European University Institute Department of Law in Redistributive Taxation We Trust Some Elements for a Democratic Theory of Tax Law

The argument moves from the justification of the general obligation to obey the law to that of the general obligation to pay taxes (it is argued that with some exceptions, tax scholarship has played down the importance of this question) by means of describing the peculiarities of the latter in reducing the cognitive, motivational and organisational demands that plague general practical discourse. This leads to the conclusion that we can consider the tax system as an autonomous subsystem within the larger legal one, and that the justification of the general obligation to pay taxes depends on three pillars. Those are the participation of legal subjects in the making of tax norms, the substantive correctness of such norms and the guarantees concerning their application. After a section devoted to a case study on the development of general principles of tax law by the Spanish Constitutional Court (that gives a sense of the specific problems at stake), I move to the enumeration of a set of liberal principles of tax law that would render legitimate a tax system if it shaped by them and correspond to the basic and aforementioned three sources of legitimacy. The essay is articulated around three basic methodological choices, summarily explained and defended in the second section: ethical constructivism (following Rawls and Nino, and also Habermas), deliberative democracy (basically following Nino, Cohen and Estlund) and post-positivism (on the basis of Alexy’s work and that of MacCormick and once again Nino). 1. The Tax Crisis and the Need for a Democratic Theory of Tax Law §1. If there was a place and a time at which the obligation to pay taxes was generally believed to be a legitimate one, that would be the Western World at the middle of this century. That was the golden era of easy finance (cf. Brownlee, 1996). The major crisis brought about by the Great Depression and the Second World War had led to the emergence of activist states everywhere (se Milward, 1984). Private freedoms were seen as dependent on the action of the state, which was now becoming a major 1 This paper is a partial outline of my Ph.D. dissertation. Special thanks to Robert Alexy, Francisco Laporta and Álvaro Rodríguez Bereijo, external members of the examining jury, who made challenging suggestions partially reflected in this essay. Many thanks to Jacques Ziller, who has supported the present publication. Once again, special thanks to Massimo La Torre, for the paper and for the long walk in Bologna, a freezing afternoon of December. The present publication would not have been possible without the always superb Marlies Becker. Many thanks to Elena and Sergio, for reading the text. 2 This is understood as referring to the countries which were to become founding members to the OECD, the Organisation for Economic Cooperation and Development, when it was constituted in 1961. Agustín José Menéndez 4 economic actor. It did not only provide basic public goods, but also implemented redistribution policies, aimed at mitigating economic hardship, so that all could regard society as a cooperative venture. Moreover, it was assigned a major role in the active management of the economy, as the guardian of basic macro-economic goals like full employment and low inflation. All such activity was funded by a revamped tax system. Not only the size and tasks assigned to the tax system had grown, but the tax mix had been radically changed. The system had moved from a real to a personal basis. If real taxes like excises or sales taxes constituted the main component of the tax system in the old days of laissez-faire, they were to be replaced by personal taxes of the kind of the income, the corporate or the estate tax. That meant that the tax burden was to be graduated according to the personal circumstances of each taxpayer. Ability to pay emerged as key factor in determining how much each individual should contribute to the public treasury, in the same way that it became relevant in determining how much she should receive from it in the form of public expenditures, (in-kind benefits or money transfers). This major transformation was facilitated by the long cycle of economic expansion that followed the Second World War, and by the feeling of solidarity derived from mobilisation for war. However, it was not long before the tax storm began to gather. Available quality data (see Schneider, 1997) on the size of the shadow economy in OECD states indicate that the latter has been on the rise since the late fifties. That means that, from such a moment, an increasing proportion of economic activities that qualify as tax bases escape control of tax authorities and do not contribute to the pool of common resources in spite of being called to do so. The trend has strengthened with the passing of time. It is important to notice that tax evasion is essentially a secretive action, and as such, it is more destabilising than forms of protest that put forward an alternative political conception of taxation (e.g. civil disobedience) (Arendt, 1972). In such a context, one can no longer assume that all citizens accept that they have a compelling obligation to pay taxes. §2. Why is this so? Two main hypothesis have been offered as an explanation of the phenomenon, namely, the theories of the fiscal crisis In redistributive taxation we trust 5 and of the increased opportunities for evading taxes derived from globalisation. Let’s analyse them and explain why they are unsatisfactory. First, the advocates of the theory of the fiscal crisis argue that we are undergoing a crisis of reproduction of the tax system (cf. O’Connor, 1973). For them, modern capitalist economy tries to achieve two conflicting goals at the same time, namely ensuring that profits are kept high for capitalists and purchasing legitimacy for public institutions. This leads to structural public deficits, or what is the same, it pushes public expenditure beyond the revenue collected through the tax system. The issue of public debt is only a temporary solution. At times of economic crisis, debt grows exponentially and the crisis unfolds. Without denying the descriptive power of this theory in some respects, the fact is that it does not correspond exactly to what we are going through. Twenty five years after the canonical statement of the fiscal crisis thesis by O’Connor, it is a fact that the tax system keeps on being able to transfer a major share of private economic resources to the system of public finance. Shrinking tax bases are still to be translated into shrinking revenue. This is so because the tax burden placed on the official economy has augmented in order to compensate the loss of revenue derived from the growing shadow economy. Thus, the crisis is not a matter of falling tax revenues but of uncontrolled and unprincipled transformation of the structure of the tax system. Second, we find authors that explain the tax crisis in terms of increasing opportunities to evade taxes and get away with it brought about by economic globalisation (see, among others, Tans and Parthasarathi, 1993). The argument goes that the reduction of the pulls and levers in the hands of the state to control the economy (among which, the key power to control the flow of capitals) would have increased exponentially the chances to evade taxes. When we couple such factual record with a Hobbesian understanding of public reason (which assumes that individuals are self-interested maximisers, cf. §10), we have a bold theory. Evasion, reformulated as non-compliance, is a matter of opportunities to do so in a painless way. The more chances individuals have to evade, the more they will do it. If technology and the complex set of developments labelled as economic globalisation have increased the aforementioned opportunities, it is evident why the shadow economy is on the rise. Agustín José Menéndez 6 Without denying that some of these arguments are far from being implausible, it is possible to contest that they provide an adequate explanation of the tax crisis. It suffices to point that it predates the emergence of large-scale opportunities for tax evasion provided by globalisation. As already indicated, the black economy started to grow significantly back in the late fifties, well before most of the opportunities to evade provided by the complex phenomenon of globalisation were available to taxpayers. Moreover, it is just not possible to establish any link or proportion between factors relevant in assessing the existence of opportunities to evade taxes (like the degree of openness of the economy) and the size of the black economy. So not even the empirical data are favourable to the advocates of the globalisation thesis when they are closely inspected. A democratic theory of law offers a different interpretation of these facts. It departs from the assumption that individuals are not only able to act on the basis of prudential reasons, but also capable of being motivated by normative ones, or what is the same, by their conception of what is right. Under this different light, the tax crisis can be explained as a legitimacy crisis. Taxpayers evade taxes because they no longer find sufficient reasons to ground their tax obligations. In turn, this can be considered as a direct consequence of the divorce between the constitutional design of the tax system (which, in general terms, has not been altered) and the reality of its implementation (dramatically transformed by the tax crisis), mediated by an increasingly incoherent set of tax norms. §3. Although a complete explanation of the tax crisis requires spelling out its social, economic and legal causes, it is argued that a key factor is the failure of legal dogmatics to reconstruct purposive tax law, or what is the same, to theorise meaningfully the tax law of the activist (and democratic) state. One could say that legal dogmatics is, in a sense, a sort of lens that allows us to make sense of the congeries of legal norms that surround us, so that we can understand their rationales and use them. If a legal theory was just a pair of glasses and we could see without them, what we call law would look as a chaotic set of norms and principles. Quite probably, In redistributive taxation we trust 7 we will not be able to make heads or tails of such amorphous set. On such a basis, we can say that what legal operators do (especially legal theorists and other dogmatics) is to elaborate on the cognitive instruments with which to make sense out of such morass of heterogeneous pieces, thus rendering them ready-made components for our practical reasoning, and more specifically, for our legal reasoning. The main tool in the hands of legal theorists is the concept of law and the companion idea of the legal system (cf. Bengoetxea, 1994). However, neither norms nor legal concepts are elaborated out of love or deference towards the cognitive tools of the theorists, but they respond to social interests and conflicts. Law is a purposive order (cf. MacCormick, 1997, esp. at 1054). In that respect, it can be seen both as a mirror of social transformation and as an active agent contributing to it. The cognitive tools with which lawyers and theorists construct and reconstruct legal norms are clearly influenced by the principles and values that underpin social arrangements. The idea of the paradigm of law, as the “image of society inscribed in the legal system” (Habermas, 1996b, 771) is quite handy to give an account of such interrelation. Paradigms can be described as sets of principles that program law, and that allow, among other things, to make sense of legal norms and to orient legal interpretation, both in easy and hard cases. Even if they are quite flexible, there are times at which the speed at which law changes is higher than the capacity of absorption of the prevailing paradigm. In such circumstances, we are witness to the demise of a certain way of understanding law and the search for a new one. Although it is far from clear when the formal paradigm of law became a rather pool tool with which to argue in legal terms about taxes, it is clear that at some point between the end of the last century and the 3 NB the difficulty involved in this mental exercise, as even in this example we have no alternative but to define the field and describe it with terms already elaborated by legal dogmatics. 4 In legal systems that have quite centralised mechanisms for surveying and monitoring the interpretation of and adjudication on legal norms (like Constitutional Courts), the paradigmatic shift gets reflected in the repeated giving of judgments that break new ground and cannot be accommodated in the old one. The concatenation of several of such judgments indicates the demise of the old paradigm, but they tend to be less determinant concerning the articulation of the new one. 5 Cf. my paper on “Three Paradigms of Tax Law”, unpublished, on file with the author. Agustín José Menéndez 8 Second World War it became plainly unrealistic to make legal tax arguments based on the formal paradigm of tax law, which conceptualised public expenditure as something akin to the expenses shared by any community of private owners. Similarly, it was clear that the question of who should pay taxes and how much should each individual pay could no longer be kept isolated from the general question concerning the legitimacy of private property, the societal obligation to take care of the victims of brute bad luck or those who could not make ends meet needs within existing socio-economic arrangements. In a few years, most Western countries had approved some types of progressive income, corporate and estate taxes. Once the new tax mix was part and parcel of the law, the strategy to consider each of its components as an anomaly was bound to fail. This opened up the opportunity and gave rise to the need of elaborating a new conception of tax law which could take into account the social principles underlying the enactment of new tax norms. But to do so, tax dogmatics would have needed to put into question its own methodology, and the foundations of its academic autonomy. Instead, most scholars sought refuge in formalism (cf. Livingston, 1998; the question is dealt in general termsthat is, without focusing specifically on tax lawin Ackerman, 1984a). This explains why what was an opportunity ended up becoming a danger. The fixation with technical tax questions gave rise to a normative vacuum which has contributed to the erosion of the limited basis of legitimacy of the tax system. In some cases, instead of exposing the dogmatic and ideological foundations of the formal paradigm of tax law, the move resulted in the infusion of new strength on the authoritarian approach to taxation. The danger was not diminished by the general prominence of prescriptivist conceptions of law and of positivism as a general philosophical doctrine. It is for these reasons that the tax crisis is to be overcome, to a certain extent and within certain limits, by means of renovating the effort 6 Several metaphors of taxation as a proxy of price have been put forward. Hume compared taxes with the costs of a ditch spread among the owners of the neighbouring meadows, Smith with the costs of paying for a lighthouse, and some Italian treatises on public finance compared them with the losses in certain kinds of shipping contracts (lex Rhodia de jactu). 7 Cf. “Three Paradigms of Taxation”, fn 5. In redistributive taxation we trust 9 at elaborating a conceptual and normative map of tax law in the activist state. §4. If the diagnosis is right, what we need is a democratic theory of tax law that allows us to reconnect tax norms with general principles that guarantee the legitimacy of the system as a whole. The present essay aims at offering some of the elements necessary in order to build such a theory. To put it in a different way, thinking tax law democratically requires revising the conception of taxation (hammering in the idea that tax law is law), the analysis of the different tax relationships (which includes not only the vertical institutionalised relations deriving from specific tax norms, but also horizontal ones which relate all members of a political community and justify public expenditure), and the principles of taxation which program concrete tax norms (which should be related to a complex conception of legitimacy). The whole reflection is to be seen as a reworking of the question of legitimacy of tax law. §5. This makes advisable to revisit some of the basic issues that have occupied tax law dogmatics by the hand of a critical citizen always ready to ask questions about the justification of her obligations, the paramount of which is, for our present concern, to pay certain amounts of money. That constitutes a sort of reversal of the basic intuition of legal realism. Instead of attempting to describe law as it looks to the eyes of the bad man who cares of nothing but prudential reasons, we will try to keep company with the reasonable citizen. This requires that we face the basic problem of the justification of the tax system. Answering this question requires us to move from the single to the plural. The argument goes that the main source of legitimacy is to be found in the fact that tax norms have been produced in a democratic way (in fashionable terms, by we the people), or to put it in slightly different words, that positive tax law is democratic and that makes it legitimate. This requires explaining in what sense and for which reasons only democracy fills the justificatory gap (or at the very least minimises it), due to the conflict between our will to autonomy and the need of partially heteronomously produced action-norms (i.e. law) in order to achieve cherished but complex social goals, like providing financing for public expenditure, which is what taxation is mainly about. But this is coupled with standards of substantive correctness of the Agustín José Menéndez 10 content of tax norms and principles guaranteeing the adequate application of tax in concrete circumstances. That leads to a complex theory of legitimacy of tax law. Far from being an indictment of technical analysis of tax law per se, the democratic theory of tax law aspires to complement it. It sees itself as building upon the major efforts undertaken by tax law dogmatics. Any summary account of the evolution of tax law makes clear to us the difficulties derived from the backward state of the discipline before concepts such as tax event or even tax rate became common currency. Instead of rejecting classic tax dogmatics, it is interested in bringing to the surface the concept of taxation and of the tax relationship that are hidden in technical constructions. It aims not at demolition, but a reconstruction that renders visible the ultimate political character of tax law. 2. Basic choices and an articulated defence against objections §6. Thinking tax law democratically is a project based on three methodological choices. Namely, ethical constructivism as a metaethical theory, deliberative democracy as the normative model of common actionnorms written in the language of law, and post-positivism as legal theory. Ethical constructivism is associated to two basic premises. First, that it is possible to argue in a rational manner about practical questions, or more precisely, to provide an inter-subjective justification of the assertions which we make on such matters. Second, that the relevant criteria for this purpose are provided by the practice of moral deliberation itself, that is, by the pragmatic assumptions which we make every time that we enter into practical discourses. Habermas provides us with a canonical statement of the yardstick against which moral judgments on action-norms are made: “Just those action norms are valid to which all possible affected persons could agree as participants in rational discourses” (Habermas, 1996a. 107). Inter-subjective standards of practical argumentation prove an essential tool to dispel the claims of the different kinds of moral scepticism. But once we take into account basic facts of human life, we In redistributive taxation we trust 11 realise that our practical reasoning cannot proceed only on the basis of moral reasoning. We need to find procedures not only to deliberate about practical problems, but also to decide them in an authoritative and conclusive manner. That explains the emergence of law as the grammar in which common action-norms are drafted. However, the mere fact of positing such norms gives rise to a burden of justification, that requires transforming standards of moral correctness into criteria of legal correctness. That results in the ideal of deliberative democracy. It basically claims that “[O]nly those statutes may claim legitimacy that can meet with the assent of all citizens in a discursive process of legislation that in turn has been legally constituted”, (Habermas, 1996a, 110). Finally, post-positivism can be seen as the legal theory that comes naturally after the previous two choices. It can be described with the help of four basic premises. First, that law is a complement of morality. As just indicated, the latter cannot discharge on its own the tasks of conflictsolving and co-ordination of action in complex modern societies. There are two main reasons that call for morality being complemented by law. On the one hand, there are many instances in which it does not have enough resources in order to determine what is the right course. That is due to the fact that the principle of universalisability is the main and almost exclusive operational criteria of moral reasoning. On the other hand, it makes considerable epistemological, motivational and organisational demands upon individuals (Habermas 1996, 118). Second, law is characterised by its autonomy vis-á-vis morality and general practical reasoning. Legal norms provide reasons that pre-empt reference to moral or general practical reasons. That can be seen as a direct consequence of the functional requirements that motivated looking for a complement to morality (cf. Hart, 1982, 253; Raz, 1985, 299, 304; Postema, 1996, 82). Third, law is to be conceptualised as a special case of general practical reasoning. This indicates that the legitimacy of law is somehow borrowed from morality. However, that does not mean collapsing law into general 8 Cf. (Alexy, 1989, 16): “Legal discourse is a special case because legal argumentation occurs in the context of a number of limiting conditions. In particular, one Agustín José Menéndez 12 practical discourse, but only that the legal norm can be justified under the limiting conditions proper of the legal institutional framework. Fourth, law is mainly a matter of practical and not of theoretical reason. Legal facts are not merely brute facts, and legal norms might not be sufficient in themselves to determine what is the right legal answer (Nino 1991a, 248 and Habermas 1996a, 107). §7. These choices are far from being exempt from challenge. A general theory of democratic tax law that presupposes the possibility of making heads or tails in practical matters (ethical constructivism), a criterion of legitimacy of common action-norms (deliberative democracy) and the autonomy of law as a social system, but whose legitimacy is borrowed from morality (post-positivism) is deemed to be controversial. Assaults would come from the many different variants of anti-normative thinking (moral emotivism, communitarianism, forms of Hobbesian practical reason, of which utilitarianism is the classic one) and from legal theories that defend a strong autonomy thesis of law vis-à-vis morality. Let’s briefly consider them and explain how they can be proven inferior alternatives to our choices. §8. Moral emotivism undermines any attempt at constructing a general theory of democratic tax law by means of challenging the possibility of making any meaningful normative judgment. The emotivist states that moral judgments are merely emotive utterances, disconnected from any objective conception of truth or correctness, and which, at most, can be linked to a notion of authenticity, or personal coherence (Foot, 1995 offers a clear outline of moral emotivism). There are five main grounds on which to reject the plausibility of moral emotivism. First, it is against what is implicit in the practice of moral reasoning. Namely, when we enter practical deliberation, we raise claims must include here its statute-bound character; its necessary regard to precedents; its involvement with doctrinal studies as developed through an institutionally organised profession of academic lawyers, as well asand this of course is not true for academic legal discourseits subjection to the requirements of procedural ordinances and regulations. The claim to correctness involved in the assertion of any legal statment is the claim that, subject to the constrains set by these limiting conditions, the assertion is rationally justifiable”. Cf. also (MacCormick 1992) and (Nino, 1995). In redistributive taxation we trust 13 to the correctness and justifiability of our statements. This can be further considered by exploring the pragmatic assumptions we make in such a context (cf. Nagel 1997a, 117). Second, it does not explain why we associate moral judgments with obligations, something that is peculiar of normative discourse when confronted with other linguistic utterances in which we express our feelings (Habermas, 1996d, 336-7). Third, it blurs the distinction between the justification and the explanation of a given norm (Nino, 1985, 126). Fourth, it seems to assume that the only criteria of correctness in practical matters must be based on the so-called correspondence theory of truth, or what is the same, in a correspondence between moral judgments and external reality. This does not take into account that we can cash objective for intersubjective acceptance, and truth for correctness without major loss (Rawls 1993, 110). Fifth and finally, it incurs in a self-contradiction, to the extent that any form of moral scepticism can be reinterpreted as a moral claim itself (that there are no objective moral claims), the assertion of which implies a pragmatic endorsement of the basic assumptions of practical deliberation (Dworkin 1996, 88; Nagel, 1997a, 128). §9. Communitarianism contests the general theory of democratic tax law on the basis that any practical argument about taxation must be based on traditionally accepted norms or values. As a metaethical standing, it can be seen as the dilution of critical into positive morality, or the claim that moral correctness must be defined by reference to local practice. For practical purposes, we can consider two different variants of the theory. According to the strong one, moral reasoning is permeated by local practice in a double sense. First, moral subjects are constituted as such through social interaction within a community. It is through education that we acquire the capacity to pass moral judgments, and it is also through socialisation that we learn substantive moral criteria that constitute the 9 The term communitarianism is a generic label applied to several authors who have addressed a number of variegated criticisms to liberal political theory. This makes the label a negative one, defined to a certain extent by reference to the target of criticism, namely the work of John Rawls and other liberals such as Ackerman, Dworkin, Nagel or Habermas. For our present purposes, the main representatives of communitarianism are Michael Sandel, Alisdair MacIntyre, Charles Taylor, David Miller and to a certain extent, Michael Walzer. Agustín José Menéndez 14 building blocks of our moral judgment (MacIntyre, 184, 10). Second, moral debate is necessarily framed by a thick conception of the good, so that it proceeds in an hermeneutic way. The possibility of conflict and disagreement is tamed by the tradition in which we have been socialised (Sandel, 1982; Sandel, 1996). This strong version can be challenged in the following way. First, the idea that we are socially equipped with standards of moral judgment is true, but it is not less true that we can use our moral judgment in order to put received criteria into question and to endorse new ones. The reflexivity of practical reason allows us to transcend socially transmitted moral criteria (Kymlincka, 1989, 254; MacCormick 1997a, 105; Nino 1989a, 177). Second, it does not explain how normative dialogue can take place between people with different conceptions of the good. It stands against plausible claims like the fluidity of cultural affiliations (Waldron, 1995). Thus, it is in serious difficulties in order to explain how we can change our convictions by means or borrowing arguments from other traditions, or how moral change is possible within the tradition itself. The weak version argues that the choice for practical reason (instead of competing ways of providing for conflict-solving and coordination within a given society) is based on traditional, not on normative arguments. Any answer to it needs to take into account that we cannot have resort to higher moral principles in order to explain the choice for practical reason understood as ethical constructivism does (this will imply falling prey to a sin of self-reference), but that, instead, we need to show that it can be better grounded than any other alternative (Nino 1989b, 8586). A possible argument could be the following. First, rational agreement and the conception of the person as reasonable are goods of such special kind that they should be seen as part of the conception of the right, and not as particular goods, equal to many others (Alexy 1994, 142). Second, we have to take into account that once we enter the practice of deliberation and we make an assertion, we presuppose the validity of assertion rules and we raise a claim to correctness concerning our assertion. This, in its turn, implies that we raise a claim to justifiability, which places upon us the duty of justifying what has been asserted, or else to provide reasons justifying the refusal of such reasons. At such point, we can see that we are within the argumentative circle, and that we have to accept other participants as equals for the purpose of the discussion, In redistributive taxation we trust 15 assuming a potentially universal circle of participants and recognising the right of all to participate in deliberation (Alexy, 1996). Third, entering into the practice of deliberation is to be defended with a two-pronged argument. On the one hand, it corresponds to the most universal form of life. As Alexy asks: “Is there any point in a form of life in which we do not assert things?”. On the other hand, even if we do not sincerely believe in the point of practical reason, the sheer number of people that opt for it constitute a strategic reason to at least pretend to accept it (Alexy, 1996). §10. Hobbesian practical reason is a generic name which refers to normative arguments that presuppose that individuals are actors with a stable set of preferences, who apply their factual knowledge in order to maximise their function of individual well-being (e.g. utility). For such theories, practical reason is not a matter of deliberation, but of aggregation of preferences (cf., among others, Buchanan and Tullock, 1962; Olson, 1965, 15-6). When applied to tax matters, Hobbesian practical reason tends to picture taxpayers as individuals who experience compliance with tax norms as a cost, so that they would not comply if they were ensured that they could get away with it, namely, that not paying will not have as a consequence being punished. It is only fear of the sanction that moves people to pay their taxes (Allingham and Sandmo, 1972; Yitzhaki, 1974). This form of moral scepticism can be challenged on three basic grounds. First, it is not possible to explain a good deal of human action as motivated by the pursuit of self-interest. This is the case when our wellbeing is dependent on that of somebody else or when we are committed to a certain course of action, even if that leaves us worse off (Sen, 1978). Moreover, we cannot explain much of political behaviour in such terms (why should people care to vote when their doing so implies a cost that is superior to the benefit derived from the infinitesimal influence they can exert on the final outcome?) (Lewin, 1991, chapter 3). In what concerns taxation, it pictures individuals as schizophrenics with incompatible preferences in matters of taxation and public expenditure. That is not only counter-intuitive, but it is clearly rebuffed by qualitative data obtained making the question in a way that allowed people to present their preferences in a related or simultaneous way (Confaloneri and Newton, 1995). Second, it is hard to explain how tax systems can be stable in the long-run, if they are merely based on coercion and self-interest. That is so Agustín José Menéndez 16 because any legal obligation that institutionalises a complex web of noninstitutional duties (like the general obligation to pay taxes) is quite fragile to high-levels of non-compliance. That means that facts have an incidence over the normative force of the obligation (i.e., if most people do not pay their taxes, the normative reasons for complying with the obligation get weakened and eventually vanish). Under such circumstances, only a more complex picture of motivation can explain how tax systems are stable in the long run. In relation to obligations like that of paying taxes, coercion discharges a triple function. Namely, it avoids that people who are willing to contribute are discouraged from doing so by the sense of hopelessness on social co-operation, it offers additional motivation to people willing to contribute but whose will might not be strong enough to do what they think right (akrasia), and it provides reasons to comply to those only moved by prudential ones (free-riding). Third, it cannot explain the present evolution of tax systems. The Hobbesian explanation of the tax crisis as a matter of increased opportunities to evade the tax burden cannot explain why the shrinking of the tax base is a phenomenon that predates the opportunities associated to the globalisation of economies and the reduction of capital controls and other devices that allowed a tighter control of tax evaders (cf. §2). §11. The last and clearly not least objection that can be raised against a democratic theory of tax law is that it is not legal. This is the same as saying that it might be relevant in normative terms, when deciding how the tax system should look like, but that is something completely different from reasoning within the domain of the existing tax system (Hart 1958, 599). This kind of objection presupposes a strong differentiation of law and morality as two different domains within practical reason. It corresponds to what we could label as the strong autonomy thesis. In more precise terms, it can be said to refer to the hypothesis that the determination of whether a norm is recognised as legally valid proceeds without any reference whatsoever to its moral correctness. In negative terms, it can be seen as a repertory of arguments against the special case thesis, which was considered as central to the post-positivist approach to legal theory characteristic of this essay. In order to defend it, it is advisable to distinguish two aspects of the special case thesis. On the one hand, it is associated with a conceptual In redistributive taxation we trust 17 connection between law and morality. On the other hand, it is related to a (limited) normative connection between law and morality. §12. The partisans of the strong autonomy thesis argue that an adequate description of law requires keeping it fully separate from general practical discourse. This is so for the very same reasons on the basis of which post-positivists propose law as the complement of morality. If the tasks of conflict-solving and co-ordination cannot be discharged by morality and we need to complement it with law, it will be plainly unintelligent to make legal norms dependent on moral concepts, because then we will have not solved the problem (Hart, 1958, 614; Raz, 1979, 86). However, it has been argued that the fact that the legal system necessarily makes a claim to correctness of the normative solutions that it provides implies a weak conceptual connection (at the systemic level) between law and morality (Alexy 1994, 41ff). This claim is quite modest, because it only requires that all legal systems raise the claim, not that they actually redeem it. Moreover, it seems just not possible to undertake moral reasoning without the assistance provided by the rules of general practical reasoning. This is reflected in the role of rules of general practical reasoning within legal reasoning. The latter are diffusely permeated by the former. Following Neil MacCormick, we can say that the norms of sound reasoning (which in most cases are non-positive) are contained by any legal system (MacCormick 1992, 120-1). Alexy has tried to make an enumeration of such norms. This phenomenon is more noticeable when legal reasoning is about legal principles in hard cases. In those cases, it becomes clear that the fact that a principle is positivised does not necessarily imply that it ceases being a moral or political principle. We can say with Alexy that the opposite is closer to truth: the positivisation of principles brings into the law a critical dimension (Alexy 1998). Moreover, it is also the case that the strict separation between law and morality can only be kept by either depicting discourses of application 10 Cf. (Alexy 1989,284-5): “General practical reasoning may be required (1) in the justification of the normative premises necessary to satisfy the different argument forms; (2) in justifying a choice between different argument forms leading to different results; (3) in the justification and testing of propositions of legal dogmatics; (4) in justifying any cases of distinguishing or overruling and (5) directly in justifying statements used in internal justification”. Agustín José Menéndez 18 of law (adjudication) as mechanic or by means of marginalising them. The strong autonomy thesis tends to exclude judicial reasoning from legal reasoning (Raz 1998a, 4). However, two arguments can be invoked against such move. First, judicial argumentation plays an essential authoritative role in relation to other kinds of reasoning about law (Postema 1996, 99-102). Second, the possibility of drawing a clear line between the two depend on the possibility of distinguishing once and for all lawcreating and law-applying functions of legal operators entrusted the task of adjudicating on legal conflicts, something which does not seem to be possible (Postema 1996, 110). §13. In normative terms, the partisans of the strong autonomy thesis argue that keeping law strictly separate from morality serves three main purposes. First, it avoids giving a blank cheque to positive legal systems. If we describe law as somehow dependent on morality, there might be an opportunity for officials to argue that law should be obeyed not only because it is law, but because it is morally sound, thus deterring people from passing a fresh moral judgment on the legal system and the concrete legal norms (Raz, 1986, 89; MacCormick, 1981, 161-2; MacCormick, 1985, 10-1). Second, it makes clear that there is no need of positivising the contents of positive morality. By avoiding the characterisation of law as the longa manus of morality, we increase the chances of social respect of personal moral choices, or what is the same, of personal autonomy (Hart, 1963). If law and morality are different entities, whether certain moral practices should be reinforced by legal protection is a question to be decided on its own merits. Third, it avoids widespread lack of compliance with the law on the basis of a direct moral assessment of its norms. The strict autonomy of law and morality implies that the question whether to obey the law is one which involves factors beyond a direct assessment of the right course of action in moral terms (Raz, 1979; Gans, 1992). However, it can be argued that there are good reasons for not keeping completely isolated law and morality due to the need of justifiability of the law in general, and because a complete isolation leaves us unarmed before cases of extreme injustice. It has been granted that the drafting of common action-norms in the grammar of law reduces the cognitive, organisational and motivational deficiencies of the language of morals. However, it is not the case that In redistributive taxation we trust 19 general legal norms can completely eliminate them. This is due to at least two reasons. First, that law is formulated in natural languages, characterised by different phenomena that render necessary creative interpretation in order to determine the concrete consequences of a norm in a specific context. Second, that the aspiration of legal norms to turn themselves into public reasons for action makes it necessary to formulate them in general terms, so that they can be sufficiently versatile as to govern action in different factual circumstances. The formulation of norms that are applicable to most cases requires paying the price of vagueness or obscurity in some contexts of application. If this is so, then some cognitive, organisational and motivational deficiencies remain, and they need to be sorted out through the process of application of norms. That in its turn implies that such process reopens the question of justification of norms, to the extent that it is a creative activity, even if in a quite limited sense, due to the massive amount of normative material that surrounds indeterminacy in a concrete context of application of the norm. At such point we are confronted by the need to find arguments that justify the creative activity. Some strategies are clearly bound to fail. On the one hand, participation is ruled out because the paradigmatic form of application, judicial adjudication, is an authoritative phenomenon (and there are very good reasons for it remaining so). On the other hand, the so called situation sense or the sense of appropriateness need to be rejected because they cannot be connected to public standards of justification. It is only by means of placing judicial reasoning within the larger domain of practical reason that we can find a solution. The creative activity of the judge derives its legitimacy from resort to general practical reasoning within the institutional context of legal reasoning. What the judge adds to the norm must be justified in the name of substantive correctness (Alexy, 1997). Finally, denying a normative connection between law and morality leaves us unequipped to deal with cases of extreme injustice. Once we adopt a point of view internal to the legal system in question, like that of somebody trying to work out the legal consequences as a committed legal operator, we realise that the issue is an extremely important one. That is especially so if we place the question within the framework of the so-