Legisprudence: A New Theoretical Approach to Legislation 9781472562661, 9781841133426

Citation preview

List of Contributors Christian Dahlman, University of Lund Svein Eng, University of Oslo Lars D Eriksson , University of Helsinki Kauko Pietilä, University of Tampere Hannu Tolonen, University of Turku Kaarlo Tuori, University of Helsinki Amaryllis Verhoeven, European Commission Wim Voermans, University of Leiden Heinrich B Winter, University of Groningen Luc J Wintgens, University of Brussels, Centre for Legislation, Regulation and Legisprudence, Brussels

1

Rationality in Legislation—Legal Theory as Legisprudence: An Introduction LUC J WINTGENS

Complaints about both the volume of legislation and its quality in most European countries have raised the question whether collaboration between legislators and legal theory can help to articulate and solve this problem. In fact, although the complaints are made with an ever-stronger voice, the solutions are by no means obvious. Legislation, it is said, is a matter of politics, and politics is not rational. Politics is a power game, which results in compromises framed into a legislative or statutory structure. This power game seems to have its own logic and, most of the time, the results outweigh any other form of logic. Legal theory, on the other hand, is considered, at least from the perspective of politics, to be a ‘theoretical’ approach to legal problems. It contributes to the description and systematisation of existing valid law. It appears, like Minerva’s owl, after the sun goes down on legislative activity. From this standpoint, there is not much hope that legal theory can usefully intervene in the process of legislation or regulation, in other words either before or during the creation of rules. Legal theory is then confined to ‘legal science’ or ‘legal dogmatics’. Some years ago, I proposed that the problem of legislation be considered from another angle.1 The premises of the problem were that, although legislation and regulation are the result of a political process, both could be the object of a theoretical study. In analogy, for example, to Hans Kelsen in legal theory,2 the main idea was not to focus primarily on the content of rules and concepts, but rather on the structure and function of legal systems. In this approach, the focus is on problems that are common to most legal systems and not on the characteristics (namely, the content of concepts that are specific for a legal system) far

1 L J Wintgens, ‘Creation and Application of Law from a Legisprudential Perspective. Some Observations on the Point of View of the Judge and the Legislator’ in A Aarnio et al (eds) Justice, Morality and Society: A Tribute to Aleksander Peczenik on the Occasion of his 60th Birthday on 16 November 1997 (Juristförtlaget i Lund, Lund, 1997) 469–89. 2 [1926–27] Revue Internationale de la théorie du droit 1–3.

2 Luc J Wintgens less even on how rules are created. Of course, Kelsen’s approach leaves legislation and regulation—apart from their formal validity aspects—out of the scope of study, because this relies on value judgments that, according to him, are not fit for theoretical study. In short, the creation of legal rules is a matter of politics, and politics is not fit for scientific study. This position is an understandable one, but it is only partially acceptable. Rule creation is a matter of choice, and this choice is legitimated because it is based upon the democratic character of the regulating process and not because it relies on a science of values. In other words, should one try to mould legislation into the frame of a science of whatever nature, we would face something like ‘scientific politics’, as Marxism propagates, and which is, for several reasons, unacceptable. A different position is to study legislative problems from the angle of legal theory. This approach I propose to call legisprudence.3 Legisprudence has as its object legislation and regulation, making use of the theoretical tools and insights of legal theory. The latter predominantly deals with the question of the application of law by the judge. Legisprudence enlarges the field of study to include the creation of law by the legislator. Within this new approach, a variety of new questions and problems are raised (including the validity of norms, their meaning and the structure of the legal system) that are traditionally dealt with from the perspective of the judge or are taken for granted by classical legal theory. However, by shifting the attention to the legislator, the same questions arise, such as in what sense the legislator has to take the systematicity of the legal order into account, what counts as a valid norm and what meanings can be created and how, to mention just a few. Traditional legal science covers many of these questions with the cloak of sovereignty. Legislators being sovereign, they decide what will count as a valid norm and what its meaning is. Whether and how it fits into the system is then a matter of interpretation—and this is the task of the judge and the legal scientist. After all, legislation and actions in the legislative field seem to be averse to theoretical inquiry. After long decades of legalism in legal reasoning, it can be said that the dominant views in legal theory that have resulted from that have actually barred the way for questioning the position of the legislator. Everything happens behind the veil of sovereignty as far as legislation (properly so called) is concerned and behind the veil of legality when dealing with the execution of legislative acts. These veils conceal a good deal of ignorance related to the possibilities of alternative theoretical reflections on rule-making. Sovereignty, in effect, creates silence about these alternatives, so that it becomes ‘sovereignty in silence’. Sovereignty of the ruler prevents his rules from being questioned in other than binary terms. Validity is a good example of this. The only question worthwhile 3 L J Wintgens, ‘La création d’une banque de données en légisprudence’, [1992] Gesetzgebung Heute 91–4.

Legal Theory as Legisprudence 3 posing is whether a given propositional content is a valid rule—yes or no. Questions as to its effectiveness, efficiency or acceptability are then not the order of the day. The claim of legisprudence is that these questions are important ones and that they can be analysed with the help of legal theory. I refer to my contribution in this book for some articulations and clarifications on this idea. In August 1998, the initiative was taken to organise the fourth Benelux–Scandinavian Symposium on Legal Theory. After successful symposia in Antwerp (1982) on the utility of legal theory, in Uppsala (1985) on the structure of law and in Amsterdam (1991) on coherence and conflict in the law, the time came for an exploration of the intersecting lines between legal theory and legislation; in short, legisprudence. This became the theme of the fourth symposium. The main papers presented at the symposium were collected and are now printed in this volume. As a brief introduction to this, I propose to sketch the main ideas of these contributions, in order to provide the reader with an idea of what will follow. Lars Eriksson’s paper on ‘Making Society Through Legislation’ starts with the idea that rational decision-making requires that we be aware in one way or another of the consequences of those decisions. When used by lawyers, rationality means more often than not ‘consistency and coherence’. Legisprudence, according to Eriksson, has to face the problem of predictability of legislation as a form of decision-making. His main thesis is that when speaking of legislation, welfarism must be criticised because most of the time the consequences of legislative decisions are not expected by the rulers. In other words, his claim is that the welfare state is a realisation of liberalism to its utmost limits. If welfarism had the ambition to overcome liberalism, then it failed. Welfare rights, he argues, have the same function as liberal rights (claim rights) and follow the same logic as liberal rights, since they share with the latter the aspect of ‘verticality’. Therefore, welfarism empties classical horizontal relationships of their meaning. This analysis, says Eriksson, contributes to legisprudence in the following sense: legislation is unavoidably the result of the conceptual framework it realises. The concepts used in this framework exclude that certain results, which are eventually desirable, could be obtained. Legisprudence therefore includes the idea that the lawgiver bases his conclusions on a detailed analysis of the consequences of his decision to avoid as far as possible undesired and undesirable results. Kauko Pietilä’s exposé is entitled ‘Rationality of Legislation in a Sociological View’. Referring to Kant, his idea is that human dignity embraces the belief that man only obeys laws to which he has contributed, be it by way of representative democracy or via representation through interest deputies. Representative democracy is analysed from the perspective of Weber’s all-pervading rationalisation and raises the question whether legislation can be rational when it results from interest representation. Pietilä suggests that interest representation is no longer representative, so that legislation is enacted without participation.

4 Luc J Wintgens Consequently, a reasonable being is no longer obliged to voluntary obedience, since human beings are treated as having only a price and no longer any dignity, to use the Kantian terminology. Like Eriksson, Pietilä insists on the predictability of the consequences of legislative decisions. In a similar vein to Rousseau, Pietilä connects interest representation to a destruction of the general will, which is split into different particular wills. If we want to discuss representation in any meaningful way, the author argues, we must be aware of the fact that interest representation causes a rift not within society, but between Parliament (and the people it is supposed to represent) on the one side and the government and civil services on the other. If both sides could interact, it would be tantamount to the underlying population interacting with itself. Moreover, if this occurs, then we face true self-legislation and selfgovernment, the ultimate goal of which is to realise the general will. Svein Eng has provided a contribution on ‘Legislative Inflation and the Quality of Law’, in which he addresses the question whether there exists a relationship between the quantity of legislation and its quality. One of his claims is that defective legislation is only one cause among others of the quantitative increase in most legal systems. However, substantially flawed legislation is among the significant causes of the great quantity of legislation. Eng reframes the problem of legislation in other terms. Are there, he asks, alternatives to legislation (such as negotiation or privatisation) and what are their motivational forces? He argues that the political and moral perspectives of legislation lose their motivational force. Means–end perspectives, on the contrary, retain the strongest motivational force. Upon this, Eng’s analysis demonstrates that technically flawed legislation seems to belong to the range of significant causes of legislative inflation. He then concludes his contribution by suggesting that the solution to the core problem of ‘good’ legislation lies in the search for and accrual of new facts, through the clarification of which questions are to be solved by legislation. Christian Dahlman, in a similar vein to Eriksson and Pietilä , refers to the predictability of legal rules but questions this in relation to acceptability in his paper ‘Predictable Rules and Flexible Principles—the Problem of Ideological Pluralism and Legitimacy’. The law’s pluralistic character, he says, reflects competition between three different ideologies: reciprocity in risk-taking, wealth maximisation and minimum safety. An evaluation of these three approaches of regulation is made from the perspective of legitimacy, or in other words, predictability and acceptability. Ideological monism is, according to the author, most favourable to predictability, but is devastating for the law’s acceptability. A pluralistic approach of rules and principles, on the contrary, avoids this flaw. From a general perspective, rules promote more legitimacy than principles, although the former are less flexible than the latter. The scope of equity is reduced when relying more on rules. A middle way argued by the author is a combination between rules and principles which creates a form of coherence that is superior to making the law ideologically homogeneous.

Legal Theory as Legisprudence 5 In his contribution, ‘Concept and Institution of the State in the European Legal Tradition’, Hannu Tolonen makes a distinction between the state as the source of the law and the state as the basis of all public legal relations. His paper deals mainly with the latter. Tolonen’s starting point is that the law is a multifaceted social and cultural phenomenon and his analysis is based on a threefold historical distinction, related to the various tasks of the state through history. The feudal state and society has a hierarchical character, focusing on the maintenance of peace and tranquillity and connected with the regimen regale and the theory of the king’s two bodies. The bureaucratic state, or the modern public power, is connected to the concepts of sovereignty and ‘Rechtsstaat’, with an emphasis on the state’s artificial nature. Finally, the welfare and postwelfare conception of the state attempts to satisfy the basic needs of all its members in an interventionist manner. The latter conception includes a critique of the overemphasis on the benefits of the market mechanism. This can be linked to Eng’s criticisms that markets do not necessarily produce morally desirable or optimal results. Like Eng, Tolonen stresses the fact that in the post-welfare conception of the state, self-regulatory social mechanisms have replaced the state’s interventionist legislative activity. According to Tolonen, who is of the opinion that the state is neither morally nor socially omnipotent, a procedural conception of law is more sensitive to the law’s connection to its social and moral environment. Furthermore, he criticises the unity of power within the state and presents arguments that frame a new way of using private and semi-public methods of the organisation of power rather than direct public intervention and legislation. Kaarlo Tuori has delivered an article on ‘Legislation Between Politics and Law’, in which he deals with the question of how legal theory, and more specifically legisprudence as a new branch of legal theory, can contribute to the process of making the law more rational. He proposes a distinction between legal norms and legal practice. Legislation, as a cluster of political practices, does not necessarily mean legal practices, except with respect to norm formulation. Legislation, in short, is a combination of two types of practices, with an emphasis on the political. Tuori’s thesis is that a distinction must be made between different levels of the law. Law is a legal order the systematicity of which is not generated by legislation. On the contrary, Tuori argues that legislation is more often a factor of disorder than of order. Legislation says Tuori, is the surface of the legal order that is systematised through adjudication and legal science. This is the way principles are articulated. The legal order is connected on this deeper level with morals, without being identical to it. The same is true for the relation between law and politics. They are connected on the surface but remain autonomous at the deeper level. Legislation, Tuori argues, is then transformed from ‘not yet law’ into ‘law’ via legal practice (legal science and adjudication). The rationality of legislation is evaluated with yardsticks other than the rationality of adjudication and legal

6 Luc J Wintgens science. Further, the question of rationality of legislation does not concern individual statutes. Finally, the integration of legislation into the legal order as a whole involves factors that are not accounted for in the purposive rationality accompanying the activity of the legislator. It is via this integration that the law evolves towards autonomy vis-à-vis politics. Amaryllis Verhoeven’s paper on ‘Legisprudence and European Law: In Search of Principles of European Legislation’ begins by establishing the polycentricity of European law-making, with a plurality of regulatory activities, actors and procedures for which the principles and norms set out in the European constitution serve as a counterweight. Her analysis of the institutional and regulating framework is supplemented with the search for principles of good legislation (duty to give reasons, legal certainty, adequate fact finding and the duty of informed choice). She goes on to criticise the regulatory, the intergovernmental and the parliamentary models of rule-making. The first and second models, she argues, are deficient because they do not take into account the political nature of the Union and because they lack democratic legitimacy. The third and most influential model considers the European Union a political enterprise, or a process of which law-making is an important aspect. Although influential, there is little hope, Verhoeven argues, that the implementation of the classical power organisation on the European level will enhance the required democratic legitimacy. This is the result, she claims, of parliamentary action on the national level being itself emptied of meaning due to the size of the European Parliament, the fictitious character of representation and homogeneity, and its connection with sovereignty. Verhoeven’s alternative relies on Habermas’s procedural and deliberative model. This model is, she explains, well suited to explain and justify the creation of norms at the European level. It requires at the same time less and more than parliamentary democracy: less, in the sense that it operates without a nation, and more since it requires an active citizenry. According to this model, sovereignty is then framed as ‘intersubjectivity’, while the constitution is considered a set of fundamental rules on rights and procedures. The constitution includes judicial review as a marginal control or a control at the borderlines of the democratic procedure: in other words, the judge controls whether the rules of the democratic game are respected. This does not rely on a binary or legalistic approach but includes, on the contrary, an overall weighing and balancing from a constitutional perspective. Wim Voermans tries to answer the question whether the use of informatics in the legislative process will enhance the rationality of legislation. His article, ‘Rationality in Legislation by using Informatics’, is a plea in favour of informaties. There are currently two approaches to the use of informatics. The first is the information-oriented approach, which supplies the information needed in the legislative process; the second is the artificial intelligence approach, where

Legal Theory as Legisprudence 7 the system can rationally solve problems based on the information that is supplied. The latter approach, according to Voermans, is not yet fully productive, although it can be used to build IT tools for specific parts of legislative drafting. Voermans has developed the Legislative Design and Advisory system (LEDA), using a checklist based on the governmental guidelines for legislative drafting in The Netherlands, which is also referred to in the paper by Heinrich Winter. Legislative drafting, Voermans argues, is an open-ended decision process, including more than purely legal knowledge. It is built on the method of the best argument. The LEDA system conceptualises, models and formalises this method. In the absence of a concrete definition of legislative quality, one has to work with the standards that have been set up, such as the guidelines provided by the Dutch government. Voermans analyses some of these guidelines, which are built into the LEDA system. This system intends to be an incorporation of these guidelines and transforms them into a computerised version so that they become more accessible. The LEDA system is a hypertext network, allowing for different types of navigation and working patterns within the system. The basic characteristics of the system are explained in the article. Legislative practice in The Netherlands shows that it is a practical tool, already used by a number of drafting agencies. The evaluation of legislation is the theme of Heinrich Winter’s contribution, ‘The Forum Model in Evaluation of Legislation’. He notes that while evaluation is fashionable and popular, the important question is what will be done with it. More precisely, he points out that as the quality of evaluation varies, evaluations themselves must be submitted to evaluation. After presenting some data on evaluation practices in The Netherlands, he formulates his thesis that, under specific circumstances, evaluation of legislation can contribute to an improvement of legislation. The evaluation of evaluation, or the meta-evaluation of legislation deals with two types of questions: the correct use of scientific methodology and the degree to which the results of evaluations are used for changes in the law. The basic issue in the field of evaluation research is to know what is meant by the quality of legislation. Evaluative practice in The Netherlands is primarily based, as we know from Voermans’ contribution on the governmental guidelines for good administration, focusing essentially on the effects of legislation and its drafting qualities. Winter has developed a new framework for the evaluation of evaluations, which is based on an argumentative and communicative model. He argues that the quality of legislation increases when more and better verifiable information is available in the circumstances under which the law is called to operate. Consequently, the quality of legislative changes depends upon the quality of the evaluations preceding the change. One of his conclusions is that highquality evaluations, because of their sobering and demystifying effect, contribute considerably to legislative improvements, since they reduce strategic behaviour.

2

Legislation as an Object of Study of Legal Theory: Legisprudence1 LUC J WINTGENS

I . INTRODUCTION

Since the publication of Noll’s Gesetzgebungslehre in 19732 greater attention has been paid to the scientific study of the phenomenon of legislation. Until that time, Noll observed, legal theory was almost exclusively focused on adjudication,3 legislation being almost completely neglected.4 Legal science was generally limited to what Noll called a science of the application of rules (Rechtsprechungswissenschaft), focusing on the person of the judge in his task of the application of rules. The creation of rules by the legislator was not the order of the day. In his opinion, however, judges and legislators, in many respects, do the same things.5 The principle of legality as the basis for judicial decisions and the lack of freedom of the judge in the application of legal rules can, from this perspective, be considered as related to the sovereignty of the legislator, who, in the creation of legal rules, enjoys a wide discretion. 1

Originally published in [1999] Rechtstheorie 11–46. P Noll, Gesetzgebungslehre (Rowohlt, Reinbek, 1973) 314. Ibid, 29; id, ‘Von der Rechtssprechungswissenschaft zur Gesetzgebungswissenschaft’ in (1971) II Jahrbuch für Rechtssoziologie und Rechtstheorie, 523 ff; W Maihofer, ‘Rechtstheorie als Basisdisziplin der Jurisprudenz’ in (1972) III ibid, 66; L Mader, L’évaluation l´égislative. Pour une analyse empirique des effets de la législation (Payot, Lausanne, 1985) 12. 4 This fact had already been mentioned, however, in J Landis, ‘Statutes and the Sources of Law’ in Harvard Legal Essays Written in Honor and Presented to Joseph Henry Beale and Samuel Williston (Harvard University Press, Cambridge, Mass, 1934) 230: ‘The interplay between legislation and adjudication has been generally explored from the standpoint of interpretation. The function of the legislature . . . has been largely ignored.’ 5 P Noll, Gesetzgebungslehre, 48 (n 2 above); id, ‘Zusammenhänge zwischen Rechtssetzung und Rechtsanwendung in allgemeiner Sicht’ in Probleme der Rechtssetzung. Referate zum Schweizerischen Juristentag 1974 (Helbing and Lichtenhahn, Basel, 1974) 249 ff. See also Landis, ‘Statutes and the Sources of Law’ (n 4 above) 233: ‘The consciousness that the judicial and the legislative processes are closely allied both in technique and in aims will inevitably make for greater interdependence in both.’ See also A Aarnio, ‘Form and Content in Law: Dimensions and Definitions of Legal Positivism’ in Philosophical Perspectives in Jurisprudence: Acta Philosophica Fennica (Philosophical Society of Finland, Helsinki, 1983) 86. 2 3

10 Luc J Wintgens In the absence of a jusnaturalistic legitimation of the law (in a liberaldemocratic state under the rule of law) the position of the judge attains its full weight. It is precisely this that makes the study of law, other than a dogmatic one—ie the description and the systematisation of existing law in view of its application—superfluous or impossible. Noll has regularly deplored this reduction of legal science to a science of the judicial application of rules.6 In this contribution, I would like to explore the presuppositions behind this conception of legal science. More specifically, I would like to investigate whether it is possible, when starting from the relationship between rule application and rule creation, to come to a theoretical perspective on legislation that allows us to articulate criteria for good legislation. At the beginning of this analysis, it should be noted that the focus will not be on legistic criteria, ie criteria that deal with the form of legislation. It is rather the aim to establish a theoretical approach that allows us to explain the absence of theoretical reflections on legislation and then make some suggestions that may contribute to the theoretical study of legislation or legisprudence.

II . THE ABSENCE OF A THEORY OF LEGISLATION

( LEGISPRUDENCE )

Legisprudence is the name for the branch of legal theory that deals with legislation from a theoretical and a practical perspective. It is especially this theoretical perspective in the study of legislation that has often been neglected, and in this section I would like to look deeper into the reasons for this. The absence of a systematic theory of legislation can be explained by the connection of positive law to natural law on the one hand, but, paradoxically, also by the disconnection of the two. Modern natural law legitimates positive law in so far as it is correctly deduced from it by the regulative use of logic. When disconnecting natural law and positive law, the latter can only be legitimated by reference to the democratically legitimated sovereign legislator. In the first case, the creation of law is based on knowledge of natural law, in the second it is based on a decision on the part of the legislator. Norm creation in a jusnaturalistic perspective is a matter of knowledge and, as a consequence, is an application of jusnaturalistic principles. In other words, the knowledge of natural law determines the creation of positive legal rules. Natural law is in this respect pre-existent to positive law and provides its cognitive foundation. Via a decision of the legislator, it acquires the value of positive law. Thus, positive law can be considered an application of the rules of natural law by the legislator. As Aarnio puts it: Due to the rationalistic nature of natural law, the question of goals in lawgiving did not arise. The idea of a purposive lawgiver as a purposively behaving creator of law

6

See references, n 3 above.

Legislation as Legal Theory: Legisprudence 11 was virtually unknown to the doctrine of natural law. The application of law was seen as a matter of knowledge.7

Aarnio points to the relationship between knowledge and application of rules. However, it is not necessary to believe that this needs to be limited to judicial norm setting, and cannot also apply to the legislator. This last point is clearly brought to light by Villey, an epigone of classical natural law, who is of the opinion that the creation of rules in general is essentially a matter of knowledge that receives its formal legal shape through the intervention of human will.8 The latter, however, is a secondary aspect of law. In this way, there is no need for a theory of legislation, distinct from a natural law theory, for epistemological reasons. Knowledge of natural law is of itself sufficient to promulgate correct legal rules. These rules then are a concretisation of natural law, or reflect a natural law conception that in its turn legitimises positive law.9 With the rise of the codification movement at the beginning of the nineteenth century in France, for example, the reference to natural law did not completely disappear.10 However, it can be argued that its legitimating influence was driven out by the French school of Exegesis.11 Aarnio, for his part, stresses the fact that an extreme rule positivism does not differ very much from a natural law conception. Even if the belief in ‘substantive natural law’ disappears, he argues, positivists adhere to the idea of a legal system as a closed whole of legal rules that are to be applied by way of logical reasoning.12 As a consequence, there is seamless connection between a natural law model of rule creation and a non-jusnaturalistic model linked to the separation of powers doctrine. In order to postulate the legal order as a closed, self-sufficient whole13 it is sufficient to disconnect the legitimation of law from its jusnaturalistic foundations. The formal legitimation of the legal system, characteristic of the liberal state, is then based only on the idea of a sovereign legislator. The cognitive aspect of law-making by the legislator in a natural law model is then pushed to a lower stage, namely to the level of the judiciary. The cognitive aspect of ruling by the legislator, in the absence of a natural law theory, does

7 Aarnio, ‘Towards a Theory of Legislation and its Role in Societal Change’ in Philosophical Perspectives in Jurisprudence (n 5 above) 247. 8 M Villey, ‘Abrégé du droit naturel classique’ in Leçons d’histoire de la philosophie du droit, 2nd edn (Dalloz, Paris, 1962) 145; id, La formation de la pensée juridique moderne: Cours d’histoire de philosophie du droit (Montchrétien, Paris, 1975) 53. 9 See O Weinberger, ‘Zur Theorie der Gesetzgebung’ in J Mokre and O Weinberger (eds), Rechtsphilosophie und Gesetzgebung. Uebrlegungen zu den Grundlagen der modernen Gesetzgebung und Gesetzesanwendung (Springer, Vienna/New York, 1976) 175–6. 10 Aarnio, ‘Form and Content in Law’ (n 5 above) 82; B Bouckaert, De exegetische school: Een kritische studie van de rechtsbronnen- en interpretatieleer bij de 19e eeuwse commentatoren van de Code Civil (Kluwer, Antwerp, 1981) 209 ff. 11 Aarnio, ‘Form and Content in Law’ (n 5 above) 82. 12 Ibid, 91. 13 Ibid, 82.

12 Luc J Wintgens not come into consideration any more, since the legislator has become sovereign. According to the principle of the separation of powers doctrine, legislative ruling is deemed to be a sovereign activity, while the application idea, formerly linked to natural law, is now passed down to the judge in such terms that the ‘application attitude’ to a rule, which was originally the attitude of the legislator with respect to natural law, is now the judge’s position towards the legal rule. In other words, the legislative position of ‘knowledge [of natural law] preceding legislative action’ has become the position taken by the judge in relation to legal rules that are the product of a sovereign legislator. However, the ‘application attitude’ is jusnaturalistic in nature, it no longer applies to the legislator, only to the judge, due to the loss of natural law as a legitimating factor for positive law and the compensating idea of sovereignty. This change of perspective can be explained by legalism. Legalism is the ethical attitude that holds moral conduct to be a matter of rule following, and moral relationships to consist of duties and rights determined by rules.14 It can be compatible with jusnaturalism, in the sense that, as an ethical attitude, it prescribes agreement through rule following, no matter where the rules come from. According to Bankowski: we can see that natural law theory is not exempt from the ambit or ambitions of legalism. It is rules that are important, not how they are arrived at. For legalism the power of natural law would lie in the rules that it generates and not in nature, God or practical reason that might be said to produce them.15

Within a legalistic form of thinking, there is no need for a theory about rules, since the only thing that is to be taken into account are rules. Legalism implies a rule-based way of looking at things. While it can go along with jusnaturalism,16 this need not be so, which means that to a certain extent, the aims of legalism and jusnaturalism are the same, and their merging merely reinforces the impact of both. Bankowski states this in the following way: Legalism doesn’t so much deny the connection between law and values as hide it and tuck it away from view. Since it concentrates on rules to the exclusion of everything else, the rules lose their sense of contingency. They dominate the entire moral universe (emphasis added).17

He thereby articulates three main aspects of legalism: (1) rules are not contingent, (2) there is no normativity outside the realm of the rules (or comprehensiveness of the normative system) and (3) the relationship between rules and 14 J N Shklar, Legalism: Law, Morals and Political Trials (Harvard University Press, Cambridge, Mass, 1986, 2nd edn.) 1. 15 Bankowski, ‘Don’t think about it: Legalism and Legality’ (1993) Rechtstheorie, Beiheft 15, 47. 16 Shklar, Legalism (n 14 above) 12. 17 Z Bankowski, ‘Don’t think about it’ (n 15 above) 47.

Legislation as Legal Theory: Legisprudence 13 values is hidden. These aspects can make clear why legalism and jusnaturalism are not opposed, but on the contrary can be compatible with each other. On the other hand, however, legalism does not need to operate in connection with jusnaturalism in order to produce its full effects. Indeed, it does not matter where rules come from, so legalism can fit equally well with positivism. The three aspects mentioned by Bankowski are linked together in the sense that the hidden relation between rules and values leads to the idea that rules are not contingent or that rules are not something to be made.18 This idea in its turn is connected to the notion of comprehensiveness of a legal system under the auspices of legalism, since there are no rules outside the system.19 The environment of the legal system has no connection whatsoever with the legal system itself, since there is a clear line between law and non-law.20 If, on the other hand, there exists a relationship between law and values, it is hidden and covered by a tendency of naturalisation of the law, as explained by Foqué and ’t Hart. They argue, in their critique of legalism in the French Code Civil movement, that the real conception of law held by the codifiers was an instrumentalistic one: law was something to be made and values were things to be realised via rules. On the other hand, as was already shown, the codifiers often referred to natural law as a justification of their ruling. This confirms the view that legalism and jusnaturalism go hand in hand, but in the Foqué/’t Hart analysis, this reference to natural law serves what they call a naturalisation tendency of positive law. As long as positive law is presented as a concretisation of natural law, it is sufficiently legitimated, although it is finally based on a decision of the sovereign legislator.21 Once this instrumentalistic approach is hidden behind the ‘thereness’ of law,22 it becomes clear that the codification movement, which serves as our starting point, contains a certain ambiguity which is important for the discussion here: an instrumentalistic approach of law goes hand in hand with a form of representationalism according to which the ‘law in the books’ reflects reality. Campbell formulates this ideas as follows: Essentially, in my view, this stems from an acceptance of law as given, which results in most subsequent thinking resting on the massive assumption that the prescriptions contained in law and laws are in some sense descriptions of the actual world.23

From this perspective, it is clear that there is no place for any theory of legislation, because the dominant ideology in legal thought, even in our day, is not yet 18 See H Hubner, Kodifikation und Entscheidungsfreiheit des Richters in der Geschichte des Privatrechts (Hanstein Verlag, Königstein, 1980) 43. 19 See for this point B Jackson, ‘Legalism’, [1979] Journal of Jewish Studies 7. 20 Shklar, Legalism (n 14 above) 2–3. 21 See R Foqué and A C ’t Hart, Instrumentaliteit en rechtsbescherming. Grondslagen van een strafrechtelijke waardendiscussie (Gouda Quint/Kluwer, Arnhem/Antwerp, 1990) 60 ff and 183 ff. 22 Shklar, Legalism (n 14 above) 3. 23 C M Campbell, ‘Legal Thought and Juristic Values’ [1974] British Journal of Law and Society 13. See also ibid, 23 and 24.

14 Luc J Wintgens completely free from this representation theory. The absence of a theory of legislation, and the limitation of legal science to a science of judicial decisions can further be explained by the reduction of law to a (complete) system of legal rules based on the pseudo-cognitive pattern of legalistic thought which conceals the idea of the legislator as an acting instance with more or less strong instrumentalistic views. In this respect, legality is the formal aspect of legal rules, and therefore a necessary and sufficient condition for the legal character of rules. On the one hand, legality confirms the legal, hence binding, character of the rule (necessary condition); on the other hand, this formal aspect of a legal rule, once it is taken as a sufficient condition, unmistakably leads to a legislative inflation, since there are no other requirements than formal ones for a rule to exist. This is a conceptual argument that complements the empirical explanations of legislative inflation proposed by Karpen24 or Baden,25 for example. Conceptually speaking, the legislative activity of the legislator is bound only by the constitution. In as far as a legal system provides a method of constitutional review of legislation, there exists a possibility of control, but here again, legalism raises its head in the sense that this control arises from the presumption of the constitutionality of the legislator’s ruling. From this perspective, the legislator is presumed to have correctly applied the constitution. In this type of constitutional review, there is only an annulation of the rule in cases of patent violation of the constitution. This means that the control of constitutionality passes via the presumption of rationality of the legislator. Only patently unreasonable rules, those that are manifestly against the constitution, are reversed. As a conclusion to this section, we can say that jusnaturalism makes a theory of legislation superfluous. Knowledge of natural law is sufficient to make good laws, and the criteria, if any, are themselves to be found in natural law. In a positivistic approach to legislation, legalism plays a similar role as in the traditional natural law theories. What legalism makes impossible, however, is a theory about law, according to which the law’s existence (or its creation) and its application are theory dependent. Legalism itself provides a unique and overall background theory of law26 that at the same time is hidden in legal discourse. It cannot be articulated, since this would imply a critique of the rule character of law. Since thinking in terms of rules is still the dominant pattern of legal thinking, legalism prevents lawyers from conceiving law in any other terms than rules. 24 U Karpen, ‘Zum gegenwartigen Stand der Gesetzgebungslehre in der Bundesrepublik Deutschland’, [1986] Zeitschrift für Gesetzgebung 5: ‘die Kürze der Wahlperiode verursacht Gesetzgebungshektik.’ 25 E Baden, ‘Kasuistik und Kommunikation’ in Sozialintegrierte Gesetzgebung: Wege zum guten und verständlichen Gesetz (Bundesministerium für Justiz, Vienna, 1973) 96. Baden argues that the growing complexity of situations that are organized by the legal system causes an inflation of rules in the legal system. 26 Cf R M Unger, The Critical Legal Studies Movement (Harvard University Press, Cambridge, Mass, 1986, 2nd edn) 8 ff.

Legislation as Legal Theory: Legisprudence 15 Any critical attitude to legal thinking will then, necessarily, be a position that is external to the legal system, since critique from an internal position or from a normative point of view is made impossible by legalism. Indeed legalism makes us hold the position that law consists of legal rules considered from an internal point of view that can no longer be questioned, due to a confusion of legality and legalism. Criticism of the legal system is then to be made from an external point of view of an observer of the system who is confronted with rules that are ‘just there’. But such critiques do not allow one to see that the internal point of view is essentially infected with the legalistic perspective on law. Such external criticism, that could eventually lead to a qualitative amelioration of the law, includes the use of legistic techniques or techniques of legislative drafting in order to obtain well-established texts, the use of deontic logic in order to obtain consistent rules, the use of sociology of law, and the setting up of pre- and post-legislative evaluation by organs of the legal system. These approaches, however, lead to detached statements of analysis and remedy for legislation, but not to internal statements, because they leave out of consideration the internal point of view, as they stick to the legalistic pattern of rule application, both by the judge and the legislator. In the end, criticism of the rule can be incorporated in a more or less enlarged judicial freedom in the interpretation of legal rules. However, this never focuses on the core of the problem, which is that rule creation is more than applying other rules under the cover of sovereignty of the legislator. With regard to the thinking pattern of the democratically legitimated legislator, it suffices to respect the principle of legality in the creation of law. The principle of legality is, as a matter of course, a necessary condition for the existence of rules, but it is at the same time a sufficient condition because it regulates both the unquestionable input (legislation) as well as the output (rule application) in legal reasoning. The consequence of this, as Noll has rightly observed, is that the theory of legal thinking is essentially limited to a theory of judicial decision (Rechtsprechungswissenschaft).

III . THE INTERNAL AND THE EXTERNAL

1. The Internal and External Point of View The remainder of this contribution will rely on the well-known distinction articulated by HLA Hart between the internal and the external point of view.27 From an external point of view, an observer sees cars stopping at red lights. Without any awareness of the normative attitude behind this behaviour, he describes it as follows: ‘A red light is a sign that they stop’. This proposition expresses a regularity of behaviour, formulated by someone who does not share 27

H L A Hart, The Concept of Law (Clarendon Press, Oxford, 1981, 11th edn) 87–88.

16 Luc J Wintgens or even know the normative attitude behind this regularity. From an internal point of view, that is the point of view of the actor who accepts a rule as a norm, the proposition made by the actor is as follows: ‘A red light is a signal [for us/for me] to stop’. The actor does not express any regularity of behaviour, he merely articulates his own normative attitude towards a rule, imposing the obligation to stop at red lights.28 A third possibility of norm-conforming action foreseen by Hart results in a proposition that includes, from an external point of view, a description of the internal point of view of the actor. This implies that the observer does not necessarily accept the rule as a norm, while, on the other hand, he includes in his description, this normative attitude of the actor. This gives the following proposition: ‘A red light is a signal for them to stop’. This last possibility was already mentioned in The Concept of Law.29 However, one finds traces of it in other places30 as well as in MacCormick’s analysis of Hart’s work,31 where it is called the ‘hermeneutic point of view’.32

2. Points of View and Theories of Law These three types of propositions from different points of view give rise to three different theoretical approaches to law. (1) The theoretical approach that takes the law from an external point of view and does not take into account its binding character for the legal actors. This is the view of Donald Black, for example, for whom law consists of observable facts, such as the observable dispositions of judges, policemen, prosecutors, etc.33 In another variant, where even behaviour does not play a role any longer, but where the scholar is content with the analysis of mere rules, the external point of view results in the study of law with respect to its (normative) consistency in deontic logic or in relation to its formal qualities required by legislative drafting. (2) The theoretical approach of law from an internal point of view results in a description and a systematisation of valid law.34 It is the position of the legal 28

H L A Hart, The Concept of Law (Clarendon Press, Oxford, 1981, 11th edn), 55 ff and 86–88. Ibid, 87. 30 Hart, ‘Introduction’ in Essays in Jurisprudence and Philosophy (Clarendon Press, Oxford, 1993) 14–15 and id, ‘Scandinavian Realism’ in ibid, 161–9. 31 N MacCormick, ‘H.L.A. Hart’ in W Twining (ed) Jurists: Profiles in Legal Theory (Arnold, London, 1981) 33 ff. 32 Hart’s description of the internal point of view is somewhat confused, when he writes: ‘they look upon it [the red traffic light] as a signal for them to stop.’ The Concept of Law, n 27 above, 87. In my interpretation, this is a hermeneutic statement, made by an observer who does not himself accept the rule as a norm. If the proposition is made by an actor himself, he would say rather, ‘the red traffic light is a signal for us to stop’. 33 D Black, ‘The Boundaries of Legal Sociology’ in D Black and M Mileski (eds) The Social Organization of Law (Academic Press, New York, 1973) 46. 34 See, for example, A Aarnio, Denkweisen der Rechtswissenschaft (Vienna/New York, Springer, 1979) 33 ff; V Kubes, Theorie der Gesetzgebung (Springer, Vienna/New York, 1987) 15. 29

Legislation as Legal Theory: Legisprudence 17 subject or the judge that is translated into theory, starting from the law itself. The legal scholar so to speak, then, behaves like a member of the legal community. This approach best corresponds to the French Ecole de l’exégèse and will be referred to as dogmatic legal theory. (3) The third theoretical approach culminates in a combination of the internal and the external point of view, that is, the hermeneutic point of view. From a theoretical point of view, this implies a description of the rule (external aspect) which includes the normative or internal aspect. In comparison with the first approach, the scholar deals with meaningful behaviour, and not just with observable regularities. In order to make a further qualification of this third possibility, it is necessary to take account of MacCormick’s distinction concerning the internal point of view. The internal aspect of a rule that finds expression in the internal point of view implies, according to Hart, a ‘critical reflective attitude’.35 This attitude is split by MacCormick into a cognitive aspect (‘reflective’) and a volitive aspect (‘critical’). The cognitive aspect—generally speaking—refers to the capacity to recognise patterns of behaviour contained in the rule and to relate them to concrete circumstances. It further refers to the capacity to make an appraisal on whether a behaviour is or is not in conformity with that pattern. The volitive aspect in its turn refers to a wish or preference that the act, or abstention from acting, be done when the envisaged circumstances are obtained.36 Drawing upon this qualification of the internal point of view, I would like to explore the further possibilities it contains for legal theory.

3. The Hermeneutic Point of View MacCormick could be taken to suggest that the cognitive aspect can be analysed from a ‘purely’ internal point of view. The consequence of this would be that rules can have meaning per se (plain-meaning theory). The distinction between a cognitive and a volitive aspect of the internal point of view makes clear that there is both a theoretical aspect and a practical aspect in rule following, but presupposes at the same time that knowing (the meaning of) a rule is something unproblematic. Following MacCormick’s line, the analysis of the cognitive aspect of the internal point of view as a description of the behavioural pattern contained in the rule, implies that the meaning of the rule originates from the rule itself or, in a broader perspective, from the system to which the rule belongs. The type of legal theory that corresponds to this approach can be found in the works of the representatives of the Ecole de l’exégèse. The exegesis of the Civil 35

Hart, The Concept of Law (n 27 above) 55. MacCormick ‘H.L.A. Hart’ (n 31 above) 33. See also id, Legal Reasoning and Legal Theory (Clarendon Press, Oxford, 1978) 275 ff. 36

18 Luc J Wintgens Code was operated from within the code itself, as is expressed by the famous dictum of the French lawyer, Bugnet: ‘Je ne connais pas le Droit civil: j’enseigne le Code Napoléon’.37 The doctrinal explanation of the authoritative texts of the rules of the code, conceptually speaking, did not allow the use of any other means than the text of the code itself (including the neutral method of logic) at the risk of finding oneself ‘outside’ the rule. It is precisely this idea of an interpreter finding himself ‘outside’ the rule that can be related to an important aspect of dogmatic legal theory. It makes clear that the qualification ‘internal’ of the internal point of view can be understood in two different ways. One the one hand, it refers to the familiar idea of the acceptance of a rule as a norm. On the other, it refers to a ‘geographically’ internal point of view, which means that the cognitive aspect of the internal point of view is limited to the explanations of the rules that can be gained from within the legal system to which the rule belongs. In this case, then, meaning is produced within the legal system. Furthermore, the plain-meaning theory provides a good support for this idea, since it claims that clear texts should not be interpreted. The so-called ‘neutrality’ of this theory, however, conceals the fact that it only relies on a possible conception of language that today is by no means the most adequate. Once the plain-meaning theory is considered the only possible conception of language, it becomes clear that it translates reality into concepts that claim to represent the latter. The legal system then becomes a mirror of reality, which means that outside the legal system there can be no relevant legal meaning. As a consequence, any production of meaning is, so to speak, ‘fed back’ to the inside of the legal system. One could say that law then produces its own meaning, which itself is just another aspect of legalism. From an epistemological perspective, this can be explained by the fact that meaning is theory dependent; however, this is concealed by legalism. Meaning is supposed to be a ‘natural’ or ‘plain’ meaning, thus hiding the fact that meaning only makes sense within the theoretical framework of an analytical theory. It is thanks to an analytical theory that phenomena can be understood as phenomena. In Kuhnian terminology, an analytical theory generates a paradigm that, in its turn, determines the types of solution for given types of problems. The latter is at least as important as the former. Since a paradigm allows a problem-solving activity, the first thing a paradigm helps to formulate is precisely the problem. In other words, a paradigm provides the potential of model solutions for specific types of problems that are also generated and determined by the paradigm. Some ‘problems’ are, in that respect, no problems at all, since the current paradigm of a science does not permit us to see them as problems.38 37 ‘I do not know civil law: I teach the Code Napoléon’: see L Husson, ‘Analyse critique de la méthode de l’exégèse’ in Nouvelles études sur la pensée juridique (Dalloz, Paris, 1974) 175. This proposition refers to the second theoretical position mentioned above. 38 See E Zuleita Puceiro, ‘Scientific Paradigms and Legal Change’ in U Kangas (ed) Essays in Legal Theory in Honour of K. Makonen, Yearbook of the Finnish Legal Society, (Helsinki, 1983), 336; Campbell, ‘Legal Thought and Juristic Values’ (n 23 above) 16.

Legislation as Legal Theory: Legisprudence 19 Once it becomes clear that legalism works as the dominant thinking pattern or paradigm of legal science, it follows that legal thinking is a matter of rule following. Legalism, however dominant as a paradigm, is nevertheless but one possible analytical theory. If that is concealed, legalism works as an ideology. In the remainder of this contribution, this form of legalism will be referred to as strong legalism, according to which legality is a necessary and sufficient condition for the existence and the meaning of law. In its strong version, law speaks for itself, since there is no meaning of law outside the legal system. A contrario, the term weak legalism will be used to refer to a conception of law according to which legality or formal validity of a rule is only a necessary and not a sufficient condition for its existence. The ontological presupposition of strong legalism that things (like legal rules) can have meaning by themselves—such that ‘clear rules ought not to be interpreted’—is to be rejected, since it conceals the fact that the clarity of meaning is itself a consequence of a preceding, though maybe unconscious, act of interpretation. A comparison may illuminate this idea. When using the word ‘day’, it could look as if the concept of day is clear in itself. This, however, is not the case. Since ‘day’ means, among other things, that it is light outside, this cannot be understood unless by contrasting it with ‘dark’. The same goes for the concept of ‘night’, which cannot be understood unless one contrasts the darkness as an aspect of the meaning of ‘night’ with the ‘lightness’ as an aspect of the meaning of ‘day’. This means that the concepts of ‘day’ and ‘night’ can only be adequately understood in contrast to each other, that is, while referring to something that they are not. If this was not the case, then, neither the concept of ‘day’ nor the concept of ‘night’ would appear in our conceptual apparatus. The same can be said for (the concepts in) legal rules. Their meaning does not automatically pop out of themselves, that is, law is not self-interpreting, but involves acts of settling meaning which also refer to what the concepts do not mean. However, the comprehensiveness of the legal order, as one of its characteristics according to legalism, precludes this type of explanation because the legal system is held to contain, in an exhaustive way, its own reservoir of meaning. Comprehensiveness of the legal order includes the idea that there are no gaps, such that all relevant legal meaning is to be found within the legal system. There is no relevant meaning to be found outside. It means that, speaking conceptually, the legal system has no need of a social environment to co-determine the meaning of the rules it contains. If this is a possible position, it does not seem to be the most satisfactory one. As the social world evolves, it sheds new light on the meaning of rules that are intended to organise the social universe, unless, of course, this interaction is barred by a strong legalistic conception of law. This can be related to the idea of the internal point of view, more specifically to its cognitive aspect. From a strong legalistic perspective, the cognitive aspect of the internal point of view will be limited to the ‘geographically’ internal domain of the legal order. Conceptually

20 Luc J Wintgens speaking, strong legalism splits the legal order between what is law and what it means (internal to the system) and what is not law (external to the system). This way of thinking is unsatisfactory, because it relies heavily on the plainmeaning theory that has long since been defeated.39 This means that the ‘purely’ internal point of view, as mentioned above, can no longer be accepted, since it limits our knowledge of law to the ‘geographically’ limited domain of the legal system itself. Where reference is made to it in the remainder of the text, it will be called ‘internalc–internalv’ point of view.40 It will not, however, be developed further. The external point of view, the position of the observer who describes behavioural regularities without including the internal point of view, will not be dealt with here either. It is not fruitful from the perspective of legal theory precisely because it excludes the normative attitude of the actor that is essential to an adequate understanding of law. All that remains, then, is what is called the ‘hermeneutic point of view’. The theoretical propositions that result from this point of view include, from the external point of view of the observer, the internal point of view of the actor. This point of view is the one a legal scholar, in the Hartian sense, has to hold, as MacCormick rightly argues.41 When used in the remainder of this chapter, it will be referred to as the hermeneutic point of views.42 This point of view expresses the idea that in order to make legal propositions, one has to rely not on objective concepts, but on legal discourse itself. This means that the scholar describes, without any necessary commitment of his own, the rules of the legal system from the perspective of those whose rules they are. The way in which Max Weber considers the relation between legal dogmatics and sociology can illuminate this perspective. Sociology, when taking law as an object of study, is not interested in the analysis of the objective meaning of legal propositions. In Weber’s opinion, law is an activity and sociology has to analyse the representations men make of the sense and the validity of legal propositions. However, while studying these meanings, the sociologist does not have to adhere to the ‘correct’ meaning from a legal point of view; these are but starting points of his own analysis, to which he will substitute other meanings, from a sociological point of view. In other words, starting from these ‘current’ meanings, the sociologist will connect other meanings to them that will help to (re)define them.43 In connection with what I have discussed, the Weberian scholar thus takes law as an object of study, taking into account the internal point of view of the 39 M van de Kerchove, ‘La doctrine du sens clair et la jurisprudence de la Cour de cassation en Belgique’ in M van de Kerchove (ed), L’interprétation en droit—approche pluridisciplinaire (Fac Univ St-Louis, Brussels, 1978) 13–50. For a general study, see M Van Hoecke, Norm, Kontext und Entscheidung. Die Interpretationsfreiheit des Richters (Acco, Leuven/Amersfoort, 1988) 273 ff. 40 The ‘c’ in subscript means ‘cognitive’, the ‘v’ means ‘volitional’. 41 N MacCormick, ‘H L A Hart’ (n 31 above) 37. 42 Where the ‘s’ in subscript means ‘scholar’. 43 M Weber, Gesammelte Aufsätze zur Wissenschaftslehre, 2nd edn (Tübingen, Mohr, 1951) 449–40.

Legislation as Legal Theory: Legisprudence 21 actors (the representations they have about what the law means) from his own (external) point of view. The combination of both points of view, called above the hermeneutic point of views, then generates new meanings that can be reintroduced in legal discourse. In this respect, there is no strict separation between pure (internal) legal meaning and sociological (external) meaning. Both are, so to speak, intertwined, in the sense that sociological explanation, while generating possible alternative meanings of legal propositions, contributes to a better understanding of law (from an internal point of view). In the following paragraphs, I propose to explore the relationship between the hermeneutic point of views and the point of view of the judge. This will further help us to investigate the point of view of the legislator.

4. The Judicial Point of View In order to elucidate the point of view of the judge, I propose to distinguish between the existence of law and its meaning. This distinction is at the basis of the concept of law advocated in this chapter. This concept relies on four theses that will be briefly articulated here, in order to show the connection between the hermeneutic point of views and the judicial point of view.44 The first and the second thesis concern the existence of law, the third and the fourth refer to its meaning. (1) The validity of law is determined by the legal system itself This means that a rule can only be a legal rule when it is created according to the legal criteria prevailing in a certain legal system. (2) The violation of a moral rule by a legal rule does not annul the latter’s legal validity This thesis is a qualification of the first in that law and morality are different normative social orders that can, however, interact with each other. In order to respect the first thesis, this interaction, when leading to conflict, leaves the legal rule unimpaired. The implication of these two theses for the existence of law is that the concept of law advocated here is non-jusnaturalistic, since (1) law is not based on morality, and by implication (2) a conflict between law and morality has no consequences for the ontological status of law. On the other hand, this concept of law is not purely positivistic either, as an implication of the two following theses. 44 These four theses are presented in a more elaborated form in L J Wintgens, Droit, principes et théories: Pour un positivisme critique (Bruylant, Brussels, 2000).

22 Luc J Wintgens (3) Morality as a social datum can serve as interpretative material in the determination of the meaning of a legal rule within the limits of theses (1) and (2) 45 This thesis follows from the position referred to earlier in this chapter that rules have no meaning per se. However, the possibility of using moral data as interpretative material must be understood in the light of the first two theses. If, on the one hand, legal validity is a matter of the legal system itself and, on the other, moral considerations can play a legitimate role within the social praxis of law,46 it does not follow that the cognitive role played by morality is the only possible cognitive path. This brings us to the fourth thesis. (4) The possible arguments that a moral explanation of a legal rule can generate have, cognitively or conceptually speaking, the same status as explanations of law on the basis of, for example, sociology, economics, etc. In this respect, and in connection with the above theses, morality cannot claim a foundational role in relation to the law. This does not, however, exclude its interpretational role. The limitation of the role of morality vis-à-vis the ontological status of law as expressed in the first two theses has, as a consequence, the effect that other explanations of law cannot be excluded, because morality has, even from the cognitive side, no preferential role in the interpretation of law.47

5. The Judicial Point of View Qualified: Another Hermeneutic Point of View? As stated above, propositions made from a hermeneutic point of view are made from an external point of view and include a description of the internal point of view of the actor. It was called above the ‘hermeneutic point of views’. Given the fact that the ‘internalc–internalv’ point of view was considered unsatisfactory because of its connection with strong legalism, the question to be explored now is whether, as an actor in the legal system, one can uphold a hermeneutic point of view. 45 This thesis reflects my interpretation of Dworkin’s theory of law as formulated in his Law’s Empire (Harvard University Press, Cambridge, Mass, 1988). According to this interpretation, Dworkin does not claim that the validity of legal principles depends on their morality, but that in the interpretation of legal rules morality can play an important role. 46 For example, in order to limit the range of application of a rule in view of its possible immoral consequences. 47 An illustration of the fourth thesis is found in the economic analysis of law. According to this approach, law is studied from an external point of view (economics) including the internal point of view of the judge. Economic analysis of law, like sociological analysis of law from a hermeneutic point of view, can determine other possible meanings of legal propositions (as Weber argues). See G Calabresi and M D Melamed, ‘Property Rules, Liability Rules and Inalienability: One View of the Cathedral’, [1992] Harvard Law Review 1089–128.

Legislation as Legal Theory: Legisprudence 23 The position I would like to argue for is that, once strong legalism is left aside, it cannot be maintained that the meaning of law is a matter for the legal system only. This refers to the third and the fourth theses mentioned above. Relying on these theses, judges make legal propositions from an internal point of view including the external point of view. This is a consequence of the distinction, from a judicial point of view, between the existence of law and its meaning. In a strong legalistic perspective, both are confused in such a way that the task of the judge is limited to the application of the rule in order to avoid, via ‘interpretative work’, the creation of new rules. In what sense, then, can it be said that there exists a link between the hermeneutic point of views and the judicial point of view? The link can be shown by referring to what was said above about the external point of view. The external point of view is the point of view of a scholar for whom law consists of observable facts in terms of behaviour or behavioural regularities.48 The hermeneutic point of views in its turn allows this behaviour to be described including the internal point of view of the actor. That is to say, the observable behaviour is described with an additional dimension—the internal point of view of the actor—as meaningful behaviour or action. From a hermeneutic point of views, the scholar does not take into account the volitional aspect of the internal point of view, in the sense that he does not need to accept the rule as a norm as a judge does. However, what the scholar does share with those holding the internal point of view is the exploration of the cognitive aspect of it. Where, for the judge, the cognitive aspect is distinct but not separated from the volitional aspect, the same is true for the scholar, be it that, from his point of view, the cognitive aspect is isolated from the volitional aspect. But in its core, when exploring the cognitive aspect of the internal point of view or the meaning of the rule, both judge and scholar are talking about the same thing. In other words, when a scholar deals with the description of meaningful behaviour, he collaborates in the articulation of the possible meanings a rule can have from the cognitive side. And as Weber argues,49 these possible meanings articulated from the hermeneutic point of views can help to (re)define what the rule means for those who accept it as norm, hence from an internal point of view. This means that judge and scholar, when dealing with the meaning of a rule, work on one and the same thing. The scholar’s task, from his perspective, is to articulate a range of possible meanings—when, for example, reading the law in the light of economics, sociology, morality, etc.—where the judge’s task is to pick up on one of those meanings in his decision which will then be linked, in an argumentative way, to the legal system. So, from a cognitive perspective, the judge takes a hermeneutic point of view that will be referred to as hermeneutic point of viewj. 48 49

N 33 above. N 43 above.

24 Luc J Wintgens The point of view of the scholar and that of the judge can then be linked in the following way: (1) point of view of the judge (hermeneutic point of viewj) = (a) internal point of view including (b) the external point of view (2) external point of view = external point of view including the internal point of view (= hermeneutic point of views) By substituting (2) for (1b), we get: (3) hermeneutic point of viewj: internal point of view including (hermeneutic point of views). As a conclusion of this, it can be shown that the cognitive aspect of the internal point of view is what links the judicial point of view to that of the scholar. In that sense, judicial decisions are preceded by an explanation of the rule that is all but a deductive operation. It refers to external aspects that can be incorporated in the understanding of the rule in the process of its application. The question I would like to explore in the next section is how these considerations concerning the hermeneutic point of views and the hermeneutic point of viewj can be connected with a legisprudential approach to law. The focus here will be on the point of view of the legislator in his rule-creating activity.

IV . THE INTERNAL AND THE EXTERNAL : THE LEGISLATOR ’ S POINT OF VIEW

1. Legalism, Sovereignty and the Internal Point of View of the Legislator Saying that judicial decisions are not reached by pure deduction from given legal rules risks stating the obvious. Saying that legislation is not a deduction from pre-established data, on the other hand, risks being qualified as silly. However, while judicial reasoning, in order to achieve its required rationality, is transformed into a deductive form,50 the same cannot be said for legislation. In legal systems with a very short history of judicial review such as the Belgian one, the rationality of the legislator and of the legal rules he promulgates was therefore presumed. In the absence of any express provision in the constitution, the Belgian Cour de cassation consistently refused to recognise any power of review by the judiciary.51 Until the attribution of the power of review to the Arbitration Court in 1989,52 there was no judicial review in Belgium. However, it goes without saying that in a democratically organised state based on the rule of law the legislator is bound by the constitution. But what 50

At least in civil law systems. See, for example, Cass, 10 January 1939, Pasicrisie, 1939, I, 1–4; Cass, 29 June 1939, Pasicrisie, 1939, I, 341–343; Cass, 20 November 1962, Pasicrisie, 1963, I, 362; Cass, 25 June 1974, Pasicrisie, 1974, I, 1114. 52 Arts 141 and 142 Belgian Constitution; Court of Arbitration Act of 6 January 1989. 51

Legislation as Legal Theory: Legisprudence 25 does the rule of law mean in the absence of judicial review? As was stated above,53 the absence of judicial review can be related to legalism, this time in the rather perverse sense that the legislator is irrefutably presumed to have followed the rules of the constitution even if it is patently clear that this was not the case. From that perspective, even an unconstitutional rule—that, as a matter of fact, is not a rule—is a rule to be applied by the judiciary. One could, however, expect that, according to the rule of law doctrine, the legislator is bound by higher norms, just as the judiciary is bound by the rules of the legislator. The problematic character of this obvious statement becomes clear once the concept of sovereignty is drawn into the analysis. What does ‘being bound by a rule’ mean for a sovereign legislator? It is quite clear that a sovereign legislator is independent of any external forces on the one hand, while on the other he is the supreme source of law within the legal system. The legislator’s ‘being bound by a rule’, in this respect turns out to be antagonistic to his sovereignty: if the sovereign legislator respects the rule of law, he does so because it pleases him; and if he is really bound by the rule of law, he is no longer a sovereign legislator. It is within this essential tension between sovereignty and the rule of law that legalism appears once again. According to strong legalism on the one hand, law is considered as something given. Hence, the full accent lies on the sovereignty of the legislator, whose ruling is presumed to be in conformity with the constitution. The absence of judicial review or the interpretation of (unconstitutional) rules as if they were in conformity with the constitution is a proof of this strong legalistic attitude. On the other hand, from the perspective of weak legalism, according to which legality is a necessary, though not a sufficient condition of law, the rule of law aspect in legislation still prevails. From this perspective, the legislator is presumed to have followed the constitution, but this time the presumption is held refutable. The latter perspective will be explored further. This can now be reformulated in terms of the internal/external point of view. The legislator can be said to have an internal point of view with regard to the rules of the constitution. This means that he accepts these rules as norms. If this acceptance were presumed in an irrefutable way, then the cognitive aspect of the internal point of view is limited to what was called above the ‘internalc–internalv’ point of view that reduces the possible explanations of the rule to a ‘geographically’ determined field of meaning within the legal system. As was said above, this limitation is due to strong legalism, which conceals the possibility that there can be meaning relevant to the law outside the legal system. The volitional aspect of the internal point of view of the legislator is then translated into the presumption that the legislator has acted in conformity with the constitution, despite any indications to the contrary. As a consequence of this, the discourse with respect to the content of legal rules which is also an aspect of 53

N 23 above.

26 Luc J Wintgens constitutionality, is limited to a tautological discourse in which the criteria of the constitutionality of the rules are determined by the legislator himself.54 A more positive approach to the internal point of view of the legislator in its turn opens a new avenue that gives fuller normative weight to the constitution. According to this approach, the ‘internal point of view of the legislator’ with both its cognitive and volitional aspects, implies that strong legalism is left aside in favour of weak legalism. Legislation can then be considered a form of action,55 and not something ‘just there’. By elucidating the internal point of the legislator from the perspective of weak legalism, it becomes clear that the rules of the constitution are more than negative limits to the legislator’s power. In this respect, generally speaking, the rules addressed to the legislator are powerconferring rules, procedural rules and rules containing rights and liberties of the citizen. In the absence of any judicial review, the constitutionality of legal rules with regard to hierarchically higher rules cannot be checked by the judiciary. A more positive development of the internal point of view of the legislator corresponding to the wish of avoiding strong legalism in the creation of rules brings us to Fuller’s distinction between the ‘morality of duty’ and the ‘morality of aspiration’.56

2. The Legislator between Duty and Aspiration The morality of duty that requires one to abide by a rule without more, best fits into the pattern of strong legalism. In a morality of aspiration, the actor is not merely asked to fulfil his duties, but also that his action realises his best possibilities. The morality of aspiration can in its turn be linked to the pattern of weak legalism. Given a minimal judicial review, not correctly applying procedural rules, for example, can be considered a ‘wrongdoing’. On this point, Dworkin’s distinction between rules or principles57 is illuminating. In most cases, it can be clearly established whether, for example, a procedural rule has been followed or not. Violation of a procedural rule in this respect is a matter of ‘all or nothing’, and the ‘rule’ resulting from such a procedure is clearly not a valid legal rule. However a constitutional rule forbidding discrimination58 is much harder to ‘apply’, since it 54 In this context, the judge is invited to interpret the legal rule, in case of unclarity (read, ‘unconstitutionality’) in a way that best fits with the constitutional rule, thus excluding interpretations that would turn out to be in conflict with the constitution. See, for example, Cass, 20 April 1950, Pasicrisie, 1980, I, 560 ff; see also Arbitration Court, 7 February 1990, no 9/90, Moniteur belge, 19 April 1990, 7363, at B.3. 55 N 23 above. 56 L L Fuller, The Morality of Law, 2nd edn (Yale University Press, New Haven, Conn, 1969) 5 ff. 57 R Dworkin, ‘A Model of Rules I’ in his Taking Rights Seriously (Harvard University Press, Cambridge, Mass, 1977) 14–45, esp 22–28. 58 Art 11 of the Belgian Constitution.

Legislation as Legal Theory: Legisprudence 27 requires a determination of what is to be considered an unjustified distinction. Its ‘application’ is therefore not a matter of ‘all or nothing’. Strong legalism, then, prevents us from asking any questions concerning law other than about its legal validity. This, according to what has been said, is due to the idea that formal validity or legality is a necessary and sufficient condition for a rule’s existence. Strictly speaking, there would even be no need for any judicial review, since a void rule does not have to be annulled. A constitutional court’s decision that such a rule is not valid would be merely a redundant confirmation of a fact, and not an annulation of the rule. Within a weak legalistic approach, however, some forms of the legislator’s action in his dealings with rules can be considered a ‘shortcoming’. So, in this respect, the space for criticism that is absorbed, in strong legalism, by concepts like sovereignty, is opened here, since formal validity is only a necessary condition of a legal rule’s existence. From this perspective, judicial review also deals with the content of rules and not merely with their formal qualities. Although fruitful, judicial review, however, risks the substitution of the judge’s appreciation for the legislator’s, in as far as judicial decisions immediately deal with the content of rules they are asked to consider in relation to the constitution. This may sound somewhat bizarre, since it can be taken to suggest that there are other ways of dealing with the content of rules than the immediate way. I believe there are. In The Morality of Law Fuller indicates this way as follows: In the whole field of human purpose—including not only human actions but artefacts of every kind—we find a pervasive refutation for the notion that we cannot know what is unsuited to an end without knowing what is perfectly suited to achieve it.59

And further: We can, for example, know what is plainly unjust without committing ourselves to declare with finality what perfect justice would be like.60

The idea exposed in these quotations is the possibility of judging without having perfect knowledge of the standards or ends that have to be achieved. Here we are faced with the idea of the principled imperfection or indetermination of human knowledge. However imperfect or indeterminate our knowledge is, though, it is not impossible to make sound judgments about what ought to be the case. Where Fuller61 points to moral reasoning, this model can also be related to legal reasoning. When connected to the position of the judge in the case of judicial review, this brings us to the following. Given the fact that most constitutions are built upon the doctrine of the separation of powers, it would be a violation of these constitutional 59

Fuller, The Morality of Law (n 56 above) 11. Ibid. 61 And others in a similar vein, see G E Langemeijer, Inleiding tot de studie van de wijsbegeerte des rechts (Tjeenk Willink, Zwolle, WEJ, 1973) 233. 60

28 Luc J Wintgens dispositions to allow the constitutional judge to substitute his own views for those of the legislator. Between perfect knowledge of the constitution’s meaning (as a kind of platonic idea) and the subjective judgment of the judge, there exists a margin of appreciation where judge and legislator can have different opinions. Where divergent opinions can coexist, according to the separation of powers doctrine, the legislator’s opinion will prevail. It will always prevail according to strong legalism; it can be struck down in judicial review according to weak legalism. It will be recalled that according to weak legalism, the legality of a rule is only a necessary condition for its existence. What, then, are the other conditions a rule has to satisfy in order to be a legal rule? Fuller says that good legislation is the work of a legislator who aspires to adhere as far as possible to the prescriptions of the morality of law. There is no need, however, to limit the scope of the criteria of good legislation to morality, or to any specific theoretical approach whatsoever. Rather, the idea of legisprudence that is sustained in this chapter aims at searching these criteria in as many available theoretical explanations as possible. A caveat must be inserted here. Any theory of legislation that pretends to establish scientific criteria for legislation would have a mistaken view of its own position in relation to the legal system, because it would miss the essential link with the system, namely the internal point of view of the actors. A legisprudential study of law is not, then, a substitution of the law-making activity of the legislator for the results of legisprudential scholarly activity, as Noll62 and Mader63 have rightly stressed. Such a form of political scientism relies on a misjudgment of the legitimation of politics that are found in the democratic organisation of the state. Although political action has no ultimate rational foundation, as Habermas rightly puts it,64 this is not a reason to transfer the legitimation of political action to a science of legislation. Such a scientification of politics (including legislative action) leads to a failure to appreciate the fact that a legal system in the praxis of creation of legal rules relies on a dynamics of its own.65 62

Noll, Gesetzgebungslehre (n 2 above) 37. Mader, L’évaluation législative (n 3 above) 31. See also B Krems, Grundfragen der Gesetzgebungslehre erörtert anhand neuerer Gesetzgebungsvorhaben der Neuregelung des Bergschadenrechts (Duncker and Humblot, Berlin, 1979) 41; W Zeh, ‘Vollzugskontrolle und Wirkungsbeobachtung als Teilfunktion der Gesetzgebung’ in Gesetzgebungstheorie und Rechtspolitik: Jahrbuch für Rechtssoziologie und Rechtstheorie, XIII (1988) 195. 64 J Habermas, ‘Scientificisation de la politique et opinion publique’ in La science et la technique comme ‘idéologie’ (Gallimard, Paris, 1968) 99. 65 This type of criticism is sometimes made by scholars who are not really enthusiastic about a theoretical approach to legislation, as for example F Schonherr, ‘Zur Oesterreichischen Gesetzessprake und—Technik’ in Sozialintegrierte Gesetzgebung (n 2 above) 337; F Perrin, ‘Possibilités et limites d’une “science” de la législation’ in P Amselek (ed), La science de la législation (Presses Universitaires de France, Paris, 1988) 21–36; S Wronkowska, ‘Das Rechtsgebungsgesetz—Instrument eines rationellen Gesetzgebungsprozesses?’ in H Schäffer and O Triffterer (eds), Rationalisierung der Gesetzgebung: Jürgen Rödig Gedächtnissymposium (Nomos, Baden-Baden, 1984) 278 ff; Zeh, ‘Vollzugskontrolle und Wirkungsbeobachtung als Teilfunktion der Gesetzgebung’ (n 63 above) 194 ff. 63

Legislation as Legal Theory: Legisprudence 29 Reference can be made here to the relationship between the judge and the scholar. The scholar, as was argued above, takes a hermeneutic point of view, that is, an external point of view including the internal point of view of the actors within the legal system (hermeneutic point of viewj). The judge, from his side, as was sustained above, takes an internal point of view including an external point of view (hermeneutic point of viewj). Both points of view deal with the same cognitive aspect of the internal point of view. What is excluded, however, is that the volitional aspect of the internal point of view of the actor would in its turn be absorbed by the cognitive explanations of the scholar. In other words, judges take their decisions on their own. All the scholar’s explanations can offer are illuminations of possible meanings of the rule, not the only true or right one to be used by the judge. A similar, though slightly more complex observation can be made of the legislator’s point of view. This is what will be done in the next section.

3. The Legislator’s Point of View: Yet Another Version of the Hermeneutic Point of View? According to the doctrine of the rule of law, any legislative activity is limited by rules that regulate the form and content of legislation. In this respect, the position of the legislator is similar to that of the judge. There is, however, the obvious fact that the legislator has considerably more freedom than the judge in his dealing with the rules to which he is subjected. Though related to the topic discussed here, the focus will be not be on the legislator’s freedom, but rather on the limitation of this freedom by rules. Where dogmatic legal theory strongly focuses on the judge’s task, there exists no such general theory of the legislator’s activity. In the first section of this chapter, a short explanation for this absence was presented. In this section, I will propose an outline to sketch the outline of a legisprudential approach to law. It will be clear that within the context of this chapter, the ambition can only be to draw the main lines of such an approach, leaving a further elaboration for another occasion. It was said above in connection with the hermeneutic point of view of the judge, that because of its relation with the hermeneutic point of views he can rely on extra-legal explanations. The legislator, by definition relies on extra-legal data, which, according to his policy insights, are translated into rules. Whatever the differences might be between the judge’s and legislator’s freedom in relation to their reliance on extra-legal data, there exists a certain freedom for both of them. Long gone now is the time that legislation was considered an art. When this term is used to describe legislative activity, I believe more things are concealed than explained. Indeed, the qualification of legislation as an art depicts it as a kind of mysterious activity carried on by wise and benevolent men who know

30 Luc J Wintgens best the needs of a community, nation or state. Such a qualification best fits with strong legalism, together with the concept of sovereignty. In such a conception of law-making, the fact that law is related to choices or decisions is not denied, it is simply hidden.66 Nothing serves this idea better than the qualification of legislation as an art. Artistic activity is in itself not rational. This does not, however, exclude that there exist some valuable theoretical explanations of art, but the artistic act in itself, has, as I believe, like political action, no ultimate rational ground. Artistic creation has its own dynamics, that can be theoretically articulated, but as in legislation these articulations cannot take the place of artistic creation itself, since it is theory and not art. When legislative activity is qualified as ‘rational’, this means that it deals with the cognitive aspect of the rules to be followed by the legislator or, more precisely, with the cognitive aspect of the internal point of view of the legislator. Rationality in legislation, then, means that the legislator does more than promulgating, in the form of legal rules, his own subjective preferences. Legislative activity becomes more rational, in as far as the cognitive aspect of the internal point of view of the legislator is taken seriously. How can this be analysed? The core idea that comes to the lawyer’s mind when speaking about validity in a legal context is legal validity. A rule is legally valid when it is promulgated in accordance with the prevailing higher, ie more general, norms of the legal system.67 This type of validity of a rule can be said to be system-related, in that the criteria for legal validity are found within the legal system. This is the first thesis of the concept of law that is sustained in this chapter.68 However, in judicial interpretation the judge (as was argued according to the third and the fourth thesis mentioned above), is not limited to legal criteria alone, due to the fact that the ‘internalc–internalv’ point of view was considered unsatisfactory. Instead, the point of view of the judge was referred to as the hermeneutic point of viewj, thus including the hermeneutic point of views. The legislator can be said to reason along a similar pattern with a stronger emphasis on his freedom. This freedom, however, is not only determined by the constraints of legality, once we face the richer texture of the concept of validity. In this respect, the determination of the cognitive content of a legal rule by the legislator in his rule creation process is complementary with giving this cognitive content a valid legal form. The latter concerns the volitional aspect of the legislator’s internal point of view, but it can by no means be reduced to that. In other words, the legislator, in creating legal rules, does more than applying rules of the constitution. It is, of course, an obvious understatement to say that he relies on social inputs that are transformed into legal outputs. The more concrete question then is: how is this rule creation activity related to the legislator’s point of view? 66 67 68

Z Bankowski, ‘Don’t think about it’ (n 15 above) 47. H Kelsen, Reine Rechtslehre, 2nd edn (Deuticke Verlag, Vienna, 1967) 200 ff. n 44 above.

Legislation as Legal Theory: Legisprudence 31 The answer, I believe, is quite simple. As with the judge, the ‘internalc–internalv’ point of view is an inadequate way in which to describe the legislator’s dealings with rules. Instead, the legislator has an internal point of view, since he accepts the rules that regulate his legislative activity as norms. At the same time, however, the cognitive aspect of these rules is replete with social data that cannot be grasped merely by looking at the rules themselves. Instead, while considering constitutional rules as norms, the legislator at the same time takes into account an external point of view in the determination of the cognitive patterns of the rules he creates. This is another similarity with the judge’s point of view, as was described above. While taking into consideration an external point of view, the legislator does not look upon the social data as raw material. Instead, they are filtered by scholarly work so that they are set up as knowledge about social reality that could be relevant for legislation. Thus, the cognitive aspect of the internal point of view of the legislator is stuffed with cognitive inputs from the external point of view. This means that, compared to the point of view of the judge, the legislator for his part takes a hermeneutic point of view. This will be referred to as the hermeneutic point of viewl. The hermeneutic point of viewl can be related to the concept of validity in the following way. Formal validity of legal rules can be connected to the volitional aspect of the hermeneutic point of viewl. It is an expression of the legislator’s will to give legal validity to a certain proposition. It is a system-internal quality of the rule created by the legislator. But according to the external point of view that is incorporated into the hermeneutic point of viewl other forms of validity can appear, connected to the cognitive aspect of the hermeneutic point of viewl. In this respect, from a sociological perspective rules can be said to be effective, while from the perspective of (social) morality rules can be said to be legitimate. Economics can inform the legislator about economically sound decisions. All these approaches, and others possible, deal with an aspect of the law’s validity from their own theoretical perspective. They produce additional approaches for the legislator about what are valid propositions according to these sciences and thus from an external point of view. The caveat mentioned above can be recalled here. It was said that a legisprudential study of law was not a substitution of legisprudential scholarly work for the ruling of the legislator. With this in mind, I agree with Achterberg69 who points to the question, how it is theoretically possible that extra-legal elements can be introduced in a legal system. On one hand, he ascertains the fact that a scientific control is a legitimate possibility of influence on legislation; on the other, he focuses on the idea that the relation between the legislator and this type of control is merely factual. In other words, the volitional aspect of the hermeneutic point of viewl remains unimpaired. 69 N Achterberg, ‘Rechtstheoretische Grundlagen einer Kontrolle der Gesetzgebung durch die Wissenschaft’ in Theorie und Dogmatik des Oeffentlichen Rechts: Ausgewählte Abhandlungen 1960–1980 von Norbert Achterberg (Duncker and Humblot, Berlin, 1980) 41–50.

32 Luc J Wintgens The important point, Achterberg stresses, is that it is only on the level of constitutional review that this factual or external control can be transformed into an internal or normative one. The consequence is that, via this transformation the criteria for good legislation can be incorporated into the constitution and controlled through the process of judicial review. I would like to illustrate this with a short observation on the way that the German Bundesverfassungsgericht has dealt with this problem.

4. The Bundesverfassungsgericht and the Rationality of Legislation As has been eloquently described by a German scholar, the legislator has a constitutional duty to reflect.70 This duty can, according to the author, be deduced from the principles of Rechtsstaatlichkeit, Demokratie and Bundesstaatlichkeit.71 This duty to reflect is more than following the rules of the constitution and thus implies more than a morality of duty in the creation of legal rules. Rather, according to the decisions of the Bundesverfassungsgericht, the duty to reflect refers to a morality of aspiration, indicating by this a scale of legislative excellence. So where the constitution does not contain explicit provisions for good legislation, this does not restrain the Bundesverfassungsgericht from imposing some constraints of this sort on the legislator. Gusy,72 in an extensive analysis of the decisions of the Bundesverfassungsgericht, has isolated five criteria that are used to measure the quality of legislation from a legisprudential perspective. These are: (1) a duty to establish the facts, (2) a duty to balance, (3) a duty of prognosis or prospective evaluation, (4) a duty to take future circumstances into consideration and (5) a duty to correct legislation at a later stage, or retrospective evaluation. A short comment on them precedes my own considerations. (1) The duty to establish the facts implies that the legislator, according to the Bundesverfassungsgericht, must show that he has made inquiries in accordance with the actual state of affairs in social research. The legislator complies with this duty eg by organising hearings or inquiry commissions, in order to make clear that: ‘der Gesetzgeber sich an dem derzeitigen Stand der Erfahrungen und Einsichten orientiert.’73 70 G Hoffmann, ‘Das verfassungsrechtliche Gebot der Rationalität im Gesetzgebungsverfahren. Zum “inneren Gesetzgebungsverfahren” im bundesdeutschen Recht’, [1990] Zeitschrift für Gesetzgebung 97–107 (99). 71 Ibid. See also H-J Mengel, ‘Die verfahrensmässigen Pflichten des Gesetzgebers und ihre verfassungsrechtliche Kontrolle’, [1990] Zeitschrift für Gesetzgebung 193–212, who pleads for a due process in legislation. 72 C Gusy, ‘Das Grundgesetz als normative Gesetzgebungslehre’, [1985] Zeitschrift für Rechtspolitik 291–99, esp 292–95. For a similar analysis, see C-A Morand, ‘Die Erfordernisse der Gesetzgebungsmethodik und des Verfassungsrechts im Hinblick auf die Gestaltung der Rechtsvorschriften’ in Jahrbuch für Rechtssoziologie und Rechtstheorie, XIII (1988) 11–29. 73 Gusy, ‘Das Grundgesetz als normative Gesetzgebungslehre’ (n 72 above) 293.

Legislation as Legal Theory: Legisprudence 33 (2) The duty to balance starts from the premise that facts do not dictate particular solutions. The central aspect of this duty is that the legislator must show clearly that the regulation he has promulgated is the result of the weighing of several alternatives. An optimal regulation is a regulation that, given all the facts in hand (see (1) ), turns out to be superior to the alternatives. (3) The duty of prognosis or of prospective evaluation is connected with the probability that certain consequences the legislator has in mind will be realised. These consequences, including disadvantageous or uncertain ones, have to be indicated in a verifiable way. Not everything can be foreseen, so this duty has to be understood as what can be reasonably foreseen, which gives rise to the two additional duties. (4) The duty to take future circumstances into consideration implies that the legislator must not remain indifferent with regard to the real consequences of his regulations. Any regulation starts its own life after promulgation and independently of the legislator. On this point a clear interaction between rule creation and rule application becomes visible, which leads to the conclusion that: ‘Gesetzgebung und Gesetzesanwendung werden so zu einem vielfach aufeinander bezogenen, symbiotischen System.’74 (5) Finally, the duty to correct or the duty of retrospective evaluation includes a duty of the legislator to recognise the imperfections of his regulations that would become manifest as a consequence of new insights according to the other criteria. What conclusions can be drawn from this analysis? The duty to reflect imposed on the legislator by the Bundesverfassungsgericht is a clear illustration of a legisprudential approach to legislation as advocated in this chapter. That is to say that the duty to reflect, as a paraphrase of the idea of the rationality of the legislator, gets its shape not by way of a presumption of rationality, but in the concrete requirement that legislation has to fulfil in order to be rational. Rationality of legislation is linked to the idea that legislation implies decision making and that decisions include choices. What the Bundesverfassungsgericht requires the legislator to do is to show how he comes to his decisions. When put into the terminology used above, this means that, according to the Bundesverfassungsgericht, the legislator must have a hermeneutic point of viewl. He cannot hide the exercise of his legislative power behind the pure volitional aspect of a supposed internal point of view—the ‘internalc–internalv’ point of view. On the contrary, the legislator has to show that his decision reflects on an informed choice. Law-making, in this respect, then turns out to be a process in which both cognitive and volitional aspects play a role, and the Bundesverfassungsgericht underlines that the legislator has to show his cards with regard to the cognitive aspect of his decisions. The requirement that these decisions are made on the basis of the actual state of social research clearly indicates the courts’ position on that point. 74

Ibid, 294.

34 Luc J Wintgens This position fits into the pattern of weak legalism, according to which legislation is, at least to some extent, an instrumentalistic relation between values and rules as a means to attain them. The decision of the legislator to use a certain means to realise a specific end however, is controlled by the court. The court thus takes a hermeneutic point of viewj that it shares with the legislator’s hermeneutic point of viewl as far as the cognitive aspect of this point of view is concerned. From a legisprudential perspective, it would be an usurpation of legislative power—and hence a clear violation of the doctrine of the separation of powers—if the court were also to deal with the volitional aspect of the hermeneutic point of viewl. Gusy’s conclusions concerning this method of judicial review are however somewhat pessimistic. According to him, the requirement of adequate fact finding is impossible to satisfy because of both the complexity of society and the fact that Parliament is the least informed instance in society. This is because there are no specific requirements for members of the legislature to have any knowledge of social sciences. And finally, Gusy states: Im Grundgesetz fehlen alle rechtlichen Vorkehrungen, die ein Erkenntnisverfahren auszeichnet. . . . In diesem Sinne ist auch demokratische Mehrheitsentscheidung nicht Herrschaft des sachlichen Richtigen, sondern des Kompromissfähigen.75

He concludes: Das Verfahren der Gesetzgebung ist ein Entscheidungs- und kein Erkenntnisverfahren. Es ermöglicht daher keine Optimierung.76

This conclusion undermines the idea of a legisprudential approach to legislation. It reminds us of Kelsen’s conception of legislation, according to which legislative decisions are pure acts of will wholly determined by politics and not by science. This rather negative view is not subscribed to here, since it focuses exclusively on the volitional aspect of the legislator’s point of view. It therefore denies the idea that legislation is a form of action according to rules, and action, in order to be rational, includes both cognitive and volitional elements. The approach argued for in this chapter, therefore, considers the legislator’s point of view as a particular version of the hermeneutic point of view. The reason for the prevalence of this point of view relies on a conception of coherence, that can be briefly summarised in the next section.77

75

Gusy, ‘Das Grundgesetz als normative Gesetzgebungslehre’ (n 72 above) 297. Ibid, 297. 77 The conception of coherence presented here is a further development of the idea expressed in L J Wintgens, ‘Coherence of the Law’, [1993] ARSP 483–519. 76

Legislation as Legal Theory: Legisprudence 35 5. The Hermeneutic Point of Viewj, the Hermeneutic Point of Viewl and Coherence Coherence of a legal system is co-determined by consistency, in the sense that coherence is a cluster of properties of which consistency is one.78 Consistency in this sense is a condition of coherence. It is a necessary condition of coherence in as far as contradictory norms can hardly be said to cohere. However, two different meanings of the concept of consistency can be distinguished. Concurrent or simultaneous inconsistency means that a legal system (including the level of individual judicial norms) contains two rules that contradict each other.79 Consecutive inconsistency implies that a legal system can contain contradictions over time—judicial decisions can be overruled, rules can change, etc.80 While the former conception of consistency is a necessary condition of rationality, and hence of coherence, the latter is, because of the principle of identity as it seems, not a necessary condition of coherence. Consecutive inconsistencies can make systems imperfectly coherent as Alexy and Peczenik argue from a theoretical perspective.81 In this view, consistency is a regulatory idea, that is, a matter of degree which, in connection with the idea of coherence, makes the latter a variable quality of a legal system. Since law or a legal system refers to a ‘form of life’, as MacCormick and Aarnio rightly puts it,82 coherence, then, is not a matter of logic alone, but a matter of ‘making sense as a whole’. This ‘making sense as a whole’ refers to ‘the whole corpus of the normative system’, and this brings MacCormick to state: ‘To put it crudely, legal decisions must make sense in the world and they must also make sense in the context of a legal system (emphasis added).’83 The requirement that legal decisions make sense in the world means that they must cohere, not only with the legal system as a whole, but with reality as well. This requirement implies that judges, when justifying their decisions, should not only make them correspond with the legal system, but also with social reality. In order to cohere with reality, the judicial point of view should be considered a hermeneutic point of viewj, including both the normative attitude of the judge

78 P W Brouwer, Samenhang in het recht: Een analytische studie (Wolters-Noordhoff, Groningen, 1990) 5 and 25. 79 For example, Cass, 27 September 1977, Pasicrisie, 1977, I, 111. (Appeal judge decided that the suspect had not respected traffic priority rules on the one hand while, on the other, he had priority according to those same rules. Therefore, the decision was reversed.) 80 For this distinction see L A Kornhauser and L G Sager, ‘Unpacking the Court’, [1986] Yale Law Journal 105 ff. 81 Cf R Alexy and A Peczenik, ‘The Concept of Coherence and Its Significance for Discursive Rationality’, [1990] Ratio Juris 130 ff. 82 N MacCormick, ‘Coherence in Legal Justification’ in W Krawietz, et al, (eds), Theorie der Normen: Festgabe für Ota Weinberger zum 65. Geburtstag (Duncker and Humblot, Berlin, 1984) 41–42; A Aarnio, On Legal Reasoning (Turun Yliopisto, Turku, 1977) 126–29. 83 N MacCormick, Legal Reasoning and Legal Theory (n 36 above) 103.

36 Luc J Wintgens towards the legal rule and its possible explanations from an external point of view.84 The conception of coherence underlying judicial decision making can be schematised as follows. (1) Simultaneous consistency is an absolute requirement for all judicial decisions. It simply means that within a judicial decision no inconsistencies or contradictions are allowed. As such, it is a requirement that is found on all the following levels of coherence. (2) First-level coherence or consecutive consistency on the level of judicial decisions implies that judicial ruling should be consistent over time. There are some possible justifications of this requirement. First, in legal systems based on precedents, the stare decisis rule is a legal obligation. Secondly, the rule character of law or its generality does not, without a sufficient reason, allow the application of a rule differently in similar cases. This would, as a matter of fact, deny the rule character of law. Thirdly, and linked to the second justification, consecutive consistency is a matter of formal justice or equality before the law. It implies the requirement that like cases should be treated alike, unless there is a good reason not to do so. (3) Second-level coherence deals with the coherence of the legal system as a whole but is also internal to that system. Judicial argumentation on the second level of coherence is needed when the judge has to make a choice between different rules available in the legal system to decide a case. Thus, a judge who has to decide whether a defendant has a right to damages under a tort rule, or to contractual damages, has to make a choice between two different regimes to come to a decision. Systematic interpretation, or the reading of a statute in the light of another, is a further example of second-level coherence.85 As such, argumentation on the second level of coherence can cause a shift in the legal system, in the sense that, as a consequence of a different rule choice or a (re)combination of legal rules, the resulting decision can be inconsistent with earlier rulings. This (consecutive) inconsistency does not, however, necessarily impair the coherence of the legal system, since without doubt it can make more sense within that system not to follow preceding decisions. But the rules of the legal system are not self reading, nor does any rule choice or systematic interpretation follow from these rules alone. That is, second-level coherence concerns coherence in the context of justification and includes more than stare decisis.

84

Which, as was stated above at n 42, is a hermeneutic point of views. See, for example, Cass 10 September 1971, Pasicrisie, 1972, 28 ff. The court decided that the plaintiff could not exercise his property right according to Art 544 of the Civil Code on the claim that the defendant’s dwelling encroached on an area of the plaintiff’s land of approximately 1 m2. The court ruled that the plaintiff was guilty of abuse of rights, which constitutes a tort in terms of Art 1382 of the Civil Code. This decision relies on a systematic interpretation of both the property rule and the tort rule, thus limiting the scope of the plaintiff’s exercise of his property right. See further L J Wintgens, ‘Law and Morality: A Critical Relation’, [1991] Ratio Juris 177–201. Other examples illustrating this thesis are given in Wintgens, ‘Coherence in the Law’ (n 77 above) 507 ff. 85

Legislation as Legal Theory: Legisprudence 37 When second-level coherence arguments are invoked to justify judicial decisions, it looks prima facie as if they are internal to the legal system. If this were the case, it would follow that judicial reasoning is a matter of logical deduction alone. Where syllogistic or deductive reasoning formalises a result and not the thinking process itself, it is an understatement to say that deductive reasoning alone is unsuitable for reaching a conclusion. Therefore, second-level coherence arguments involve non-legal elements which are transformed into legal arguments. These extra-legal arguments constitute a third level of coherence. (4) The third-level coherence is involved when extra-legal arguments are transformed into legal arguments. Referring to the case quoted above, the Cour de cassation in Belgium decided that abuse of (property) rights is a tort in the sense of Article 1382 of the Civil Code. When reading the Belgian tort rule, one only discovers that it requires a defendant to pay damages for harm caused by his fault. What the court in fact did, was to start from the premise that the exercise of one’s right in such a way that the gain is disproportionately smaller than the loss for a defendant was so unacceptable that it was analogous to a tort. The latter argument is not, however, a legal one, but relies on a conception of morality, that was in turn transformed into a tort. One could come to a similar conclusion on the basis of economic reasoning in that it would be more efficient to rule for the defendant, who is compensated by the plaintiff, while the latter is served with an injunction prohibiting the exercise of his property right. This type of conclusion is by no means logically compelling and refers to an extra-legal context of discovery that allows the articulation of alternative explanations of the rules of a legal system. In this respect, the judge takes a hermeneutic point of view: his decision is linked to the rules of the system that are accepted as norms (internal point of view) while at the same time this solution can be linked to an explanation from a hermeneutic point of views.86 It is via the hermeneutic point of viewj that judges reason on the third level of coherence, whereby the meaning of coherence as ‘making sense as a whole’ includes arguments that refer to a larger whole than the legal system itself. Legislators can be said to reason in a similar vein, according to the legisprudential approach argued for in this chapter. The conception of coherence underlying legislative action can be schematised as follows. (1) Simultaneous consistency is, as in judicial decisions, an absolute requirement, in the sense that within a legislative decision, no contradictions are allowed. Rules permitting one to drive when the traffic light is red, while at the same time obliging one to stop, do not fulfil this requirement.87 (2) First-level coherence on the part of the legislator requires him, in analogy with the judge, to be consistent with earlier rulings. Where the judge’s decisions 86 See also J Dukeminier and J E Krier, Property, 2nd edn (Little, Brown and Co, Boston/ Toronto, 1988), 785 ff; Calabresi, and Melamed, ‘Property Rules, Liability Rules and Inalienability’ (n 47 above) 1089–128. 87 Cf Fuller, The Morality of Law (n 56 above) 65–70.

38 Luc J Wintgens must, as much as possible, be in accordance with precedents (for moral or legal reasons), first-level coherence in the domain of legislation sets negative bounds to it. Although a legislator has a large freedom to change the rules he makes, first-level coherence requires him not to change these rules too frequently.88 This can also be considered a prerequisite of formal justice, as Campbell rightly states: If it is formally unjust not to apply the same rule to similar cases, then it must follow that it is formally unjust to change a rule, since this will involve treating persons who are in essentially the same category differently on the irrelevant grounds that the cases occurred at different times.89

Although the legislator enjoys a larger freedom than the judge in his rule setting, his freedom is limited by the first level of coherence, so that he must give good reasons for changing an existing rule. (3) Second-level coherence in legislation includes the obligation to respect the existing legal system as far as possible. This means that in the creation of new rules, the legislator should take care not to make hay of the legal system. Indeed, some branches of a legal system present a high degree of systematicity (like Belgian contract law, for example). Any change in legislation has the potential to impair this systematicity or systemic coherence. If the legal system is to make sense as a whole (on the intrasystemic level), then second-level coherence requires of the legislator that his new rulings fit as well as possible with this whole.90 This requirement of systematic legislation is related to the structures within the legal system as a whole that have to be respected by the legislator.91 (4) Argumentation on the third level of coherence on the legislator’s part links the legal system to social reality. Indeed, the legislator translates social data into legal rules and on this level of argumentation, he gives arguments for his decisions, for example by referring to the actual state of affairs in the social world. The Belgian Court of Arbitration ruled, for example, that the distinction made between workmen and clerks cannot be justified any longer, given the current legislation and given the actual socio-economic circumstances.92 Knowledge of latter, in order to make the legal system more coherent as a whole, including social reality, can only be gained when the legislator takes a hermeneutic point of viewl. 88

Cf Fuller, The Morality of Law (n 56 above) 79–81. T D Campbell, ‘Formal Justice and Rule Change’, [1973] Analysis 113. 90 See Court of Arbitration, 23 May 1990, no 18/90, Moniteur belge, 27 July 1990, 14.767 ff. The court ruled in this case that the legislator violates the principle of equality if his ruling misjudges the basic principles of the Belgian legal system. 91 See Court of Arbitration, 21 January 1993, no 3/93, Moniteur belge, 5 February 1993; Court of Arbitration, 1 December 1993, no 82/93, Moniteur belge, 19 December 1994; Court of Arbitration, 16 June 1994, no 46/94, Moniteur belge, 6 July 1994. 92 Court of Arbitration, 29 September 1993, no 69/93, Moniteur belge, 27 October 1993; Court of Arbitration, 1 December 1993, no 82/93, Moniteur belge, 16 December 1993; Court of Arbitration, 22 June 1994, no 50/94, Moniteur belge, 7 July 1994. 89

Legislation as Legal Theory: Legisprudence 39

V . SUMMARY AND CONCLUSION

In this contribution on the rationality of legislation, some ideas have been put forward on the relationship between the two types of ruling action, that is judicial and legislative decision-making. After explaining the absence of legisprudential research as a consequence of strong legalism, and after exploring the potential fruitfulness of Hart’s distinction between the internal and the external point of view, I have focused on the hermeneutic point of view. Three variants of the hermeneutic point of view were analysed. From a hermeneutic point of views, a scholar investigates the legal system from an external point of view, including the internal point of view of the actors. The judge was said to take a hermeneutic point of viewj, thus including in his normative acceptance of a rule as a norm some possible alternative explanations made from the hermeneutic point of views, while the legislator takes a hermeneutic point of viewl. The latter means that the legislator takes the rules of the constitution as norms to follow, but at the same time he gains possible explanations of the meaning of these rules from an external point of view. This was but a theoretical way of saying how law is linked to social reality. This opens an avenue for a legisprudential approach to law, that is, the study of rational legislation. Legisprudence has both a theoretical and a practical aspect. The theoretical aspect involves questions on the concept of sovereignty, the relation between the legal system and social reality, both from a judicial and a legislative perspective (and the similarities between both). This relation, as was argued, is based on a conception of coherence of a legal system. The practical aspect of a legisprudential approach of law concerns the elaboration of concrete criteria of rational legislation. They are gained from within the legal system relying on its own dynamics according to the hermeneutic point of view of authoritative actors.

3

Making Society through Legislation LARS D ERIKSSON

I . INTRODUCTION

In order to be rational social actors we have to be able to predict what we are doing; we have to know the consequences of our decisions. But do we necessarily have to be consistent, and do our decisions necessarily have to be internally coherent? I admit that as lawyers and as legal scholars we should aim at consistency and coherence, but does this imply that the lawgiver, too, has to act in the same way as the judge, the lawyer, and the legal scholar? I am not at all sure about that. Legisprudence is not, and should not be, jurisprudence. As a matter of fact we are never able to predict all the consequences of our decisions. Because of that the rationality of what we are doing and what we decide is often highly questionable. It is, indeed, much easier to act consistently and to make decisions which are internally coherent than to act or make decisions the consequences of which are predictable. I think, therefore, we usually prefer a definition of rationality which emphasizes consistency and coherence. It is more secure to choose a definition of rationality which insists upon consistency and coherence than a definition laying stress on the predictability of the consequences of our decisions. In my view legisprudence—if it wants to be a serious theoretical approach— must deal primarily with questions of predictable consequences. Legisprudence has to deal with the most focal problems of the lawgiver, not with the most focal problems of the judge or legal scholar. In this chapter I am going to challenge one of our most common understandings of the welfare state. I am going to claim that the welfare state—in its consequences—should be considered not a Sozialstaat, but an outcome of the classical liberal project. When saying that I am not, of course, saying that the liberalism of the welfare state should be regarded as a deliberate conspiracy of the liberals. I am only trying to say that the conceptual tools used by even the most convinced welfare state proponents (be they legislators, politicians or lawyers) have been derived from the liberal tradition. I do believe that the sincere intentions of the proponents of the welfare state were to frame a state and a society based not only on individual freedom and property, but also on social justice and equality.

42 Lars D Eriksson Nevertheless, what we are seeing today is a welfare state based on an extremely individualistic morality of rights and on a legal system in which individual rights constitute the most fundamental aspect of the system. We do not see today a welfare state based on relations of mutual responsibility and reciprocity, neither in morality nor in the legal system of the welfare state. Justice and equality are, in the last resort, collective values. A society based on them would be a society and a legal order based on at least some communitarian principles. But the logic of individual freedom and individual rights do not allow for any communitarian morality or communitarian legal institutions. The liberalism of the welfare state has not only resulted in a morality of rights and an institutionalised order, the intrinsic principle of which is the safeguarding of individual legal rights, it has also resulted in a pulverization of social bonds and social structures which could possibly have functioned as strongholds for collective values and alternative organizational and societal principles. The paradox I shall be examining is that the legislative intentions of the proponents of the welfare state never materialised in the way the lawgivers presupposed. The consequences of welfarism were not the consequences expected by the lawgivers.

II . THE LIBERAL STATE AND THE WELFARE STATE

Since the Second World War the liberal state and the welfare state have been regarded as two qualitatively different state formations. The liberal state has been considered an expression of unfettered market economy and capitalism: the scope of state power should be minimized, the state should not intervene in social or economic processes except when absolutely necessary in order to guarantee personal freedom, order and security. The diverse subsystems of society should be autonomous orders functioning according to their own internal logic. The welfare state, on the other hand, has been regarded as a state formation in which crude market economy is restrained and modified through state interventions in the economy and through the extension of still more comprehensive social security networks. Liberals claim that the fundamental value of the liberal state is freedom: the individual should be guaranteed optimal freedom to realise his or her personal desires and goals. The proponents of the welfare state maintain that the fundamental value of welfare is social justice: individuals should be guaranteed equal real possibilities to live a tolerable life, ie, everyone should be guaranteed a reasonable standard of living irrespective of age, health and work. The welfare state thus presupposes social equality and a fair distribution of income. So we are used to thinking, but I believe it is time to question these conceptions now, when welfare states experience serious crises. I am asserting that the welfare state does not constitute a qualitatively new state formation, and I am maintaining that it does not differ essentially from the liberal state. On the contrary, I am convinced

Making Society through Legislation 43 that the welfare state is an expression of a liberalism developed to its utmost limits. I will try to be more concrete. It is the existence of civil and political rights that distinguishes the liberal state from all earlier state formations. The characteristic feature of these rights is that they confer on the citizens and individuals rights against the state. The state is obliged to respect the freedom of expression of the individual citizen, to respect their property rights, etc. This means that the liberal state in its initial stages of development first and foremost played a passive role: civil rights set limits to what the state was allowed to do and not to do. The rights gave the citizens certain concrete claims against the state. So far but no further! The rights functioned vertically: they gave the citizens claims against the state and created corresponding obligations for the state. But the rights did not function, and do not even today function horizontally, ie they do not give the citizens any rights in their relations with one another. The rights relation is essentially a relation between the citizen and the state. Individuals are allowed to do whatever they want in their mutual relations. The welfare state introduced social, economic and cultural rights, for example, the right to a job, to education, to a fair standard of living, to healthy working conditions, to a safe old age, etc. It is precisely the existence of these rights which is considered characteristic of the modern welfare state and which— according to the prevailing view—constitutes it as a new type of state. But let us have a closer look at the question of whether the social, economic and cultural rights actually offer anything qualitatively new. It is often said that the purpose of these rights is to establish actual equality among the citizens, whereas the purpose of traditional civil rights is only to safeguard the freedom of the individual. That might be true, and I do not contest it. It is also often argued that social, economic and cultural rights presuppose active efforts and measures by the state: the state is obliged to actively create new jobs, to produce educational facilities for its citizens, to develop health care, to guarantee the subsistence of elderly people, etc. In other words, social rights presuppose active interventions by the state in contradistinction to traditional rights which, as we saw, explicitly forbid the state to intervene in constitutionally protected spheres of liberty. To sum up, first the purpose of social rights is to promote social justice, whereas the purpose of traditional rights is to safeguard individual liberty; and second, while traditional rights oblige the state not to act, social rights directly require it to do so. It is unquestionable that welfare rights have conveyed new tasks on the state authorities and that they have made life more tolerable for a great number of the citizens. I do not deny that. What I am denying is that they have overcome liberalism. First: in fact, welfare rights also function vertically, in other words in the same way as traditional rights. The individual citizens have to make their claims for the fulfilment of their social and economic rights to the state authorities. Social and economic rights do not function horizontally. I cannot demand a job from

44 Lars D Eriksson my employer if I am unemployed, and I cannot demand that my family, my friends or my neighbours take care of me if I am sick or old. Secondly: welfare rights, too, are individual rights. I only have an individual right to my child benefit, to my unemployment benefit, to my sickness benefit or to my superannuation benefit, etc. I am able to implement these social and economic rights only as an atomised, isolated individual. The logic of social and economic rights in this sense exactly as individualistic as traditional rights. Thirdly: despite the fact that social rights imply, at least indirectly, active state intervention in societal and economic processes, they have nonetheless left intact those social and economic structures that in the last resort generate social injustice. I am primarily thinking of the structures of the market economy which social rights have so far not upset in any important way. From this perspective the welfare state seems to be only a variant of the old liberal project. The crisis of the welfare state today is at least partly a result of the fact that there are in principle no limits to how far welfare rights can be stretched. These rights as a matter of fact are limitless: when new individual needs appear, or when we become conscious of what our needs actually are, we are always able to put our demands that they should be fulfilled in the form of legal rights. As long as there is guaranteed economic growth, these expansive demands on the state do not cause any problems; but in a time of economic decline the system will face the imminent risk of collapse.

III . SOME PROBLEMS RELATED TO WELFARE RIGHTS

Individual social rights imply that the state authorities have specific obligations and responsibilities for their fulfilment. a) The assertion that this rights-centred system has resulted in a decline of moral responsibility in the citizenry is hardly totally unfounded. Social rights, as we saw, only create individual claims against the state, but they do not impose any social obligations on the citizens and individuals themselves. The fact that I have a right to sickness benefit does not impose on me any obligations to take care of those sick persons who are closest to me. And the fact that I receive unemployment benefit does not oblige me to assist other unemployed persons to get a job. The system as a whole compels me to concentrate solely on my own rights, and undertakes that my rights will be respected and taken account of. I have a right to demand a bed in a hospital when I am sick, irrespective of whether my best friend would be in greater need of it. I have a right to unemployment benefit, irrespective of whether my neighbour would be in still greater need of a job than I am, etc. In the long run, all forms of social solidarity and fellowship will diminish. b) There are more problems, however. Social rights are only rights for the living generation. They lack a historical dimension. The logic of the rights does

Making Society through Legislation 45 not demand that we consider future generations, ie they do not impose on us any obligations as to the future. I have a right to my pension which I have partly financed myself, but I do not have any obligations to promote the safeguarding of a reasonable living standard for those elderly people who come after me. c) And further, the fact that as individuals we have rights without corresponding duties too easily results in passivity in our relations with the political system in general. I can confine myself to making my claims either orally or in writing, and—if for some reasons they are not accepted—appeal to the appropriate higher authority. I do not need the help or support of others, except perhaps of a lawyer. If I am unemployed, for example, and do not receive my unemployment benefit, I should not in any case turn to other unemployed persons in order to mobilise a protest: I should lodge an appeal against the decision that offered me too little. Is it, therefore, so strange that there are no widespread actions of protest among the unemployed? The welfare state has created a system of individualistic rights that as a matter of fact demobilise the citizens instead of mobilising them. d) In other words, welfare rights contribute to a general depoliticisation of social questions. This depoliticisation does not, however, only express itself through general passivity. It also manifests more specific expressions. Social rights often offer the authorities extensive powers to interfere in traditional civil rights. In order to receive a social right that I am entitled to, I can be forced to renounce some civil right. It is evident that this kind of conflict of different rights has consequences for individuals’ conception of themselves as independent and adult persons. The purpose of civil rights was to create independent and autonomous individuals, but that purpose can be totally frustrated by social rights to which conditions are attached, that deprive the citizens of their civil rights.

IV . AN EXEMPLIFICATION : WHAT HAPPENED TO THE JOB ?

Work, in the same way as the family over the centuries, has functioned as a welfare institution. Work has not only been a place, but a social community, where those who worked together remained in at least some kind of mutual relationships of responsibility and solidarity. What has happened during the liberal era—and I include in it the epoch of the welfare state—is that work has tended to develop into an individual right, ie a right to salary from a certain place of work, or if a job is nowhere to be found, a right to unemployment benefit. The social functions of work have more or less fallen into the background. The right to work, which ideally might be regarded as a right to a social community guaranteeing, among other things, my subsistence, has in practice been reduced to an individual right to salary for a job done or, alternatively, to a right to be entitled to some social benefits where no job is available. In these circumstances, is it so

46 Lars D Eriksson odd that, for example, the trade union movement today faces never-before experienced problems to engage and organise its members? Within the framework of a totally rights-centred and individualistic work concept it is, of course, difficult to mobilise the workers for joint actions in the name of common interests, which are, as a matter of fact, the common interests in a world where the only societal function of work is that it creates individual entitlements to a salary or to some economic benefits. I admit that what I have said so far sounds unforgivably reactionary. What I am trying to say, however, is that the very strong rights-centredness of the classical liberal state as well as of the modern welfare state has had as its consequence the disappearance of many social structures of solidarity and reciprocity which at earlier stages functioned as welfare institutions on the so-called micro levels of society. In this process many horizontal relationships, including rights as well as obligations, have dissolved. We also see the consequences of this development at the macro levels of society. Discussions and debates about common interests are exceptional phenomena today. Legislative welfare projects based on reciprocity and solidarity for those involved are rejected as ‘unrealistic’ or ‘unacceptable’. A great part of the political debate deals with disputes about how conflicting special interests should be balanced. Debates, for example, about ‘the good society’ or ‘the common good’ are conspicuous by their absence, but debates abound about our special interests and how we would be able to implement or strengthen our individual rights.

V . IS THIS REALLY A CONTRIBUTION TO LEGISPRUDENCE ?

Yes, I think it is (provided that there exists any elaborated theory of legisprudence). It is not only important for theoretical reasons, but also for practical ones, that we analyse the basic conceptual tools we consciously use when we think about law or when we make law. These conceptual tools are often, in the last resort, ideological concepts. In this chapter I have tried to uncover the liberal paradigm of legal rights. What characterises these rights is, in my analysis: first, that they are individual rights, and exclusively individual rights; second, that they are vertical rights; and third, that they are rights without corresponding individual obligations. Within this conceptual scheme we cannot arrive at goals which themselves are non-liberal, eg communitarian. I hope, therefore, that my analysis sheds some light on the relationship between law and power. I believe that no one doubts that the law often functions as an instrument of political power, but if my analysis is correct, there also exists a more profound relationship between law and power. The conceptual scheme in the framework of which law has to operate— a scheme which in the last resort is based on certain political and ideological conceptions of how the state and society should be organised—‘forces’ the law

Making Society through Legislation 47 to produce only a certain kind of legal norms and rules. These norms and rules cannot contradict what they are supposed to confirm. The power in law, in other words, is the power of its politico-ideological presuppositions. When the basic legal concepts in the law are based on a liberal conception of society, the consequences of the legislation for society cannot be anything other than liberal. The lawgiver, who perhaps aspires to frame a society of solidarity, reciprocity and equality, will, through the law he or she produces, give rise to a society of extreme individualism and social fragmentation. The lawgiver will be a prisoner of his or her legal concepts. This is a fact which is often neglected in the theory of legislation. It is neglected because those who are actually preparing the statutes for the lawgiver—mostly lawyers—are thinking in terms of legal consistency and coherence. They prefer logic to politics. They prefer legal tradition to political and social change. I can see only one way out of this dilemma, which in fact makes the lawgiver simply an executor of the will of the administration: the lawgiver has to base his or her legislative decisions on detailed analyses of what the consequences will be of the legislative decisions made. The lawgiver has to be able to foresee the effects of his or her decisions. The coherence and consistency of law is not the primary problem of the lawgiver. The lawgiver’s primary task is to produce laws the consequences of which correspond to the aims he or she aspires to. This may, of course, cause new problems for the judges, lawyers and legal scholars, but that is another story. We are not talking about jurisprudence in this volume; we are talking about legisprudence.

4

Rationality of Legislation in a Sociological View KAUKO PIETILÄ

I . AN OVERTURE IN PHILOSOPHY

Kant views legislation in relation to the dignity (Würde) and the reasonableness (Vernünftigkeit) of human beings. He combines the three things into a single complex by speaking of: ‘the dignity of a reasonable being which obeys no other law than the one that it, at the same time, itself gives.’1 The quotation says that dignity, reasonableness and the practice of obeying only self-imposed law go inseparably together. They constitute a wholeness in which the components presuppose and endorse one another. We can also read Kant’s dictum—that a being has reason and dignity if it obeys no other law than the one it enacts itself—from the other end: if laws are enacted without the being to whom they apply taking part, they are not reasonable. If the rational equals the reasonable, such legislation is not rational. Finally, if we are committed to following our reason in our action, which is recommendable since there seems to be no better ultimate authority to guide it, then the laws so enacted should not be obeyed.2

II . THE PROBLEM ITSELF

The difficult question is how Kant’s dictum could be realised in people’s actual mundane social practices. Kant speaks of his reasonable being in the singular, 1 I Kant, Grundlegung zur Metaphysik der Sitten in Werke, IV, (Georg Reimer, Berlin, 1911/1785), 434. I owe this starting point to E Garzón Valdés, ‘Legal Security and Equity. Otto Brusiin Lecture 1992’ in A Aarnio, K Pietilä and J Uusitalo (eds), Interests, Morality and the Law (Research Institute for Social Sciences, University of Tampere, publication no 14, 1996). Two viewpoints are implied, one observing price (Preis), the other dignity (Würde). These are, says Kant, mutually exclusive: ‘In the realm of purposes everything has either a price or dignity. That which has a price, that can be also replaced by something else as its equivalent; that which in contrast is elevated above all price and allows, hence, of no equivalent, that has dignity.’ 2 My purpose is to avoid situations in which laws exist that I, following my good reason, should not obey. This is because in such conditions some authority must try and make me obey.

50 Kauko Pietilä yet the law can never be a private, an individual concern. A dictionary says that the law is ‘a rule or set of rules, enforceable by the courts, regulating . . . the relationship between the organs of government and the subjects of the state, and the relationship or conduct of subjects towards each other.’3 The law regulates relationships between agents. An agent cannot have a law just for itself. I cannot materialise Kant’s fundamental requirement by simply commanding myself to behave consistently from this moment on in a definite way and by conscientiously following this order. We therefore have a practical problem with Kant’s construction of dignity and reason in relation to legislation: if I say that you have an obligation to obey the law you have legislated yourself, but no other law, then the personal pronouns you and yourself must be understood, whatever the composition of Kant’s vernünftiges Wesen, to refer to a number (greater than one) of people, and actually be written as ‘you’ and ‘yourselves’.4 The law’s empire always involves more than one individual. The practical problem is to find for those included in the ‘you’ and the ‘yourselves’ procedures through which they can decide what rules to obey and what not to obey. In agreeing about that, they participate in enacting their laws, and consequently have reason and dignity. What might these procedures be? I suppose that not infrequently large numbers of people become subject to laws without being in any demonstrable sense partners to the enactment of those laws. Such condition is rather the rule in today’s world. The exciting and sensational contingency, the exceptional case—and yet the standard and the yardstick—is the occasion in which the many succeed in making themselves the lawgiver itself. How can this be? How is self-government to be practically organised in conditions in which the ‘self’ of this self-government is composed of a myriad of more or less independent and separate human beings? The problem was not rendered obsolete by the adoption of the so-called democratic procedures to appoint the sovereign. This can be seen, from what Habermas calls his materialistische Grundfrage: How might a differentiated social system that lacks both summit and centre still organise itself, once one can no longer imagine the ‘self’ of self-organisation embodied in the form of macro-subjects such as the social classes of theories of class, or the people of popular sovereignty?5

3 Collins English Dictionary and Thesaurus, Version 1.0 (WordPerfect © and HarperCollins Publishers, 1992). 4 The whole problematic originates from Rousseau’s idea of the general will which cannot be easily used, says Chapman, ‘for the purpose of legitimating constitutional democracy on the scale of the national state’. See J W Chapman, ‘General Will’ in D L Sills (ed), International Encyclopaedia of the Social Sciences, VI (Macmillan/The Free Press, New York, 1968), 86. The same difficulty can be recognised in Kant’s idea. 5 J Habermas, ‘Die nachholende Revolution und linker Revisionsbedarf: Was heißt Sozialismus heute?’ in his Die nachholende Revolution: Kleine Politische Schriften VII (Suhrkamp, Frankfort on Main, 1990) 195–96.

Rationality of Legislation in a Sociological View 51 Yet in some sense the solution to the problem I am considering here must be in the idea and practice of democracy. Therefore, let us briefly reflect on how representative government can be organised in practice.

III . MODES OF REPRESENTATION

Consider two forms of representation, presented and analysed by Max Weber.6 One is free representation (freie Repräsentation), the other representation through interest deputies (Repräsentation durch Interessenvertreter). Free representation is embodied in the representative assembly of the modern state, the Parliament, whose: function is not understandable apart from the voluntaristic intervention of the parties which present the candidates and programmes to the politically passive citizens. They also, by the process of compromise and ballot in the parliament, create the norms which govern the administrative process, control the administration itself, support it by their confidence, or overthrow it by withdrawal of confidence if, by virtue of commanding a majority of votes, they are in a position to do this.7

Free representation is the very model of democracy. It sees the Parliament as the centre of the political process in which the principal political forces of society assemble in order to legislate durch Kompromiß oder Abstimmung. The endless variety of conditions and life forms in the constituencies is compressed through parties into debatable dimensions, into few principal alternative programmes.8 6 M Weber, Wirtschaft und Gesellschaft. Grundriss der verstehenden Soziologie, 5th edn (J C B Mohr [Paul Siebeck], Tübingen, 1980/1922), 172–76. Weber has five forms of representation; two of them deserve a closer look here: ‘Genuine parliamentary representation with the voluntaristic play of interests in the political sphere; the ensuing plebiscitary party system with its consequences; and the modern idea of rational representation by interest groups, these are all peculiar to the western world . . .’ (176, emphasis added). Weber reserves the attribute ‘rational’ for the latter form only. 7 Ibid, 172. For all his criticism of Weber, Beetham’s liberal democracy comes fairly close: ‘The principle of popular sovereignty is carried through consistently in electoral terms, in that not only the legislature but the head of the executive also is chosen directly or indirectly by ballot, on the basis of universal suffrage. The formal guarantee of the freedoms of belief, expression and association, exercised within an autonomous sphere of civil society, ensures that, in principle, electoral outcomes reflect the “popular will”, and that public opinion can be brought to bear on government in a continuous manner.’ (D Beetham, The Legitimation of Power (Macmillan, Basingstoke/London, 1991), 163–64. Beetham adds another dimension by stating that, in liberal democracy, ‘[e]conomic activity is based upon the private ownership of property, and upon private decisions about investment, employment and consumption within a formally free market’. I suggest doubt in relation to Beetham’s definition. First, what is brought about in elections is perhaps not popular sovereignty but a ritual that confirms the balloting people’s next-to-complete political impotency. Secondly, it is not entirely absurd to think that modern capitalism has long since outgrown the conceptual livery of ‘private ownership of property’, of ‘private decisions within a formally free market’. (I shall give you an example of what may happen if hypermodern capitalist economy is doctrinally conducted with the outdated tools of free privatism.) 8 The precondition for such compression is that the constituencies can see themselves divided into few distinct groups on some grounds essential to them. The grounds used to be the ones that gave reason to speak of classes in the sense Marx defined them: ‘The owners of labour power, owners of capital, and land-owners, whose respective sources of income are wages, profit and ground-rent, in

52 Kauko Pietilä These relatively few parties/programmes struggle for votes in elections, and they struggle and negotiate about legislation and policy in the Parliament. Now this gives me, under the premise of universal suffrage, the citizen’s opportunity to participate, albeit indirectly, through my representative, in legislation. Such arrangement already approaches self-government and participation in lawgiving under the conditions of immense modern society.9 It is thinkable that liberal democracy became materialised to some extent in some place at some time. In Finland liberal democracy may have been at its most vigorous in the period between the end of the wars in 1944 and the end of Dr Kekkonen’s presidency in 1981. Since then a radical reorganisation—a silent revolution, in fact—has taken place. In terms of the forms of representation, the change is from free representation to interest deputation. Weber describes the latter: Representation through interest deputies refers to such representative bodies to which the members are not selected freely, without taking into account the candidates’ affiliations to occupation, estate, or class, but to which people choose representatives who are of their own kind with respect to occupation, social estate, or class position, and who assemble to form—as the phrase currently most often has it—a ‘representation of occupational estates’. . . . [T]he professional advocate of an interest group can only be a person who can devote his whole time to his function; among the non-propertied strata this task hence devolves upon the paid secretaries of interest organisations.10

Weber says that ‘[t]he chances for representation based on “occupational estates” are not insignificant’.11 They involve, for example, the following rational prospects: Under these modern conditions, the thoroughly rationalised system of professional ministers and prefects, as in France, offers significant chances for pushing the old forms everywhere into the background, probably supplemented by interest deputies, normally in the form of advisory bodies recruited from among the economically and socially most influential strata. This practice [. . .] is becoming increasingly frequent and gradually may well be ordered more formally. This latter development, in particular, which seeks to put the concrete experience of the interest groups into the service of a rational administration by trained specialised officials, will certainly be important in the future and further increase the power of bureaucracy. It is well known how Bismarck sought to make use of the plan for a ‘National Economic Council’ as a weapon against the parliament, accusing the opposing majority—to whom he would have never granted the right to parliamentary investigation as practised in England—

other words, wage-labourers, capitalists and land-owners, constitute then three big classes of modern society based upon the capitalist mode of production.’ (K Marx, Capital. A Critique of Political Economy, III, F Engels (ed) (Lawrence & Wishart, London, 1977/1894) 885. 9 Barber suggests ways to improve the democratic performance of representative governments. See B R Barber, Strong Democracy: Participatory Politics for a New Age (University of California Press, Berkeley, 1984) 261–311. 10 Weber, Wirtschaft und Gesellschaft (n 6 above) 175. 11 Ibid, 176.

Rationality of Legislation in a Sociological View 53 of trying to prevent officialdom, in the interest of parliamentary power, from becoming ‘too knowing.’12

This shows what is at stake when free representation and interest deputation become involved in a conflict—the legislative power of the Parliament. The latter form rising to predominance, Parliament’s formal position may remain intact, but legislation is substantially no longer the business of the Parliament but primarily the prerogative of the government—the ‘thoroughly rationalised system of professional ministers and prefects’—supplemented by their advisory and consultative extensions invaded and occupied by deputies of interest groups. Weber rather celebrates this as one more conquest of the all-pervading rationalisation13—which gives me the opportunity to play Weber and Kant against each other. Can we deem interest deputation as conducive to rational legislation if we build our notion of the rational on Kant’s dictum quoted at the beginning of this chapter? We can boil this down to the simple question: Am I, through representation based on interest advocacy, a participator in the enactment of laws? Are you? To my knowledge I am not a member of any interest organisation; but I might be. Some such organisations may nevertheless count me as one of their clientele. Through such organisations—and through representation based on interest deputation in general—I am oriented to my interests and to the interests of the groups of which I am a member. I am not oriented to legislation the way I used to be when I participated in general elections in the bloom of liberal democracy: when I acquainted myself with the party programmes and their background philosophies, weighed them and debated them, chose the party and the candidate, cast my vote, followed the action in the Parliament, and so on. Every reasonable human being knows now that today such interest in parliamentary politics is without sense: —the programmes of the parties no longer differ in any principled respect; —the differences between the policies the parties are committed to, and pursue, are even more negligible: all parties practise—would practise if they could clear their way to the Cabinet—exactly the same policy designed and dictated to them by the government, the civil service, and the organised interests behind them; —the only real policy followed by all parties is to win enough votes in the general election to be invited to form a Cabinet; questions of policy are so utterly trivial that the recent Cabinets of Finland, for example, contained in absolutely 12

Ibid, 576. Bleaker views see the classical republican model, Jean Bodin’s ‘puissance absolue et perpétuelle d’une République’, to be faring badly in modern times. Views of decay see the autonomous state capitulating to society: ‘The modern state would be largely occupied and instrumentalised by social power complexes; it would no longer bring about its central services of order as a neutral body standing above the separate interest groups, but as an executive of particular societal forces.’ H Dreier, ‘Souveränität’ in Staatslexikon, IV, 7th edn, (Herder, Freiburg, 1986), 1207. 13

54 Kauko Pietilä frictionless co-operation the former conservatives (National Coalition Party), the former left socialists and communists (now they call themselves Left Alliance), the environmentalists (Green Party), and the social democrats (who can no longer remember what it meant to be social and democratic); —all parties appeal to the same general public, and advertise themselves primarily as so many groups of nice people who espouse all the good things the middle-class imagination can ever invent; —the only figure in general elections having any societal significance at all is the overall turnout percentage reflecting the credibility or believability (trust as Giddens puts it14) enjoyed among the common public by the economic and administrative system gathered around the government, the civil service, and the organised interests. The constitution granting me the right to use my vote in parliamentary elections once in every four years, secures me a right that is practically null and void. The conclusion is, in the terms used in this chapter, that I can have legislation as my aim and purpose no longer in any conceivable way, neither under free representation nor under interest deputation.15 In these circumstances: —the laws are enacted without my participation; —hence, I am not, as a reasonable being, obliged to grant voluntary obedience; —such obedience would be unreasonable; in other words, recent legislation is not reasonable = rational; —therefore, I have no dignity (even though I may have a price, presumably cheap).

IV . THE SENSELESS LAWGIVER I : THE CASE OF REDUCED TAXES

As I have no dignity, I must be a clown and a fool. Let me therefore behave like one and present two examples of the quality of legislation—of the performance of the legislature—in the present conditions in which interest deputation is the dominant form of representative government. A legislature that has sunk into insignificance can have, we may presume, no reasonable idea of what it means in the first place to legislate. Let me try and corroborate. In Finland, taxes are imposed through the enactment of laws that ordain the duties; taxation requires legislation. It is the diet that legislates.16 14

A Giddens, Consequences of Modernity (Polity Press, Cambridge, 1992). Interest deputation is now the dominant form. Yet the interest deputies are not delegated to legislate; they are delegated to work for interests, a quite different thing. However dense you make the network of interest advocacy, and however closely you tie it to the clienteles on one hand, and the government on the other, you cannot provide a channel to legislation this way. Examples below will show what may ensue when laws are enacted without the understanding that this activity is legislation. 16 That is, of course, de jure, according to the constitutional rule. 15

Rationality of Legislation in a Sociological View 55 Now, in order to stimulate the general economy, the Finnish diet expressed a willingness to relieve the burden of income taxation; they were especially concerned to mitigate the situation of people with small and moderate incomes. This was at the beginning of 1996. A legislative process ensued, with the result (see the report in the Aamulehti newspaper, 19 February, 1996) that in spite of the pronounced intentions and explicit promises, taxation of the low- and medium-income groups remained the same or was tightened. Only the high-income bracket profited from the reform (even it only nominally, the newspaper hastened to add, because ‘the seeming reductions in income taxes were hoovered back into the treasury to the last penny through tightening other taxes and tariffs’). In connection with this episode, some journalists gave evidence of exceptional energy and published a cluster of interviews (Aamulehti, 20 February, 1996). They contacted three parliamentary representatives and asked how it was possible that the result of the reform was the exact opposite to what was originally intended. It turns out that the representatives had no idea what they were doing when they enacted the laws in question. Representative Esko Helle (Left Alliance, Cabinet party) said that there must be ‘something wrong’ in the Parliament’s decision now that the taxes were, after allowances, even more onerous for the low-income group. The decision of the Parliament was, however, his own decision; he himself voted for it. Representative Kimmo Sasi (National Coalition, Cabinet party) announced himself to be astonished. The newspaper reports: ‘Chairman to the taxation committee of the diet, he had had, together with the other members, a “hunch” that the taxation of most people would be reduced by one per cent.’ Representative Sasi is startled and taken aback by his own decision (notice that the diet legislates on the basis of ‘hunches’, ‘näppituntuma’ in Finnish). The chairman of the Green Party’s diet group (also Cabinet party), representative Paavo Nikula said that ‘the changes in taxation do not meet the rightful expectations of the people’. Even this man speaks, like his colleagues, of the legislation, which is his own doing, as if it were enacted not by himself but by some entirely different person.17 But of course the representatives are right, and I, with my obsolete expectations, am wrong. The laws are enacted, de facto, not in the diet, but somewhere else, by some other people. They are enacted by the government and the civil service, in this particular case by the Ministry of State Finances, in consultation with their advisory bodies. The diet has the simple task of accepting the government’s 17 Representative Nikula makes the legislature a cruel parody with his words. In saying that the citizens have a right to something better than what they actually got, he imitates Groucho Marx, who indignantly refused to be a member of a society that accepted people like himself to be its associates. Representative Nikula, who seems not to have understood what he did as a legislator, and did not take the responsibility for it, was subsequently elected the highest legal authority of the country, the Attorney General of Finland. Representative Sasi was made a minister. Only representative Helle from the Left Alliance was not made anything particular.

56 Kauko Pietilä Bills in a rubber stamp fashion. Interest advocacy prevailing over free representation, the former relations are overturned: now the Parliament is responsible to the government, the government to the civil service, the civil service to interest groups, and interest groups to their clienteles.18 We saw that the Parliament is outmanoeuvred and consequently out of touch with the events, bewildered, knowing nothing about the things it is ordered to legislate.19 The example demonstrates the shift in the centre of legislative gravity away from Parliament to the thoroughly rationalised system of professional ministers supplemented by advisory bodies recruited from among the economically and socially most influential interest groups.20

18 It should come as no surprise that citizens are seen to be responsible to the diet: people behave badly in abusing and deriding the assembly and the representatives; they ought to look at the mirror and repent, admonishes editor Erkki Pennanen of the Helsingin Sanomat newspaper (14 August 1998). 19 Not for the first time in world history a Parliament manifests weakness and servility in relation to the government. Just listen to the ‘disgusted verdict’ of Denzil Holles concerning the English Parliament of 1647 (A Fraser, Cromwell: Our Chief of Men (Arrow, London, 1997/1973) 208): ‘The Army now did all, the Parliament was but a Cypher, only cry’d Amen to what the Councils of War had determined. They make themselves an absolute Third Estate . . .’ (Cypher was the seventeenth century’s term for nonentity.) Accusing the legislature of weakness, I imply not that the representatives are weak persons and that things could be better if only stronger characters were elected. Parties and representatives have deliberately chosen to appear on the stage as primarily nice people. They would never identify themselves as legislators; none of the parties has anything even remotely resembling a legislative programme. The party programmes are something the representatives could not care less about. Recent political science only encourages the easygoing attitude towards principles: ‘[A] politician who is up to what changing situations demand or open up as opportunities, does not stay and grieve about what the party programme might say. He/she understands that you can always explain your changed policy line as a response to changed circumstances.’ (E Aarnio and K Palonen, Puolueiden periaateohjelmat 1995 (Party programmes 1995). University of Jyväskylä, Department of Political Science, publication no 68, 1995, 8.) On the other side of the coin official party politics has become something the common people could not care less about. A Swedish survey of what people in general know about party politics (Helsingin Sanomat, 20 August 1998): half the population do not know which agency makes the laws (but do the pollsters know?); half the population do not know how many parties have portfolios in the Cabinet, etc. 20 The two most important rationalities for Weber are goal-rationality and value-rationality. In the ‘thoroughly rationalised system of professional ministers and prefects’ the emphasis is on goalrationality which is, even in other respects, the better rationality for Weber. Action based on it ‘is oriented to a system of discrete individual ends, such that the end, the means and the secondary results are all rationally taken into account and weighed’ (G D Mitchell, A New Dictionary of the Social Sciences (Aldine, New York, 1979) 2) while ‘Wertrationalität . . . involves weighing the means towards an end rationally against each other, but the end itself is not questioned’ (S Mennell, ‘Action Theory’ in M Mann (ed), The International Encyclopedia of Sociology (Continuum, New York, 1984) 3). Goal-rationality is the rationality that allows putting prices on things; ‘that which in contrast is elevated above all price and allows, hence, of no equivalent, that has dignity’, and it is here that you find value-rationality. When interest advocacy is substituted for free representation, valuerationality becomes replaced by goal-rationality. What we gain is a price; what we lose is our dignity.

Rationality of Legislation in a Sociological View 57

V . THE SENSELESS LAWGIVER II : THE CASE OF FINANCIAL LIBERATION

Nyberg and Vihriälä list the legislative and administrative measures with which the domestic and foreign financial market was liberated in Finland in the 1980s and early 1990s21: 1980 1983 1984 1985 1986

1987

1988 1989 1990 1991

Banks free to cover commercial forward positions Relaxation of lending rate regulation: pass-through formula Equal entry for foreign banks to the call money market Limited currency operations allowed to authorised banks Call money deposit rate separated from the credit rate Abolition of regulation of lending rates Free long-term foreign borrowing for manufacturing and shipping companies Floating rates allowed on some loans Cds exempt from reserve requirement Open market operations introduced Helibor rates introduced Free long-term borrowing for other companies Credit guidelines discontinued Floating rates allowed on all loans Free direct investment abroad for non-financial companies Foreign regulations relaxed except for households and short-term capital movements Prime rates allowed as reference rate Free household foreign investment Free short-term capital movements Free foreign borrowing for households.

The list shows that the financial market truly was liberated in Finland to the extent that no agent from outside of the said market was left with any significant legal instruments to regulate processes taking place within the market. The market was deregulated; material non-financial constraints were abolished. The list shows that the liberation was a consistent political programme carried through systematically, persistently and deliberately in a period of some ten years. Legislation, to which the diets were ordered (or asked or seduced . . .) to give their blessing,22 created the legal preconditions for, as the researchers of Bank of Finland say, ‘the liberalization process’ that very soon led to a ‘subsequent credit boom’ which in its turn rapidly developed ‘into a recession of 21 P Nyberg and V Vihriälä, The Finnish Banking Crisis and its Handling (Bank of Finland, Discussion paper no 8, 1993) 11. 22 Four diets were elected—and four coalition ministries were formed—during this period; in 1979, 1983, 1987 and 1991. The last diet and ministry was left, to be sure, just to sort out the chaos the three preceding ones had created.

58 Kauko Pietilä unprecedented depth’.23 Recession is the official flat expression; the actual result was an economic catastrophe that threatened the monetary system of the country with total ruin; the whole banking system was on the verge of collapse. The state stepped in with the citizens’ pecuniary means to save the disintegrating system. The catastrophe became visible and pressing during 1991; by the end of 1993, the state’s intervention had cost the citizens some 83 billion FIM or ‘almost 18 per cent of nominal GDP’.24 Legislation produced a catastrophe for a large majority of the population—and an equally paramount, brilliant success for a tiny fraction—through a de facto warrant to confiscate and redistribute the society’s wealth. It is evident that the sequence of events—libertarian legislation, credit boom and collapse—was not governed by fate, but by human agents’ action. The rules in the list above were enacted; they were positive laws; they were not given to us as the tablets of law were handed down to Moses: ‘And He giveth unto Moses, when He finisheth speaking with him in Mount Sinai, two tables of the testimony, tables of stone, written by the finger of God.’25 But of course, things are not as simple as that. All people did not participate in the enactment of the financial laws; to some people the laws were given the way the tablets of law were given unto the people of Israel. Who participated? And who just received what they were given?

23 Nyberg and Vihriälä, The Finnish Banking Crisis and it Handling (n 21 above) 3; emphasis added. 24 P Nyberg and V Vihriälä, The Finnish Banking Crisis and its Handling (an update of developments through 1993) (Bank of Finland, Discussion paper no 7, 1994), 41–3; emphasis added. The Bank of Finland, being one of the principal agents of the story, is likely to underrate the cost of the catastrophe. Characteristically it only sees the banks’ enormous losses that were ultimately covered with the citizens’ money and wealth. The researchers hardly say a word about the other side of the coin, the equally enormous profits made by some sharks. Furthermore, the researchers see the whole episode as an accident, an inadvertent event, something without premeditation and deliberate action behind it. They see things the same way as, for example, Giddens, Consequences of Modernity (n 14 above), 131 (emphasis added): ‘For the high consequence global risks which we all now run are key elements of the runaway, juggernaut character of modernity, and no specific individuals or groups are responsible for them or can be constrained to “set things right”.’ Giddens must enjoy inestimable respect even in Finland’s political circles; after all, he gives an unconditional absolution to the diets that enacted the legal devices which were used to produce the next-to-fatal economic and social disaster for Finland. It is also worthwhile to notice the Bank of Finland’s confused and contradictory notions as to what should be achieved with liberation and regulation. Nyberg and Vihriälä examine the reasons why the liberation brought such devastating consequences (Nyberg and Vihriälä, The Finnish Banking Crisis and its Handling (n 21 above), 11): ‘The liberalization of the domestic credit market and foreign exchange transactions was not accompanied by other reforms that would have significantly dampened the expansionary effects of liberalization. The incentives to borrow remained unchanged, economic policy in general did not adequately restrain domestic demand or price and income expectations, and banking supervision was not tightened.’ But, of course, dampening, discouraging, restraining and supervisory measures of economic policy are exactly what the liberators wanted to get rid of in the first place. It is no liberation if you give opportunities with one hand and take them back with the other. You cannot deregulate by first demolishing regulatory practices and then establishing new equipotent practices in their place. 25 Exodus 32:18. The financial laws were enacted years before Prime Minister Lipponen assumed office in 1995—and the sobriquet Moses.

Rationality of Legislation in a Sociological View 59 It is hardly possible to deny that the liberation of the financial market in Finland was a consistent line of policy, followed steadily throughout the 1980s up to the Big Bang at the end of the decade (it has been followed undeviatingly even after the crash). I already said that three or four elections were held during the period. The idea of liberal democracy, said Beetham, requires that a ‘formal guarantee of the freedoms of belief, expression and association, exercised within an autonomous sphere of civil society, ensures that, in principle, electoral outcomes reflect the “popular will”, and that public opinion can be brought to bear on government in a continuous manner’ (well-placed words those ‘in principle’; with them Beetham effectively waters down the whole idea). Economic policy, centred on liberating the financial market, was the issue that made the period, and times ever since, an era, an epoch. Yet, in spite of the ‘formal guarantee of the freedoms of belief, expression and association’, the epochal issue was never publicly debated; it was not a central or even a peripheral political question in a single general election. In spite of the guaranteed freedoms, belief, expression and association were not ‘exercised within an autonomous sphere of civil society’ with the effect of making the financial liberation an outstanding political issue. Why? Why are questions—debates!—of policy in general absent from general elections? The idea of free representation à la Weber and Beetham rests on the assumption that structuration takes place in civil society that leads to contending political programmes, to struggles over policy in parliament, and to legislation durch Kompromiß oder Abstimmung (through compromise or by vote), and that this is the way the general interest—or the general will—can be approached in practice. Perhaps such politically momentous structuration is no longer there? Meyrowitz identifies modernity with the medium of print which ‘leads to an emphasis on stages, levels, and ranks. The world comes to seem naturally layered and segmented.’ Post-modernity equals electronic media, and this makes a difference: while print tends to divide its audience into groups based on education, age, class and gender, a great deal of electronic information and experience is shared across demographic categories.26 I lean on Meyrowitz’s observation (to see the disappearance of demographic differences as a result from the change from print media to electronic gadgets is simplification and exaggeration)27: no structuration any longer. Yet the polity is founded on the assumption of essential structuration and the corresponding political organisation. No wonder the Finnish diet, for example, is out of its 26 J Meyrowitz, ‘Shifting Worlds of Strangers: Medium Theory and Changes in “them” versus “us” ’, [1997] Sociological Inquiry 65. 27 What Meyrowitz presents rather sketchily is developed more thoroughly in S Crook, J Pakulski and M Waters, Postmodernization: Change in Advanced Society (Sage, London, Newbury Park and New Delhi, 1992). In their argument (p 111), social class is diluted; there is ‘a submersion of the divisions of class . . . into more fluid cultural patterns of social differentiation in which social membership derives not from underlying material determinants or even from socially constructed groups but rather from symbolically specified associations which are simulated within mass information media.’ Within mass information media and not the Parliament.

60 Kauko Pietilä wits, at a loss, completely bewildered, caught unawares by its own action (which is what happened in both cases above, in the diet trying to lower taxes and actually raising them, and the same assembly liberating the financial market and nearly destroying the economy).

VI . THIS IS A CONSTITUTIONAL PROBLEM

The interpretations of Rousseau’s idea of the general will, as well as Kant’s idea of rational will/reasonable obedience, have rightly tended towards some kind of elective arrangement, such as Weber’s free representation or Beetham’s liberal democracy. Yet there are taxing conditions in Rousseau’s view for elective procedures to be approximations of the general will: ‘There is frequently much difference between the will of all and the general will. The latter regards only the common interest; the former regards private interest, and is indeed but a sum of private wills.’28 Elections no doubt build up sums of private wills; they approach the general will on this multiple condition: ‘If, when the people, sufficiently informed, deliberated, there was to be no communication among them, from the grand total of trifling differences the general will would always result, and their resolutions be always good.’29 That is, if the people were sufficiently informed, and if they would not communicate (divide into parties and coalitions?), then their deliberation (voting?) would produce a grand total of views in which the innumerable trifling difference would cancel one another out and leave the general will behind. But, Rousseau moves on: when cabals and partial associations are formed at the expense of the great association, the will of each such association, though general with regard to its members, is private with regard to the State: it can then be said no longer that there are as many voters as men, but only as many as there are associations. By this means the differences being less numerous, they produce a result less general.30

The conclusion is that it is ‘of the utmost importance for obtaining the expression of the general will, that no partial society should be formed in the State, and that every citizen should speak his opinion entirely from himself’.31 Now, if the post-modernist impression holds and we have homogenisation and dwindling divisions on the aggregate level, increasing choice, variety, and idiosyncrasy on the individual level,32 are we not approaching conditions in which the general will is truly expressed in elective outcomes? Does the diet, chosen in general elec28 J-J Rousseau, The Social Contract or Principles of Political Right (C Frankel (ed) (Hafner, New York, 1947/1762) ), 26. 29 Ibid. 30 Ibid. 31 Ibid, 27. 32 Meyrowitz, ‘Shifting Worlds of Strangers’ (n 26 above) 66.

Rationality of Legislation in a Sociological View 61 tions without ‘cabals and partial associations’, not truly represent the general will? And is Kant’s problem of rational will/reasonable obedience not thereby finally settled? Yes and no. Yes; there is the opportunity. But no; the opportunity ought to be seized first, and that has not occurred. Let us examine this a bit more closely. What does the Parliament, no longer entertaining party division, factional strife, disagreement and dispute in their midst, encounter in their dealings with the government? They come upon Bills propounding measures the likes of those in the list referred to at the beginning of Section V, and the one that was advertised as a reduction of taxation. From the diet’s point of view, the Bills come from the other side—from the side of the government, the civil service, and the organised interests. What do we find on this side? In Rousseau’s terms we meet once again the will of all, once again sums, aggregates, and compromises of private interest. This means that legislation no longer proceeds from the Parliament and the people who elected it. They no longer know what to do. They have no use for political programmes which have all become obsolete; yet there is nothing in their stead. And yet again, the diet is supposed, according to the constitution, to find their raison d’être in parties and their programmes. The essential paragraph (Para 2) of the Finnish constitution, from which everything else follows directly or indirectly, prescribes this: State power in Finland belongs to the people who are represented by the diet in session.33

33 The paragraph, dating from 1919, anchors ultimate political power inalienably in the people; it is the people who possess the decisive state authority, and this can be changed only through amending the constitution. Such rigid formulation is impracticable in the modern swiftly changing conditions. And so it is that some more up-to-date constitutions, those of the Kingdom of Sweden and the Federal Republic of Germany, for example, invented a more dynamic formulation for the same basic rule. In the Swedish case (from 1974), the regulation reads as follows: ‘All offentlig makt i Sverige utgår från folket’ (All public power in Sweden emanates from the people). In the German instance (from 1949), the rule has an almost identical wording: ‘Alle Staatsgewalt geht vom Volke aus’ (All state power emanates from the people). The Swedish and German rules are in better harmony with post-modern conditions than the Finnish one—for the following reason: we can read the verbs in the Swedish and German formulations as saying that state authority moves away from the people, leaves them, and finds a new locus somewhere else. The verbs utgå från and von etwas ausgehen mean, literally, to go out from something. This is the meaning of the verb emanate, too: to issue or proceed from or as from a source. Yet the English verb does not manage to reproduce all the possibilities the original verbs utgå från and von etwas ausgehen allow for, especially the possibility that the people do not, after all, have the role of a source (in the sense of a well that keeps oozing water and is never exhausted) but they have the role of storehouse that can be emptied completely and definitely, in the sense that the things that used to be stored in that house get out of it, move away, leave the place, and never come back. Yet there is wonderful magic in these things. The German language recognises the act of casting one’s vote with the verb abgeben, literally, to give away; Koch also uses the verb entäußern, to cede, another expression for giving something away, meaning literally to externalise, to turn out. (See U Koch, Das Gewicht der Stimme: Die Verteidigung des Nichtwählers (Rotbuch, Nördlingen and Leck, 1994) 8.) Englishmen and Englishwomen do exactly the same; they cast or give their vote, alienate it from themselves. And then, a miracle eventuates: ‘The vote that was given away, however, grows again to be back there in the run of a few years, like an indefatigable weed’ (ibid, 9).

62 Kauko Pietilä The rule notwithstanding, the diet cannot know what it means to represent the people without discussing the matter; yet they cannot discuss anything without either dividing among themselves or finding an outside partner to dialogue with. Now it seems that they can no longer can partition themselves in their role as the people’s representatives. They could, as an undivided single body, discuss and debate matters with an outside partner. But in the constitutional horizon, there can be no such. The people are the sovereign; in political practice the sovereign is the diet; and in their own affairs sovereigns do not dialogue with outside partners. The fact is, however, that laws are made outside the diet,34 in the complex composed of the government, the civil service, and the organised co-opted particular interests. While the diet are unable to discuss among themselves how to represent the people, and as they cannot acknowledge any outside partner to their sovereignty, they are entirely at a loss and feel compelled to counterfeit and simulate representation by taking the government’s Bills as if they were their own ideas and resolutions. In this deadlock, legislation assumes such absurd forms as shown in the examples above. What could be done, then, to restore the legislature to its senses, so that it would at least know what the palaver is all about? I cannot see that anybody could do much about the civil society to make it divide once again in a politically significant way. Reformation of the party system would be farcical. The contradictions of post-modern society do not run across the civil society. Yet contradictions, and this means even structurally representable contradictions, have anything but evaporated from the social life of human beings. Their location has changed. The whole story above about the change from free representation to interest deputation insinuates that we should situate the contradictions, not between classes of civil society, but between the diet and the government. We would have on one side the diet, the people (the people to which the Finnish, Swedish, and German constitutions refer), and the general interest; and on the other side the government, the civil service, and the organised co-opted private/particular interests. What we have on both sides can ultimately be traced back to one and the same basic population, so that, if we were able to make the two sides interact and encounter one another, it would be just one and the same underlying population interacting with itself. This could be, as it would be the population interacting with itself, a precondition for what I have been looking for all along in this chapter: to self-government. In that self-government, the population’s aggregated and compromised particular interests (ultimately represented by the principle of co-option and the government) would contend with its general interest (represented by the principle of 34 But not in the form of legislation, not with an eye on Kant’s conception of the rational will ‘as the will to act so that one’s action may be taken as a universal rule of conduct’ (C Frankel, ‘Introduction’ in Rousseau, The Social Contract (n 28 above) xxviii) ), but as aggregations and compromises of interests, that is, as sums of private interests—which express not the general will, but the will of all.

Rationality of Legislation in a Sociological View 63 general election and the diet). This would require rewriting the constitution, particularly its central all-important paragraph. A sociologist might suggest the following formulation: State power (in Finland, Sweden, Germany, etc.) belongs to the people who are represented by the diet in session and by the government in office. The diet stands for the general interest of society, the government for particular interests. What would follow from this remains dormant in this chapter.

5

Legislative Inflation and the Quality of Law SVEIN ENG

I . INTRODUCTION

In the invitation to the Fourth Benelux–Scandinavian Symposium on Legal Theory I was asked to talk about legislative inflation and the quality of law. In the invitation the subject was described as follows: Legislative production in the (post)modern society has reached critical quantitative limits. The causes of legislative inflation are multiple both in number and in nature. In order to face alternative forms of regulating human cooperation an analysis of these causes should be undertaken in order to avoid simply replacing abundant legislative regulation with alternatives that will turn out to be affected by the same or very similar deficiencies. Thus it can be asked whether the overproduction of legislation is due to its decreasing quality in so far as the defects of bad legislation are compensated by introducing a new legislative intervention, resulting in an ever accelerating growth of legal systems.

The passage quoted touches upon, inter alia, the following questions: —What are the main causes of legislative inflation? —To what extent is bad legislation a significant cause of great legislative quantity? —To what extent do the flaws referred to by the phrase ‘bad legislation’ also affect alternative techniques for regulating behaviour? These questions are obviously too comprehensive for it to be possible to give them a substantially serious treatment in the time at my disposal. I have interpreted my task as that of presenting an inventory of some problems, not as giving solutions. Roughly speaking, the task falls into two parts. The first part of the task is to clarify problems and intuitions regularly placed under such headings as ‘legislative inflation and the quality of law’. This clarification is partly of a conceptual kind; and it is partly of an empirical kind, in the form of some reminders and examples. The second part of the task I have set myself is to point to some directions for further systematic investigation.

66 Svein Eng I use the term ‘legislation’ in a broad sense. The paradigm case is statutes enacted by a democratically elected national assembly. But I also use the term to refer to other kinds of rule, first and foremost delegated legislation and European Community law. Legislation in these senses is just as important for our subject as parliamentary legislation. In addition, the term, may in many contexts, fruitfully be read as also covering enactment of general rules at the constitutional level and in international law; see especially the steadily increasing interest in a particular blend of ‘constitutionalism’ and human rights thinking, as instantiated, for example, by discussions of whether to incorporate the European Convention on Human Rights in national legal systems.

II . IS GREAT LEGISLATIVE QUANTITY PER SE A BAD THING ?

1. Legislative Quantity as Product The invitation to the symposium expressed a critical evaluation of the present situation with respect to the quantity of statutes. It is said that ‘legislative production in the (post)modern society has reached critical quantitative limits’; and one speaks about ‘overproduction of legislation’. I think one should ask whether this evaluation represents an intersubjectively held value-premise, and thus, an appropriate starting point for discussion, or whether this evaluation should itself be a topic for discussion. I will give an example to illustrate this. Let us assume that the person charged with handling the waste of a company is presented with two alternative statutes. Both statutes aim at protecting the environment from pollution, and both provide for sanctions in the form of criminal liability and liability in tort. However, there is the following difference with respect to the content of the statutes. The first alternative is ten pages long, of which nine consist of a legal definition of waste, giving detailed information in an objective form, for example by listing concrete kinds of waste. This alternative may be illustrated by the central legal definition of ‘waste’ in European Community law. The main part of this definition consists of a list of waste, the ‘European Waste Catalogue’, as it is called.1 The second alternative takes one page. It is framed in vague and value-laden terms, allowing much discretion as to the understanding of the concept of waste. As an illustration one can imagine the following legal definition of ‘waste’: ‘any substance or object significantly harmful to human beings, animals, or their environment’. Which of these two alternatives would be preferred by a person in charge of handling waste? One relevant consideration is the value of foreseeability. This lends support to the first and longer alternative. Another consideration is the 1 Commission Decision 94/3 of 20 December 1993, pursuant to the Council Directive on Waste of 15 July 1975 (75/442), as amended by Council Directive 91/156 of 18 March 1991 .

Legislative Inflation and the Quality of Law 67 negative value of having to spend time on reading. This lends support to the second and shorter alternative. However, when there is some kind of readily seen system, and when one is not too unfamiliar with the words and phrases used, one may without too much effort be able to obtain a rough idea of where to find the answers to concrete questions; for example, where to find out whether the concept of waste covers toner cartridges. In such cases the first and longer alternative would probably be preferred. This example illustrates two things. First, it illustrates that there is no simple connection between great legislative quantity per se and a negative evaluation on the part of the addressees. Secondly, the example illustrates that criteria for bad–good in this connection are not, or not only, the abstract amount of legislation, that is, the present amount of legislation seen by itself. One must relate this amount to each particular person subject to the legal system, and evaluate the more concrete combinations of legislative amount and kind in these relations.

2. Legislative Quantity as Process One would think that the law-maker, like all of us, would have limited time and resources at his disposal, and that there would be a limit to legislative quantity which, when exceeded, would produce bad legislative results. However, this limit is of a somewhat reduced importance in societies where the function of the law-maker has been split up, as in the societies of the participants in this symposium. The formal law-maker, paradigmatically the democratically elected national assembly, is only one link in the chain of law-making. The most important political impetus usually comes from the government; and the drafting of the Act is done in the administration. There is in principle no limit to how many civil servants, departmental or interdepartmental committees, and so on, one can involve in the process of lawmaking. However, there are political, financial and other practical limits to the size of the departmental apparatus. And in the national assembly itself, one necessarily comes up against the time constraint: there is only one national assembly, and, given some rules of parliamentary procedure, there are some purely physical limits to how much legislation the national assembly can process. To find out what the departmental and parliamentary constraints are, one would have to investigate the particular legal system.2 Thus, not on this point either is there any simple connection between great legislative quantity per se and what one would deem to be bad legislative results. The analysis illustrates that the criteria for bad–good in this connection are not, or not only, the abstract amount of legislation, that is to say, the amount of 2 For an example which seems to give a representative illustration of the main points, see the description of the English system in M Zander, The Law-Making Process (Weidenfeld and Nicolson, London, 1994, 4th edn) 1–104, esp 22–24.

68 Svein Eng legislation seen by itself. One must relate this amount to each separate link in the law-making process, and evaluate the more concrete combinations of legislative amount and kind in these relations.

III . SOME MAIN CAUSES OF LEGISLATION — AND OF LEGISLATIVE INFLATION

The main causes of legislation are at the same time potential causes of legislative inflation. If one is confronted by an amount of legislation which one deems to be in some respect disproportionately large, it may often be a plausible hypothesis that this excess is caused by one or more of the main causes of legislation ‘running wild’. I shall simply mention some of the main causes of legislation—and thus (see the preceding two sentences) of legislative inflation. First, legislation serves as an expression of values that change with time or that are held by some but not all sections of the population. Secondly, legislation is used as a means to shape society in accordance with political ideas. Thirdly, legislation is used as a means to solve problems created by social, economic and technological change. Fourthly, legislation may fulfil a legal duty to legislate. The prime example in Europe today is legislation as implementation of European Community law.3

IV . IS BAD LEGISLATION AMONG THE SIGNIFICANT CAUSES OF GREAT LEGISLATIVE QUANTITY ?

I have inserted the qualification ‘significant’, since I assume that most people who reflect on the matter will agree that bad legislation in some situations is a cause of new legislation. The problem to be discussed is whether, and if so how and to what extent bad legislation is a relatively stable and forceful cause of legislative quantity. Above, when we discussed whether great legislative quantity per se is a bad thing, it was not necessary to analyse the criteria of ‘bad’. It was sufficient for the corroboration of a negative answer to point to some simple counter-examples and ‘counter-facts’. Now, when asking whether bad legislation is a significant cause of legislative quantity, it is necessary to say something about what we could reasonably mean by ‘bad’ in this context. I distinguish between substantial flaws, technical flaws and means-end flaws. I use the term ‘substan3 The impact of EC law is, in psychological terms, probably most strongly experienced in a legal system such as the English system, where the courts do not have the power to invalidate legislation by reference to the constitution. The power to invalidate legislation by reference to EC law is thus a new element in the system; for the first time the courts are invested with the power to invalidate legislation by reference to a legal level above the level of statute law. However, psychology aside, the actual impact qua cause of new legislation is broadly the same in all member countries (as well as in my own country, Norway, through the European Economic Area Agreement).

Legislative Inflation and the Quality of Law 69 tial flaw’ in much the same sense as ‘material flaw’, ‘flaw in substance’, ‘flaw relating to content’, and the like. For the purposes of the present discussion it is not necessary to define the tripartite distinction between substantial, technical and means-end flaws any further in a general way. It is sufficient to consider the more concrete types listed and specific explanations given in what follows (sections 1.1, 2.1 and 3.1 respectively).

1. Substantial Flaws 1.1 Briefly on Substantial Flaws Legislation may be viewed as substantially flawed on the basis of several sets of criteria. Among the most regularly invoked are the following: (a) Moral criteria. One sees legislation as an expression of values and criticises legislation on the basis of moral values, contending that the legislation is unjust, unfair, or the like. (b) Political criteria. Legislation is seen as a means to shape society in accordance with political ideas and legislation is criticised on the basis of a particular political ideology. (c) Legal criteria. The legislator may have a legal obligation to legislate. As mentioned, the prime example in Europe today is legislation as implementation of European Community law. (d) Practical disharmony. The criteria (a)–(c) may be used to criticise a particular statute or larger parts of the legislation of a legal system. When the perspective is widened to include more than one statute, one soon encounters the problem of practical disharmony. I use ‘practical disharmony’ to refer to situations where one statute leads to behaviour which another statute seeks to counteract. For example, the consumption of alcoholic beverages may be taxed in order to lower such consumption. The resulting difference between the costs of producing alcohol and the market price of alcohol is a strong motive for the illicit production and smuggling of alcohol. That is to say, there is practical disharmony between rules in tax law and in criminal law. 1.2 Is Substantially Flawed Legislation Among the Significant Causes of Great Legislative Quantity? (a) The answer is obviously in the affirmative. The view that the content of legislation is morally or politically flawed constitutes a strong impetus towards new legislation. One particular combination of the political and the moral deserves special attention: the combination of, on the one hand, politically motivated changes in statute law, and, on the other, the moral principle that like cases should be treated alike. A constant experience of legislators is that a politically

70 Svein Eng motivated change in statute law, which by itself seems innocent and well motivated, may lead to extremely extensive and complicated regulations when the principle that like cases should be treated alike is taken into consideration; and by most people this principle is seen as just as binding on legislators as on judges. For example, in tax law one has the basic rule that the ordinary wages or salary paid by the employer to the employee are taxable. In combination with the principle that like cases should be treated alike, this seemingly simple rule leads to a steadily increasing mass of complicated tax law. This mass of law concerns, first, the extent to which all the different kinds of benefit from the employer to the employee should be placed on a par with the ordinary wages or salary, and consequently be taxable; and second, how the tax-relevant value of the benefit should be assessed. Examples of such benefits are the free or subsidised use of car, bicycle, phone, newspapers, working clothes, personal computers, kindergarten, holiday resorts, etc, the receipt of gifts on the occasion of the employee’s birthday, marriage, etc, and so on. (b) When it comes to legislation, I do not think that it is to any significant extent possible to change, especially not to reduce, the motivational force of moral and political considerations (see for example the illustration from tax law just outlined). (c) Another question is whether moral and political considerations have different motivational force in relation to purportedly alternative techniques of regulating human behaviour.4 (i) Some main examples of such alternative techniques are: —Negotiations, settlements, contracts, etc, whether they are between public bodies and private legal persons, or between different private legal persons; whether they are on a small scale, or concern large projects (for example an agreement between an oil company and the state relating to the extraction of an oil deposit) or concern whole sectors of the economy (in Norway, for example, agreed documents in the sectors of transport and insurance). —Financial instruments, such as subsidies, or such as taxes, duties or the like, used to incorporate the external costs of an activity in the market price of the products of the activity. —Privatisation of services traditionally treated as public in post-war Europe, such as medical care, postal services, telecommunications, electricity, water, etc; and further, deregulation of highly regulated services, whether public or private, such as financial services. Services traditionally treated as public as well as highly regulated services are often 4 I insert the qualification ‘purportedly’, since the term ‘alternative’ is often read in a substantive sense, and since it is often questionable whether a non-legislative regulatory technique is in fact a substantive alternative: that is, whether its use can replace legislation at all, and if so, to what extent its use effects a decrease in legislation (see, for example, below on privatisation, deregulation and local government autonomy).

Legislative Inflation and the Quality of Law 71 granted or do in effect have an aspect of statutory monopoly, and political debates concerning such services often centre on these statutes. In this perspective, privatisation and deregulation are presented as alternatives to legislation. In effect, however, privatisation and deregulation often necessitate a great deal of new legislation, in order to establish the new market and safeguard its proper functioning.5 —Local government autonomy may also be mentioned. However, a reduction of legislation passed by central government and directed to local authorities is often accompanied by a corresponding or larger increase in legislation passed by local authorities. In practice, many of the alternative techniques are combined with each other and with legislation: in relation to the alternative techniques just mentioned we find in Norwegian practice, for example, the combination of separate legislation for state-owned companies, tax legislation, and a varied array of regulatory techniques through different types of mandatory and non-mandatory rules of contract law. Several of the combinations found in practice, demonstrate the graduated nature of both the legislative element and the alternative elements. Finally, one should note that when the perspective is broadened to include the implementation of legislation, one meets the practice of delegating the implementation to private bodies.6 (ii) Does the motivational force of moral and political considerations differ between legislation and such possible alternative techniques of regulating human behaviour as mentioned in item (i) above? Logically speaking, the alternative techniques for regulating human behaviour are all susceptible to judgments in the perspectives of morals and politics. Normatively speaking, I see no general reasons why the alternative techniques should not be constantly judged in the moral and political perspectives, just as legislation is. Factually speaking, there is, however, a possible difference; that is to say, there is a possible difference between legislation and alternative techniques of regulation when it comes to the motivational force of morals and politics: a critically evaluative attitude towards legislation is deeply ingrained in the mentality of the media and the citizens, and they employ a rather standardised inventory of language and arguments. 5 I submit that the following observations concerning the English experience in the nineteenth century have more general validity: ‘The road to the free market was opened and kept open by an enormous increase in continuous, centrally organized and controlled interventionism. . . . [T]he introduction of free markets, far from doing away with the need for control, regulation, and intervention, enormously increased their range. Administrators had to be constantly on the watch to ensure the free working of the system. Thus even those who wished most ardently to free the state from all unnecessary duties, and whose whole philosophy demanded the restriction of state activities, could not but entrust the self-same state with the new powers, organs, and instruments required for the establishment of laissez-faire’, K Polanyi, The Great Transformation: the Political and Economic Origins of Our Time (Beacon Press, Boston, 1957/1944) 140–41. 6 S Eng, ‘Plassering av offentlig kompetanse hos private—noen hensyn for og imot’ (‘Public powers exercised by private legal persons—some considerations for and against’) (1992) 31 Lov og Rett 544 ff.

72 Svein Eng There is no such habitual attitude towards, and there is no such standardised inventory of language and arguments concerning, many of the alternative mechanisms for regulating human behaviour. Take as an example the actions of privatised public services. At present there is no standardised inventory of language and argument with which to judge their actions. How does one argue today when the local post office is closed down and the decision is said to be a result of the competition in the market? To put this in general terms: when an activity is defined as an activity in a market, how does one argue today against the market? The perspective which first seems to lose motivational force is the political one. At present, many people again argue as if a market were a non-political entity, that is, an entity outside the realm of political considerations. The task of politics is seen as that of establishing markets to the greatest possible extent, and of guaranteeing that a market functions as a market.7 Instantiations of this would be the growth of the New Right, as politically epitomised by the incumbencies of Thatcher and Reagan, and as academically epitomised by the sudden disappearance of Marxism from mainstream social science, be it a Marxist perspective, a confrontation with such a perspective, or simply some standard reference to Marx’s works; the answer from the social democratic parties of Western Europe to the growth of the New Right: their ‘modernising’ themselves, that is to say, their acceptance of the terms of the debate as set by the 7 For a recent justification of this view of the relation between politics and market, see F A Hayek, The Constitution of Liberty (University of Chicago Press, Chicago, 1960), passim, and id, Law, Legislation and Liberty (a collection of three volumes published in 1973, 1976 and 1979 respectively) (Routledge & Kegan Paul, London, 1982), passim. The historical roots of this view, and indeed of making a clear distinction between government and society at all, go back to the beginnings of liberal economics. See paradigmatically, A Smith, Wealth of Nations (P. F. Collier & Son, Harvard Classics, New York, 1937/1776) 446 (penultimate paragraph of Ch IX of Book IV): ‘According to the system of natural liberty, the sovereign has only three duties to attend to; three duties of great importance, indeed, but plain and intelligible to common understandings; first, the duty of protecting the society from the violence and invasion of other independent societies; secondly, the duty of protecting, as far as possible, every member of the society from the injustice or oppression of every other member of it, or the duty of establishing an exact administration of justice; and, thirdly, the duty of erecting and maintaining certain public works and certain public institutions, which it can never be for the interest of any individual, or small number of individuals, to erect and maintain; because the profit could never repay the expense to any individual or small number of individuals, though it may frequently do much more than repay it to a great society.’ Quite often the distinction between government and society is associated with something negative and positive respectively. See paradigmatically, T Paine, Common Sense (Penguin Books, Harmondsworth, 1976/1776) 65: ‘Society is produced by our wants, and government by our wickedness; the former promotes our happiness positively by uniting our affections, the latter negatively by restraining our vices. The one encourages intercourse, the other creates distinctions. The first is a patron, the last a punisher. Society in every state is a blessing, but government even in its best state is but a necessary evil’ (Paine’s emphasis omitted). Among authors traditionally called ‘philosophers’, Hegel was the first to make a theoretically significant distinction between elements similar to what Smith and Paine termed ‘sovereign/government’ and ‘society’. See Hegel, Grundlinien der Philosophie des Rechts, dritter Teil, zweiter Abschnitt (‘die bürgerliche Gesellschaft’), esp §§ 189–208 (‘das System der Bedürfnisse’). However, Hegel’s concept of and views on the state (‘der Staat’) and its relations to society were quite different from the concepts and views in liberal economics.

Legislative Inflation and the Quality of Law 73 New Right and their adoption of many of the New Right’s policies; the European Community’s enactment of the Single Market: an enactment with legal force to invalidate contradicting law in the national systems, at the constitutional level as well as at the level of statute law; and the introduction and nature of market policies in Russia and Eastern Europe. The next perspective which seems to lose motivational force is the moral one. At present, many people again argue as if the market by definition were a moral entity. There are usually two stages in such an argument. First, it is admitted that many of the results of a market arrangement seem immoral. Second, it is asserted that this is only an outward appearance; it is asserted that ‘in reality’, which most often means ‘in the long run’, the market also produces the best results in a moral perspective.8 The perspective which today seems to have the strongest motivational force in relation to the alternative techniques of regulation is the means–end perspective. When it comes to the element of ends, the value premises, these are often not thematised. When explicated, they seem to be close to the seventeenthcentury foundations of what the Canadian philosopher C B Macpherson termed ‘possessive individualism’.9 One could applaud this development, see it as a benefit of the alternative techniques, and as a confirmation of the view that legislation is flawed as a technique for regulating human behaviour. The view would then be that legislation fooled us into a centralised regulation of questions which it should be left to smaller units to decide; units like the consumer, the family, the company or other private organisations. Or one could deplore this development, see it as a flaw of the 8 F Hayek, Law, Legislation and Liberty (n 7 above) II, 72, 116, 144–47; for the ‘long run’ qualification, see eg ibid 114–15, 122. The historical roots of this view of the relation between morals and market are much the same as for the view that the market is an entity outside the realm of politics, see the preceding footnote. It is beyond the scope of this chapter to examine the epistemological and ontological status of the outlined view of the market, a view which plays a central role in political and academic discussion in Europe at present. Suffice it to point to an argumentative similarity between adverse positions, see ibid, 113: ‘[The market order is] the only overall order that comprehends nearly all mankind . . . the economist is therefore entitled to insist that conduciveness to that order be accepted as a standard by which all particular institutions are judged’, and Trotsky: ‘None of us wants to be or can be right against his own party. The party in the last analysis is always right, because the party is the only historical instrument given to the proletariat to resolve its fundamental tasks. . . . I know that it is impossible to be right against the party. One can be right only with the party and through the party, for history has not created any other way of determining what is right’, here cited from R Medvedev, Let History Judge—The Origins and Consequences of Stalinism, rev edn (Oxford University Press, Oxford/New York, 1989) 127. 9 C B Macpherson, The Political Theory of Possessive Individualism (Oxford University Press, Oxford/New York, 1964/1962). See the summary of the concept of possessive individualism, 263–64. The main points are (somewhat rephrased): (i) No human being stands in the relation of subordination to any other human being, since and in the sense that every human being is independent of the will of any other human being; (ii) Every human being stands in a relation to himself very much like the ownership relation, since and in the sense that every human being has exclusive control of rights in his own person and capacities, and proprietorship is the general form of such exclusive control; (iii) The basic relation between human beings is the contractual relation; (iv) The state (the political structure) is a device to protect the individual’s property in his person and goods, and his contractual relations.

74 Svein Eng alternative techniques. The view would then be that the alternative techniques fooled us into taking important questions outside the realms of moral and political discourse. Which stance to take is a question of choosing basic positions in moral and political philosophy.

2. Technical Flaws 2.1 Briefly on Technical Flaws The subject of technical flaws in legislation is an important one, but at the same time a little-discussed subject, at least in Scandinavia. I shall mention some main types of technical flaw, with brief comments on some of them: (a) Logical inconsistencies. (b) Difficult language. (c) Too casuistic or determinate legislation, or too general, indeterminate or discretionary legislation. Depending on which values attention is focused on— for example material justice or legal predictability—a given statute may be seen as being too casuistic or determinate, as being too general or indeterminate, or as providing too much discretion concerning how to apply it. (d) Legislation through travaux préparatoires or through delegation to judicial practice. Legislation may be viewed as flawed because the legislator does not answer important questions in the text of the statute, but instead gives the solution in the parliamentary or pre-parliamentary materials (‘forarbeider’, ‘legislative history’, ‘travaux préparatoires’), or leaves the questions to be solved in judicial practice. (e) Lack of system. Law may be seen as a system of information with two characteristics. First, the amount of information which the practitioners of the legal profession are expected to be able to handle is very large. Second, this amount of information is constantly changing in parts and increasing in sum. These characteristics make the system feature important to law: important to the teaching, learning, mental storing and application of law. Concerning the different kinds of systematic flaw I shall simply point to one distinction. On the one hand, a particular statute may be unsystematic in itself. Compare, for example, the casuistic Scandinavian legislation on contracts with the systems of the American Uniform Commercial Code and the German Bürgerliches Gesetzbuch. On the other hand, a particular statute may be systematic in itself, but by using key words or phrases in one or more new senses, the statute tends to fragment the legal system, that is to say, to turn a system of norms into a fragmentary collection of norms. For example, the person in charge of handling the waste of a company has to see to it that the firm complies

Legislative Inflation and the Quality of Law 75 with several different statutes relating to waste. His application of these statutes is made respectively easier or more difficult according to whether the different statutes use the same or different concepts of waste and the same or different terms for these concepts. By analogy with the problem of introducing new software into a working computer, one can speak of the problem of ‘the compatibility’ of new legislation; and by focusing on the source of the compatibility or incompatibility in question, namely the definitions constituting the concepts and correlating the concepts with terms, one can speak of the problem of ‘the compatibility of definitions’.10 2.2 Is Technically Flawed Legislation Among the Significant Causes of Great Legislative Quantity? (a) The answer to this question is not so clear as the answer to the question whether substantially flawed legislation is among the significant causes of great legislative quantity. One would have to investigate carefully the particular legal system to see whether, and if so to what extent, the intent to remove technical flaws was a motive behind new legislation. In Scandinavia, the intent to remove technical flaws, especially a wish to collect and systematise scattered rules, has in recent years quite often been among the more important motives for new legislation. Also, if new legislation is initiated on other grounds, the technical aspects are quite often taken up as well.11 (b) The occurrence of the different kinds of technical flaw may vary between different legal systems. One example is logical inconsistency. This occurs more rarely in jurisdictions where legislation is prepared in a centralised, highly professionalised and homogeneous manner, as in the English jurisdiction, than in jurisdictions where legislative drafting lacks these features, as in the American jurisdiction.12 Another example is the legislator’s failure to answer important questions in the text of the statute, instead providing the solution in the parliamentary or pre-parliamentary materials (‘forarbeider’, ‘legislative history’, ‘travaux préparatoires’). This is quite usual within the Norwegian legislative tradition. A necessary condition for this technique to work is that counsel and courts be allowed to take the statements in the parliamentary and preparliamentary materials into account.13 This condition is only to a very small 10 I use the word ‘definition’ in a broad sense. See my theory of definition, presented in S Eng, U/enighetsanalyse—med særlig sikte på jus og allmenn rettsteori (Analysis of dis/agreement—with particular reference to law and legal theory) (Universitetsforlaget, Oslo, 1998) 55–266 (Ch II B). The book is being translated into English for publication by Kluwer. 11 On legislation in commercial law, see J Hellner, Lagstiftning inom förmögenhetsrätten (Legislation in commercial law) (Juristförlaget, Stockholm, 1990) 170. 12 On the comparison between the English and the American systems, see P S Atiyah and R S Summers, Form and Substance in Anglo-American Law (Clarendon Press, Oxford, 1991), 63–66, 315 ff. 13 On the Norwegian doctrine on this point, see T Eckhoff, Rettskildelære (The doctrine of the sources of law) (Tano, Oslo, 1993, 3rd edn) ch 3.

76 Svein Eng extent fulfilled in the English legal system.14 A third example is the comparative lack of legislative system. See on the one hand the English and American jurisdictions, on the other the German jurisdiction, and more in the middle the Scandinavian legislative tradition. Whether or not the intent to remove technical flaws is a motive behind new legislation, it is probable that legislation can be made more effective by avoiding some technical flaws, for example by avoiding difficult language, or by avoiding legislation through travaux préparatoires or delegation to judicial practice. And to the same extent as technical flaws are avoided, these flaws are removed from the significant causes of great legislative quantity; it is then no longer necessary to engage in continual repair of technically imperfect legislation. (c) Whether the intent to remove technical flaws will have different motivational force in relation to alternative techniques of regulating human behaviour, is a subject on which it is difficult to say anything general; the question will have to be discussed in relation to each particular alternative.

3. Means–End Flaws 3.1 Briefly on Means–End Flaws In pointing to means–end flaws, one takes some moral, political or other value for granted, and criticises legislation for being an insufficient means to the attainment of that value. Such means–end considerations have an aspect of technicality, in that it is not contended that there is any flaw in the values strived for (compare section 2 above). At the same time the means–end considerations have an aspect of substantiality, in that by definition they presuppose attachment to an end, that is, to a value (compare section 1 above). This character of being something ‘in between’ the purely technical and the purely substantial justifies treating means–end considerations as a separate category for the purposes of the present discussion. Other descriptively adequate terms for what I call ‘means–end (flaws/considerations, etc)’ are ‘effectiveness’ and ‘instrumental’. One should notice that although means–end considerations by definition presuppose attachment to an end, that is, to a value, this value is not necessarily thematised in the argumentation, or even 14 See R Cross, Statutory Interpretation, (Butterworths, London, 1987, 2nd edn) ch 6 (esp 154–62); J W Harris, Legal Philosophies (Butterworths, London, 1980) 145; Z Bankowski/D N MacCormick, ‘Statutory Interpretation in the United Kingdom’ in D N MacCormick and R S Summers (eds), Interpreting Statutes: A Comparative Study (Dartmouth, Aldershot, 1991) 380–82. The differences between the Norwegian and the English legal systems concerning the relevance and weight of parliamentary materials were somewhat diminished, but at the same time consolidated, in Pepper v Hart [1993] AC 593; see esp at 640, where Lord Browne-Wilkinson, speaking for a majority of six to one, enumerates the conditions for permitting reference to parliamentary materials.

Legislative Inflation and the Quality of Law 77 psychologically present to the participants in the argumentation. It is not unusual to find that the presupposed ends hover in a state of such ‘taken for granted-ness’ that the ends may appositely be characterised as ‘pre-conscious’. For example, this holds good for much means–end reasoning in relation to the idea of a market (section 1.2 (c) above). 3.2 Is Means–End Flawed Legislation Among the Significant Causes of Great Legislative Quantity? (a) Means–end considerations are omnipresent in moral and political criticism of legislation. However, it is doubtful whether means–end considerations motivate as strongly as suggested by their frequent verbal occurrence. In moral and political contexts it is often the expressive aspect of legislation which motivates most strongly.15 For example, until 1972 in the Norwegian statute book, it was a crime for males to engage in homosexual acts, or for a man and woman to live together without being married.16 In 1972 these statutes had long since ceased to be enforced. Only nominally were they part of Norwegian law. However, the repeal of these statutes evoked strong objections. These objections showed that the important motivating factor was not the causal effectiveness of the statutes, but their role as symbolic and public expressions of value. In political contexts it is often said that a disagreement does not relate to the ends, but to the means (‘we are all adherents of . . . democracy/the principle of the equal worth of all human beings/equal chances for men and women, etc; we are all opposed to . . . human misery/war/the destructive effects of drugs, etc’). And political discussion often takes the form of pointing to different aspects of reality, as if the problem were that the political opponent was not fully aware of the significance of that aspect. The discussion is couched in terms of choosing means to supposedly given and common ends. At the same time political discussion is often not very receptive to existing research in the relevant field of social reality. I believe it to be a fruitful working hypothesis that much political argument on means is more a way of trying to establish or keep a certain freedom of political decision and action in relation to the different interests involved, than a real search for the most effective means. (b) It is a tautology that legislation can be made more effective by avoiding means–end flaws. And to the same extent as means–end flaws are avoided, these flaws are removed from the significant causes of great legislative quantity; it is then no longer necessary to engage in continual repair of ineffective legislation.

15 On the expressive aspect of legislation, see V Aubert, Om straffens sosiale funksjon (On the social functions of punishment) (Universitetsforlaget, Oslo, 1979/1954), 178–93; id, ‘Some Social Functions of Legislation’ (1966) 10 Acta Sociologica, 98 ff, esp 109 ff; id, Rettens sosiale funksjon (The social functions of the law (Universitetsforlaget, Oslo/Bergen/Tromsø, 1976), 154–55. 16 The Norwegian Criminal Justice Act of 22 May 1902, no 10 § 213 and § 379 respectively.

78 Svein Eng (c) Whether the intent to remove means–end flaws will have different motivational force in relation to alternative techniques of regulating human behaviour, is again a question which will have to be discussed in relation to each particular alternative (compare section 2.2 (c) above). (d) In relation to actual occurrences of discussion of legislation, there is an important difference between, on the one hand, considerations concerning means–end flaws and, on the other, considerations concerning substantial flaws and technical flaws. The latter two kinds of argument are most often limited to the comparison of legislative alternatives. The first kind of argument, concerning means–end, is very often extended to a comparison of the legislative alternative/s with alternative techniques of regulating human behaviour. This feature of discussions of legislation is intimately connected with two other features. The first is that many people discuss alternative techniques of regulating human behaviour mainly in the means–end perspective. To give a paradigmatic example: the alternative techniques are seen as variants of the market, and the market is seen as an entity outside the realm of political considerations, and as an entity which by definition produces moral results (see section 1.2 (c) above). The second is the tendency to substitute means–end considerations for substantial considerations in discussions of legislation.

V . WORKING FROM BELOW : LEGAL DEFINITIONS AS A STARTING POINT FOR REFLECTIONS AND AS A TEST CASE FOR GENERAL PROPOSITIONS

Above, we have, roughly speaking, worked from above; we have analysed propositions framed in rather indeterminate language with a view to obtaining propositions closer to our interests. It may be fruitful, in addition and in parallel, to work from below, from parts of the concrete reality which are central to our inquiry. One such part of reality is legal definitions. The subject of legal definitions has been surprisingly little discussed in the literature. This is an aspect of the fact that definitions outside logic and mathematics have been little discussed in the literature. In another work I have analysed definitions framed in everyday language, including legal definitions.17 In this chapter in a couple of instances I have used legal definitions at the level of example and instantiation (sections II 1 and IV 2.1 (e) above). In the present context, suffice it simply to state that the subject of legal definitions may serve as a starting point for reflections on, and as a test case for general propositions concerning, legislative inflation and the quality of law. The field of legal definitions instantiates many of the concepts and problems presented above.

17

S Eng, U/enighetsanalyse (n 10 above), 55–266 (Ch II B).

Legislative Inflation and the Quality of Law 79

VI . CONCLUSION

The emphasis in this chapter has been on an analysis of actually occurring language and argumentation concerning legislation, with a view to mapping what we could reasonably mean by some stock phrases such as ‘legislative inflation’ and ‘bad legislation’. The aim has been to operationalise somewhat the phrase ‘legislative inflation and the quality of law’. It is difficult to say something on this topic which is both general and interesting. The subject must be split up in order to reach adequate concepts and questions. What are the criteria of adequacy here? Partly receptiveness to facts, but not only that. Our interest may lie not only in the search for and accrual of new facts, but also in the clarification of a question and its relations to other questions.

6

Predictable Rules and Flexible Principles—The Problem of Ideological Pluralism and Legitimacy CHRISTIAN DAHLMAN

I . INTRODUCTION

In a well-known distinction, made by Ronald Dworkin, rules and principles are logically different because rules have an ‘all or nothing fashion’ while principles have a ‘dimension of weight’.1 In this chapter I will use this distinction to investigate how different ways of dealing with ideological pluralism affects the legitimacy of the legal system. The legitimacy of the legal system concerns to what extent people accept the law from an internal moral point of view (and not merely accept the fact that the law is in force). One of the most important features of legitimacy is that it demands that the law satisfy two needs at the same time: it must be predictable as well as acceptable.2 The law’s predictability is equivalent to the degree that people (or rather, their lawyers) correctly predict what rights and duties courts will accord them. That the law is fairly predictable is very important, beacuse it gives people stable expectations on which to base their actions. That the law is acceptable means that people feel that their values and interests are taken account of to such an extent that they can accept the legal system as a whole. In a value-pluralistic society this means that you cannot expect to agree with every ruling. You can, however, expect that your values are taken under consideration just as much as the values of other people.

1 R Dworkin, Taking Rights Seriously (Harvard University Press, Cambridge, Mass, 1977) 24 and 26. 2 A similar twofold understanding of the concept of legitimacy can be found in J Habermas, Faktizität und Geltung: Beiträge zur Diskurstheorie des Rechts und des demokratischen Rechtsstaats (Suhrkamp Verlag, Frankfort on Main, 1992) 243.

82 Christian Dahlman

II . TWO WAYS OF DEALING WITH PLURALISTIC LEGAL CONCEPTS

It is hardly controversial that the law is ideologically pluralistic, in the sense that it contains rival theories about morality. Some cases for example, are solved to promote the maximisation of aggregate wealth (the ‘moral theory’ of Richard Posner) while others are solved to promote the interest of the least advantaged (the moral theory of John Rawls).3 This pluralism extends to the content of single legal concepts. Let me take the concept of negligence as an example. There are (at least) three different definitions of ‘due care’ in tort law: 1. ‘Due care is the care taken by normally careful citizens in the defendant’s community’. 2. ‘Due care is taking all precautions which are economically efficient’ (also known as Learned Hand’s formula). 3. ‘Due care is not to expose people to unacceptable danger’. These three definitions rest on three different ideologies about what the law should promote: 1. reciprocity in risk-taking; 2. wealth-maximisation; 3. minimum safety (social security). This ideological plurality is often pointed out by tort law scholars.4 It is often described as a serious problem that ‘negligence’ does not mean the same thing in one case as it does in another. Lawyers fear that tort law is losing its legitimacy due to this pluralism. It is said to be in a crisis.5 As far as I can see, there are basically three different ways of dealing with this kind of ideological pluralism. The first method is to make the concept ideologically monistic. This means choosing one of the definitions and making it into a rule to be used in all cases (and to abolish the other definitions). This method is rather brutal, and the only lawyer I have come across advocating it is Richard Posner. In his economic theory of law, Posner (at least in his early works)6 wants 3 R Posner, ‘The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudication’, [1980] Hofstra Law Review 487; J Rawls, A Theory of Justice (Oxford University Press, Oxford, 1971) 60. 4 G Fletcher, ‘Fairness and Utility in Tort Theory’, (1972) Harvard Law Review 537, 540; H Steiner, Moral Argument and Social Vision in the Courts. A Study of Tort Accident Law (University of Wisconsin Press, Madison, Wi, 1987) 18; M Kotler, ‘Competing Conceptions of Autonomy: A Reappraisal of the Basis of Tort Law’, [1992] Tulane Law Review 347–49; N Weston, ‘The Metaphysics of Modern Tort Theory’, [1994] Valparaiso University Law Review 919–22. 5 G Priest, ‘Satisfying the Multiple Goals of Tort Law’, [1988] Valparaiso University Law Review 646; S Smith, ‘The Critics and the Crisis: A Reassessment of Current Conceptions of Tort Law’, [1987] Cornell Law Review 765–68. 6 Lately Posner claims that he has refrained from this fundamentalism, having become ‘profoundly sceptical of efforts to construct coherent moral systems’. See R Posner, ‘Wealth Maximization and Tort Law: A Philosopical Inquiry’ in D Owen (ed), Philosophical Foundations of Tort Law (Clarendon Press, Oxford, 1995) 101.

Predictable Rules and Flexible Principles 83 to make the law ideologically homogenous around the single goal of wealth maximisation.7 The second way of dealing with ideologically pluralistic legal concepts is to treat the different definitions as conflicting principles that have to be weighed against each other. No ideologies are thrown out. Their conflicting values and goals are all recognized as relevant (prima facie) in every case, and the skilful judge has to weigh them against each other. This method is advocated (among others) by Jürgen Habermas, who acknowledges the existence of rival ‘paradigms’ within the legal system (each of which interpret the world according to a different ideology). Habermas asserts that these paradigms are autopoietic in nature. They tend to close within themselves and develop a one-eyed interpretation of the world that has lost contact with the world-views in the other paradigms.8 Habermas sees this as deeply problematic. He wants the pardigms to ‘open to each other’ and become a single paradigm.9 According to Habermas’s discourse theory of law, the legal norms are legitimate if they could have been the result of a hypothetical discourse between all of those affected by them. Arguments based on conflicting ideologies shall be confronted with each other. The force of the better argument shall decide the case (and the law should thereby be made ideologically coherent). The third way to deal with an ideologically pluralistic legal concept is to turn the definitions into rules (instead of principles) and create meta-rules that determine which rule is applicable in what situation. The three definitions of due care stated above could, for example, be transformed into the following set of rules and meta-rules: Rules: 1. It is negligent to refrain from a precaution that would have been taken by normally careful citizens. 2. It is negligent to refrain from a precaution that would have been economically efficient. 3. It is negligent to refrain from a precaution that would have prevented the plaintiff’s exposure to unacceptable danger. Meta-rules: III) Rule 1 is applicable in cases where the defendant is an individual who has caused the accident in his private life. III) Rule 2 is applicable in cases where both the plaintiff and the defendant are corporations or government bodies. III) Rule 3 is applicable in cases where the plaintiff is an individual and the defendant is a corporation or a government body. 7 R Posner, ‘The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudication’, [1980] Hofstra Law Review 487. 8 Habermas, Faktizität und Geltung, (n 2 above) 271. 9 Ibid 272.

84 Christian Dahlman With a system of rules (and meta-rules) the judge does not have to weigh values and goals from one ideology against those of another. Operations of that kind are no longer necessary in every case. They have been made in the considerations underpinning the meta-rules. We will now evaluate these methods from the point of view of legitimacy. As we have seen above, this means that we will examine their abilities to make the law predictable as well as acceptable. The first way of dealing with ideological pluralism greatly promotes predictability. Principles that express different ideologies are incommensurable. They are too different to be compared with each other and it is therefore very difficult to predict how a judge will choose between them. This problem is eliminated if the ideological pluralism is eliminated, ie if one of the rules formulated above is used in every case. Say, for example, that the rule that states that it is negligent to refrain from a precaution that would have been economically efficient would be applied in all cases (without considerations of other aspects of negligence). This would make adjudication very predictable. This rule also contains evaluations, but these evaluations are empirical and therefore less arbitrary. To be able to determine whether a precaution would have been economically efficient, the court has to evaluate the magnitude of different costs. Such evaluations are naturally not 100 per cent predictable, but they are far more predictable than choices between rival ideologies. The problem with this method is that it is devastating for the law’s acceptability. When the law is made ideologically monistic it celebrates one value (in the example above, the promotion of economic efficiency) and is made blind to all others. For many people (probably just about everyone except Richard Posner) this would mean that values that are very important become completely disregarded and it would be impossible to accept the law from a moral point of view.10 As we have seen above, legitimacy requires that the law is made both predictable and acceptable. A more than sufficient amount of one can not compensate for a less than sufficient amount of the other. Making the law monistic is a poor way of dealing with ideological pluralism. With a plurality of rules or a plurality of principles, this flaw is prevented from occurring. These two methods are actually not so very different: they both maintain ideological pluralism and are therefore faced with making choices between rival ideologies. The difference is that with a plurality of due care principles these choices are made in each case, while with a plurality of due care rules they are made (in advance) on a meta-level. The consequence of this difference is that rules are more predictable and principles are more flexible. Rules are more predictable for the same reason that monism is more predictable. The system of rules and meta-rules described above reduces the number of evaluations where the judge has to choose between incommensurable ideologies, and replaces them with evaluations that are more empirical. It is less arbitrary to decide whether a precaution is ‘normal’ in the defendant’s community 10

J Raz, ‘The Relevance of Coherence’, [1992] Boston University Law Review 311.

Predictable Rules and Flexible Principles 85 (rule 1) or whether the accident was caused in ‘private life’ (meta-rule I), than it is to decide whether economic efficiency is more important than social security in a certain case. In this respect rules promote legitimacy more than principles. Rules are, on the other hand, less flexible than principles. When an evaluation is structured to become more empirical, the scope for equity is reduced. The problem with a system of rules and meta-rules is that the choices between rival ideologies are made on simple criteria, such as ‘the defendant is an individual’ vs ‘the defendant is a corporation’. This works well in most cases, but there will always be cases where the meta-rules force the court to apply an ideology, although the court would have found the values promoted by another ideology more important (in the case at hand) if it had been allowed to weigh rival principles against each other. This problem can be reduced by modulating the metarules, but it can never be eliminated. Thus, when it comes to acceptablity, principles promote legitimacy better than rules.

III . CONCLUSION

In conclusion, ideological pluralism can be achieved in different ways with different consequences for the balance between predictablity and acceptability that is crucial for the legitimacy of the legal system. Using rules makes the law predictable, but undermines acceptability if the rules are badly articulated. Using principles avoids this problem, but—on the other hand—makes the law less predictable. However, both these methods of creating coherence are superior to making the law ideologically homogenous.

7

Concept and Institution of the State in the European Legal Tradition HANNU TOLONEN

I . POINTS OF DEPARTURE

In the European legal, political and cultural tradition the state and its legislative function is an essential concept in two senses. According to the first, the state and public power are understood as an external source of all law. In this sense of legal positivism, all forms of law are volitions of this sovereign power: law is the product of the will of the legislator. The law is constituted by sovereign legislative power which also creates the unity of legal order. But there is another concept of the state which is more specific and more essential to the legal system. It can be found especially in constitutional and administrative law. According to this concept, the state is the basis of all legal relations and forms of organisation in public law. State power and the legal characteristics of sovereignty are based on and expressed in constitutional law. Thus, it seems that nowadays the concept of state has two aspects. On the one hand, the state is the source of all law, on the other, the state is the basis of all public legal relations. This chapter is concerned with the latter conception. Public legal relations differ qualitatively from private ones. Public law relations are characterised by an inequality of relations (taxation, administration), by public interest and a single subjectivity. Private law, in contrast, is dominated by an equality of legal subjects, private benefits and interests, as well as a plurality of legal subjects. Thus, we can speak of different subject structures, interest structures and different legal acts. As various authors have pointed out, this formerly clear-cut distinction between public and private law can no longer be made absolutely at the end of the twentieth century, when a special kind of public contract mechanism, new informal forms of legal activity and even the so-called polycentric modes of law are receiving increasing emphasis. Unified, sovereign state power has been called into question by increasing supranational law from without, and by the new legal mechanisms from within. My point is that the law is a multifaceted, social and cultural phenomenon as it is presented and analysed in many recent theories: the law is characterised by

88 Hannu Tolonen a specific institutional and discursive form as well as by an (institutional and formal) activity and structure of its own. Not only does the law develop through its own applications existing socially and juridically as a system to be applied. In this sense it therefore exists and develops as a practice. But the law also directs and regulates society and social life; it therefore exists and develops as normative institutions. The state is undoubtedly a legal institution—even a preeminent institution in a legal, political and cultural sense. Much has been said recently about the normative, social and philosophical nature of institutions. As examples, we can refer to Neil MacCormick’s and Ota Weiberger’s institutional theory of law, and further to John Searle’s, Peter Berger’s, Thomas Luckman’s and Erik Lagerspetz’s social and philosophical theories of institutions. There are some common features in these various forms, all of which emphasise the linguistic nature of institutions. Further, all of these writers underline the interplay between institutions and (institutional) activities. Institutions are abstract entities which are realised in activities and practices which in turn reproduce, formulate in concrete form or create new features of these institutions. Because institutions are linguistic reproductions, they have always been influenced by cultural (including social, philosophical and economic) ideas. The basic legal and social institutions especially, such as property, contract and state, have their own cultural traditions of ideas. In this sense each legal institution must also be seen as a cultural structure. All three aspects—institution as practice, normative structure and cultural sedimentation—are vital, especially in their various mutual interconnections. Legal praxis is always an application of norms and normative institutions; however, it also formulates, modifies and reproduces these norms and their structure. In the history of legal ideas, development is always to be understood as different stages which continue, modify or disrupt the basic conceptions of the previous stage. When examining these phenomena there are two ways of looking at the history of the state. We can examine the state as a conception and focus on its abstract conceptual features and ask how they originated doctrinally, culturally and socially. But we can also view the state as a set of concrete activities and examine it as an institution. The latter focus, the state as institution, can be further connected with the interplay between institutions and organisations partly following Douglas C North’s historical studies of legal and economic ideas. In his book Institutions, Institutional Change and Performance, North argues that every legal, real and living institution (including the state) must basically consist of an interplay between institutional and organisational levels. Institutions have constitutive and limiting norms, but these norms must also be realised in concrete questions and be performed by real actors.1

1 C North, Institutions, Institutional Change and Performance (Cambridge University Press, Cambridge, 1990).

The State in the European Legal Tradition 89 The institution and conception of the state has always been based on a unique level of human cooperation which has been characterised inter alia as the perfect society, public power or the common interest. In a more detailed way we can distinguish historically between three main stages of the institution of the state and its constitutive concepts. Each of these stages has existed in a different social environment and philosophical atmosphere. However, certain features of all are relevant to our actual conception and institution of the state. Accordingly, I shall draw a rough threefold historical distinction which serves as a starting point. The institution called the state has had various basic tasks throughout its history. The maintenance of peace and order was the earliest function of the state. The next stage was the emergence of a modern bureaucracy, which was necessary for standardising and directing economic and societal life, as well as for the waging of large-scale war for which permanent and formal taxes were needed. The third stage, which currently exists, is a state producing and distributing public goods, as well as ensuring all its citizens a basic standard of living. With these roughly sketched features in mind we can talk about the feudal, the bureaucratic and the welfare state and examine the characteristics of each in a more detailed way. The latter two may both be subdivided into two separate forms following each other chronologically: the bureaucratic state can be divided into the sovereign state (3.1) and the Rechtsstaat (3.2, for which there is no accurate corresponding English term); and the welfare state into the welfare state (4.1) and the post-welfare state (4.2).

II . FEUDAL STATE / SOCIETY

Feudalism is a specific formation of state and society. The most distinctive feature of this society consists of the intertwining of public and private areas and the functions of state/society. Private property and public power were inseparably interwoven. This meant that there were no institutions of the state in the sense of public law relations. In the Middle Ages, the state was understood and conceptualised as analogous to property and/or the family. This is the exact meaning to which the terms patrimonial state and/or patriarchal state refer. Further, this is partly the reason why the formation of the state had an informal character in which the social, juridical and moral spheres were interconnected. This is the Weberian view of feudal society. The recent research on feudalism has emphasised the hierarchical character of the feudal state, whose main functions were the maintenance of peace and tranquillity, to provide military forces, to maintain the order of estates and to perform its cultural functions. The feudal state performed vital societal functions: literary culture, church, universities (since the eleventh century), monasteries, taxes, money, etc. When the feudal state performed these ‘public’ tasks, the character of a status society (a society divided into noblemen, priests and peasants, in Latin bellatores, oratores and

90 Hannu Tolonen laboratores) was its most typical organisational principle. This, in turn, implied an openly unequal society. Modern research on feudalism has thrown light on many other more specifically juridical features of the activities and ideas of the Middle Ages. Therefore, with some reservations we can speak of a conception of the early European state with a juridical, moral and social character. The various terms applied to the state is a good starting point for examining the medieval legal thinking. I have found as many as nine typical medieval concepts which refer to state power, in the sense of dominant power. Imperium, civitas and res publica were typical terms used in the early Middle Ages. The basic terms for dominant power in the twelfth, thirteenth and fourteenth centuries consists of the concepts of status (from which the ‘state’ originated in various national languages, eg Staat, estado, l’état, etc.) and regimen, to which we can add the Greek word polis (used also in Latin), societas civilis, principatus and dominium (meaning both dominant power and property). The legitimising function for these various terms for the state is essential. Briefly stated, the concept of res publica contains the characteristics of affairs common to all. The basic organisational principle is concerned with the concepts of status and regimen. Referring to the latter, we can make a basic organisational and legitimating distinction between the following ideas. (1) Regimen regale (status regalis) concerns the person of the king as the dominant power and uniting societal principle. This conception was connected with the idea of what Ernst H Kantorowics called ‘the king’s two bodies’. There were several facets of powers in the king’s persona: his private, political and universal life, or vita civilis/vita politica/vita universalis. The dominant power and public sphere were identified with the political person of the king. This ‘body’ of the king was immortal and, according to Kantorowics’s detailed analysis, it was provided with spiritual and superhuman capabilities.2 (2) Regimen politicuum (status politikon) conceptualised the public dominant power as an organisation of the people. This was originally presented by Aristotle in his Politics which was translated into Latin in the middle of the thirteenth century. As ‘regimen politicuum’ the state was understood as an instrument of the whole population, but it did not include the sense of equal participation in power or the state as a nation. Rather, this term was understood in terms of the medieval conception of an unequal status society. These ideas give this conception an Aristotelian meaning in the Middle Ages: the state was understood as a balance of those different estates mentioned above, or forms of government which, in turn, referred to the Aristotelian doctrine of forms of government (monarchy, aristocracy and timocracy and their corrupted forms of tyranny, oligarchy and democracy). These ideas were summarised in the 2 H Kantorowics, The King’s Two Bodies (Princeton University Press, Princeton, 1981/1957); W Mager, Zur Entstehung des modernen Staatsbegriffs (Akad d Wiss und Lit Abh d D geist Und soz Kl, 1968).

The State in the European Legal Tradition 91 concept of the ‘mixed government’, politeia mese, translated into Latin as politia bene commixta (as found, eg, in Summa theol. I II, quaest. 105.2 of Thomas Aquinas). It was a principal solution for the problem of the best possible form of government.3 The distinction between regimen regale and regimen politicuum is explained by Walter Ullman as two basic organisational principles: ‘descending thesis’ and ‘ascending thesis’.4 The distinction made by this terminology reflects the basically different conceptions of the source of power: power descends from the monarch or ascends from the people. The former is a universal principle, whilst the latter (which is not the same as present-day democracy) pays primary attention to various circumstances and to the right balance between social groups (estates) as a basis of the best possible government. In the late Middle Ages, we can find several features of democracy (as it is presently conceived) in the principle of ‘quod omnes tangit’, in the concept of ‘universitas civium’, and in the universal consent of citizens represented by Marsilius Padova in the early fourteenth century.5 But the influential ideas originate in another source. The modern conception of the state and public power derives mainly from the power of regimen regale, from the king’s absolute power (compare the controversy between monarchists (‘absolutists’) and republicans).

III . THE MODERN PUBLIC POWER : THE BUREAUCRATIC STATE

The watershed is to be seen in the features of the nation-state and its conception of public power. This decisive turning-point consists of the following features: —the separation of public and private relations both morally and juridically as distinct areas; —the origin of the notion of sovereign power; and —the origin of the idea of legislative power, as well as the distinction between legislation and applying the law. In the Middle Ages there was no genuine legislative power. Instead there existed a scheme of natural law and the idea of contractual relations as a source of natural obligations and rights.

3 See in detail H Tolonen, ‘Valtion juridisesta käsitteellistämaisetä’ (‘The juridical conception of the state’) in Suomalaisen oikeusajattelun perusteista (Viljanen, Turku, 1982). 4 W Ullmann, Principles of Government and Politics in the Middle Ages (Methuen, London, 1961). 5 ‘Quod omnes tangit abomnibus est approbari’ means that what concerns all must be approved by all. See Post, Studies in Medieval Legal Thought (Princeton University Press, Princeton, 1962). See also H Tolonen, Valtion juridisesta käsitteellistämaisetä (n 3 above).

92 Hannu Tolonen

1. The Concept of Sovereignty All three characteristics mentioned above are united in the concept of sovereign power. At the beginning of the Modern Age, ie in the sixteenth and seventeenth centuries, this power was for the first time seen as qualitatively different from private (power) relations. It was superior to all other social relations and was able to create and modify them as it willed. The distinction between private and public law and the modern idea of legislative power derives from this historical source. In this new sovereign power we see the following new characteristics: —the unity of all public power of the state; —absolute public power, ie the highest, inalienable and indivisible power; and —legally unlimited power which is the source of all law but cannot itself be legally limited. These are the basic juridical characteristics of the state and public power.6 They are united with various ideas about the state’s legitimacy. In these we can see the profiles of the descending and ascending theses. Accordingly, we find the conception of rex dei gratia (power legitimated by God), ratio status (the public task of the state) or the basic theory of social contract with its various nuances. In the new conceptions the emphasis is on the artificial nature of the state body. In this the vital concept of representation is implied, which derives its historical origin from the medieval conciliarism of canon law.

2. Rechtsstaat Sovereignty has been a new way of looking at the power of the state and its constituent parts ever since the sixteenth or seventeenth centuries. When, at the beginning of the nineteenth century, Ludwig Haller presented his old-fashioned and openly feudal theories about the patrimonial state, he met with systematic opposition.7 His views about factual dependence and contractual or property relations as the basis of the state were systematically criticised by proponents of the so-called organic theory of the state (Julius Friedrich Stahl was perhaps its main representative). According to the organic theory, the state was a Gemeinwesen (commonwealth in nature standing above particular and individual interests. But the state is also ‘eine Anstalt’, ‘eine Institution’. This concept comes close to the conception of the legal personality (Persöhnlichkeit, juristische Persöhnlichkeit) of the state. It was, in turn, the starting point of two

6 7

For details see H Tolonen, ‘Julkinen ja yksityinen’ (‘Public and Private’), [1979] Oikeus . C L von Haller, Restauration der Rechtswissenschaft, I–IV, 2nd edn (Winterthur, 1820).

The State in the European Legal Tradition 93 different central views of the Rechtsstaat doctrine. The first concerns the unity of the state (later the unity of the will of the state). The second implies the concept of juridical representation, the capacity of the unity of the state to be represented, to have different acting state bodies. Through the further development of these ideas, we come to the Rechtsstaat doctrine which is a specific legal way of looking at the state and public power as distinct from natural or factual points of view. In public law it is a parallel conception to Begriffsjurisprudenz in private law: it aims at a specific legal construction through which it can examine the real and factual phenomena of increasing state bureaucracy. It brings all public legal relations under a uniting principle, one of which is a specific public law relationship constituted by a state person and its will, Staatspersöhnlichkeit and Staatswille. We can summarise two aspects of this legal quality of the state. First, public law or ‘Staatsrechtswissenschaft’ must be separated from natural (as von Gerber says) or social (as Jellinek says) points of view. We must find a specific legal point of view, ‘eine rechtswissenschaftliche Betrachtung’, ‘eine rechtliche Construction’. This constitutive legal principle is the unified will of the state will. The state is a system of capabilities of the will, ‘ein System von Willensmöglichkeiten’ as von Gerber says. The state is a public legal personality creating rights and duties for its citizens. Second, the Rechtsstaat doctrine also had a clear idea of state activities and their basic organisational principles. When the state interferes with a citizen’s freedom and property a legislative act is required to legitimise its action. This must, in turn, be an abstract norm postulating the equal treatment of all citizens (the so-called Rechtssatz). Parliament must participate in this legislative enactment. These legal features form the state’s legislative activities which are also called external state activities (state vs citizen). But there is also an internal organisational legal structure of state bureaucracy which can be realised, not in legislative form but in administrative procedures. The latter are called the internal relations of the state and they are not conceived as genuine legal relations because the state is only a single legal personality. This external/internal relationship (state vs citizen/state organ vs state organ) as presented by Otto Mayer, for example, parallels intervening/executive government (in German ‘Eingriffsverwaltung/Leistunsverwaltung’).8

IV . WELFARE STATE AND POST - WELFARE CONCEPTION OF THE STATE

The Rechtsstaat doctrine is a specific juridical way of conceptualising and looking at the relations of public law. The core and basic message of this conception 8 Mayer, ‘Zur Lehre vom öffentlich-rechtliche Vertrage’, [1888] Archiv des öffentlichen Rechts 37 ff. See also K Tuori, Valtiohallinnon sivuelinorganisaatiosta, I (Administrative organs of state government) (Vammala, 1983).

94 Hannu Tolonen is legal certainty or Rechtssicherheit. The task of public law is to examine the various ways in which citizens are provided with a legal remedy in their private legal relations, as well as protected from the abuse of public power. In this sense the Rechtsstaat doctrine realised (perhaps in a German form) the paradigmatic liberal model in separating private and public law and in guaranteeing the rights of property and liberty of private citizens.

1. Welfare Doctrine In relation to the latter-mentioned point of view, the logic of welfarist conception of the state is a different one. It sees the task of the state in a wider perspective. The welfare state is essentially connected with the basic social and legal capabilities of citizens. It aims at material equality and attempts to satisfy the basic needs of all its members. These needs refer to basic existence, protection from exceptional events (eg, illness and unemployment) and the provision of public services (such as education and the health service). The basic character of the state is interventionist, ie, it interferes with such legal positions which the paradigm of the Rechtsstaat called ‘private’. This amounts to a distributive state regulating the unequal positions of citizens, creating new protective roles as, for example, in consumer law and in the rearrangement of a private person’s debts.

2. Post-welfare State: Welfare and Post-welfare The fundamental and favourite theme of many social scientists and philosophers has been the moral corruption and economic breakdown of the welfare state. This theme has many aspects in neoliberal economic philosophy, in the idea of the breakdown of collective contracts in labour law and in an overemphasis on the benefits of the market mechanism in various fields of social life. These writers are trying to reverse past developments by attempting to revive old liberal doctrines, as well as having an uncritical belief in the regulative invisible hand of market mechanisms. I do not share their point of view. I believe that we live in a complicated post-industrial society with new challenges for which the old liberal doctrine of self-regulating market forces is no solution. We need the state and state regulation. But what kind of state and legislative regulation? I refer to this problem as the welfarist/post-welfarist dilemma and believe that it is the great theoretical as well as practical challenge of our time. For the theory of state and legislation this problem is basic. There is a remarkable set of new phenomena which is organisational and practical on the one hand, and conceptual on the other. Since the early 1980s there have been various forms of the so-called theory of reflexive law which concern the area of state

The State in the European Legal Tradition 95 activity and the nature of public law. Nowadays these questions are categorised as legal pluralism and post-modern discursive theories.9 One key problem is the new organisational mechanisms mediating between the traditional private and public legal areas and which are increasingly dominating the institution of the state. This new phenomenon was also emphasised in the idea of reflexive law and its later versions.10 Its message can be summarised as the idea of self-regulating social mechanisms which have replaced the simple state interventionist legislative activity (collective agreements between employers and employees being the paradigmatic example). Instead of intervening, public power only regulates the procedure and competence; the consensus of the parties is the direct objective, not some intended direct social effect. Instead of state authority and legislation, the relevant social interest groups are the agents of this social engineering. Theories of reflexive law parallel Jürgen Habermas’s thinking on procedural justice which, according to his view, replaces the interventionist state. Both theories emphasise the flexibility of this new method: it is more open and sensitive to the social environment and its requirements. I shall attempt to discuss these relevant questions further. First, welfare state activity and the methods of reflexive law are not such opposites, as both Teubner and Habermas think. Rather, I would underline their complementary nature: the basic capability and equality of citizens may also be achieved by the methods of reflexive law. That the legislator is legally unlimited and omnipotent has been the basic truth in sovereignty theories, in the Rechtsstaat conception and in interventionist theories. But, whether he/she is socially and morally omnipotent is another matter. In this question we undoubtedly have become more sensitive to procedures than direct legislative methods. The law exists in a social and moral environment in a way which remotely reminds us of the old feudal state/society conception with its contractual relations and open moral structure of law. Secondly, this question also implies an essentially wider field of legal phenomena which has become very topical in the 1990s. The common characteristics of these new phenomena are the limits of sovereignty and their positions between private and public areas of law, as well as the private and public methods of 9 The first thematic analyses on reflexive law were the studies of Niklas Luhmann and Gunther Teubner. See N Luhmann, Legitimation durch Verfahren (Suhrkamp, Frankfort on Main, 1969); G Teubner, ‘Substantive and Reflexive Elements in Modern Law’, [1983] Modern Law Review. The second emphasis of many of Teubner’s articles in the late 1980s and 1990s is the so-called autopoetic law. Most recently, increasing emphasis has been given to various ideas of legal pluralism and postmodern discourse theories. See G Teubner, ‘The Two Faces of Janus’ in K Tuori and Z Bankowski (eds), Law and Power (Deborah Charles, Chippenham, 1997) 119–40. The term ‘relational theories’ also concerns the same phenomena. See H Wilke, ‘Three Types of Legal Structure. The Conditional, the Purposive and the Relational’ in G Teubner (ed), Dilemmas of the Law in the Welfare State (de Grutter, Berlin/New York, 1986); P Fitzpatrick, ‘Relational Power and the Limits of Law’ in Tuori and Bankowski (eds), Law and Power above, 85–97. 10 N 8 above.

96 Hannu Tolonen directing societal relations. We can refer to four features which concern the organisational and institutional conception of state and legislation. (a) The unity of public power has been called into question. We find in state administration new organisational situations in which contractual relations between state bodies have become legal in nature. This is an anomaly and an exception to the Rechtsstaat doctrine which also prevailed in Finland after the Second World War. This is because, according to its own basic principle, the state is only one legal personality: within the state there are no legal relations. Expanding state activity simply needs some kind of contractual capability. In his book on administrative contracts, the Finnish researcher Olli Mäenpää has formulated a concept of weak legal subjectivity of state organs.11 (b) The sovereignty of the state or legislative omnipotence, the most traditional established doctrine of Western legal thinking, has become problematic in the light of new phenomena. According to this doctrine, public power is legally unlimited which, in turn, implies the idea that the public power is not able to bind its will. I refer to a Finnish example (also taken from Mäenpää’s research): according to a ruling of the Finnish Supreme Administrative Court, the city of Helsinki was obliged to pay (limited) compensation for the damage caused by its public town planning activity; earlier, in a private contractual relationship the city had raised expectations of certain building possibilities for the other party. (c) New administrative methods based on contract or property. In the 1990s a new tendency has been to use a combination of administrative power and contractual method. An example from Finnish legislation is the new Unemployment Act introduced at the beginning of 1998. It makes possible and even usual a combination of public power and private freedom of contract connected to an employment plan. The result of this combination is an employment contract which binds the public power to try to provide work and obliges the citizen to provide and maintain his working capability. This example is interesting because the basic rights of citizens (the right to work) even in a welfarist (not to mention the Rechtsstaat) doctrine have always been outside the sphere of contractual possibility. (d) Private and public risk management. The new risk theories of Luhmann, Ewald and many others provide a new way of looking at the legal system: we can speak of social risks (Ewald), environmental risks (Luhmann), contractual risks and criminal law can be examined as a method of sharing risks (eg, traffic regulation). We can distinguish between the profiles of public and private risk management: the former consists of state regulation, whilst the latter uses contractual methods and insurance. I shall limit myself to referring to various cases of social risk management in which the private method of insurance is used. A Finnish example is the insurance system for patients being treated for injury, introduced in 1987. Several things are essential: in a society of risks and 11

See Mäenpää, Hallintosopimus (Administrative Contract) (Vammala, 1988).

The State in the European Legal Tradition 97 catastrophes (as it is often described) public power is required to participate in managing risks; in this the state uses insurance and contractual methods. The sharing and management, especially of social risks, has a deep social and moral character as emphasised and analysed by François Ewald, in particular.12 All of the cases (a)–(d) concern the welfarist function of the state, but represent a new way of using private or semi-public methods instead of direct public interventionism and legislation.

V . CONCLUSION

Looking back at the relevant historical types of the state, we can return to the conception on the one hand, and to the institution which is realised on the organisational level in practical questions, on the other. Viewed in this perspective the state with its regulative legislative function has always been an organisational and practical question. In different forms it has been an institution which has had to face new challenges in its differing social environments. But we also need a legally coherent view of the state and its activities. Further, we need state legitimacy, a moral and philosophical justification in answer to the following question: how is the state an institution concerned with common affairs and interests in moral and legal principles? In answering this question we need a coherent and a morally justified concept of the state. In both senses we are living in an age of change and new questions. In the new post-welfarist thinking we must re-evaluate many organisational and practical questions. Furthermore, I believe that the common characteristics of these new organisational questions concern the semi-public nature and private organisational methods. But they are more than purely technical means to an end: they essentially concern the legal and moral nature of the state and public law.

12

F Ewald, L’Etat Providence (Grasset, Paris, 1986).

8

Legislation Between Politics and Law KAARLO TUORI

I . THE PROBLEM OF LEGISPRUDENCE

It is a well-known fact that legal theory, as well as legal science in general, usually approaches the law from the perspective of the judge, and not from that of the legislator. Legal theory is interested in the decision-making and argumentation of the judge, not in the decision-making and argumentation of the legislator. The rationality of law has been the central issue for legal theory, but this rationality has most often almost spontaneously been equated with the rationality of the judge and her decisions. The twentieth century has been a century of positive law, an era of the sovereignty of the legislator. In the course of it, especially in the wake of the social welfare state, legislation has increasingly been resorted to in an effort of conscious management of society, both of its economic system and the sphere of the everyday life of the citizens, ie their Lebenswelt. Our bureaucrats and politicians particularly have seen in the law mainly a means of social engineering, a medium, to use Jürgen Habermas’s vocabulary. In the closing decades of the twentieth century, however, the problems to which this instrumentalisation of the law gives rise have become more and more conspicuous: legislative inflation and the juridification of social relationships (Verrechtlichung); the alleged decrease in the quality of law drafting and, consequently, of legislation itself; and the lack of legitimacy of state intervention through law. These problems relate not to the rationality of adjudication, but to the rationality of legislation. But if these problems concern the rationality of legislation, what exactly do we mean by rationality in this connection? Can legal theory contribute anything to the elucidation and solution of problems related to the rationality of legislation? Is legisprudence, or jurisprudence from the perspective of the legislator, a viable alternative to dominant legal theory and to the perspective of the judge it has adopted?1 In order to approach these questions, I shall first try to analyse the role legislation plays in the overall context of the law. 1 The term legisprudence has been introduced by Luc Wintgens. See L J Wintgens, ‘Creation and Application of Law from a Legisprudential Perspective: Some Observations on the Point of View of the Judge and the Legislator’ in A Aarnio et al (eds) Justice, Morality and Society: A Tribute to

100 Kaarlo Tuori

II . LEGISLATIVE ACTIVITIES AS A COMBINATION OF POLITICAL AND LEGAL PRACTICES

In our era of positive law, legislation is by far the most important source of the law; but the source of the law does not mean the same as the law itself. In fact, I would like to defend the thesis that, in a very important sense, legislation, ie the end product of legislative practice, is to be considered ‘not-yet-law’ rather than ‘already-law’, raw material for the law rather than ‘law itself’. In some previous papers, I have made a distinction between two aspects of the law—law as legal norms, as a legal order, and law as certain social practices, as legal practices.2 These aspects of the law, of course, are closely interconnected. Thus, the legal practices are responsible for the continuous production and reproduction of the law as legal norms. What, then, are these specific social practices that we can call legal practices? I have proposed two defining criteria, one of which refers to the agents of these practices, and the other to their outcomes. Accordingly, legal practices in a strict sense consist of social practices whose main agents are legal professionals and which contribute to the continuous production and reproduction of the law as a legal order. The most obvious activities falling under my definition of legal practices are those of adjudication and legal science. But what about legislation, the activity which, in our contemporary society, bears the main responsibility in the production of new legal norms? Is not legislation, too, to be included in the legal practices producing and reproducing the law as a legal order? My answer would be yes and no. Remember that, according to my definition, the main agents of legal practices consist of lawyers, legal professionals. By this criterion, legislation seems to fall outside the law understood as specific social practices. In a democratic political system, legislative activities are not reserved for lawyers, but are carried out by politicians. This is true especially of the final phases of legislative processes, when the government makes its decisions on submitting Bills to the Parliament and the Parliament, after due deliberations, decides on the acceptance or rejection of the Bills. The debate preceding these decisions, both in the official bodies and in the public sphere surrounding them, does not proceed through legal, but mainly political arguments. What we seem to have here is not legal, but political practices. The legislative process, however, includes even phases, where lawyers and legal arguments occupy the main stage. Thus, law drafting involves a phase of norm formulation, where legal expertise is dominant and where the main concern is the maintenance of the internal consistency and coherence of the legal order. In some countries, even the parliamentary procedure may contain Aleksander Peczenik on the Occasion of his 60th Birthday on 16 November 1997 (Juristförtlaget i Lund, Lund, 1997) 469–89. 2 See, eg, K Tuori, ‘Towards a Multi-layered View of Modern Law’, in ibid, 427–42.

Legislation between Politics and Law 101 elements where juridical argumentation comes to the fore. This is the case, for example, in countries where this procedure includes a mechanism for controlling the constitutionality of the Bills under deliberation. In Finland, for example, the Constitutional Committee of the Parliament, aided by constitutional experts, bears the main responsibility for this control. Thus, legislative activities can be characterised as a combination of political and legal practices, with, however, political aspects having the final say. Ever since Max Weber, we are accustomed to seeing one of the main characteristic features of modern law in its autonomy. The role legislation plays in the production of the legal order and its peculiar nature as a combination of political and legal practices indicates that we have here the main channel through which, in our modern society, the law maintains its openness towards politics. The autonomy of modern law does not amount to autarchy, at least not in relation to politics. This combination of political and legal aspects typical of legislation already implies that its criteria of rationality cannot be equated with those of adjudication.

III . LEGISLATION IN THE OVERALL STRUCTURE OF THE LEGAL ORDER

By ‘legislation’ is meant not only certain polito-legal practices, but also their end products, that is, statutes, by-laws and other kinds of explicit legal regulations. Let us now turn to the examination of this aspect of legislation. It is time to justify the thesis that legislation, in the normative dimension of law, is to be considered ‘not-yet-law’, raw material for the law, rather than ‘already-law’. Here I take recourse to another distinction with which I have tried to elaborate a view of law I call critical positivism: the distinction between different levels of the law. This distinction concerns the law as a normative phenomenon, ie the law as a legal order.3 My basic idea is that the law as a legal order is not exhausted by what can be observed on its surface, that is, by such normative phenomena as individual statutes and court decisions. The law involves even deeper layers which, for example, confer upon it its systematic nature, its consistency and coherence. These deeper layers can be called the legal culture and the deep structure of law. The legal culture includes the general legal concepts and the general legal principles of different fields of law, that is, of private law, penal law, constitutional law, administrative law, etc. In the vocabulary of continental European legal science, the general legal concepts and principles constitute the general doctrines (allgemeine Lehren) of various fields of law. It is largely due to this element of the law, this component of the level I have called the legal culture, that the legal order receives its systematic nature, is transformed from an agglomerate of individual statutes and other legal regulations, as well as individual court 3

Ibid.

102 Kaarlo Tuori decisions, into a coherent whole. This coherence, in turn, is an indispensable precondition for the realisation of such elementary legal and, at the same time, moral principles as legal security and formal equality and justice. And it is in this transformed, systematic shape, and not as an unordered agglomerate of norms, that the law functions in, for example, adjudication. What has to be emphasised here is that, in the division of labour between legal practices, this transformation of the legal raw material to be found on the surface level of law does not take place through legislation. The legislation produces individual statutes, which have yet to be interpreted and brought into systematic relationships with each other, before the law can fulfil its functions in the society. Interpretation and systematisation are tasks for which other legal practices, adjudication and legal science, are responsible. In different legal cultures, say in common law and continental European countries, the respective emphasis on the courts and on legal science varies. Roughly put, in common law countries, the courts play an important role in furthering and maintaining the coherence of law, whereas the contribution of legal science is of minor significance. In continental Europe, in turn, it has been university-based legal science, which has, ever since the emergence of modern law, played the major role in the systematisation of the legal order. Reference can again be made to Max Weber. On his analysis, one of the main aspects of the formal rationality of modern law was its autonomy, and, in continental Europe, this autonomy was largely due to the systematising work accomplished by university-based legal science. Whatever the respective contribution of adjudication and legal science, what is important to our analysis is that, with respect to the coherence of the law, legislation is most often a factor of disorder rather than order; the coherence of the law is mainly produced through other legal practices than legislation. We have, of course, examples of the codification of law through legislation, such as the Bürgerliches Gesetzbuch. But even BGB is more justly to be considered rather as a triumph of legal science than of the legislator. The legislator, in fact, only gave a formal recognition to the work of German private law which science had accomplished during the nineteenth century. IV . LAW AND MORALS

The law stands between politics and morals, so we have learned from, for example, Jürgen Habermas.4 The law has specific functions in relation to both politics and morals: it is a medium for achieving collective, political goals, but it also has its role to play in guaranteeing the morally mediated social integration of the society. Modern law is autonomous law, but it still maintains links to both politics and morals. Legislation, as we have already seen, functions as the channel through which political influences enter the law. What about morals, then? 4 See, above all, J Habermas, Faktizität und Geltung: Beiträge zur Diskurstheorie des Rechts und des demokratischen Rechtsstaats (Suhrkamp, Frankfort on Main, 1992).

Legislation between Politics and Law 103 We can point to legislation which can be seen as direct reinforcement of prevailing moral conceptions, in the field of, say, criminal and family law. These are instances Habermas has had in mind when speaking of ‘law as an institution’.5 But, in contemporary society, the legislator is most often moved by political motives; what he usually aims at is collective political goals. This also means that the law maintains its connections to morals mainly through other legal practices than legislation. Let us examine more closely how this takes place. The systematisation of the law, which, according to my previous analysis, is due to adjudication and legal science, is not only the effect of the conceptual network, which is thrown over the legal raw material and where individual statutes issued by the legislator find their locations as part of private law, criminal law, tax law, etc. The systematisation also concerns the normative contents of the law. Neil MacCormick, among others, has distinguished between the consistency and the coherence of the law.6 The consistency of the law is equal to the logically non-contradictory nature of legal norms. Reference can be made to the standards for solving norm conflicts which are familiar to every lawyer and which form an integral part of our expert legal culture, such as lex superior, lex posterior and lex specialis. Their main function can be seen in the assurance of the logical consistency of the legal order. If consistency is a property which is attached to the law as a set of rules, coherence concerns its principled nature: the law receives its coherence from legal principles. Some legal principles, which confer normative coherence on the various fields of law, have been formally confirmed by the legislator, but for the major part, these principles are brought into play by the legal practices of adjudication and legal science. In general, the legislator is motivated by policyoriented considerations, which, adopted as they are in different political constellations, may even be in contradiction with one another. The interpretative and systematising activity of adjudication and legal science, instead, is mainly guided by legal principles. These principles, such as nulla poena sine lege in penal law or pacta sunt servanda in civil law, to take two conspicuous and familiar examples, are clearly morally laden. And if we try to penetrate into the normative deep structure of modern law, what we find are principles in which moral and legal aspects are intertwined in an inseparable way; let us only think of the fundamental principles related to human rights. It is through legal principles that the law maintains its openness towards the moral sphere of society. If my analysis hits the mark, the links to politics and morals, respectively, are located on different levels of the law and mediated through different legal practices. Political considerations, mediated through legislation, enter the law on its surface level. The links to morals, instead, are maintained on the levels of the legal culture and the deep structure of the law, and mediated primarily through the legal practices of adjudication and legal science. 5 6

J Habermas, Theorie des kommunikativen Handelns, II (Suhrkamp, Frankfort on Main, 1981) 536. N MacCormick, Legal Reasoning and Legal Theory (Clarendon Press, Oxford, 1978) 152 ff.

104 Kaarlo Tuori

V . EXCURSION : THE AUTONOMY OF THE LAW

Through our analysis of the law’s location between morals and politics, we obtain a fresh view on the autonomy of modern law, as well as on the Dworkinian discussion of the relationship between principles and policies.7 When examined in the framework of the multi-layered view, the degree of the law’s autonomy looks different in the directions of politics and morals. Through legislation, the law on its surface is closely connected to politics, but gains more and more autonomy on the deeper levels. In relation to morals, we obtain a contrary picture of the degree of the law’s autonomy: in the fundamental normative principles in the deep structure the law is intimately entwined with morals, whereas its autonomy with respect to moral norms increases when approaching the surface of individual statutes and decisions. Hans Kelsen could claim the total separation and independence of legal and moral orders only by restricting his gaze on the normative surface and by ignoring the levels of the legal culture and the deep structure. As to the relationship between policies and principles, we can speak of a normative censorship to which policy-oriented legislation is subjected in the transformative practices of adjudication and legal science, and where the role of censor is played by legal principles. This censor enters the stage every time a policy-based statute is interpreted in the light of a morally laden principle. This censorship receives its most conspicuous expressions in instances of judicial review. Modern law is autonomous, but it is a question only of relative autonomy, to resort to an expression somewhat outdated after 1989. It may sound paradoxical that in relation to politics the law claims its autonomy through elements which relativise its autonomy in relation to morals. Policy-based legislation finds its limits in legal principles, which keep open the channels between law and morals.

VI . THE RATIONALITY OF LEGISLATION

Thus, according to the preceding analysis, legislation as a social practice is a peculiar combination of political and legal aspects, with an emphasis on the political side. As to the outcome of legislative practices, in turn, legislation is to be considered ‘not-yet-law’, legal raw-material, rather than ‘already-law’. The road from ‘not-yet-law’ to ‘already-law’ passes—so I have claimed—through the transformative, that is, interpretative and systematising processes of adjudication and legal science. In these processes, morally laden principles play a central role, restricting the influence of the political factors introduced by legislation. 7

R Dworkin, Taking Rights Seriously (Harvard University Press, Cambridge, Mass, 1978).

Legislation between Politics and Law 105 What, then, about the rationality of legislation? By what kind of criteria should it be assessed? To the lessons to be learned from Max Weber belongs the insight to the polyvalence of the notion of rationality, also in the context of law. The rationality of the law can be appraised in different dimensions and resorting to different criteria. Thus, it is possible that the rationality of legislation should be assessed with different yardsticks than the rationality of adjudication and legal science. In the appraisal of the rationality of legislation, a distinction between three dimensions of rationality appears to be fruitful. These dimensions are the object rationality, the internal rationality and the normative rationality. The object rationality of the law is, in the final instance, measured by its ability to fulfil its social functions, ie to further the achieving of collective goals and to secure morally mediated social integration. Internal rationality relates to the internal consistency and coherence of the legal order. Finally, normative rationality of the law is equal to its (normative) legitimacy. When exploring the rationality of legislation, the focus can be either on single statutes or on the state of legislation in general. The object rationality of legislation has been under examination especially in the strand of the sociology of law called implementation research. Most often the studies have concerned individual legal reforms, the degree to which the political goals attached to these reforms have been achieved and the causes for eventual failures. In contrast, the debate about legislative inflation, the juridification and, consequently, bureaucratisation of either the economic system or the life world (Lebenswelt) of the society, has not so much concerned individual statutes or legal reforms as legislation in general. However, the criteria implied by the critical tone in this discussion have also been drawn on the dimension of object rationality. The critics have claimed that legislation, particularly in the era of the welfare state, has overstepped its appropriate domain and disturbed the internal, spontaneous mechanisms through which the economic system and the everyday life world, respectively, maintain their integration. What often afflicts especially implementation research is an instrumentalist view of the relationships between political goals, legislation and the social effects of law: the intervening transformative processes, which I have emphasised, often tend to be bypassed. These processes aim at maintaining the internal rationality of legal order, ie, its consistency and its morally related coherence. In modern society, the legislator usually acts according to the purposive rational pattern, where new legislation is considered a means to achieve certain politically determined goals; these, in turn, are defined as the desired social effects of the legislation. When new legislation is integrated into the systematic whole of the legal order and, for example, submitted to the normative censorship of legal principles, we witness the intervention of factors not accounted for in the purposive rational pattern and affecting the ability of legislation to achieve the political goals of the legislator. In its strive for internal rationality, the law asserts its autonomy in

106 Kaarlo Tuori relation to politics, to the political goals of the legislator. However, internal rationality is already a relevant dimension in the assessment of legislation. Thus, individual statutes should, of course, meet the criterion of logical consistency, and legislative practice should, in its legal aspects, pay attention to the internal rationality of the legal order as a whole, for instance, try to avoid outright contradictions between old and new legislation. One explanation for the alleged decrease in the internal rationality of legislation in, eg, Finland, may lie in the fact that law drafting takes place increasingly elsewhere in the state machinery than in the Ministry of Justice, where the expertise required by the monitoring of the internal rationality is concentrated. Whatever the juridical quality of law drafting, in legislation the dimension of internal rationality is in a way subjected to object rationality, which the legislator tries to ensure through conceiving of the legislation in light of the purposive rational pattern. The main responsibility for the maintenance of internal rationality remains on the shoulders of judges and legal scholars. Their efforts in this direction, in turn, constitute a constant source of disappointment for politicians and bureaucrats, who see their political goals sacrificed on the altar of the internal rationality of the law. Finally, we have the third dimension of rationality, that of the normative rationality or the legitimacy of the law. In modern society, extrapositive normative yardsticks, derived from objective or subjective nature, have increasingly lost their credibility and the criteria for legitimacy have more and more acquired a procedural, instead of a substantive nature. Under these conditions, legislative processes play a crucial role in securing the overall legitimacy of law. When legislation constitutes the most important source of law, a necessary requirement for the legitimacy of law consists of a democratic legislative procedure, where deliberative processes within the official bodies are surrounded by debates in the public sphere of the life world. However, it also has to be emphasised that the democratic nature of lawmaking procedures does not suffice for assuring the legitimacy of law. Taking into account the significance of such legal practices as adjudication and legal science for the production and reproduction of the legal order, their procedural conditions should also meet specific requirements for legitimacy. What these requirements are is an important topic for legal theory and legal philosophy. Its exploration here would, however, take us too far from our main theme. In comparison to other legal practices and their outcomes, legislation seems to be characterised by one dimension of rationality which lacks its counterpart in adjudication and legal science. Internal and normative rationality are also highly relevant in the assessment of adjudication and legal science. Thus, for example, coherence and appropriateness, which Klaus Günther has defined as the validity criteria for adjudication,8 can be understood as specifications of the requirements of internal and normative rationality, respectively. Object ration8

See, especially, K Günther, Der Sinn für Angemessenheit (Suhrkamp, Frankfort on Main, 1988).

Legislation between Politics and Law 107 ality, instead, where the focus is on the relationship between law and (the rest of) society, is not directly relevant in the context of these practices. This claim seems justified in spite of the spread of consequentialist argumentation in both adjudication and legal science, ever since the second half of the nineteenth century. Generally speaking, argumentation in adjudication and in legal science, which usually adopts to law the same perspective as the judge, escapes the purposive rational pattern typical of legislation. If adjudication in general can be analysed in purposive rational terms, the goal it tries to achieve is in a way internal to law: the realisation of law. The significance of the dimension of object rationality reflects the dominance of the political aspect in legislative practices: legislation is used by political decision-makers as a means to political goals and, from that perspective, its success or failure is measured in the dimension of object rationality. It is also along this dimension that—and this is my final thesis—legisprudence, ie legal science that adopts the perspective of the legislator, confronts the limits of its enterprise. Object rationality is affected even by factors which can be regarded as internal to law; for instance, factors related to its internal rationality. Primarily, however, in the division of labour between empirical or descriptive social and behavioural sciences and predominantly normative legal science, both the specification of the conditions for the object rationality of legislation and the exploration as to how these requirements have been met in particular instances belong to the domain of the former. It is true, though, that in these tasks, social and behavioural sciences need the aid of legal science. I would, however, also like to defend an anti-imperialist stance in this connection: I do not see any reason for an imperialist striving of legal science to conquer new terrain in the examination of the object rationality of legislation.

9

Legisprudence and European Law: in Search of the Principles of European Legislation AMARYLLIS VERHOEVEN

I . INTRODUCTION *

For some years, institutions and political actors in the European Union have expressed a growing interest in the quality of European legislation. Better lawmaking, it is hoped, will render the European Union more transparent and closer to its citizens and thus lessen the ‘democracy deficit’ from which it is perceived to suffer.1 This chapter addresses the concerns for better law-making in Europe. Rather than approaching the matter from a legalistic point of view, it focuses more broadly on principles and constraints European law-makers must respect. The first part offers a brief description of the European law-making process. Although parliamentary involvement in European law-making has increased over the years, the condition of European law-making is far from one wherein a Parliament, vested with sovereign powers, enacts laws that are deemed inviolable. European law-making is a dispersed and diffuse process that involves a multitude of actors and procedures and is subject to a series of constraints enforced by judges. The second part attempts to draw a conceptual model that can explain the European law-making process and justify it in democratic terms. Three different models prevail in current academic literature (the regulatory, the intergovernmental and the parliamentary), none of which, in my opinion, is entirely satisfactory. I propose instead to borrow from the procedural/deliberative model of democracy advocated by, among others, Jürgen Habermas. European * This text has been finalised in January 2000 and does not take account of more recent, punctual events. 1 Cf Declaration no 39 on the quality of legislative drafting annexed to the final text of the Treaty of Amsterdam, signed on 2 October 1997. The Treaty of Amsterdam revised the Treaties on which the European Union is founded, among others the Treaty on European Union (hereafter ‘TEU’) and the Treaty establishing the European Community (hereafter ‘EC’). See also the report of the European Commission entitled ‘Better Lawmaking 1998’ (COM (1998) 715 final).

110 Amaryllis Verhoeven law-making can be analysed as a process of democratic will-formation under conditions of difference. Constitutional constraints are to safeguard the democratic nature of the process and, in doing so, safeguard the legitimacy of its outcome.

II . LAW - MAKING IN EUROPE : AN UNCONVENTIONAL SCHEME

1. Who ‘Legislates’ in the European Union? Contrary to national legal systems, which follow from parliamentary models of law-making, there is no concept or theoretical framework underpinning European law-making. The condition of European law-making is one of a conceptual void—or rather one of a conceptual Tower of Babel, since different models and visions on rule-making compete at the European level. Starting at the level of textbook description, it is difficult to grasp who in the governmental framework of the European Union is a legislator and which norms, in the bulk of European law, are legislative acts. Article 249 EC,2 a key provision governing European law-making, states that: [i]n order to carry out their tasks and in accordance with the provisions of this Treaty, the European Parliament acting jointly with the Council, the Council and the Commission shall make regulations and issue directives, take decisions, make recommendations or deliver opinions.

A few things may be remarked here. First, there is no reference in this provision—as in fact hardly anywhere in the European Treaties—to the terms legislator or legislation.3 European institutions may produce a wealth of measures: the Treaty refers to regulations, directives, decisions, recommendations and opinions but in practice other instruments (such as resolutions, declarations, inter-institutional agreements) are also used. Significantly, however, EU institutions cannot adopt measures properly called ‘laws’ or ‘statutes’, although many instruments (in particular regulations and directives) may contain binding measures of a general nature. Further, while national legal systems vest legislative powers in a Parliament (or rather in what the British call the ‘Queen in Parliament’, since some form of co-operation of the executive branch is usually required), there is no single legislature in the European Union. Legislative activity is dispersed over mainly three European institutions (the Council, the European Commission and the European Parliament).4 Moreover, there is no single procedure for law-making. 2

Throughout this chapter, the new numbering that results from the Amsterdam Treaty is used. Since the Treaty of Amsterdam, however, Art 207 EC refers to the ‘legislative capacity’ of the Council. 4 In addition, the Economic and Social Committee and the Committee of the Regions have certain consultative functions. 3

Legisprudence and European Law 111 Which procedure is to be followed depends on the subject matter of the legislative measure at issue. Contrary to national legislatures, the European institutions have only attributed powers. They can only act in a certain field if a specific Treaty provision allows them to do so. That Treaty provision, the so-called legal basis, prescribes the objective and nature of the measure that can be adopted, but also the procedural requirements for its adoption and the institutions competent to adopt it. The Council, an intergovernmental body consisting of representatives of national governments, enacts legislative measures by a unanimity or qualified majority vote, now most often acting jointly with the European Parliament. Parliamentary involvement in European law-making is, however, relatively novel (it is a result of Treaty amendments that aimed at curing the so-called democracy deficit) and is still not guaranteed in a uniform and general manner. The legislative powers of the European Parliament range from a mere power to be consulted to genuine co-legislative powers under the so-called co-decision procedure. The European Commission, a body of civil servants led by appointees of the Member States (since recently, a parliamentary assent is required), has the exclusive power to initiate legislation. No legislative acts can be adopted without a Commission initiative. The Commission has, as a rule, no powers to enact legislative measures by itself, but enjoys in practice broad powers to implement European law pursuant to a delegation by the Council. Delegated law-making often takes place in various committees of national representatives (ie, the socalled comitology committees). A further complication in describing the European legislature is the fact that there is no genuine trias politica or separation of powers at the European level. While a certain distinction between legislative and executive activity can be drawn, these activities are not clearly located in separable bodies. As a general matter, executive rule-making powers (ie, powers to implement norms adopted by other EU institutions) are vested in the Commission, whereas the Member States are to ensure that European norms are properly implemented and executed on their territories.5 The dispersed processes of law-making in Europe are streamlined and rationalised, somehow, by the principles and norms set out in the European constitution. That constitution does not take the form of a formal text, but is a rather flexible whole of norms flowing from different sources. It includes the European Treaties, acts of a fundamental nature (eg, concerning the elections for the European Parliament) and a range of judge-made rules and principles. The European Courts (ie, the European Court of Justice and the Court of First Instance) are called to sanction respect of those constitutional principles.

5 K Lenaerts, ‘Regulating the regulatory process: “delegation of powers” in the European Community’, [1993] EL Rev 23.

112 Amaryllis Verhoeven

2. Constitutional Constraints on European Law-making National legislatures typically enjoy wide margins of discretion. They are, of course, bound to respect constitutional principles such as fundamental rights and due process elements. In continental legal thinking, however, the legislature is traditionally considered better placed than courts to judge whether laws comply with the constitution.6 The national administrative process, on the other hand, is usually subject to a series of constraints that are open for judicial review. National administrators typically must respect the legality principle (ie, they can only act if there is a proper legal basis for it) as well as various ‘principles of good and proper administration’. Contrary to national systems, European law-making is tied to relatively tight constitutional constraints that touch on matters of (i) competence (whether the European institutions can act): (ii) content (what objectives they are to pursue, what values and what rights they are to respect and promote); and (iii) procedure (which institutions must act according to what processes).7 All these constraints are subject to judicial review. Further, the same constitutional constraints apply, in principle, to the legislative and administrative activity—as mentioned, there is no clear-cut separation between law-making and administrative rule-making within the European Union. (a) Principles of Competence European institutions have only attributed powers: they can act only if and when there is proper ‘legal basis’ allowing them to do so. However, powers have, over the years, been extended considerably. It has been observed that ‘[t]here simply is no nucleus of sovereignty that the Member States can invoke, as such, against the Community.’8 In order to counterbalance the increasing output of European legislation in ever-broader fields, the principle of subsidiarity has been introduced in the EC Treaty.9 According to that principle, the Community must take action only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can be better achieved by action on the part of the Community. Subsidiarity is a dynamic concept. It works in two ways. It allows Community action (within the limits of its powers) to be expanded where circumstances so require and, conversely, urges it to be restricted or discontinued 6 Contrary, in Marbury v Madison, the US Supreme Court decided that it is the judiciary’s task to judge the constitutionality of laws (5 US 137). 7 See, for a detailed account, H Schermers and M Waelbroeck, Judicial Protection in the European Community (Kluwer, The Hague, 1991); C Timmermans, ‘How can one improve the quality of Community legislation?’, [1997] CML Rev 1229. 8 K Lenaerts, ‘Constitutionalism and the many faces of federalism’, [1990] Am J Comp L 220. 9 Art 5 E.C.

Legisprudence and European Law 113 where it is no longer justified.10 Subsidiarity is a flexible device that allows balancing the requirement of efficiency and democracy in large and complex societies. Importantly, subsidiarity should not only be understood as a federal principle of division of powers between the European institutions and the Member States. It also has another meaning, rooted in Catholic social theory,11 according to which rule-making should be kept at the minimum in order not to hamper the free development of associational life. This version of subsidiarity calls for alternatives to legislation, such as self-regulation.12 It also links subsidiarity to proportionality. Proportionality requires Community action not to go beyond what is necessary to achieve the objectives of the Treaty. At the occasion of the Amsterdam revision of the Treaties, Member States agreed that proportionality should mean the following: The form of Community action shall be as simple as possible, consistent with satisfactory achievement of the objective of the measure and the need for effective enforcement. The Community shall legislate only to the extent necessary. Other things being equal, directives should be preferred over regulations and framework directives to detailed measures. . . . Regarding the nature and the extent of Community action, Community measures should leave as much scope for national decision as possible.13

Although open to judicial review, the Court of Justice has hitherto subjected respect for subsidiarity and proportionality to a minimum control only. The Court does not venture to second-guess political institutions’ judgements as to whether and to what extent to legislate. It requires, however, that the institutions concerned state the reasons why, in their opinion, the principles of subsidiarity and proportionality are met and submits the stated reasons to a (minimum) rationality test.14 (b) Principles of Content (i) Objectives Community institutions must act not only within the limits of the powers conferred upon them by the Treaties, but also within the limits ‘of the objectives assigned to (them) therein’.15 Contrary to national legislators who, in principle, can legislate as they please (or at least within very broad margins of what can be 10 See Protocol on the Application of the Principles of Subsidiarity and Proportionality, annexed to the Treaty of Amsterdam. 11 See the Encyclical letter of Pope Pius XI entitled ‘Quadragesimo Anno’. 12 In the social field, the European Treaties allow the social partners to draw up agreements which become law when ratified by the Council. 13 See Protocol cited, at n 10 above, at point 6. 14 See, eg, Case C–84/94, United Kingdom v Council, 1996 ECR I–5755 (Working Time Directive) and Case C–359/92, Germany v Council, 1994 ECR I–3698. 15 Art 5 EC.

114 Amaryllis Verhoeven deemed rational), EU institutions are bound to further specific objectives set forth in the Treaties. The Treaties define, first, general tasks and objectives. These include socioeconomic objectives (such as economic growth, reduction of employment), political objectives (such as the assertion of a common identity on the international scene) as well as more value-oriented objectives (such as equality between men and women).16 Further, specific objectives can be found in individual Treaty articles empowering Community institutions to take action in certain fields. Thus, Community institutions, for instance, are called ‘to contribute to the attainment of a high level of consumer protection’17 and to pursue an environmental policy on the basis of a list of objectives such as protecting and improving the quality of the environment, protecting human health, promoting a prudent and rational utilisation of natural resources as well as international co-operation.18 The Treaty provisions on culture are particularly interesting in this regard. Article 151 EC calls for action by the Community in the field of culture. Such action should ‘contribute to the flowering of the cultures of the Member states, while respecting their national and regional diversity and at the same time bringing the common cultural heritage to the fore’. But cultural issues are also to be taken into account when the Community acts outside the cultural area: Article 151, paragraph 4 EC requires the Community ‘to take cultural aspects into account’ as a general policy matter, ‘in particular in order to respect and promote the diversity of its cultures’. The aim of that provision is to shy off fears of a ‘melting-pot Europe’, wherein particular identities are sacrificed on the altar of a blind common market ideology. The message of Article 151, paragraph 4 resonates in the general obligation on the Union to respect the national identities of its Member States.19 The Court of Justice in principle reviews whether the institutions respect the objectives and policies set forth in the Treaties. However, such review remains in practice rather marginal, as the Court leaves quite ample room for discretionary judgement by the political institutions of the Union on how to achieve the Treaty objectives. Discretion is especially large where different policy objectives are at stake and the political institutions have to reconcile and balance between them.20 (ii) Principles of Good and Proper Legislation Over the years, the Court of Justice has developed (on the basis of the Treaties) a series of principles of good and proper legislation. These include the principle 16 17 18 19 20

205.

Arts 2 and 3 EC; Art 2 TEU. Art 153 EC. Art 174 EC. Art 7, para 3 TEU. Cf, for instance, Case 37/83, Rewe, 1984 ECR 1229; C–233/94, Germany v Council, 1997 ECR

Legisprudence and European Law 115 of proportionality (discussed above), the duty to give reasons, the principle of legal certainty and the duty to make an informed choice. Pursuant to Article 253 EC, every binding Community act must state the reasons on which it is based. That obligation applies to legislative and executive acts alike (whereas at the national level, usually only administrative acts must comply with that duty). The Court of Justice has ruled that the obligation to give reasons is an essential requirement and in case of insufficient reasoning will annul the act on that ground. The duty to give reason is a formidable tool of legislative control. The Court of Justice justifies the obligation as follows: In imposing upon the [European institutions] the obligation to state reasons for [their] decisions, Article [253] is not taking mere formal considerations into account but seeks to give an opportunity to the parties of defending their rights, to the Court of exercising its supervisory functions and to Member States . . . of ascertaining the circumstances in which the [institutions have] applied the Treaty.21

It has been argued before the Court that less reasoning is required when the Community institutions enjoy a wide discretion in taking their decisions (in analogy to the fact that at the national level, legislatures normally do not have a duty to give reasons). However, the Court of Justice has not accepted that view. On the contrary, the Court finds that Community institutions must always carefully document the measures they take and particularly so when they enjoy extensive freedom in reaching their decisions.22 The principle of legal certainty embraces not only the principle of nonretroactivity and respect for legitimate expectations, but also imposes certain standards of legislative drafting and presentation. The Court has, for instance, intervened in cases where conditions governing and levels of financial burdens imposed by Community legislation were too unclear to be understood.23 In another case the Court has mentioned the requirement of ‘legal clarity’ as imperative in a sector in which any uncertainty may lead to incidents and the application of particular sanctions.24 As discussed above, various initiatives are now being taken at the level of the Community institutions in order to improve the quality of Community legislation. Finally, Community institutions are under an obligation to decide only after having informed themselves of all the facts and arguments of a given case. In the field of environmental policy, such a duty is explicitly enshrined in the Treaty. Thus, the Treaty explicitly requires that law-making in the environmental field takes available scientific and technical data, as well as the potential benefits and costs of action or lack of action, into account.25 As a general matter, the Court 21

Case 24/62, Germany v Commission, 1963 ECR 69. See, eg, Case 17/74, Transocean Marine Paint, 1974 ECR 1080, consideration 16. 23 Case 137/85, Maïzena v Balm, 1987 ECR 4603; Case T–172/89, Vandemoortele, 1990 ECR I–4690. 24 Case 32/79, Commission v United Kingdom, 1980 ECR 2445, consideration 46. 25 Arts 74, para 3 EC. 22

116 Amaryllis Verhoeven of Justice has required that Community institutions are to base their decisions on a careful and impartial examination of all aspects of a case.26 The duty to base decisions on an informed choice is linked to issues of transparency and direct democracy. As said before, the Commission retains the monopoly to initiate law-making. The Commission, when drawing up proposals, is increasingly making use of techniques that allow for greater openness and participation. For instance, it publishes discussion and consultation documents (so-called Green and White Papers) with the aim of fostering discussion with interested parties. Legislative proposals are put on the Internet with the same aim. Moreover, the Commission regularly enters into dialogue with interest groups.27 Such interaction is designed to lead to better law-making (it increases the possibility of reaching an informed choice) as well as to strengthen the social acceptability (and thus legitimacy) of European rules. In this context, it may be remarked that although all principles of good and proper legislation apply to legislative and executive measures alike, the Court of Justice has in practice adopted a differentiating approach. Review of acts of the Council tends to be rather deferential, whereas acts of the Commission have been controlled more strictly.28 That difference is explained on the one hand by the fact that the democratic accountability of the Commission is deemed weaker than that of the Council. On the other hand, it is linked to the fact that the Council tends to deal with legislative-type acts, whereas the Commission is more often in charge of technically complex but politically less visible matters. (iii) Fundamental Values With the gradual broadening of economic integration towards political integration, principles and values have increasingly joined policy objectives as constraints on European law-making. The TEU now reads that [t]he Union is founded on the principles of liberty, democracy, respect for human rights and fundamental freedoms, and the rule of law, principles which are common to the Member States.29

This statement is, to a large extent, declaratory of a long-standing practice of judicial review of Community action in light of these principles. In particular, there is a wealth of cases on respect by the institutions of fundamental rights and freedoms. However, time and space prevents us from dwelling further on these matters.

26 Case C–269/90, Hauptzollamt München-Mitte v Technische Universität München, 1991 ECR I–5495; see also Case C–212/91, Angelopharm v Freie und Hansestadt Hamburg, 1994 ECR I–200. 27 The Commission keeps a directory of interest groups covering some 600 organisations. 28 Compare the Court’s review of Case C–331/88, Fedesa and others, 1990 ECR I–(4023) 4063 with its review in Technische Universität München, n 26 above. 29 Art 7, para 1 TEU.

Legisprudence and European Law 117 (c) Principles of Procedure As mentioned above, the legal basis for each Community action determines which institution must act according to what procedure (eg, voting majorities). The choice of legal basis has, therefore, important institutional consequences. The European courts review whether the law-making processes prescribed by the Treaties are duly respected and appear particularly vigilant where issues of democracy are at stake. One recalls that the involvement of the European Parliament in decisionmaking at the European level is still a fairly recent phenomenon. However, the Court of Justice has, in a series of cases, proven to be a reliable supporter of the Parliament’s voice on the legislative scene. It has defended the prerogatives of the Parliament by making non-respect of these prerogatives by the other institutions (in particular the Council) a ground for annulment of the act at issue. In doing so, it has referred to democracy as a fundamental principle on which the European legal order is based.30 Further, it has granted the Parliament a right to defend its prerogatives in court (it has done so even contra legem, ie, notwithstanding the fact that the Treaty did not—yet—accord the European Parliament a ius standi).31 Judicial control on respect by the institutions of principles of democracy also takes place outside the parliamentary context. In UEAPME, for instance, it was held that social policy agreements concluded between social partners may only be turned into European law (through a ratification by the Council) if the social partners that concluded the agreement are sufficiently representative of labour and industry.32

III . IN SEARCH OF A CONCEPTUAL MODEL

1. Three Competing Models of Law-making Above, an attempt was made to describe the rather unusual nature of European law-making. Here, our concern is with how to ‘conceptualise’ that process and, more importantly, to justify it in democratic terms. Over time, various conceptual models have been proposed. I have grouped them here as the regulatory, the intergovernmental and the parliamentary model.33 Each of these models has a descriptive component (they attempt to explain what European law-making is 30

Case 138/79, SA Roquette Frères v Raad (Isoglucose), 1980 ECR 3333. Case C–70/88, Parlement v Raad, 1990 ECR I–2041. The EC Treaty has since been amended. Case T–135/96, UEAPME v Council, 1998 ECR II–C335 33 See also J Caporaso, ‘The European Union and Forms of State: Westphalian, Regulatory or Post-Modern?’, [1996] JCMS 29 ff; R Dehousse, ‘European Institutional Architecture after Amsterdam: Parliamentary System or Regulatory Structure?’, [1998] CML Rev 595; G Majone, ‘Europe’s Democratic Deficit: the Question of Standards’, [1998] ELJ 5 ff. 31 32

118 Amaryllis Verhoeven and how it functions) as well as a normative one (they try to justify European law-making in democratic terms). The scope of these models is not limited to European rule-making sensu stricto but embraces the nature of European integration itself. Each model entails a different (constitutional) blueprint for the European integration project based on particular ideas of democratic legitimacy.34 Whereas the models have throughout the history of the Union exerted a varying influence, each model retains its advocates today. As will be discussed, none of them is, in my view, adequate or sufficient to explain and legitimise Europe’s law-making process. The first, which has been called the regulatory model, views the Union as a special purpose organisation (Zweckverband) or an agency, the purpose of which is to address a number of issues over which it can achieve a greater efficiency than the Member States acting individually.35 In that view, the goal of European integration is primarily economic, not political. The task of the European institutions is not to enact norms on broad, political matters but to develop regulatory solutions for a series of socio-economic problems. In this model, rule-making in Europe is both limited in scope and highly instrumental in nature. It is a form of problem-solving and seeks in (real or presumed) expertise the primary ground to legitimise itself. Typically, this model will avoid the term ‘law-making’ and use the term ‘rule-making’ instead. The second model stresses the intergovernmental features of European integration and law-making. It views the Union as an international organisation, although a rather integrated one, and stresses the fact that national states and governments are the main actors on the European forum. According to this view, rule-making in Europe is ‘political’ but not ‘legislative’ in nature: the term political here is used to denote power politics and diplomacy rather than the pursuit of the public interest.36 The regulatory and intergovernmental models are, in my opinion, flawed mainly for two reasons. First, the regulatory and intergovernmental models fail to take account of the truly political nature of the Union (note that I use the term ‘political’ here not in the diplomatic sense, but in order to indicate that the European Union sets norms for a political community of European citizens). The regulatory and intergovernmental models are reductionist in that they tend to concentrate solely on one type of rule-making, ie, technical, problem-solving 34 The word constitutional is bracketed here since the term constitution is regarded by some authors as exclusively linked to the institution of the state (among others, D Grimm, ‘Does Europe Need a Constitution?, [1995] Eur LJ 28). Hence, speaking of a constitution for Europe would imply, in their opinion, promoting a statal vision for Europe. 35 See, among others, G Majone, ‘From the Positive to the Regulatory State: Causes and Consequences in the Mode of Governance’, [1997] Journal of Public Policy 139 ff; R McGowan and W Wallace, ‘Towards a European Regulatory State’, [1996] JEPP 560 ff. 36 See, for instance, A Moravcsik, ‘Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach’ in S Bulmer and A Scott (eds), Economic and Political Integration in Europe: Internal Dynamics and Global Context (Blackwell, Oxford, 1994) 29–80; W Wallace, ‘Europe as a Confederation: the Community and the Nation-state’, [1983] JCMS 57–68.

Legisprudence and European Law 119 regulation in the Commission and committees on the one hand and international diplomatic-style bargaining within the Council on the other. European law-making goes well beyond the scope of technical, problemsolving activity. Today’s European law-making differs in no significant degree as regards content, scope and effect, from national legislation. It enters many fields and implies often difficult policy choices. At the same time, European rulemaking no longer follows the pattern of diplomatic negotiation. Since 1987, the Council decides in many instances on the basis of a qualified majority vote (which means that an individual Member State that does not agree with the proposed measure can be bound against its will). Further, the European Parliament now has broad powers of ‘co-decision’, ie, of co-enacting legislative measures on an (almost) equal footing with the Council. The second reason is related to the first. Since the regulatory and intergovernmental models are no longer adequate to describe the process of rule-making at the European level, their account of democratic legitimacy is insufficient. The regulatory model seeks to avoid problems of democratic accountability by stressing the technical, problem-solving nature of the process. Where no political, but only technical choices must be made, it makes sense to delegate rulemaking to non-elected agency-type bodies that offer a certain expertise. However, forms of delegated rule-making can hardly be called democratic once they leave the strict confines of technical market regulation. In reality, European rules have policy consequences for virtually all aspects of life—and such policy choices ought, in current understandings of democracy, to be made by elected (or at least otherwise representative) bodies. In the intergovernmental model, the democratic legitimacy of European integration and the European institutions proceeds from the democratic legitimacy of the Member States. National Parliaments are supposed to control their national ministers operating at the European level. Yet such controls are, in reality, minimal, and do not allow for a real policy formation at the European level. The third, parliamentary model, is probably most influential nowadays. It avoids the mistakes of the other two models. It sees the European Union as a political enterprise, an ongoing process towards a genuine political community, and considers European law-making to be an essential element of that process. Again, its concern is with how to ensure the democratic nature of European lawmaking. On this point, many advocates of the parliamentary model argue that the best way to cure the EU’s democratic deficit is to reproduce, at the European level, the guarantees of parliamentary democracy found at the national level. Hence, it is argued that the European Parliament should have stronger legislative functions as well as better means and powers to control the implementation of legislation by the Commission and various committees.37

37 See, among others, the resolution of the European Parliament on the democratic deficit in the EC of 17 June 1988, OJ 1988, No C 187/229.

120 Amaryllis Verhoeven The quest for more parliamentarism at the European level is not new and has already led to significant and mostly beneficial reforms. As mentioned above, Parliament and Council in many instances jointly enact legislative measures on the basis of a procedure that largely follows bicameral parliamentary models of law-making. In addition, the European Parliament’s control on the Commission has been enhanced. Yet the parliamentary model of law-making has its limits in the European context. Streamlining European law-making processes on the basis of statal models will not necessarily enhance the democratic legitimacy of European law and, concomitantly, will not necessarily lead to better lawmaking. Here is, tentatively, a list of reasons why. First, the parliamentary model, however much we are used to it, is not necessarily attractive. One cannot but see that at the national level also, parliaments have lost much of their aura and powers. Laws and regulations, at the national level, are also more and more made through alternative channels (‘special power’ legislation is but one example). Secondly, the European parliament can probably not play, within the European Union, the central role that national parliaments play within their own system. To begin with, there is a problem of size. As the European Union expands, the European Parliament comprises an ever-increasing number of members. Recently, a ceiling of 700 members of Parliament has been introduced,38 but that number is both too high and too low. Too high to lead to an efficient communication of the public interest. As James Madison already observed, the more members a parliamentary assembly comprises, the more passion prevails over reason. Too low, on the other hand, to be truly representative. At present, the number of parliamentary representatives elected in Finland, for instance, is sixteen; that is, Finland chooses just above 2 per cent of the members of the European Parliament. That number will have to be reduced after the next round of new accessions, if the ceiling is retained. Further, in classical democratic theory,39 Parliament is thought to represent the nation, a presumed homogeneous body of people. Yet there is no European nation—if Europe forms a political community at all, it is a community characterised by diversity and pluralism. Significantly, the European Parliament does not represent the European people, but rather the ‘peoples of the States brought together in the Community’.40 Third and most fundamentally, parliamentarism is traditionally bound up with a specific idea on sovereignty. In traditional thinking, sovereignty is deemed to be vested in the nation represented in parliament. Parliamentary legislation is deemed supreme, whereby this supremacy is justified on either jusnaturalistic or voluntaristic grounds. The inviolability of the law follows from 38

Art 189 EC. The lineage of which runs through Jean-Jacques Rousseau and Sieyès (who ‘re-read’ Rousseau in a considerable manner). 40 Art 189 EC. 39

Legisprudence and European Law 121 this thinking: since the lawgiver is supreme, laws must be applied but cannot be criticised by the judiciary. Further, the lawgiver is thought to control all other organs of the state, including the executive. Although the European Parliament produces, together with the Council, legislative-type norms that take precedence over national law—even over national constitutions—41it cannot (and does not) claim to represent a European ‘sovereign.’ A European ‘body politic’—as peoples of a state are traditionally considered to constitute—simply does not exist. While the Member States have, through the European Treaties, entrusted the European institutions with the exercise of certain sovereign rights, they retain original powers and therefore consider themselves still ‘sovereign’ in the sense of the ultimate source of the law, both national and European.42 Significantly, the European Treaties, which form the ‘constitutional charter’43 on which Union governance is based, are adopted and must be amended by a unanimous vote involving all Member States.44 The lack of a sovereign mandate accounts for the differences between European and national law-making described above. European law-makers enjoy a significantly smaller degree of initiative and political discretion than their national counterparts. As discussed, they are bound by an impressive array of restrictions as regards competence, content and procedure. Respect of these constraints is subject to judicial review. Clearly, traditional views on legislation developed in the context of state forms of parliamentary democracy cannot account for such far-reaching legislative limits and judicial review on legislation. In fact, the acceptability of judicial review of legislation is traditionally questioned. Of course, liberal thinkers have argued that it is the courts’ duty to protect individual rights against the tyranny of the majority imposing its will in the national parliament. But judicial review European-style is not only (not even primarily) about protecting individual rights—it is rather about controlling that the European institutions use the powers that have been entrusted to them in a proper and democratically accountable way. The existence of constraints in terms of objectives, for instance, is nothing but the logical flip side of the fact that the European Union is not endowed with original powers. The Member States have delegated well-defined and well-circumscribed powers to the European institutions in order to carry through certain tasks and objectives concerning European integration. The powers accorded to the institutions of the European Union are precisely ‘the means necessary to attain its objectives and carry through its policies’.45

41

Case 11/70, Internationale Handelsgesellschaft, 1970 ECR 1125. Below, I will criticise the classical conception of sovereignty by reference to the works of Jürgen Habermas. 43 Case 294/83, Les Verts v European Parliament, 1986 ECR 1339. 44 Art 48 TEU. 45 Art 7, para 4 TEU. 42

122 Amaryllis Verhoeven It could be argued, on the other hand, that European law-making is not legislative in nature but only ‘administrative’ and hence, should be analysed and structured by reference to national models of administrative decision-making. Admittedly, the constraints imposed on European law-makers are rather similar to those imposed on national administrators (legal basis, duty to give reasons, etc.). Also, judicial review on European legislation follows, by and large, the pattern of judicial review of administrative acts at the national level. For instance, European legislative norms can be annulled on grounds that are very similar to the grounds contained in French administrative law.46 The lack of a real separation between legislative and administrative activity points in the same direction. On the basis of these findings, asserting that the European law-making process is purely ‘administrative’ in character is only half true, however. True in the sense that European law-makers, like national administrators, yield no original, sovereign powers; false, however, if it would lead us to ignore the essentially political nature of the European law-making process (as, for instance, the regulatory model does). Within the limits of their powers, and of the objectives set forth in the Treaties, European law-makers do make important policy choices. The European Courts duly take account of the political nature of the law-making process by using a rather large degree of deference when, for instance, reviewing whether the objectives set forth in the European Treaties are furthered. Again, the essence of judicial review of European law-making activity is not to see whether the institutions stay neatly within the frame of their powers, but rather—so at least I argue—to see whether political choices are taken in a truly democratic manner, taking into account the decentralised nature of the European Union and the claims of its Member States to ‘original’ powers. 2. Law-making European-style: Beyond Parliamentary Sovereignty The condition of European law-making is law-making devoid of sovereignty. It is a legislative activity closely tied to a web of constitutional requirements enforced by courts. The fact that no European nation exists as a discernible body or subject is both troublesome and liberating. Troublesome, because it makes it at first sight difficult to locate the seats of democracy in the European venture; liberating, because it frees law-making of the carcass of (unitary) parliamentarism in which it was traditionally vested. In this respect, the European Union might offer unique opportunities to democratic theory and practice. Rather than through parliamentary models, European rule-making can in my view be better analysed on the basis of a procedural/deliberative model of 46 Ie, lack of competence, infringement of an essential procedural requirement, infringement of European law and misuse of powers (Art 230 EC). Cf M Lagrange, ‘La Cour de Justice des Communautés européennes: du Plan Schuman à l’Union européenne’, [1978] Rev trim dr eur 2 ff.

Legisprudence and European Law 123 democracy that is indebted among others to Jürgen Habermas. Habermas advocates a proceduralist conception of democracy. Rather than on the state (the political society), on institutions (parliaments) or on rights, his conception focuses on an ideal procedure for deliberation and decision-making.47 The procedural/deliberative model appears well suited to explain the creation of norms at the European level. At the same time, it has normative consequences—it provides an angle from where to test, criticise and improve European law-making. First, in a procedural conception of democracy, democracy no longer presupposes the existence of a nation as a more or less homogeneous social whole. The very structure of communicative action is hoped to be constitutive of the political society in which communicative action takes place, as well as of the normative frame of rules which govern that society. Second, in order to be democratic, law-making no longer needs to be centred in a parliamentary body (deemed to represent the nation). Procedural democracy allows for a decentralized structure of decision-making, which contains both formal (institutional) and, crucially, informal channels: Discourse theory works instead with the higher-level intersubjectivity of communicative processes that flow through both the parliamentary bodies and the informal networks of the public sphere. Within and outside the parliamentary complex, these subjectless forms of communication constitute arena’s in which a more or less rational opinion- and will-formation can take place.48

Deliberative democracy requires both less and more than traditional parliamentary democracy. Less, in the sense that no ‘nation’ (as a specific set of a priori cultural, historical, bonds, etc.) is required. More, in the sense that deliberative democracy requires an active citizenry genuinely entering into debate in the public sphere. Viewed against the procedural/deliberative model, the dispersed and diffuse picture of rule-making perceived at the European level does not necessarily appear in a negative light. The so-called ‘institutional balance’, ie the balanced interactions between the European institutions, might be explained as a mechanism allowing different constituencies which make up Europe’s civil society to enter into a debate with each other. The constraints imposed by European law on this decision-making process could be explained as institutional devices encouraging the various actors in the legislative process to treat their preferences not as simply rigid beliefs but rather as choices which are open to debate and alteration. The Commission’s exclusive right of legislative initiative (which stands at odds with parliamentary models of law-making) can, for example, be explained in this light. The Commission, a body of independent experts, is 47 Cf in particular J Habermas, Between Facts and Norms (Polity Press, New York, 1997). Habermas’s political theory rests on his theory of communicative action he developed earlier in The Theory of Communicative Action (Beacon Press, Boston, 1984). 48 J Habermas, ‘Three Normative Models of Democracy’ in S Benhabib (ed), Democracy and Difference (Princeton University Press, Princeton, 1996) 28.

124 Amaryllis Verhoeven arguably best placed to voice the interests of the Community as a whole; hence it should retain agenda-setting powers.49 Typically, for matters where a large consensus among Member States and European opinion exists or matters of a ‘technical’ nature, decision-making is entrusted to relatively autonomous institutions (Commission, agencies) that can use relatively quick and efficient procedures. However, more ‘sensitive’ matters require the involvement of more, and in particular more accountable, institutional actors (Council, European Parliament) according to processes that are unavoidably less efficient. Third, Habermassian theory invites us to reconsider our traditional notions of sovereignty. Popular sovereignty is no longer considered an attribute of a particular group of people (a ‘Nation’). Rather than being vested in one collective body, popular sovereignty is re-interpreted in intersubjectivist terms. It ceases to be connected to a person, and becomes proceduralised: [T]he self of the self-organizing legal community here disappears in the subjectless forms of communication that regulate the flow of deliberations in such a way that their fallible results enjoy the presumption of rationality. . . . Popular sovereignty . . . retreats into democratic procedures and the legal implementation of their demanding communicative presuppositions only in order to make itself felt as communicatively generated power. Strictly speaking, this communicative power springs from the interactions between legally institutionalised will-formation and culturally mobilised publics which constitutes a society and its legal order.50

Fourth, the procedural/deliberative model allows us to view the European ‘constitution’ (which, as mentioned above, is not a formal text but rather a set of fundamental rules pertaining to procedure and rights) in a novel light. From a deliberative point of view, the concept of a constitution is not something secondary. The democratic process is not a spontaneous activity, but an institutionalised activity. It is the product of rules that are designed to maximise the value of the democratic process in terms of uncovering the best solutions to questions of public life. Since such constitutional rules are the prerequisite of the democratic process, all acting on the public arena must follow them. Hence, a procedural–deliberative conception of democracy requires some forms of judicial review of rule-–making activity in light of the requirements of the constitution. However, contrary to what liberals argue, the purpose of judicial review is not to protect individual rights that somehow belong to a realm outside the democratic process. The purpose is rather to guarantee that the rules and conditions that make democratic discussion and decision possible are respected.51 The above-mentioned

49 P Craig, ‘Democracy and Rule-making within the EC: an Empirical and Normative Assessment’, [1997] ELJ 105–30. 50 Habermas, ‘Three Normative Models of Democracy’, (n 48 above) 29. 51 For an excellent account of the relationship between democracy and the constitutionalism in a deliberative concept of democracy, see C Nino, The Constitution of Deliberative Democracy (Yale University Press, New Haven, 1996).

Legisprudence and European Law 125 constraints of competence, content and procedure to which European legislators are bound can be analysed in this light. It is often argued that the democratic justification of the European constitution and of judicial review in light of this constitution rests on shaky grounds. That is only so, however, if one approaches constitutions and constitutionalism from a traditional, state perspective, ie, if one wishes to view the European constitution as the highest expression of the will of a people. In that sense, the European Union has no constitution, or at least no democratically valid one. Yet the legitimacy of the European constitution and constitutional review is not based on any such view. The European constitution is legitimate since and to the extent that it can be viewed as the frame enabling European citizens, groups and institutions to make use of communicative power within a European ‘public sphere’ in such a way as to lead to democratically justified results. It follows, also, that the actual European constitution (ie the current set of rules and conditions) has no validity or any sacrosanct character per se. The actual constitution is valid only to the extent it meets the requirements of the ‘ideal’ set of constitutional norms required for a proper functioning of the democratic society. As a result, the European constitution appears as a flexible, dynamic and evolutionary whole of rules that enjoy a certain presumption of a ‘higher’ validity, a presumption that can be rebutted, however, if the rules do not meet the requirement of democracy. It should finally be stressed that our argument—that the procedural/deliberative model provides the most accurate account of European law-making—does not aim to glorify the existing institutional and legal arrangements by reference to this model. On the contrary, our hope is that this model can be used as a standard against which to critically assess the European normative process and the constitutional frame that underpins it. In particular, we hope that this model will allow us to ‘test’ whether and how the various legislative constraints that were briefly described above can contribute to a more democratic process of decision-making at the European level.

IV . CONCLUSION

Today’s European Union struggles with the question of how to make European legislation more rational and democratic. I have argued that reforming European law-making on the basis of parliamentary patterns at the national level does not offer a solution in the absence of the fundamental preconditions of state parliamentarism. In particular, EU law-makers do not enjoy original powers (that come, traditionally, with the idea of sovereignty), but only delegated powers which they are to use within a set of relatively stringent constraints (if compared to national law-makers). The way ahead lies, rather, in re-thinking European law-making in light of a procedural–deliberative model of democracy. Viewed from this perspective,

126 Amaryllis Verhoeven European law-making and in particular the European constitution, which contains a far-reaching set of principles and constraints European law-makers are bound to respect, appear in a different light. Judicial review, in order to enforce compliance with the European constitution is legitimate, then, since (and to the extent) it furthers the process of democratic will formation and decision-making at the European Union level.

10

Rationality in Legislation by Employing Informatics? WIM VOERMANS

I . DRAFTING ASSISTANCE AND COMPUTERS : LEGIMATICS

Using informatics to further the drafting of legislation, ie the field of legimatics, is a discipline which—though very young—can already claim a modest tradition. In legimatics basically two approaches towards the development of legal IT systems currently exist: the information-oriented approach and the Artificial Intelligence (AI) approach. In the information-oriented approach, legal and legislative problem-solving processes are considered to be information problems. IT systems built according to the information-oriented approach assist system users by processing and providing accurate information to solve the information problems which arise in solving a certain problem. Information-oriented IT systems supply the information needs. For the development of IT systems like these, an accurate insight into the information needs of a problem-solving process is required. In the AI-based approach, legal and legislative problem-solving are considered to be reasoning processes which require knowledge. In systems built according to the AI-oriented approach, attempts are made to represent the knowledge needed to solve a certain legal or legislative problem and model it in a way which allows a computer system to ‘reason’ with it. Legal AI systems, therefore, can (partly) solve legal problems by ‘machine processed’ legal reasoning. Building AI-based systems requires accurate insight into how specific legal problems are solved and what kinds of specific knowledge are used during the problem solving. In the recent past some authors like myself have argued that, given the characteristics and open-ended nature of many legal problem-solving processes, like the legislative process or the policy processes leading to decision in public administration, the AI-based approach is as yet not productive for the building of automated IT drafting systems assisting legislators for the duration of their decision processes. Legislative drafting involves far too many different and too complex sorts of reasoning and knowledge to be representable by a computer system. This does not mean, however, that the AI-based approach cannot be productive in

128 Wim Voermans building IT tools for specific parts of legislative drafting- or decision-support systems for the application of legislation. In this contribution I will discuss the development, motivation and functionalism of one drafting-support information system in particular: the so-called LEDA system. This system was built to support Dutch legislative draftsmen during the drafting process. LEDA (Legislative Design and Advisory System) is designed to offer easy access to the Dutch Directives for Regulations (Aanwijzingen voor de regelgeving). It guides users through an interactive drafting checklist and checks legislative drafts to see whether or not important drafting requirements are met.1 The LEDA system is currently being used within Dutch ministerial departments. The Belgian federal government is considering a similar system, called Solon, to support legislative drafting.2

II . LEGISLATIVE DRAFTING AND LEGAL PROBLEM SOLVING

The Dutch LEDA project started out with a theoretical survey of the legislative process in The Netherlands. In order to be able to assess the possibilities of computer-supported legislative drafting, an in-depth insight into the nature of legislative processes is fundamental. Considered closely, legislative drafting appears as a complex and open-ended decision process which differs quite substantially from many strictly ‘legal’ decision processes. Legislative drafting, for instance, involves far more sorts of knowledge than mere legal knowledge. Furthermore, legislative processes and legislative problem solving are only partly determined by legal rules. If we examine the legislative decision-making process more closely, we see, for instance, that legislative draftsmen do not use merely legislative methods and legal rules to tackle legislative problems. During this process they constantly make all kinds of legislative decisions. These decisions can never claim to be perfect, or legally valid decisions. Legislative decisions or solutions can only claim to be ‘relatively appropriate’ solutions3 in view of all the (factual, societal, political, legal, and socio-economic) circumstances involved. Legislative decision making is therefore not a process of application of fixed legal standards, but an open process in which a legislative draftsman weighs different possible solutions in light of their relative appropriateness. The best solution relatively is the solution which is substantiated by the 1 W Voermans, Sturen in de mist . . . maar dan met radar. De mogelijkheden van de toegepaste informatica bij het ontwerpen van wetgeving (W E J Tjeenk Willink, Zwolle, 1995). 2 R Van Kuyck, S Debaene and B Van Buggenhout, ‘Solon. A Computer Aided Statutory Drafting System for the Flemish Government’ in Conference Proceedings of the 5th International Conference on the Law in the Information Society (CD-Rom, Florence, 1998). 3 R Hotz, ‘Strukturierung des Vorverfahrens der Gesetzgebung. Erste Schritte zu einem allfälligen Einsatz von Computern bei der Schweizerischen Gesetzgebung’ in T Öhlinger (ed), Gesetzgebung und Computer (C H Beck, Munich, 1984) 164–91.

Rationality in Legisprudence by Employing Informatics? 129 most convincing arguments. The most convincing arguments will be the arguments which rate highly in the legislative discourse in which legislative draftsmen find themselves together with their departmental superiors, politicians, members of parliament, interested parties, lobby groups, etc. Very convincing arguments, or authoritative arguments, in this discourse will be the arguments upon which almost everyone agrees. In this sense, legal (eg, constitutional) arguments or generally accepted legislative methods and techniques constitute strong authoritative arguments to back up a solution, while mere personal or political opinions or beliefs have a much lower-ranking status. The appropriateness of a draft is largely dependent on the quality and the status of arguments which sustain the solutions held in it. In the legislative decision process, legislative draftsmen will always try to find and use the strongest argument possible to substantiate a solution and in choosing between equivalent solutions he or she will choose the solution which is supported by the most convincing arguments within the legislative discourse. This searching for and weighing of—especially—authoritative arguments is a process which can be conceptualised, modelled and formalised.4 The LEDA system harbours a modelling of this ‘argumentative strategy’.

III . LEGISLATIVE QUALITY STANDARDS AS A MOTIVATION FOR THE DEVELOPMENT OF LEDA

Legislative drafting is—as I pointed out in the previous paragraph—not primarily aimed at achieving legal validity, but rather on reaching the highest possible quality standard of legislative decisions. Legislative quality, in its turn, is largely dependent on the span of the considerations underlying the decisions in a Bill. But what do we mean when we want to discuss legislative quality? There is no general definition, but legislative quality will always concern questions related to the way in which legislation meets generally accepted legislative quality standards. These standards are not universal. They vary in nature and content according to the legal system to which they belong. They may involve constitutional, legal, political, societal and administrative standards alike. The Dutch Directives for Regulations are an example of a heterogeneous collection of legislative quality standards. In the midst of the 1980s the Dutch government became increasingly concerned with the quality of legislation due to serious problems regarding its quality and effectiveness. To improve the overall legislative quality, different policies were pursued and enacted.5 One of the main results of these governmental efforts and policies was the adoption of a general legislative policy, which consists of a set of measures aimed at the lasting improvement of 4 5

S Toulmin, The Uses of Argument (Cambridge University Press, Cambridge, 1958). Zicht op Wetgeving (Legislation in Perspective) (Dutch Ministry of Justice, The Hague, 1991).

130 Wim Voermans legislative quality by setting quality criteria. A substantial part of these measures concerns the fundamental drafting stage. 1. The Directives for Regulations In The Netherlands, the increasing complexity of this assignment resulted in a crisis in the legislative quality of Bills in the latter part of the 1980s. As a reaction, legislative quality policies were adopted and laid down in the Directives for Regulations.6 These Dutch Directives are quite elaborate. They are a comprehensive legislative technique handbook, but also contain substantial legal and policy-related legislative issues. As a result, the Directives are a voluminous set of drafting guidelines (372 in total), accompanied by much secondary information (for example, explanations, illustrations, model clauses, etc.) which are to be observed by all government officials and public servants when drafting Bills.7 Deviation from the Directives is allowed only if application of them would lead to ‘unacceptable results’ (Directive 5). The Directives constitute a voluminous ‘Draftsman’s Handbook’ dealing with every important activity within the drafting process. They concern methodological and substantive legislative issues, for example, how to prepare a draft, how to adopt elements of public policy into proposed legislation, how to implement European legislation, what kind of legislative instruments to use, how to delegate legislative powers, how to attribute administrative authority, what kind of quality considerations are to be made, etc. Directive 7 offers a good example of these ‘methodological’ Directives. It states: Before deciding to introduce a regulation, the following steps shall be taken: a. knowledge of the relevant facts and circumstances shall be acquired; b. the objectives being aimed at shall be defined in the most specific, accurate terms possible; c. it shall be investigated whether the objectives selected can be achieved using the capacity for self-regulation in the sector or sectors concerned or whether government intervention is required; d. if government intervention is necessary, it shall be investigated whether the objectives in view could be achieved by amending or making better use of existing instruments, or, if this proves to be impossible, what other options are available; e. the various options shall be compared and considered with care.

Other Directives concern the more technical aspects of drafting, such as the structural design of a draft (arrangement of the elements in the draft) and even more strictly legislative technique-oriented Directives concerning the phrasing and terminology of a draft (including the use of model clauses, model presentation letters, etc.). Finally, there is a group of Directives that concern all kinds of drafting-related (legislative) procedures. In this section a number of model letters and style requirements are incorporated. 6 Aanwijzingen voor de Zegelgeving (Dutch Directives for Regulations 1993), Official Journal 1993, 230 (Stert. 1993, 230). 7 The Directives have been amended several times since 1993. The last revision dates from 2002.

Rationality in Legisprudence by Employing Informatics? 131

2. Handling the Directives The Dutch Directives are voluminous. There were 372, but their total number has increased due to different amendments that have taken the form of A–Z Directives. The total number of Directives exceeds 410 at the moment. On top of that, nearly every Directive has a separate explanatory memorandum at the bottom, which contains an explanation, and—in many cases—some illustrations. The bare text of the Directives covers more than 200 pages. The sheer size of the Directives limits their accessibility and constitutes a serious obstacle for their users. This circumstance makes it difficult for legislators (even experienced ones) to find their way through the new Directives during the drafting stage. An information system, it was felt, could be the way out of these problems. This signified the start of the LEDA project.

3. The Goals of the LEDA project The main goal of the LEDA project was to make the information in the Directives themselves accessible in concordance with the information need during the different stages of the drafting process. A secondary goal was to make the information, referred to in the Directives (secondary information), available to users. Many Directives, as it happens, do not prescribe what the solution should be in a certain situation—as is often the case with ordinary legal rules— but rather, which activity should be undertaken at a certain moment, and what kind of information is needed to be able to perform that prescribed activity. The third goal of the LEDA project was to offer knowledge-based drafting support on the basis of the legislative knowledge within the Directives, pursuant to the knowledge-based access of the information from the Directives. To be able to do this, an analysis of the drafting process itself was made, and an analysis of how the Directives should be used during the different drafting stages (a so-called activity and information analysis). The model of the drafting process as the result was subsequently programmed into the system. The modelling of the drafting process constitutes the backbone of the LEDA system. All functions and attributes of the system are connected to it.

IV . THE LEDA - SYSTEM : HOW DOES IT WORK ?

1. The LEDA system’s Functions: General Features The LEDA system offers three major functions: methodological support, document drafting and document assembly support, and knowledge-based information retrieval. The combined functions make LEDA an integrated authoral

132 Wim Voermans system, ie an IT system which assists users in solving legislative problems on the basis of legislative information; and, moreover, the system supports its users in drafting a legislative document which meets with the requirements of the Directives. Technically, the LEDA system is a hypertext network which allows for different kinds of navigation and working patterns within the system. The support offered by the LEDA system, though practical, is very modest in nature. LEDA assists in the prestructuring of a draft by offering a drafting method to the user which consists of a set of drafting levels. These drafting levels—which act as transgressional layers in an edit field—contain important information, mostly derived from the Directives for Regulations, about legislative quality requirements to be considered within a particular drafting level. The information levels correspond with different possible substantial and structural elements within a draft. For instance, LEDA harbours levels such as ‘definitional clause’, ‘attribution of administrative authority’, ‘prohibition-permit systems’, ‘supervision (model) clauses’, ‘sanction systems’, ‘transitory regime’, etc. LEDA in its present form consists of 54 of these levels for the drafting module alone. Users do not need to use all of these levels: they are invited to make choices which alter the number and order of the LEDA information levels. By tailoring the information environment, LEDA tries to address the particular information need within a particular drafting project. Combined, the drafting levels constitute a semantic network which can be navigated at random. By progressing through the network of levels a LEDA user is confronted with documentary information and active checklists which, when used or filled out, procure the main building blocks for a draft. These building blocks can be edited at will while working with LEDA: the system is designed as a plug-in in MS ® Word © 97. The LEDA system is mainly an informational skeleton, which guides its users through drafting new legislation. The system functions predominantly as an elaborate legislative guide, for it contains a number of Directives that should be observed during the different stages of the drafting of a Bill. In addition, LEDA possesses a functionality which makes it possible to analyse a draft text dynamically in order to see which Directives are relevant. To be able to do this, LEDA is able to recognise drafting concepts in the draft text (eg, formulations used to delegate or sub-delegate powers). Once these concepts are recognised by LEDA, the systems connects (by means of hypertext) the analysed text fragment to information leaflets corresponding to the drafting concept in question. LEDA’s functions are integrated throughout the system and organised in two major modular components called the Preparatory (policy) Module (Pmodule) and the Basic Design Screen (BDS).

2. The Preparatory Module The Preparatory Module in LEDA offers knowledge-based access to the Directives concerning substantive, methodological and structural design issues,

Rationality in Legisprudence by Employing Informatics? 133 in a way consistent with the chronology of events in the drafting stage. The Pmodule permits the user not only to draft a preparatory document (eg, a policy memorandum), but also supports the creation of a skeletal form of a draft, which can be used as the basis for the actual structural design and formulation of a draft for which the BDS module is the dedicated platform. To this end, the

P r e p a ra t o r y M o d u l e Methodological step 1 —N1 —level text —selection —template/edit-field

Methodological step 2 —N2 —level text —selection —template/edit-field Methodological step 3 —N3 —level text —selection —template/edit-field

Level Information —N1 level dependent a) relevant Directives b) level information c) analysis scheme

General Information level dependent 1) Directives (all) 2) Kluwer ADW 3) Kluwer legal library 4) database gateways 5) clipboard

Basic Design Screen Structure element N − 1 eg inscription —template/edit-field Structure element N + 2 eg definition —template/edit-field Structure element N + 3 —template/edit-field

Structure element N + 4 —template/edit-field

Level Information —N1 level dependent a) relevant Directives b) level information c) analysis scheme

General Information level dependent 1) Directives (all) 2) Kluwer ADW 3) Kluwer legal library 4) database gateways 5) clipboard

Figure 1: Illustration of the interconnection of the Pmodule and the BDS module

134 Wim Voermans Pmodule guides the user through a hypertext network of semantic hierarchical and referential links. To offer guidance, the hypertext network of the Pmodule is divided into different levels, corresponding with the different methodological steps of the design-step model derived from the Directives. The levels in their turn serve as a checklist, expressing important points of attention regarding methodological and substantial aspects and the structural design of a draft. Figure 1 shows how the Pmodule and the BDS module are interconnected. The Pmodule consists of various methodological and consecutive levels. These methodological levels are referentially linked with level information. The level information component consists of (access to) the relevant Directives, access to relevant secondary information (as referred to by the relevant Directives), and a graphic template scheme for user analysis of certain options. Level information changes according to the level which is active (ie the level in which the user is working). The methodological levels themselves consist of fields containing information (about what is to be done within a certain level) and knowledge-based templates. The level-template documents which mainly serve to insert (or draft) text, also support the identification of important sub-items, and the choice between options. Both on the basis of the choice of the user and automatic analysis of text input in the template, the system makes inferences regarding the arrangement of levels further down the network’s path (eg, in the Basic Design Screen). From the point of view of the user, the levels form an interactive word processor which methodological guidance and provides relevant (semantically interlinked) information, in the form of authoritative arguments. The user may progress randomly through the level-structured hypertext network. This fundamental openness of the system is necessary as the user legislator is always free—when drafting a legislative text without the use of the system—to deviate from the Directives themselves whenever there is a good reason.8 To accommodate reluctant users, there is even a possibility of shutting down the levels altogether. What remains is a word processor linked to information in a single defaultinformation level explaining the methodological approach of the Directives, and providing (links to) the relevant Directives and secondary information. To prevent becoming lost in the hypertext network, user guidance is provided by the levels themselves, together with easy backtracking procedures and a step tracer, which consist of a major and minor active compass which visibly records the path hitherto followed in the network. On top of this, the Pmodule is provided with a General Information component to offer non-hypertextual access to various internal or external databases. Users can retrieve text from these databases while working in the different levels. The text in the internal databases, however, is hypertextually linked. 8 See Directive 5 of the Directives for Regulations, issued by the Prime Minister on 26 November 1992, [1992] Staatscourant 230.

Rationality in Legisprudence by Employing Informatics? 135 3. The Basic Design Screen The Basic Design Screen Module (BDS) is developed and structured in a way very similar to the Pmodule. Like the Pmodule it consists of a level structure, linked with level information (see Figure 2). The levels contain templates mainly consisting of free-text fields, which allow system-supported insertions (eg of model clauses or examples). The templates within the levels of the BDS do not, however, express points of attention with regard to the preparation and structural design, but important phrasing, terminology and terminology-related (substantive) issues regarding the structural elements of a draft. The arrangement of the levels in the BDS is both based on knowledge (gained from the Directives) and knowledge-based inferences made by the Pmodule. The BDS itself can be regarded as one large knowledge-based template which is shaped and directed by the Pmodule. The BDS represents the preferred structure of a draft, modelled to the needs of the user. Like the Pmodule the BDS has a very open structure: the user may progress randomly, do away with the levels altogether and receive default information,

Figure 2: An example of the BDS module level information

136 Wim Voermans and delete or add certain levels. The user-guidance function is similar to that in the Pmodule. The BDS has, however, one distinctly different feature compared to the Pmodule: it possesses a conceptual dependency parser.

4. The CD Parser When a user has finished the drafting of a text (within a certain level of the BDS), he may be interested to know whether he has overlooked a relevant Directive. In other words, did he/she overlook an authoritative or high-ranking argument? To accommodate this interest, LEDA possesses a conceptual dependency parser (CDP). This CDP automatically analyses (parses) the user-inserted text in a BDS level and dynamically creates links to a particular concept in the database or the text of a Directive if the text analysis indicates the relevance. To be able to do this, the CDP not only detects key words and key-word combinations and matches them with patterns in the database (string matching), but also analyses concepts in text sentences (by using the linguistic conceptual dependency method) and matches them with concepts in the database (so-called automated conceptual information retrieval). The CDP functions as a sophisticated legislative spell-check. However, instead of finding misspelled words, marking them and offering alternative, correctly spelled words, LEDA’s CD only marks points of attention in a draft Bill and offers Directives and other information that can be of use to the user. This form of conceptual dependency parsing, combined with automated conceptual information retrieval, is very powerful because both the concepts in the level-related text and the concepts in the database can be quite accurately defined. In combination, these functions offer a dedicated and semi-intelligent legislative proofreader.

V . CONCLUSION

The information-oriented approach to the development of practical legimatic systems seems to work. LEDA is being used in the actual departmental drafting process in The Netherlands. It has been evaluated in 1997, and both the Ministry of Justice and the Ministry of Health, Housing and Environmental Affairs have indicated at the moment that they are interested in continuing to work with LEDA. A commercial version of LEDA was developed in 2000. It is now being distributed among several Dutch ministerial departments. The commercial version operates under MS Access and MS Word. The LEDA approach is also being adopted in similar projects in Australia (Enact),9 Italy (Lexedit)10 and Belgium (Solon). 9

See T Arnold-Moore, Information Systems for Legislation (Melbourne, 1998). See C P Biagoli, G Mercatali and G Sartor, Legimatica, informatica per legiferare (Ed Scientifice Italiane, Padua, 1995). 10

Rationality in Legisprudence by Employing Informatics? 137 Pre-structuring the draft process and offering knowledge-based access to relevant (authoritative) information systems like LEDA are first steps on the way to really intelligent drafting-support systems that will mimic legislative reasoning in its full complexity by using AI techniques. AI-based tools will, in the near future, dramatically improve the functionality and the quality of existing legimatic information systems like LEDA. The AI approach bears a lot of promise when it is combined with the results of the information-oriented approach. Legimatic AI tools, suitable for consistency checking and considering the deontological consequences of a draft, will not only be able to improve the quality of drafting-support systems, they can also initiate a new way of thinking about legislative quality and launch new approaches to legislative drafting.11 This combination of drafting support and purely AI-based legislative analysis and review systems is for the moment, however, blocked by the necessity of (user unfriendly) complex knowledge representation and formalisation of natural (draft) language to accommodate analysis and review systems. Systems, like the LEDA system, in their turn may affect the drafting process in numerous ways. First of all, through its functions, the system accelerates the pace of legislative drafting and may indirectly contribute to the improvement of the quality of legislative drafts by way of forcing attention on the requirements of the Directives for Regulations. Furthermore, systems like LEDA can contribute to the emancipation of legislative expertise for Members of Parliament or legislative laymen by making legislative drafting, hitherto the realm of professional legislative draftsmen within ministries, transparent and easily accessible. Legislative knowledge itself will benefit from this. The drafting knowledge in LEDA will more and more become a mutual point of reference for those concerned with legislation. Using LEDA as a platform for the how and what of legislative drafting, legislative actors will feed and extend the body of legislative knowledge by their inevitable discussions on legislative topics. New experiences and insights in legislative drafting can very easily be added to the ‘knowledge base’ of the system. Working with IT drafting systems like LEDA may in the long run even give cause to re-think and re-model the legislative process itself. Legislative processes until now have been very paper-oriented and sometimes cumbersome, due to the need for communication in writing between the legislative partners involved. The introduction of IT systems may well alter the paper-based rationale of the legislative process and replace it by a much faster digital process. In any case, the rationality of legislation may improve by using IT drafting systems, but whether this will be the case is, as before, totally dependent on the legislative operator at the other end of the keyboard.

11 R W Van Kralingen and W Voermans, ‘A “Down to Earth” Approach to IT Support for Legislative Drafting’ in C-A Morand (ed), Légistique formelle et matérielle (Presses University AixMarseille, Aix-en-Provence, 1999).

11

The Forum Model in Evaluation of Legislation1 HEINRICH B WINTER

I . INTRODUCTION

In the last few decades in The Netherlands, evaluation of legislation has become a popular and almost standard activity in the legislative process. In various governmental documents, the implementation of evaluation research was stimulated.2 In governmental guidelines for regulation, a model was presented for an evaluation article in a law.3 The growing amount of evaluation studies almost automatically led to the obvious question of utilisation of the research findings. A meta-evaluation showed that utilisation was relatively high.4 A follow-up was aimed at the explanation for this outcome.5 This explanation will be presented here as ‘the forum model of legislation’. First, I will make some remarks on the growing importance of evaluation research concerning legislation. Among other things, the need for evaluation research thrived on the discussion on the quality and quantity of state regulation in The Netherlands. In this chapter the mainstream of this discussion will be presented. After discussing the orientation of meta-evaluations, I will turn to the central thesis of the chapter: the explanation of the utilisation of evaluation research concerning legislation. In section VI I present the forum model of legislation. After concluding this chapter, I will point out some drawbacks and risks of evaluation research concerning legislation.

1 This chapter is a revised version of a paper that was presented at a symposium held in Turku, Finland, in August 1998, entitled ‘Rationality and Legislation’. An earlier version was published in [1999] Associations 237–50. I would like to thank Michiel Herweijer for his helpful responses to a draft of that paper. 2 For instance: ‘Zicht op wetgeving’, Kamerstukken II, 1990–1, 22 008, nos 1–2, 39–40. 3 Aanwijzingen voor de regelgeving, Ministerie van Justitie: aanwijzing 164 (Sdu, Den Haag, 1998). 4 H B Winter, M Scheltema, M Herweijer, Evaluatie van wetgeving (Kluwer, Deventer, 1990). 5 H B Winter, Evaluatie in het wetgevingsforum (Kluwer, Deventer, 1996).

140 Heinrich Winter

II . THE GROWING IMPORTANCE OF EVALUATION - RESEARCH

The first three law evaluations in The Netherlands were published in the early 1980s. The first evaluation concerned a law (1970) establishing ‘democratic’ administration in state-funded universities (Wet universitaire bestuurshervorming, Evaluation Report issued in 1980). The second evaluation concerned a law (from 1978) that established extended access rights to government information (Wet openbaarheid van bestuur, Report published in 1983). The third evaluation concerned a law (from 1979) regulating the process in which industrial companies obtain environmental permits for their factories (Wet algemene bepalingen milieuhygiëne, Report in 1983). These three laws had recently been established at the time of evaluation. The three laws aimed at substantial reform. We may call them ‘modifying’ laws, in contrast to ‘codifying’ laws which only introduce legal rules that already existed before enactment in the form of customs, administrative practices or jurisprudence. The three evaluations were commissioned by Parliament and were supervised by committees of independent scholars. The research was carried out by both sociologists and jurists. The jurists studied the legal text, its specification in guidelines and the ensuing jurisprudence. The sociologists studied the law’s application and its immediate effects. Since these early attempts, more than one hundred law evaluations have been carried out in The Netherlands. In the late 1980s I drew a sample of 50 law evaluations completed 1983–1987.6 All sampled evaluation studies were commissioned by ministries and were predominantly carried out by independent researchers (university-based or research firms). Very recently I have tried to determine the amount of evaluation studies of legislation that have been undertaken during the last five years.7 Evidently, the amount is much higher now than ten years ago: the five-year period from 1993–1997 shows a strong increase in evaluation studies. The amount of studies can be estimated at some one hundred and fifty. Several explanations can be given for this development. First, evaluation of legislation is simply following the development which is apparent to policy evaluation in general. The last decades show an increase in studies on the effects of policies. In most cases, financial motives were the reason these studies were undertaken. Besides, there is growing awareness that government activities do not always reach their targets. Budgets ran out and the necessity for budgetary reforms led to a need for evaluation studies, but more in general, the growing awareness that a government which has to look carefully at its spending has to know exactly what the results are of its activities. This is not a new phenomenon. ‘More bang for a buck’, did become a more important 6

See n 4 above. N Struiksma and H B Winter, Inventarisatie wetsevaluaties, ministerie van Justitie (Pro Facto, Groningen, 1998). 7

The Forum Model in Evaluation of Legislation 141 parole during the last decades, hand in hand with increasing government interventions. The need for a strict budgetary policy only stressed the importance of effective government activities. In 1991 the Dutch General Accounting Office (Algemene Rekenkamer) completed a survey on which it reported that the fourteen ministries together initiate more than three hundred evaluation studies each year; the yearly state budget for evaluation research totals more than f85m (that is, about euro 38m, or US$42m).8 Most evaluation studies are concerned with issues of policy, budget or organisation. Law evaluations form only a small segment of the total evaluation effort, but this segment is growing, maybe even at a higher pace than evaluation in general. Secondly, the growing importance of evaluation of legislation can be contributed to a changing attitude in the legal departments of the ministries. In circles of lawyers and legal departments of the ministries, for a long period of time a simple, legalistic approach dominated. Following this approach, legislation was the most authoritative way of expression of a government authority. When a Bill, after a careful process of preparation, was enacted in Parliament, the proposition was that the implementation would be conducted fairly automatically. The vanishing of this firm belief in the authority of legislation took a long time. From several studies, evidence grew that faithful implementation of legislation is not automatic. This led to a growing amount of implementation research and the increasing quantity of evaluation research.9 Thirdly, during the last decade, skills in performing evaluation research have grown rapidly. This also gave way to a growing evaluation practice: evaluation researchers became more experienced with several methods for the studies, with the sometimes complex relationships with commissioners and with policy implications following the research. Commissioners became more skilled in the selection of adequate researchers, with the implementation of research findings and with the infrastructure needed to accompany evaluation research. Evaluation research became ‘business as usual’, which also meant less universities and more private evaluation firms.

III . LEGISLATIVE POLICY : FROM QUANTITY TO QUALITY

The increase in the number of evaluation studies carried out in The Netherlands during the last few decades has been accompanied by a development in the legislative policies which I would like to discuss, because it relates to the central thesis I will turn to later. Since the beginning of the 1980s, in The Netherlands, 8

Algemene Rekenkamer, Verslag 1990, Kamerstukken II, 1990–1, 22 032. See, for instance, a book with an ominous subtitle: J L Pressman and A Wildavsky, Implementation: Why Great Expectations in Washington are Dashed in Oklahoma (University of California Press, Berkeley, 1973). 9

142 Heinrich Winter quantity and quality of state legislation has been under discussion. The Netherlands, of course, is no exception in this respect. All over the western world a similar development took place under different labels, like deregulation, Thatcherism, or Reaganomics. At first in The Netherlands, this critical approach to state regulation aimed at reducing the number of laws and the complexity of these laws. At the end of the 1980s, the conviction gained ground that these goals had not been reached. During the years following, at the beginning of the 1990s, legislative policies changed ambitions. From a quantitative approach, policies changed to a more qualitative orientation. The Ministry of Justice developed a policy for the quality of legislation. In an influential government document called ‘Sight on Legislation’ (‘Zicht op wetgeving’, 1990), a set of quality standards was formulated that is used to both improve and determine the quality of a law. Quality is defined in terms of the well-known criteria of effectiveness and efficiency, but also in terms of feasibility and enforceability, and in terms of textual quality like the accessibility, the clarity and simplicity of the legal text. A couple of years ago, the policy aimed at improving the quality of legislation was structured in a more systematic approach in an operation under the complex title ‘Functioning of Markets, Deregulation and Quality of Legislation’. The labelling of the operation makes it clear that an economic approach is predominant. At the beginning of the first Cabinet under Prime Minister Wim Kok, a selection of laws was made that was subjected to a close examination in order to improve their. The question is relevant if evaluation of legislation can make a contribution to such an improvement of legislative quality. My thesis is that it can, under specific circumstances. To support this thesis, I made a study of several evaluation studies that were carried out in The Netherlands in the last ten years10: a meta-evaluation. Before turning to the central thesis of this chapter, I will elaborate on the nature of meta-evaluations. As shown in the next section, meta-evaluations often concentrate on the methodology, praxeology and utilisation of the evaluations which in their turn were subjected to an evaluation. In my opinion, the most obvious question, to what extent the results of the evaluation study led to improved policies or legislation, is almost never asked.

IV . THE ORIENTATION OF META - EVALUATIONS

Evaluation of evaluations, meta-evaluations, have basically been oriented to two questions. The first question that used to occupy the scientific forum is whether the evaluation was conducted following the rules of applied scientific research, specifically methodology and praxeology. The evaluation of the research methodology is concerned with determining if collection and analysis 10

See n 5 above.

The Forum Model in Evaluation of Legislation 143 of data meet the well-known standards of validity, reliability and generalisation. A test of the praxeology involves questions concerning the relevance and usability of the research information for the policymaker. During the 1970s the initial distrust of the scientific forum towards policyoriented evaluation studies diminished, and the focus of attention shifted to another question: to what degree are the results of the evaluation studies used for changes in the object of the study, the policy, or the law? There is a massive amount of literature on research utilisation. Conclusions of these studies, however, vary wildly. Some researchers, like Weiss and Van de Vall, conclude that utilisation is simply insufficient,11 while others reach contrary conclusions.12 Part of the explanation for these different conclusions lies in the various definitions of research utilisation that are applied. The use of research findings will be higher, if a broader utilisation concept is applied. Weiss mentions seven utilisation models, ranging from direct, visible use to conceptual use of research findings.13 Secondly, the object of utilisation studies varies as well. It is plausible that the use of commissioned research is more widespread than the use of academic, non-commissioned evaluation research. Thirdly, in relation to this, it is plausible that evaluation of legislation offers better chances of utilisation because the legislative process is strongly structured in an institutional and procedural sense. I will return to this aspect later, in section VI. Meta-evaluations are pointed at questions of utilisation, methodology and praxeology. In this respect it is striking that the focus of attention of these metaquestions never lies at the actual heart of the matter: this being the contribution of evaluation studies—in the long run—to the quality of the policy or the law that has been studied. To this empirical question the central thesis of this chapter is addressed: to which degree and in what way can evaluation of legislation lead to a change in the quality of this legislation?

V . CENTRAL THESIS : SYSTEMATIC EVALUATION CAN LEAD TO IMPROVED QUALITY OF LAWS

In earlier research in the field of evaluation of legislation,14 I focused on the utilisation question. To which degree were evaluation studies actually used? My findings were somewhat surprising because, contrary to what I expected on the basis of other research—of which I have referred to Weiss and Van de Vall— I found that the majority of the evaluation studies undertaken were actually 11 C H Weiss (ed), Using Social Research in Public Policy Making (Lexington, 1977); M Van de Vall, Sociaal beleidsonderzoek (Samsom, Alphen aan den Rijn, 1980). 12 G van den Berg, Effectief evalueren (Publisher?, Lisse, 1987); see also, H-P Mulder, G Walraven et al, ‘Gebruik van beleidsevalatie-onderzoek bij de rijksoverheid’, [1991] Beleidswetenschap 203–27. 13 C H Weiss, ‘The Many Meanings of Research-Utilization’, [1979] Public Administration Review 426–31. 14 See n 4 above.

144 Heinrich Winter used in the legislative process. In a large number of the cases I analysed, the law itself was changed on the basis of the research findings. In most cases, changes were made in the implementation-process and in the organisation. Subsequently, I wanted to go further and answer the question just mentioned: in what respect did utilisation of the research findings also mean an improvement of the quality of the evaluated law? To trigger this question and to measure the improvement of the quality of a law, one has to know what is meant by ‘quality’, therefore, a set of standards is needed. For these standards, I turned to the government document ‘Sight on Legislation’ (‘Zicht op wetgeving’, 1990), referred to earlier. The relevant criteria are concerned with practical and textual aspects of the law. Practical aspects include the presumed ex ante effectiveness of a law, the expected efficiency, the prospective feasibility and the enforceability of the law. Textual aspects are the accessibility, the clarity and simplicity of the legal text. To determine the change in quality of a law on the basis of an evaluation study, an intensive study has to be made. The research I undertook was a case-study research, limited to three cases that were selected from the known evaluation studies on legislation. In the selection of the evaluation studies, I took into account a distinction I made between systematic evaluation based on empirical research on the one hand and subjective evaluations, based on ‘quick-and-dirty’ research and on informed insiders on the other hand. In the cases, utilisation of the research findings was a predominant condition, because the central question was focused on the improvement of the quality of the evaluated law. In the two cases (the Freedom of Information Act and the Noise Control Act) in which the quality of the evaluation undertaken was high, there seemed to be an obvious improvement of the quality of the law. On the other hand, in the case where the quality of the evaluation was questionable (the modification of the Housing Act), the quality of the law diminished. I like to underline that these findings are interesting, but not necessarily convincing. The amount of cases carefully studied was limited, so generalisation of the findings is hazardous. Formulating a theoretical framework in which these findings can be placed and understood, can therefore contribute to a more tenable conclusion.15 In the following section I will present this theoretical framework, which I have labelled the forum model of legislation. Beforehand it is necessary to stress that the forum model is not a theory familiar to the concept of legisprudence.16 Rather, it is about the legislative process, about the way in which Parliament discusses legislative proposals, establishes new legislation and changes existing laws. 15 Case-study research, according to Yin, is seen as an explorational research strategy, suited for the development of theories. R K Yin, Case Study Research (Sage, Beverly Hills/New Delhi, 1984). 16 The concept of developing a theoretical framework for legislation was labelled ‘legisprudence’. As such, the concept was discussed amply and—in my opinion—meaningfully, during the symposium on ‘Rationality and Legislation’ in Turku, August 1998. See also: Luc J Wintgens, ‘Legisprudence: Elements for a New Theory of Legislation’, [1999] Associations 185–209 and B Dorbeck-Jung, ‘Realistic Legisprudence: A Multidisciplinary Approach to the Creation and Evaluation of Legislation’, ibid, 211–35.

The Forum Model in Evaluation of Legislation 145

VI . THE FORUM MODEL

The legislative process can be conceptualised in different ways. To start with, the conceptual framework of two classical interaction models are described: the barrier model and the garbage can. The barrier model, originated by Bachrach and Baratz, indicates power as the predominant factor in determining the outcome of a policy process.17 This model divides a policy process into sections between which barriers can prevent an issue from attaining agenda status. To start a policy process, a societal need should be translatedto in a demand to the political system. This demand should reach the political agenda, whereupon a decision should be taken, and finally, implementation should commence. Predominant beliefs, values and coalitions can hinder an issue in overcoming the barriers. A second interaction model focuses on the unpredictable aspects of policymaking processes. In an application of the Cohen, March and Olson garbage can model of organised anarchies,18 Kingdon depicts the policy-making process as an unpredictable coupling of streams.19 The policy stream, the political stream and the problem stream converge whenever a policy window opens up. These windows can open as a consequence of political developments or whenever a serious crisis arises. The central premise of this model concerns the predominant role of chance in policy-making processes. Both these models turned out to be unsuitable for the theoretical conception of the legislative process that was needed. The purpose of my research was to understand the relationship between evaluation of legislation on the one hand and improvement (or worsening) of the quality of legislation on the other. The legislator receives information on the functioning of law in practice. On the basis of this information, he may contemplate a revision of the law. The legislator wants to know if he reached his goals. He wants to predict and evaluate the effects of his work. On the basis of the information from the evaluation, the legislator may contemplate a revision of the law. The resulting decision cannot be explained as simply the outcome of a power game or the consequence of mere chance. In the legislative process arguments seem to carry a great deal of weight. Besides, for the process of changing a law in the wake of an evaluation, outside initiatives, as the barrier model stipulates, are hardly relevant. There is also no unpredictability there: the timing of the evaluation is known, the problems are described and solutions are formulated. Thus, the forum model is not an input–output model, nor a chaos model, but an argumentation and communication model. On the basis of the forum model 17

P Bachrach and M S Baratz, Power and Poverty (Oxford University Press, New York, 1970). M D Cohen, J G March and J P Olsen, ‘A Garbage Can Model of Organizational Choice’, [1972] Administrative Science Quarterly 1–25. 19 J W Kingdon, Agendas, Alternatives, and Public Policies (HarperCollins, Boston, 1984). 18

146 Heinrich Winter lies the assumption that through the evaluation study, information became available, which creates possibilities for a public debate. Ideally, a law is the result of such a debate, based on arguments related to the situation as determined in the evaluation. On these premises, the forum model of the legislative process is established. The forum model of the legislative process assumes that the quality of the laws produced by the legislative process increases when more and better verifiable information is available on the circumstances under which the law has to function. According to this assumption, the quality of the evaluation must be an important factor influencing the quality of legislation. The legislative process creates laws following strictly regulated procedures, by participants that are known beforehand and through a discussion which takes several rounds. This legislative process serves at least two purposes. First, the legislative process is a democratic, open process. The rules of the process are set up with this intention. For example, the rules according to which selection of participants takes place, are set up so as to ensure that legislation will be the embodiment of commonly shared values. Furthermore, democratic legislative procedure is intended to grant legitimacy for the laws of the land. Secondly, the legislative process aims at establishing good decisions. The rules of the legislative process are designed to provide for careful deliberations. Thus, in The Netherlands the State Council submits a formal recommendation on all Bills pending, a large part of the deliberations is in written form and the debate takes place in several rounds. Through the intermediate substitution of parliamentary participants, a certain degree of reflection is acquired. The open civic nature of the legislative process is an essential prior condition. It enhances outside influence in the process—also through an active public opinion and free press—and promotes the quality of the arguments that provide the foundation for it. Do the results of my case studies corroborate the outlines of this forum model? The studied cases indeed underline the formulated assumptions of the forum model. The proposals which were formulated as a reaction of the systematic evaluations led to a serious debate, which clearly contributed to the determined increase of quality of the two laws. The parliamentary debate on the proposal which was founded on the subjective evaluation of the Housing Act however, was not very substantial, but displayed strong strategic overtones. As a consequence, the changes in the Housing Act did not improve the quality of the law at all. Evaluation, the study showed, introduces images of reality in the legislative process that support the debate on legislation. Following Habermas’s procedural, deliberative model of democracy, discussions on legislation can be characterised as a combination of communicative argumentations and strategic argumentations.20 Communicative argumentation can be defined as an open 20

J Habermas, Theorie des kommunikativen Handelns (Suhrkamp, Frankfort on Main, 1981).

The Forum Model in Evaluation of Legislation 147 exchange of information and arguments, while strategic argumentation gives way to power games and strategy and is the result of a much less open process. High quality evaluation limits the opportunities for strategic argumentations. Evaluation of legislation should be aimed at strengthening the position of communicative argumentations in a debate on the preparation of modification of a law. Of course, one should remain realistic about the possibilities of evaluation. Many participants with often very different interests take part in discussions on a Bill (as well as in all policy development). Results of evaluation research cannot replace the considerations made by these participants in view of their respective interests. But, on the other hand, it is not necessary to be very modest about the role evaluation results do play in the legislative process. The results of an evaluation can—and in fact will—inspire the deliberations of the participants in the legislative process. The results of an evaluation at least set boundaries for the possible positions which the participants in a debate can choose. Evaluations supply information on the effects of a certain law and on the state of affairs in a certain field of government action. The possible options the legislator has at his disposal are influenced—and often diminished—by this information. Some people seem to suggest that the results of evaluations, and research in general, are not, or not very, relevant to the political debate. Research findings can be found on the street; qualified civil servants, administrators and politicians know which way to go without research. Exactly this view was found to be the opinion of the lawyers of the ministries of the Dutch central government, according to a study conducted by the Dutch General Accounting Office.21 The data of my cases contradict these views. In this respect it is remarkable that the results of the evaluation of both the Freedom of Information Act and the Noise Control Act diverge strongly from the expectations many had formulated on the functioning of these laws, sometimes before, sometimes after the coming into force of the evaluated law. The evaluation of the Freedom of Information Act demonstrates that this law did not have the destructive effect on the functioning of the civil service and government authorities that was feared beforehand. Because of this fear, Parliament accepted the law reluctantly and only after a sharp debate. The evaluation proved that the law did not lead to an abundant amount of requests for information. For the Noise Control Act a similar discrepancy between the original expectations and the outcome of the evaluation was noticed. It is remarkable that in the case of the Noise Control Act, even after the coming into force of this law, complaints were raised on a large scale and at the highest administrative levels. The evaluation shows that things were actually not as bad as they initially looked. After some time, bottlenecks in the implementation were solved—also thanks to the support the ministry gave to the local governments. 21 Algemene Rekenkamer, ‘Wetgeving: organisatie, proces en produkt’, Kamerstukken II, 1993–4, 23 710, nos 1–2.

148 Heinrich Winter The case of the Housing Act also shows that data concerning the implementation of the law can lead to new views. Research findings on the functioning of the Housing Act 1991 indicate that assumptions on which the change of the law was based in 1991 were partly incorrect. In fact, some of the conclusions of the evaluation that was undertaken as part of the deregulation operation and on which the reform was grounded proved to be wrong.

VII . CONCLUSION

In sum, empirical, systematic evaluations can have a sobering and demystifying effect. They bring realism. Evaluation adds value to the legislative process because it tests the assumptions on which a law is based. It supplies the legislator with information on the functioning of a law and on the circumstances that influence the law’s performance. This information enlightens the legislative process and contributes to the quality of the debate on legislation. That information appears to be valuable, since utilisation in the legislative process is relatively high. One of the consequences of the increased complexity and changeable nature of society is that individual experiences cannot serve as a solid ground for adjusting legislation—if ever that was the case. Neither can random observations of departmental lawyers or implementation officials constitute a solid foundation upon which the legislator can build a case for change. The legislative process seems receptive to the reports of evaluations if these reports produce reliable information on the functioning of the law. This receptivity can be explained by modelling the legislative process as a forum event. Laws are created within the bounds of strictly regulated procedures, by participants that are known beforehand and through a discussion which takes several rounds. Among other things, the rules of the legislative process are set up to establish careful deliberations. In this respect it is relevant that the State Council advises on all Bills pending. Furthermore, a large part of the deliberations is in written form and the debate takes place in several rounds. Through the deliberations of different parliamentary participants (in The Netherlands, Eerste Kamer en Tweede Kamer, comparable to Senate and Congress in the US), a certain degree of reflection is acquired. Of course, the open civic nature of the legislative process is an essential prior condition. It enhances outside influence in the process—also through an active public opinion and free press—and promotes the quality of the arguments that provide the foundation for it. Evaluation researchers and policymakers, including the legislator, have in common that they must justify themselves in front of a forum. For the legislator this means decisions have to be taken—and eventually laid down in laws—which can be defended in public. The legislative process is receptive to information from an evaluation because serious argumentations can be founded on it. The better the quality of this information, the more seriously the proposed law can be tested. Proposals can

The Forum Model in Evaluation of Legislation 149 be built on a much more solid basis, which means a growing chance that the proposals will hold in a debate. To its origins, evaluation suits very well with the legislative process. The careful preparation of decisions on legislation is promoted by the supply of information on the functioning of the law. Best decisions are informed decisions. However, the preparation of laws will always be a mixture of communicative and strategic actions. In a democracy the role of (scientific) research is not predominant. Research findings are always subject to political judgement, but evaluators can certainly contribute to a carefully considered judgement. From this perspective, evaluation aims at delivering a positive influence to the debate on legislation. In this respect, the publicity of that debate forms a prior condition. The publicity of the debate makes it verifiable: the results of an evaluation cannot be ignored. The three cases under study gave different indications for this function of evaluation in the legislative process. In the discussions on the three proposed laws, decision-makers, politicians and officials repeatedly referred to the results of the evaluations. Proposals of ministers were compared with these results. With the Freedom of Information Act and the Noise Control Act, this happened very explicitly on those points on which the ministers disagreed with the advice of the Evaluation Committee. Parliament then asked for further justification of the proposal with the results of the evaluation taken into consideration. In this respect, it is striking that the changes to the Housing Act led to critical remarks on the soundness of the grounds on which the proposed new law was based. The debate on the new Housing Act had a strong ideological connotation. Because empirical data were not available, opportunities for strategic action became manifold. There was more room for wishful thinking due to the lack of objective hard facts. Strategic actions were also found in the two other cases, but the debates in these cases were clearly less prejudiced. Of course, one should be careful in formulating these rationalistic conclusions. Legislation does not always result from a process, characterised as an open, democratic debate between decision-makers that are forced to react to each others’ arguments, in which the quality of the arguments and the information is dominant. The debate on legislation is often influenced by political rationality. Indeed, there is no unconditional relation between the quality of the evaluation-results and the quality of law. On the other hand, however, evaluation results mark the boundaries within which the political debate can take place. The options chosen by the decisionmakers are influenced by the evaluation study and by the several control mechanisms which are related to the public debate which takes place in the legislative forum. This results in the thesis that empirically founded, systematic law evaluations help to strike a good balance between interest, power, values and ideology on the one hand and objective, factual information that is not contested, on the other. This balance favours the chance of improving the quality of the evaluated law.

150 Heinrich Winter

VIII . DRAWBACKS AND RISKS

Perhaps after the foregoing one should expect a warm plea for systematic evaluation of all legislation. This plea will not be delivered, however, because both the decision to evaluate a law and the use of the results of an evaluation have certain drawbacks. First, law evaluation as a political institution may lead to the neglect of ex ante evaluation and the lowering of the barrier to new legislation. The emphasis in the political debate is on the desirability of an evaluation in the long run, not on the serious testing of the proposed law at hand. Secondly, my research shows that the utilisation of evaluations is relatively high. Thus, evaluation may speed up the revision of recently enacted laws. The stability of the legislation can be threatened by this, and as is known, stability is a predominant condition of ‘the Rechtsstaat’. Thirdly, evaluations create new policy-making opportunities for the adversaries of the evaluated law. Agenda status is not obtained because of the results of the evaluation, but because of the simple fact that a momentum is created. When these effects occur, there is no reason to believe evaluation will contribute to the improvement of legislation. The use of evaluation findings sometimes poses a threat to the quality of the law. In general, laws are evaluated very shortly after the moment they were enacted.22 Most evaluations are carried out at a time when the law is still being introduced in the field. Formulating definite answers concerning the effects of the evaluated law is impossible most of the time. When evaluation reports pretend to give more than provisional information on the effects of the law, the evaluation findings are misleading and of no value in improving legislation. The second threat an evaluation can constitute for the quality of laws has to do with the pressure imposed on evaluators to produce usable recommendations. Politicians sometimes show more interest in simple, straightforward propositions for new legislation than in understanding the current process of law application. In their search for usable recommendations, evaluators may be tempted to violate the ‘condition of symmetry’ by formulating recommendations that cannot be founded on the research findings.

22

H B Winter, M Scheltema and M Herweijer, Evaluatie van wetgeving (n 4 above).