VI. The interest in rules *
It is always of great interest to see how distinguished scholars of a field of research interpret their own basic methodological position. Geoffrey Brennan's and James M. Buchanan's the reason of rules contains such a discussion. Subsequently I will outline and critically discuss this important book chapter by chapter.
The first chapter discusses what the authors call "the constitutional imperative". Several examples of game matrices are given which aptly show how rules may be helpful in reaching individual aims. Special stress is laid on prisoner's dilemmas and coordination games (including those with partly conflicting interests). The problem of enforceability of rules is not neglected.
In addition to the by now standard analysis of the "rationale" of rules and institutions in certain game situations, the authors make some points which are not standard. Notably, they insist on the distinction between institutional design and institutional reform. The former takes place in a situation where no rules exist whereas the latter takes place if rules are changed from a starting point of preexisting rules. The preexisting rules are a kind of capital which, at least in part, will be destroyed by a change. Therefore, it may happen that some rule changes, which from a global point of view of rule design, might seem superior to the "local optima" reached under preexisting rules, will seem inferior to existing solutions if looked at from the point of view of stepwise reform starting from the status quo. I think that this argument if not entirely new is well put and important.
The same applies to the thesis that all institutions, especially those of markets and of the political order, will have to face the same adaptive behavior of human individuals. Human individuals will react on the exigencies of those situations they are confronted with. Even those who want to stir society towards some preferred outcome by specific directives, will have to take into account that individuals will adapt to directives according to their own interests and not according to the interests of the authors of the directives. Therefore the political as well as the market process is viewed as a "system of interacting individuals from which outcomes emerge as equilibria" (15).
The authors insist that in evaluating alternative institutional arrangements one has to concentrate on the equilibria that presumably will arise from individual adaptive behavior. Only such outcomes are feasible which can be generated as equilibria by some institutional process or other. This is a very concise and convincing, if -- with respect to equilibria which may not always be reached -- somewhat "strong", way to introduce the basic public choice perspective on social institutions. It immediately draws attention to the evaluation of alternative rules rather than outcomes. Though, still, rules are evaluated with respect to the equilibria -- i.e. outcomes -- they give rise to. But, the authors insist, there is another reason for concentrating on rules, which precedes any evaluation relying only on outcomes: the evaluation of an outcome may essentially depend on knowledge of how or under what kinds of rules it came about. In this respect they are quite close to Robert Nozick's criticism of end state criteria. Brenanan and Buchanan prefer process criteria too. But they subscribe to a different brand of contractarianism than does Nozick.
The second chapter contains an elegant exposition of the "contractarian vision". The authors regard this vision as central to their enterprise. I think, however, that this is partly a misinterpretation of the rational basis of their own views. I would suggest that their position would much better be reinterpreted as a "right based theory of political institutions".
A right based moral theory, as the reader might suspect, would be my own approach. It should not be misunderstood as a kind of natural rights approach which I would reject (and so do Brennan and Buchanan). According to the view suggested here to have rights is in the interest of individuals. The political and legal order literally produces these rights as a kind of commodity which enters individual utility. The legal staff can produce alternative settings of protection and its activity is valued accordingly. Rights are the kind of commodity people want to get from the institutional setting. The justification of this setting therefore is to be based on interests of the individual consumer of rights directly. What other consumers get is of but indirect interest to that consumer who is the addressee of such a justification. He may attach high value to the rights of others but he will value them according to his own utility function (cf. for the same point made by the authors in a different context chapter 3, 111 last sentence).
Viewed this way constitutional political economy becomes complementary to John L. Mackie's vision of a right based moral theory (cf. 1985 c) which starts from given ends of individuals to which "rights producing institutions" supply means. To be sure, such a theory can provide a justification of rights only to those who share certain ends. Whether a system which respects the rights of all individuals arises depends on empirical circumstances. There is no theoretical guarantee that a universalistic system will be justified respecting the rights of everybody. "Rights particularism" cannot be excluded a priori.
To accept what has been said before would be coherent with Brennan's and Buchanan's tenet that there is no truth in practical affairs. Nevertheless, the authors prefer to stick to some universalistic basic value premise which evidently is -- to use Buchanan's favorite phrase borrowed from F. Knight -- a "relatively absolute" to them. But, instead of introducing the norm that existing rights must not be altered except for the case that agreement of all parties involved can be reached as a kind of normative working hypothesis, they try to incorporate their study into a contractarian vision -- or, for that matter, the contractarian vision into their study. That is, they try to launch their theory in a boat that drowned some two hundred years ago and several times afterwards. To be sure the ghostly ship of contractarianism never looks more attractive than when it is manned by James M. Buchanan and crew. I concede that it is always interesting to see the contractarian argument under full sail. But, at least according to my own reading of the evidence, eventually the ship is doomed. (At least if we use contractarianism as a direct justificatory argument and not only as a method of persuasion or a rational public ideology; cf. on the latter also below and 1.5. above.)
I will not discuss the classical criticisms of contractarianism in any detail here. Suffice it to note first that a fictitious contract can only bind fictitious individuals. Therefore the social contract cannot directly bind real individuals who were not party to it. The authors themselves partly agree: "It is not clear exactly how a person can be bound by promises he has not made (103) Second, to apply the notion of a possible social contract merely as a criterion of legitimacy does not help much either. Some reason must be given why exactly this criterion is going to be applied and not a different one. Then, the social contract is superfluous from the point of view of legitimation because those reasons justifying the application of the criterion could as well have been used directly as standards of evaluation. These are the most general classical criticisms. They apply to the book as well as to any other genuinely contractarian theory. (Rawls's theory is an exception only because in relying on his "decision procedure for normative ethics" it is basically noncontractarian.) But, the criticism can be made much more specific as far as the book presently discussed is concerned.
"The critical normative presupposition on which the whole contractarian construction stands or falls is the location of value exclusively in the individual human being ... There is no apparent means of discriminating among persons... Consistency requires that all persons be treated as moral equivalents ... From these presuppositions, and these alone, it becomes possible to derive a contractarian 'explanation' of collective order." (21-22) Then it is argued that a state would be justified as far as it can be conceived as emerging from a social contract among individuals. "Note that in the conceptual derivation of the origins of the state just sketched, there is no resort to any source of value external to the expressed preferences of individuals who join together in political community." (22)
This last remark, I should like to stipulate, is simply not true. The flaw in the argument arises from not distinguishing sufficiently between two levels of respect for individual preferences. On the one hand, it is true that in the contractarian argument only individual preferences are taken into account. According to the argument all social states that arise from agreement are legitimate. It is also true that they are based on purely individual values insofar. But, that only individual agreement should be of importance does not derive from agreement. Further, and more precisely, why should only those states that, at least conceptually, could have been brought about by agreement be legitimate? It may be that some individuals as a matter of fact will share the value that they should not intrude into the realm of other people without a preceding concession of those who otherwise would simply be victims of "power". Nevertheless, this acceptance of a basic right of autonomy of the person is clearly an external value. The individuals in the process are not asked whether they would take that stance too, or not. To subscribe in this way to the ideal of the autonomy of the person and a kind of normative individualism imposes the process of conceptual agreement, unanimity and all the central libertarian values on the social process from outside.
To be sure, the authors do not claim to know that the values that describe the autonomy of the person are "true". They might, as I said, also claim that they accept it as a kind of "working hypothesis" or "premise" without further argument. Then, they should be reminded that this is functionally equivalent to taking a natural rights' stance as the starting point of argument. But they insist at the same time that "the state does not emerge to protect natural rights." (22)
Still, they might want to argue that the social contract may be helpful as a kind of "thought model" which proves instrumentally helpful in the process of finding out preferred institutional solutions to which individuals could switch without violating the presupposed system of libertarian basic rights (and values). Now, first, this would be of value only to those who are already subscribing to libertarian values. Second, it may be doubted whether the contractarian argument is really helpful even for the libertarian. Perhaps to entertain contractarian arguments might cause more harm than good to the libertarian vision. -- This somewhat strong claim is raised here only as a conceivable possibility. But even then, it might seem to merit some further comment.
The contractarian argument is meant to be a model for developing criteria of legitimate action within large numbers' settings. But the concept of a contractual agreement between large numbers of individuals without any institutions external to their interaction -- is in a sense entirely misleading. The notion of a contract in such a setting is based on the existence of an enforcing agency. The very institution of the contract has to be artificially produced in every social setting but in the large numbers' setting it has to be produced by a specialized agency (or some equivalent). Economists absorbed with what appears to be ordered "anarchy" of a market forget about this point too easily. They forget that in every market exchange within a large numbers' setting there are at least three and not only two parties involved: the direct contractors a and b and the enforcing agency c which provides the legal umbrella (cf. on this also paper IV. above).
The market is an institution and institutions as well as the rights they provide have to be "produced". Therefore the magic number of any more developed form of exchange is n+1 (n>1). This holds for the social contract as well as for any other one. Compare this with the statement of the authors: "The presence or absence of other buyers or sellers will, of course, affect the terms of trade between any two traders, but 2 remains the magic number for the economic analyst." (28)
To be sure, Buchanan himself (cf. 1965) was one of the first to remind us of the essential relevance (well known to Hume already) of the large numbers' dilemma. And he certainly is generally aware of the central role of the referee in such settings. The authors even state that "political order ... must be antecedent to economic order" (26). That they seem to forget about these aspects when it comes to the problem of social contract seems to be due to the fact that they assume that libertarianism and the respect for the autonomy of other persons would fare better if contractarianism be accepted.
Sidestepping the psychological questions raised by the last remark I would presume that libertarianism and libertarian respect for the rights of persons can do fairly well without any notions of social contract. We can directly argue from the libertarian and individualistic values without any need to take recourse to a contractarian argument. If for instance it can be shown -- and this should be quite plausible -- that individual rights are more secure if the rule of 5/6 majority should prevail than under simple majority then the libertarian who subscribes to the ideal that individual rights should be as secure as possible has good reason to accept this measure as a superior alternative to majority rule. No contract is needed here nor is it of any help. If, now, the libertarian would be asked further why people should stop short of unanimity, then again no recourse to a contract is needed. To be sure, the contractarian would answer that individuals would unanimously agree on accepting just this rule because of the costs that they will face. But, how does he reach this conclusion? In "predicting" that all would agree he will refer to individual interests and argue that they are best served by such a rule (cf. the classical discussion in Buchanan and Tullock 1962). This reference to individual interests is the central point of the argument and not the contractual agreement. It is the crucial step in the theoretical exercise. The agreement of all as such does not necessarily add to the justification. But there is no need for it either. One could apply the argument from interest directly without any contract and justify certain measures to each individual separately.
That other individuals will agree too is an "extra" which may or may not be important to individuals who are party of the conceptual contract. The importance to them can only be guaranteed by imposition from outside. The economist in this respect would do what Buchanan rejects as "playing God" should he disregard the fact that some individuals might not care for the agreement of others but instead of this are genuine ethical particularists. As a final remark to this critique I would like to repeat that I have no quarrel with libertarianism. On the contrary, I would like it in as strong a position as can be and I therefore would suggest that it jump off board from the beautiful but doomed ship of contractarianism.
Still, I am not sure whether contractarianism on the level of persuasion might not be of great value. But this leads to an entirely different argument about the "economics and psychology of conviction systems or ideologies" which enter institutional settings. Conviction systems as a matter of fact influence institutional incentives and outcomes of the institutional process. The authors might be partly right as far as some function of social contract as a kind of "civil religion" is concerned. In this respect I feel quite uneasy as a critic of contractarianism. Nevertheless, for reasons of space I have to leave the important point of contractarianism as public ideology aside here, though not completely (cf. also 1. above). Some related questions about the psychology of different world views and their consequences are dealt with in the chapter following the introduction of the contractarian vision. To this chapter I will turn now.
The main line of argument of the third chapter is devoted to the problem of how individual behavior and, as a consequence of this, the development, reform, or design of institutions will be influenced if people believe in the existence of solutions to certain problems which are "right" or "correct" independent of their own interests, goals, aims etc. The thesis of the authors is that the belief in truth in matters practical or in the existence of "correct" social solutions will influence attitudes toward basic social institutions. This will especially hold good for a social scientist who believes that he can know what correct solutions to basic social problems would look like. "As he takes the role as adviser to princes, whether imaginary or real, the scientist-philosopher must oppose all suggestions for constitutional reform that involve additional limits on governmental authority, and, for like reasons, he must support all proposals for relaxation of existing constraints." (42)
Now, the truth of this statement certainly depends as the authors themselves acknowledge in a footnote on what the adviser believes to be true. If the adviser believes to know that it is right to protect individual rights and that it is therefore correct or right to put limits, on governmental authority then the conclusion would not follow. Thus, one may coherently stick to the premise of knowing right from wrong in practical affairs and at the same time be a libertarian believing in the necessity of constitutional limits on political and collective institutions. But this, as a mere possibility would not completely undermine the argument. It does not show that there might not be some psychological tendency or bias towards authoritarianism which is regularly brought about by the belief in truth in practical affairs. I think that the authors are quite right in insisting on this possibility. For, even if the belief in truth as such is in no way sufficient to precommit people towards mingling into others' affairs this belief still may be necessary or indispensable for such a commitment from a psychological point of view. If no own interests are at stake why in the world should the individual care for otherwise merely self regarding acts unless a belief in the "objective correctness" of certain standards of decent behavior would prevail?
Our enthusiasm for the good of somebody else in whom we have no other interest can be fuelled only by our belief that we know what is right for him. Even if we grant for the sake of argument that knowledge as such can motivate acts in which we otherwise would have no interest it would not make much sense to become meddlesome if we would accept that only the individuals themselves can know what is good or of value to them. Though this latter statement taken descriptively is much too strong to go without qualification it should suffice here to indicate the general line of argument that should be adopted to support the thesis of the authors. This is the one case in which knowledge is the relevant dimension of the problem and therefore the acknowledgement of ignorance might be sufficient to suppress meddlesome attitudes. -However, against meddlesome preferences based on genuine individual interests in others' behavior -the more plausible case -- the "epistemological" conviction about one's own ignorance would have no force still.
The authors try to point out the influence that the belief or disbelief in truth according to their own logic have on social organization. They compare science and politics. "Our point is that in science, I what exists' is wholly independent of any scientist or scientists." (38) As far as "internal questions" (cf. Carnap 1956, Empiricism, Semantics, and Ontology -- appendix) are concerned which may be raised within a scientific linguistic framework it belongs to the central logic of science as a social enterprise that agreement as such is not constitutive for truth.
Insofar I would completely agree with the authors. Nevertheless they do not spell out what this idea of independent truth implies for the social organization of science. At least to my mind it gives rise to what may be called the "declaration of rights of science" stating that the individual scientist is under no obligation to subscribe to any thesis simply because others subscribe to it. According to the "logic of the assumption of independent truth or independent reality" it would be futile to try to alter the truth or the reality as such by agreement. This does not make good sense if the course of affairs is regarded as independent of agreement.
On the one hand, the belief in objective truth can support a "tolerant form" of social organization as far as science is concerned. On the other hand, it has been claimed tha t in practical matters the same belief at least psychologically leads towards more "intolerant" forms of organization. This apparent incoherence, I think, points to the root of the problem. One, if not the basic principle of rational decision making and, for that matter, rational choice behavior requires that we discriminate between those aspects of the world that can be influenced by our decision and those which cannot be influenced by it. Without this distinction a meaningful self -interpretation of human action would
not be possible. An actor who decides does not try to predict his choices. He decides. The very notion of choice and decision implies that at least the deciding individual himself learns something by the -choice that he did not know himself before (cf. also essay V.). Otherwise the individual would not engage in the activity of deciding -- i.e. making a decision -- but rather in that of predicting. And this argument about the different meaning of the activities of deciding and predicting to the individual himself is independent of the question whether determinism as an "ontological phenomenon" holds good or not. (The authors express essentially the same view of decision making though in a different context to which I will turn in my discussion of chapter 5.)
Basically, the distinction between those realms that can be influenced and those realms that cannot be influenced by a decision is of the same relevance for science and politics. But in science any decision to accept or to refute a hypothesis is controlled by a reality which according to the internal logic of the decision is regarded as independent of the decision. It is assumed that reality leaves us with no choice what it should be like. We do not decide about reality itself but -if at all - only about our view of it. We assume that this view will be most improved if we grant autonomy of decision to any individual about what he regards as true. In this sense it is exactly the assumption of the existence of an independent reality that brings about a tendency towards tolerance in science.
Brennan and Buchanan, therefore, cannot claim that the belief in an objective reality as such induces intolerance. The thesis must be confined to practical problems. As far as these problems are concerned they make use of the example of a jury to illustrate their point- The decision of the jury is an indicator of something which, by the logic of the set up, is. meant to be independent of the decision of the jury itself. No par-tic pant would imagine that the very decision of the jury would have any influence on the past events its decision refers to. Contrary to that the political process is not an indicator of something that is regarded as independent of the decision as such. No collective value exists with which a collective decision could be compared. The decision is simply generated.
The implicit awareness that there is no independent reality of value makes people invent a kind of secondary reality of value. This invention has mislead Borda and Condorcet who explicitly discussed the merits of alternative rules of voting in terms of the probability of finding the truth. These theoreticians started, for instance, from the premise that every participant in an election has an equal chance to be right or wrong in the assessment of the common good and then computed the probability that the result of voting would indicate the true public interest if alternative voting schemes were used (cf. for an overview Nunni 1983).
In practical problems emerging in the context of science the practical decision appears as a twin of decisions. If we want to act on the basis of scientific evidence we first have a kind of jury decision of some experts about the acceptance or rejection of some hypotheses. Then, in the next step, we decide on how to act practically under the premise of that jury decision. In the Rousseau (Borda, Condorcet) interpretation of democratic procedures this distinction is missing. It is not seen that two decisions are generated: one about the acceptance of a judgement and one about acting on that judgement. If the distinction between the two decisions is not made it seems to be quite natural that scientists want to eliminate limits on both. The first decision is simply a device to indicate the truth. In science this characteristically falls into the autonomy of the individual scientist or the community of scientists. Why then, use the judgement of people who are not experts and why impose any limits on the decision of the experts? If all people are experts why, then, should we put any limits on the electorate? To leave the second decision with the electorate is a constraint on the scientific community. But the experts could still be convinced that they only have to give "the world the best menu from which to choose" whereas the second decision about taking action rests with the electorate.
All this being said, it should be obvious that the following point made by Brennan and Buchanan at least is not a matter of logic. "It is the objective quality of what is sought that differentiates the noncontractarian from the contractarian position; it is not a postulated difference in motivation." (37) Instead of Brennan's and Buchanan's discussion of epistemological questions I would prefer a strictly normative argument in favor of a libertarian position which views individuals as bearers of a right to decide about how they shall live. This normative individualism requires that they be treated as if they knew best what is good for them. This requirement itself is not based on an insight that it is right that they be treated this way. It is based on the interest of those individuals who want to have this right. At the same time they need neither insist that rights never be changed nor blur the difference between rules and outcomes. Criteria for both may under certain circumstances be derived from individual interests.
Brennan and Buchanan have a lot to say about how the evaluation of social institutions in terms of individual interests may be accomplished. This holds good in a twofold manner. First, they argue that the individual should be modelled as pursuing his own interest in describing the workings of social institutions. Second, the normative evaluation of institutions and the underlying rules is accomplished by arguments alluding to the interests of individuals. The additional justificatory requirement of conceivable agreement of all individuals on the ultimate justificatory level can be dropped from the point of view of an interest based moral theory without influencing the relevance of the interest oriented considerations of the authors. To these I will turn now.
In the fourth chapter of their book Brennan and Buchanan try to show that the model of homo oeconomicus is a fruitful tool of non-normative analysis. In addition to their former argument from symmetry that they used in the analysis of market and non-market institutions they argue that homo oeconomicus can be useful in developing what might be called "worst case scenarios". Here it must be asked in turn why worst case scenarios should matter. Several arguments are offered. First, attention is drawn to the fact that social relationships contain a potential for conflict. The potential antagonism may be deeply buried beneath the surface of cooperation but it may rise all the time.
The authors use the helpful example of a legal contract that is elaborated only after two parties have generally agreed on a transaction. The parties involved may basically trust each other. Nevertheless a detailed contract may be set up. This contract is a kind of contingency institution which never may be needed. Indeed the better the contract makes interests coincide, the better will it guarantee that the transaction will be accomplished smoothly without invoking any legal authorities. This, the authors argue convincingly, will be reached best by using the model of homo oeconomicus in setting up the contract and a good lawyer will exactly have this model in mind.
A second argument, which is closely related to the first, might be called the argument from suspicion. Even if people overwhelmingly want to do what they regard as morally good beyond their own narrow self interest, they will not like it, should their "altruism" be exploited by others. Altruistic motivations immediately will break down if people get the impression that a considerable number of individuals have a good time exploiting the altruism of others. Therefore, even those who basically intend to behave in ways different of their own narrow self interest, will often have a strong motive to avoid the risk of being exploited and therefore to behave as if they were selfish themselves. To use Urmson's lucky phrase "saints and heroes" are rare (cf. on this 1958). Few of us can stand the thought to be exploited by those, who by their exploitative behavior, show that they are the last to merit our altruism. Therefore not only exploitation which actually takes or took place but also a mere thread that exploitation may occur can make all acts which are not narrowly selfish appear as supererogatory. This gives rise to "Gresham's law in politics" as the authors have nicely put it (60). Bad behavior drives out good behavior, trust is substituted by suspicion -- and even the suspicion that this might occur because everybody knows that institutional or contractual arrangements are not fool prove against opportunistic behavior may bring about the undesired result.
The foregoing considerations are discussed in a closely related third argument about what the authors call "quasi-risk" aversion. "The harm inflicted on his fellows by a person who behaves 'worse' than the average person in the community is greater than the benefits provided by another person who behaves 'better' than the average person. Accordingly, the average person model understates the average harm done." (55) The argument is supported by the model of N "local" monopolists of whom N/2 behave as if conditions of perfect competition would prevail whereas N/2 extract maximum rent. A calculation starting from the average price would underestimate the welfare losses. For, half of the consumers face a much higher price than the average one. Therefore the actual welfare loss must be determined as half of the welfare loss N consumers would have incurred under the higher price.
This is an interesting argument which would merit further development. The authors are quite short about its generalization to what they call "the general case" (59). On the other hand, one of the great merits of the book in general is that it induces the reader to speculate himself on further extensions of the argument. One such extension seems to be close at hand here. The very notion of a "right" seems to imply the idea of a guarantee. Having a right we can have our way independently and even against the interests of other individuals. Insofar as we have the right we do not depend on others -- neither their agreement nor their benevolence. Thus, the value of a right can best be explored by assuming that the bearer of the right faces purely selfish individuals with possibly opposing interests. Therefore, within an interest based moral theory which tries to evaluate any social institution (moral, political, legal) in terms of how well this institution serves individuals in their "pursuit of happiness", it seems quite natural to rely on the model of homo oeconomicus. If individuals take an interest in having rights they will be rationally inclined to use such a model in assessing the value of any proposed set of rights. This value can be fully assessed only under the assumption that one has to get one's way against opponents. This brings us back full circle to the illuminating model of the contract between two parties from which the authors started.
The chapter on how to model the individual for constitutional analysis offers some fine reasons to apply " good old homo oeconomicus" for that purpose. The argument can hardly be better presented. It makes quite evident that the. model of homo oeconomicus has to be a central tool in any analysis which tries to evaluate basic institutions of society. Nevertheless, from my own experience in propagating the model I doubt whether especially "continental" social philosophers will be easily convinced. They seem to prefer to live comfortably and speculatively without homo oeconomicus (though of course there was Macchiavelli and even Fichte saw the value of worst case scenarios in his 1807 piece on Macchiavelli).
Still, I have one reservation myself. It may be that homo oeconomicus is not really adequate to set up worst case scenarios. It may be too optimistic. Historical experience shows that the creation of the "real public bads" usually strongly depends on some form of altruism. This observation certainly belongs to the wisdom of the ages. The authors seem to feel that way too when they insist that there is no truth in deciding problems of value. They are well aware that enthusiasm for a just cause is the root of nearly every public bad arising from concerted action on a larger scale. These bads do not come about by engaging in self-interested actions they arise from transgressing the confines of self-interest. Here, clearly, homo oeconomicus is inadequate as a means to construct a worst case scenario.
However, homo oeconomicus may be helpful for designing social institutions such that the human tendency to become engaged in "just causes" is controlled by self-interest. If participation in some collective behavior becomes too costly then zealots will become rare. The general thesis would be that only self-interest can prevent people from switching to potentially harmful activities of expressive behavior which involve asymmetric costs: low to those engaging in the activities and high to those suffering from the activities. Therefore rational individuals might want to insure themselves against such "weaknesses of their nature". As human nature cannot be changed these incentives must -- as David Hume would have said -- be "artificially" provided through the creation of social institutions which raise costs of "irresponsible" behavior. This leads to the problem of rationally binding 11 oneself" either individually or collectively which is elaborated in chapters five and six of the book.
Amongst all features of human nature David Hume regarded one as most central for an adequate understanding of the whole structure of social institutions: the natural inclination of human beings to prefer the near to the more remote. This inclination, argued Hume in his famous treatise on human nature, has a temporal and a social dimension. Both are of equal importance. Both may prevent us from getting preferred results by tempting us to try to take "free rides" as far as longrun private or as public interests are concerned.
Bertrand Russell made essentially the same point on several occasions. He entertained a never failing awareness that human civilization is built on the ability to postpone immediate satisfaction in favor of longterm interests and that it therefore is under some emotional pressure all the time. To be sure, economists have to say something about these themes, too. There seems to be a growing awareness of the problem of decision over time. For the purely private part of this problem it may suffice to point to the meanwhile broad literature on endogenous changes in tastes or preferences and Schelling's use of the term "egonomics" for the theory of prudent self- management. It seems not too far-fetched to regard the term constitutional economics simply as the counterpart of egonomics applied to the social realm provided that we are aware of the fact that there is no superindividual "ego" but only individuals engaging in concerted self-management or collective action. Using the term this way it will become free of communitarian and perhaps also of some of its more doubtful contractarian overtones.
Brennan and Buchanan start with a premise that is closely related to what has been said above about the distinction between prediction and decision. "We confront options now, not later, and although the action that we take now may influence the choice of options available to us later, along with our possible orderings of these options, the fact remains that we cannot, in the present, make choices in future time. Nonetheless, the choices that we make now must embody the recognition that we will face choices at some later date." (68) Present decision involves both the knowledge that we cannot "predecide" our later decisions as far as they are "real" decisions and the knowledge that present choices may influence and may deliberately be used to influence future decisions.
At a first glance it may seem strange that individuals could be interested to make precommitments. If they are rational they should be able to be rational all the time. The most basic interest of human beings, after all seems to be to be free to choose as they "go along" (cf. here Mackie's use of such a principle as a kind of working hypothesis in 1985 c). This is aptly shown by Nozick's example of the experience machine which offers the possibility to "predecide" future experiences by choosing a pleasant course of life from a menu (cf. 1974). Why are we reluctant to enter such a machine, say for the next ten years, if we know that we would have pleasant experiences all the time?
Regardless of all problems of risk aversion it seems that we do not want to know in advance how we shall live. It seems that we want to engage in genuinely intertemporal choice. We want to make choices in future time.
Still, this is not the end of the story. Already David Hume remarked (cf. 1948, 87 ff. "of the origin of government", treatise, book 111, chap. 7) that at the distance of a year we will almost certainly choose the better of two alternatives whereas in the choice situation itself we will suffer from temptation. Therefore there seem to be some parts of decisions where it may be in our interest to change at least the incentives of future decision situations or to restrict the range of genuinely intertemporal choice. This raises the problem of how this can be done.
Brennan and Buchanan discuss the example of Robinson Crusoe who could lie down close to the sea in order to wake him up early in the morning. (Today we have the less wet method of the alarm clock. At least some things improve!) This would be a kind of purely private precommitment. But, Crusoe after the arrival of Friday might have made him do the job instead of the sea. This would have been the most simple example of an extension of the division of labor to the provision of commitment facilities. The authors do not put it this way. Nevertheless, it seems to be one of the most central features of human society that the division of labor can be extended to the "production of rules". This is exactly the reason why people can get locked in with certain social rules which provide a possibility to commit society "as a whole" to certain courses of action. And this, again, is a reason of rules which is at the very root of civilized society. (Of course individuals may want to accept commitments only in exchange against likewise commitments of other individuals and thus may not be interested in commitment as such. But this explanation why they might agree on entering binding commitments does not explain how they can commit themselves or -- more technically speaking -how commitment as a strategy can be subgame perfect in the game of life.)
The classical insight that a -- if not "the" -- reason of rules ties in enabling us to bind ourselves and thus to insure ourselves socially against acting merely on the exigencies of the moment, has been subject to erosion in recent years. Brennan and Buchanan are convinced that this process should be stopped and if possible be reversed. Therefore they engage in an enterprise which best may be described as "rational persuasion". As they themselves remark, as a matter of fact their whole "book is an extended essay in persuasion". (134) To my mind, what they have to say in chapters five and six is the most persuasive part of it. First, as mentioned already, genuine intertemporal choice of an individual is discussed. Regardless of the considerable role of the temptations of the moment, the "continuity" of individual decisionmakers over time is quite strong. At least decisionmakers form one person. The "shadow of the future" becomes an important motive for each of them. The authors show convincingly that intertemporal decisionmaking in a social setting will suffer more severely from a bias towards short-run considerations than purely individual decisionmaking. This goes beyond the common statement of scholars of public choice that the frequency of elections will push politicians towards thinking in terms of election periods (inducing a political business cycle and all that). Instead of this, or more generally, the individual who participates in a sequential process of social decisionmaking will have much less foresight of the next decisionmaker in the sequence than in a purely individualistic setting.
Above all the individual who is involved in such a process will not be able to know what others intend by making certain intermediate decisions in a longer sequence. For example, two individuals a and b might both vote in favor of higher savings in period to. But, a regards this as a step towards a long range program of debt retirement whereas b considers it as a step to even higher spending in period t1. Here party ideologies may be quite helpful in reducing uncertainty. Indeed this may have been one of the more rational reasons for such ideologies and party programs. They are a kind of commitment that imposes some constraints on choice. Still, except for very rare circumstances the possibility of a change in the ruling "majority coalition" of voters would be very real and bring about considerable uncertainties for individual public choice. "A more general way of stating the central proposition of the analysis is to say that the individual participant in collective choice will find it difficult to make instrumental use of the collectivity in the furtherance of genuinely long-term objectives. 'Investmentlike' collective alternatives will tend to be placed lower than 'consumptionlike' alternatives in individual orderings." (79)
As one will expect the authors do not intend to avoid the policy issue of institutional checks for the bias towards 'consumptionlike' alternatives. In chapter 6 of their book, Brennan and Buchanan offer reasons and evidence that show that it may be in the best interest of individuals to engage in certain forms of constitutional reform which would amount to binding the collectivity stronger than it is nowadays bound. They hint especially at three traps which might be removed by such measures: the high-tax trap, the inflation trap, the public-debt trap. The argument is basically the same in all three cases. It is pointed out that the decisionmaker who has to decide in period to has no guarantee that the long-term policy he wants to initiate would not be revoked in period t1. There is at least high uncertainty involved whether after a first step there will be a second one in the intended direction. Making the distinction between the aspects of reality that can be and those that cannot or can be less controlled by a decision, the rational actor will concentrate on the former rather than on the latter aspects. This, gives rise to a tendency to "overgraze the intertemporal commons" or, in more Humean terms, to prefer the more near unduly to the more remote.
After all, one could still argue that this is what people themselves want. Why, then, should we bother? -- We can treat them as if they knew best what is good for them and still explain what it means to prefer the near "unduly" to the more remote. Not the presence of "bad" motives but rather the rationally motivated absence of "trust" gives rise to obstacles for reaching results which may be desired by the same people. Therefore, when Brennan and Buchanan propagate a change in the rules of the game they need not presuppose that they know better which ends people or, for short, 'we' should pursue. Instead of this they can argue that it is in our own interest that we should bind ourselves to long-term rules -- like for instance that of a balanced budget -- so that we can resist the temptations of the moment and develop a certain amount of trust. -- Let us hope that we at least cannot resist this temptation and thus will agree on some such restrictions.
The seventh chapter turns to the problem of justice. "Justice" is one of the key concepts of social phi losophy. It may be approached from different angels. One may try to describe what people mean by "justice". The analysis may also aim at an explication of "justice" (in the technical philosophical sense of explication). Both ways of discussing the problem would as such not logically imply the thesis that bringing about justice or just states of affairs -- as the counterpart of the concept would be normatively justified. This re quires a second and independent logical step. This step is necessary in a straightforward normative approach which tries to suggest what "justice" should mean according to some normative conception or other and also tries to lay down criteria which allow for an as sessment of justice. But, clearly, here the step is a small one. The third type of analysis is based on some normative theory of "right and wrong".
The authors do not explicitly state what the, aim of their analysis is. But it seems to be quite clear that they are approaching the problem of justice from the point of view of normative theory. They intend to lay down criteria of justice which represent their view of justice 11 as a good thing" and how that term should be used accordingly. To be sure, this is a legitimate approach. Nevertheless, it should be noted that the criteria of justice then either are a simple expression of the preferences of their authors or have to be justified from other normative sources.
This draws our attention to the underlying normative theory. It has to be asked whether this theory seems to be acceptable or not. In a next step we then may ask whether it is appropriate to use the term "justice" for the criteria stemming from the normative theory proposed as a theory of justice. Though, from a logical point of view, one is free to choose ones terminology as seems fit there are opportunity costs involved in such a choice. It may be more or less similar to ordinary usage, more or less illuminating, misleading, precise, etc.
The normative theory of the authors can be described very shortly. They insist that just conduct is conduct according to rules. Injustice reduces to the violation of rules. Behavior that stays within given rules is just and so are the results of such behavior. They insist "that the injustice springs from a violation of rules -- rules that are legitimately expected to be applied by participants or affected parties. Such rules isolate certain considerations that are relevant to the outcome of the event and to the determination of rewards, such rules isolate other considerations that are either explicitly or presumptively irrelevant." (99) This, after all, seems to amount to relativism implying that whatever comes about according to rules is just. But, clearly, Brennan and Buchanan are well aware that besides the problem of formal justice or, as they call it, "justice within rules" (97) there is the problem of the material justice of the rules themselves or, in their terminology (which departs from the standard usage following Chaim Perelman), of "justice among rules" (97).
This leads to the problem of whether the rules themselves are just or not. "(I)t deals with justice as a criterion for evaluating alternative sets of rules" (97). In this context the authors draw attention to the fact that for many purposes of analysis "justice as a means of evaluating rules can be usefully viewed as an example of justice within rules". (97) This is a very important point which much too often is neglected. A rule may be just because it originated from a just rule. Nevertheless, the process must eventually get to an end. Somewhere we must start with a rule which cannot be evaluated according to a rule (cf. for this point in a more general setting Vanberg's assessment of Buchanan's contractarianism, 1986 a). Here the authors rely on the notion of agreement. As far as the ultimate rules are concerned those rules are just which, at least conceptually, could have emerged from unanimous agreement. It certainly would become boring to repeat my reservations against contractarianism. But in the present context two points might nevertheless merit further discussion. First, a sufficient condition or material criterion of justice certainly may be that people unanimously agree.. If somebody gets what he wants then this cannot involve any injustice to him. But, as has been pointed out frequently, the voluntariness of choice depends on opportunity costs. Here a hidden material criterion of justice enters the picture. The authors discuss this and related points at some length. Nevertheless, I do not think that they convincingly answer the problem. Second, if we need some material criterion for assessing the starting situation and the most basic rules would it then not be possible to apply these criteria on all levels of the argument and to all sorts of rules and not only to the most fundamental ones' ? (The latter point has been more extensively developed by Vanberg, cf. again 1986 a.)
The foregoing somewhat Humean argument can be extended again. One may insist on the thesis that the basic justification of any social institution in the last resort must steem from given interests (in the widest sense of that term including individual ideals). if appropriate interests do not exist then no justification exists as far a's the addressees of the justification are concerned. Rules, too, are justified as far as they serve our interests. (To be sure, this does hold good only for the justification of the rules. Application of rules towards addressees of the rules is possible without assuming that these individuals have been addressees of the justification of the rules.)
The better rules serve our interests, the less does it seem counterintuitive that "justice" is reduced to behavior within or relative to rules. On the other hand, if certain individuals or groups will not regard it as being in their best interest to live under a certain set of rules then they will cease to regard the violation of the rules as "injustice". This also applies to "meta-rules" (105) governing the enactment of rules. If a meta-rule leads to results from which our interests suffer this will count against the meta-rule. This is a kind of falsificatory process against meta-rules. Here as well as in other branches of rational action rationality strongly depends on a process of trial and error -- though, admittedly, the time cycles needed might be extremely long (cf. for a more extended discussion of such views Albert 1985).
From the point of view of an interest based moral theory, besides the observation of given rules not much clearance is left for the term justice at least as far as the term is used as a theoretically legitimate term. Justice has content only relative to rules. In this respect I fully agree with the consequences implied in Brennan's and Buchanan's approach. Nevertheless, by insisting on agreement as fundamental justification of any institution they seem to be reluctant to go down the line of relativism or value skepticism to its logical end. Still, they might claim that their normative theory seems to retain more similarity with everyday usage of the term "justice" than any completely relativistic concept of justice could. This is an advantage of the Brennan and Buchanan view. But it is not much of an advantage. All rule oriented types of theories of justice which try to reduce that notion as far as possible to the behavior within or relative to rules exclude a lot of other usages of the concept of "justice" which are quite common in everyday life.
People frequently use the term "justice" in everyday life to describe what they think is an objective normative structure of reality. This goes beyond justice relative to rules. If it is pointed out that it is even less compatible with the concept of justice proposed by a skeptical ethical theory this is correct. But, within the framework of a skeptical ethic based on the pursuit of given interests everyday usage can be interpreted in the sense of Mackie's notion of an error theory (cf. esp. 1977). The claim that something -- especially a rule -is unjust may still indicate then that individuals have the feeling that some objective norm is violated. But this feeling is explained as an illusion and thus is according to this view not describing an objective feature of the structure of social reality.
Clearly Buchanan accepts the latter. Still, the thoroughgoing skeptic will insist that unanimous or any other form of collective agreement ultimately cannot serve as a foundation for the legitimation of rules. On this Buchanan does not agree. As Buchanan himself insists on value skepticism it seems to me, however, that this is not consistent. To my view he should go down the line because he is committed to an interest based theory of just rules without noticing it.
If we cannot show the justice of the most basic rules of the social game -- at least not in any sense which does not depend on an interpretation of "justice" in terms of interests -- then we are restricted to persuading people of other reasons of rules. In this context, examples of the kind given in the two foregoing chapters of the book are of greater value than any abstract discussion of justice. Examples of how rules work in the interest of individuals -- even if it seems to be counterintuitive at first sight that they do so -- are presumably of the greatest value for reaching the task of rational persuasion. Knowledge of such examples will provide a better common basis for the discussion of constitutional reform than any abstract considerations of justice.
Again, I fully agree with the authors: "Without a shared 'constitutional mentality', without some initial common ground from which discourse can proceed, all argument on design comes to naught. Persons must be cognizant of the reason of rules before they can enter into dialogues devoted to questions concerning choices among rules." (xi) I should like to add here that this will be even of greater importance as far as the question of strengthening or weakening rules is concerned. Here the choice may be quite evident to most people most of the time if they are cognizant of the reason of rules. For such people it may also prove helpful as a mode of thought, as a theoretical model, or as a crutch to their imagination if they entertain some contractarian notions. In this I would join the contractarian camp. Still, I would presume that models as those developed in the fifth and sixth chapter of the book are of greater value in persuading people that it is reasonable to live under a regime of rules than contractarian notions of the justice of rules as stemming form agreement ever could be.
The eighth chapter offers some additional reasons for looking at society from a strictly institutional or constitutional point of view. It starts from raising questions like the following two: If we take it for granted that the income distribution amongst people should be as close to equality as possible "(a)re there any reasons for believing that the political order will generate a more nearly equal distribution of income than a free market would generate (as seems to be commonly assumed)? What 'rules' of political order are likely to generate more 'equitable' outcomes?" (112) "Traditional discussions of distributive justice are hopelessly remote from the real world of distributive politics.. Indeed, crucial political constraints are typically ignored altogether, and the institutional feasibility of distributive justice is brushed aside as an irritating minor technicality. This may be an acceptable procedure, for the moral philosopher, but for the social analyst it is totally inadequate." (133)
Being a moral philosopher myself I agree with everything except for the statement that moral philosophers may ignore the questions raised. Contrary to Brennan and Buchanan, I think that they are much too polite to my own profession. Those moral philosophers who refrain from engaging in social analysis of the kind proposed by the authors themselves become 11 professional good men" but cease to produce moral precepts of any value whatsoever. To think otherwise would not only overextend the separation between means and ends it also would misguide our hopes of what can be accomplished socially. This would, at least to my mind, be the worst outcome for moral philosophy. Sound moral philosophy is about what we can rationally hope for that it become true. Forming rational hopes, we must take into account the fact that individuals will not simply follow precepts but will adapt their behavior to the precepts of morals in some sense or other which cannot easily be foreseen. Without enforced rules one could not trust that, while observing precepts oneself, one would not be exploited. And even under such rules we must concentrate on the equilibria likely to emerge after adaptation, and not on the intentions of those who inacted the rules. It is simply mistaken to delegate these questions to the realm of policy implementation for they strongly influence the set of possible worlds that we can rationally hope to bring about by and within human institutions.
All these points have been made by the authors in one way or other. These truths are simple enough. Still, they are frequently neglected by philosophers, economists, and the layman. This negligence is pointed out persuasively as a major fault of most current discussions about distributive justice by Brennan and Buchanan: "We cannot adequately characterize governments as 'choosing' among alternative distributions. Rather, distributions emerge from the complex interaction of autonomous agents, all making individual choices under a given set of rules." (132) If we take into account both the "adaptation axiom" -human beings will adapt partly autonomously to the situations that they are confronted with -- and the "symmetry axiom" human beings will basically behave the same way in all social institutions then our normative perspective will shift. "The normatively relevant comparison is between ... imperfect institutions ... Or, to put the question most starkly, is movement toward equality institutionally feasible? ... The virtue of the constitutional perspective is that it places this question firmly at center stage. It does so by shifting the domain of normative inquiry from the set of imaginable income distributions to the set of feasible institutional arrangements from which income distributions will emerge" (117).
The authors illustrate their convincing general argument by the simple model of three different tax schemes under simple majority rule. They make use of the well-known instability properties of such an interaction in the case of three actors engaged in a simple share-out game. They analyze how these properties will influence expected many-period and within-period income under alternative tax schemes. This shows "constitutional analysis at work" and therefore is of great value in itself. The tentative conclusions which are drawn by the authors subsequently are at least as interesting.
The most interesting conclusion is that "efficiency considerations" under certain circumstances -- like the persistence of rent seeking incentives in the political realm -- may favor what the authors call an "explicit constitutional transfer package" (130) After this they continue with commenting on that notion: "(A)Ithough it seems natural to refer to such an arrangement as 'constitutionally determined transfer', there is something rather misleading about such terminology. There is, after all, no sense in which these policies would 'redistribute' resources or income among persons. Rather, the constitutional order would involve a set of property rights in which each individual genuinely 'owned' only a portion of the product that the labor embodied in his person generated." (130)
As I have mentioned before my own preferred view of a moral theory would be a right based one. This theory would in the last resort still remain an interest based theory insofar as all rights supplied by a social order are viewed as means to given individual ends.
This fits very well into the institutionalist and constitutionalist framework. For, rights are what people want of institutions or constitutions. This is the reason why they place value on them. As the very meaning of a right reduces to a guarantee which is artificially provided by the social order it should be quite clear that the rights people can have are as dependent on institutional viability as everything else. Attention must be shifted from conceivable sets of rights to rights producing institutions. What rights people will have will depend on the institutional equilibrium which emerges from the social process. Individuals do not enter the process with predefined rights. It should also be clear that no distinction between participatory rights and merely defensive or negative rights can be made from this point of view. Brennan and Buchanan seem to be very close to this position with their notion of a 11 constitutional transfer package". This package is not bringing about "attenuated ownership" but may be rather the only way to bring into existence some kind of guaranteed and socially viable ownership. Transfer via a "constitutional transfer package" is not diminishing preexisting rights but a price or precondition for creating constitutionally enforced rights. Without constitutional transfer there would be no constitution. In that sense it is clearly not redistributing constitutional rights held in advance. A quite different topic is transfer on the level of post constitutional choices. This is redistributive and violates "the reason of rules".
The final and ninth chapter of Brennan's and Buchanan's book is devoted to the question of how suggestions like those made by the authors of this work can influence social reality. After all they cannot address a benevolent dictator. They have to face squarely the problem of how their own suggestions will if at all influence the social process. They are well aware of this problem. In response they state first as already cited above that their book is an "extended essay in persuasion" (134) and second that their "efforts ... are aimed largely at the academic constituency" (149). They also consider the problem of how their theory it self would match the premise of an homo oeconomicus. (It seems not entirely frivolous to ask why they them-
selves are engaging in persuasion.)
The homo oeconomicus model certainly cannot be the only building block from which the whole story of institutional analysis is construed. At least I myself would regard this as a mistaken position. I am fairly sure that Brennan and Buchanan would agree. At least it was their own pioneering work (along with that of Tullock and Lomasky) that pointed out the relevance of low cost behavior in such situations as voting. If an individual knows that a decision has to take place behind a "veil of insignificance" (cf. Kliemt 1986 b) then he will at least tentatively make expressive use of the vote. (I would presume that even the expression of self-interest in such a situation would involve a concept of expressing "legitimate" self-interest. There will be a self-interpretation as being entitled to certain things.)
In the realm of expressive low cost activities people's belief about what they can rationally hope for in "foro interno" will exert influence on social results and thus on behavior in "foro externo". It may well be that contractarianism is just that type of "civil religion", which in comparison with other philosophical forms of argument which start from more collectivistic or more sceptical premises, might lead to best results. Whatever people believe to be real has real consequences, and in this, what they believe about the justification of institutions forms no exception.
The last remarks lead quite naturally to my final comments on the whole Brennan and Buchanan enterprise. In general, there is a certain internal tension between the model of homo oeconomicus and a perspective insisting that we should stay close to a realistic view of individual adaptive behavior in institutional design. On the most basic level, the game of social interaction will be "uncooperative" in that there are no prior "artificial" commitment facilities. There will be no external referee enforcing rules nor will there be externally provided institutions. Institutions cannot, so to say, "be imported from the outside". Therefore all institutions and especially the "power of binding oneself" must be produced within the interaction itself.
Within the purely interest based approach to social institutions the so-called Hobbesian problem of social order arises. Homo oeconomicus as a theoretical model makes it very hard to explain how social order in the first place can be produced. Stable social organization without some amount of non-opportunistic behavior would simply be not viable (cf. on this Vaughn 1989). How could anybody actually have any incentive to "produce" those rules which are constitutive for social order? The authors leave this question completely aside. Even if everybody would know that it would be in his interest to live under a certain set of rules why should he not try to take a free ride? Those who would be expected to follow other rules requiring to punish free riders will have the same incentive (cf. in this context Mackie's remarkable essay on "Morality and the retributive emotions" 1985 b).
These arguments can easily be developed further as an additional critique not only of the homo oeconomicus model but also of contractarianism (cf. Kliemt 1980). As far as the social contract is applied to the basic structure of society I see no way to avoid that problem. But I have also argued elsewhere that the problem of social order can in principle be resolved within a homo oeconomicus perspective (cf. for an abridged English version, Kliemt and Schauenberg 1984). The basic result of this analysis is that the backbone of stable social interaction must be a structure of small groups either naturally or artificially separated and internally locked into equilibria which are linked to the provision of order to the game outside. Nevertheless, this possibility to explain the deep structure of social interaction -- and how the process of institution formation could get started in the first place -within a broadly game theoretic framework does not exclude that the surface structure of social interaction will be influenced strongly by different phenomena, and therefore cannot be understood without recourse to behavior deviating from the model of homo oeconomicus. This in turn will influence the deeper incentive structure.
Experience shows that rules in society exist. Without them no society is viable. Notably commitment rules as those forming the institutions of promise and contract exist. They are provided within the game of social interaction itself. That this is possible in a stable form depends on the fact that some individuals take an "internal point of view" (cf. Hart 1961) to the rules. They accept them as standards of their own behavior and not only as an indicator predicting future (negative or positive) sanctions.
"It is important to make this point because, in some economists' discussions of the law, one obtains the impression that choosing whether to abide by the rules is like selecting a drink at a soft-drink machine; that is, one either abides by the rules and pays no penalty or fails to abide by the rules and simply pays the price of so doing, as reflected in the rules." (101) Here the authors get very close to Herbert Hart's classical critique, of a purely sanction-oriented concept of normative obligation (widely shared by theorists of the empiricist tradition in the theory of law like John Austin, Theodor Geiger or -- at least in part -- Thomas Hobbes himself). Therefore they should be quite happy with the notion of a "rule of recognition". This notion was developed by Hart to explain the characteristics of a legal order which is not only exhibiting regularities of behavior but is also based on the acceptance of certain rules by certain groups of individuals. No legal order would exist -- at least not in a civilized society -- if people would not use the rules as accepted standards of correct behavior beyond their indicative function as predictors of sanctions. Therefore the explanation of the phenomena observed in a rule governed society requires that we take recourse to more than a case-by-case calculation of interests. Individual actors will be motivated too by their, as Hume would have said, "imagination" of rules which they accept as standards of own behavior (regardless of what have been the motives of adopting the rules in the first place).
It is important to note that Hart's fundamental analysis of the role of rules in society would only require that a "sufficiently large and powerful" fraction of all individuals in a society will accept the basic rules of the game (this point was made already by Marsilius of Padua in his "defensor pacis"). Within Hart's fundamental theory of the rules governing society there is no necessity for unanimity -- and rightly so. This also holds true for the basic constitutional rules of society to which the rule of recognition as an identifying device refers. Whatever they may be it is sufficient for the existence of these rules that they can be, identified by the "rule of recognition". Afterwards the "game" -at least the game of rule enforcement -- is played "by" these rules.
The most important rules of all are those which allow for deliberate changes in rules. On a deep level of analysis one might state that this is possible because individuals taking an internal point of view to the rules of change may transfer their allegiance to the rules enacted according to the rules of rule change. Accordingly Herbert Hart describes "law" generally as a unity of secondary and primary rules. The former governing the process of rule creation and the identification of valid law, the latter at least in general addressing the subjects of law. In one of my favorite phrases of his famous book on the concept of law Hart compares the invention of rules which allow for deliberate rule change with the invention of the wheel (cf. 1961, 41). Again the authors come very close to Hart's position when they introduce their notion of meta-rules within their discussion of justice.
Though I would not like to be misunderstood here as raising the objection that somebody has reinvented the wheel -- I have quite a record in doing just that myself -- I nevertheless think that it is really almost tragic that Brennan and Buchanan in their analysis of the reason of rules seem to be unaware of Hart's analysis of the working of rules in setting up legal institutions. It should be hoped that in the future somebody will combine both approaches and offer further reasons for rules. In any case the book discussed here is a huge step forward in refreshing one of the foundations of a free western society: our understanding of the reason of rules.
<< References >>
|
Kliemt, Hartmut: Papers on Buchanan and related Subjects. – Munich (Accedo Verlagsgesellschaft), 1990: p. 123 – 161. (Studies in Economics and Social Science (SESS), Vol. 1 (1990)) |
[Top]