themselves in the shoes of other actors in the environment, and they must have a complementary ability to consider how their own action frames may contribute to the problematic situations in which they find themselves.’ And it is exactly this kind of perspectivism which fosters the pragmatic problem-solving attitude advocated by the social logic of responsive regulation.