I. On economics and ethics *
Economists are notorious for uttering such statements as that there is no such thing as a free lunch. They claim to economize on love as the scarcest of all resources. Hard nosed as they are, they insist that public rhetoric in favor of any cause serving the good of mankind is only concealing a deeper structure of more mundane and personal interests. People may act as if they were moral men, but, after the actors' masquerade is taken away, rational economic man will eventually emerge. After centuries of criticism of the subjectivistic and individualistic Hobbesian approach to moral science, economists stubbornly stick to the heritage of Thomas Hobbes who insisted that interest rather than reason ultimately governs human action and consequently the world. As their great ancestor they spoil the green lawns of benevolence described by professional good men with the crab grass of their skepticism -- and rightly so.
If professional good men never liked the theories of Thomas Hobbes then they do not like the world view of most modern economists. But, raising the voice of skepticism against the enthusiasms and superstitions of opinion serves an extremely important function in our intellectual life. I personally hope that in the long run the warnings of the skeptics might diffuse sufficiently to capture the minds of all social scientists and ultimately of the general public.
Using the behavioral model of homo oeconomicus as a guidance we can hammer home many important insights into the workings of our social and political institutions. Nevertheless, having a hammer we should not expect everything in the world to be a nail. There is more to say about the relationship between ethics and economics than modern economic imperialists might be willing to concede. On the other hand, there is also more to learn from economics than most practical philosophers -- and, as should be added here, many other social scientists -- may expect.
The moral science of Adam Smith and his contemporaries did not yet make a sharp distinction between philosophical and empirical questions as some philosophers do nowadays. The British Moralists in particular were interested in all questions of a moral science including the normative ones. This does not mean that they necessarily were blurring the distinction between normative and descriptive issues. At least since David Hume's famous remarks in the treatise many practical philosophers were well aware of the justificatory gap between is and ought (Singer 1973 gives a balanced view of the debate over is and ought). Nevertheless, the names of such philosopher economists like Jeremy Bentham, John Stuart Mill and Henry Sidgwick illustrate that disciplinary specialization within moral science was only rudimentarily developed until the end of the 19th century.
In some thirty years from the turn of the century to Lionel Robbins' essay on the nature and significance of economic science (1935), the situation almost completely changed. Robbins was preoccupied with making a sharp distinction between economics and ethics. He regarded it as essential for the progress of economics that certain classes of value judgements were banned from the field of economics. At the same time philosophers started to define their own subject as directly opposed to empirical sciences. What later on would be called "the linguistic turn" was well under its way (cf. Rorty 1970). Logical positivism and the thesis that value utterances neither had a definite meaning nor would lend themselves to rational justification was in its aggressive adolescent years.
As a result, moral philosophers as well as economists progressively regarded it as a professional vice rather than a virtue to discuss problems from an interdisciplinary point of view. Nowadays the situation has changed again, and almost completely so. Leading figures in economics as well as in normative ethics -- including normative political philosophy -- gain wide spread attention by applying methods of the one to questions of the other subject. A preference for interdisciplinarity has captured the minds of many who work in the fields of ethics and economics. But quite in line with the general outlook of economics some skeptical warnings about the chances of cooperation should be issued.
The relationship between given ends and scarce resources that have alternative uses is regarded as the only proper subject of normative economics by most modem economists. They thus follow Lionel Robbins' famous statement (cf. 1935, 16). Within economics as a science there is no clearance to argue about ultimate ends, aims, or values. One can know whether certain ends, aims, or values are there but the economist cannot scientifically scrutinize whether rightly or wrongly so.
A minority of ethical theorists would also accept that all rational justifications of norms are strictly hypothetical in the technical Kantian sense of that term. As Alisdair Maclntyre states in after virtue (1981, 52): "Reason is calculative; it can assess truths of fact and mathematical relations but nothing more. In the realm of practice it can speak only of means. About ends it must be silent".
Any such position can be classified as non-cognitivistic. "Moral scientists" who argue along the lines of the Hobbesian logic of social philosophy almost automatically are leaning towards non-cognitivism. Therefore, it comes as no surprise that modern normative economics clearly entertains a non-cognitivistic metaethics.
Contrary to that, any meta-ethical position that maintains that there exist non-hypothetical justifications of norms may be called cognitivistic. It is assumed that these norms are justified for the addressees of the justification independently of their own factual aims, ends, or values. To put it differently, categorical rather than merely hypothetical validity is claimed for at least some norms. Such a categorical justification claims to hold good not only relative to particular interests of particular individuals. It raises a universal claim independent of such contingencies as factual aims, ends, or values. Therefore, within a cognitivistic approach reason is not merely instrumental. This, is the majority view among ethical theorists but a minority view among economists.
How we look at the relationships between economics and ethics depends crucially on whether we accept non-cognitivism or cognitivism on the justificatory level of normative argument. If ethical cognitivism is true cognitive normative results of ethical theory that go beyond means -ends -relationships should be adaptable to economics like those of sociology or social psychology. Importing such results from ethics will not undermine the cognitive character of economics then. Further, it might even be asked why normative economics should be confined to instrumental or technological questions whereas the scrutiny of other cognitively accessible practical results is left to ethics. Such self-restraint would seem quite arbitrary, should these results incorporate intersubjective knowledge.
If, on the other hand, non-cognitivism holds good as a general meta-theoretical norm then a sharp distinction between economics and ethics on methodological grounds would not make much sense either. Then, rational discussion of practical questions even in ethics could not go beyond instrumentally valid hypothetical imperatives guiding the pursuit of given aims, ends, or values. There would be no ethical problem that could not be tackled by methods of standard normative economics.
Thus, in both cases, if non-cognitivism should be true as well as if cognitivism should be true, the claim of a principal methodological distinction between ethics and normative economics could not be maintained. As far as I know this important, elementary point is not explicitly acknowledged by most scholars who have studied the relationships of ethics and economics. Most of them assume that there is a wide methodological gap between the two subjects. This lead economists astray in their assessment of the relationship of economics and ethics. For instance, when Henry Hazlitt in his foundations of morality (1964), at least implicitly, came very close to the view proposed here and tried to discuss ethical issues within the framework of economics his efforts were, quite surprisingly, greeted by Frank Knight (1966) with the complaint that he took abstract economics as absolute ethics. But, according to the foregoing argument there either is no absolute ethics, then economic methods might well be appropriate to analyze all ethical problems that can be subject to rational discussion, or there is a legitimate claim of ethical theory to provide knowledge of categorical claims of absolute ethics, then these claims about absolute ethics should be incorporated like other parts of knowledge provided by scientific neighbors into economics.
Recently many theoreticians have tried to incorporate parts of "cognitivistic ethics" into economics. Notably Armatya K. Sen has argued in favor of bringing cognitivistic ethical claims back in into economics. Others, as for instance James M. Buchanan, have taken quite the opposite stance and tried to conquer some of the citadels of ethics with the gloomy methods of noncognitivism. The endeavors of these two eminent economists illustrate quite well some essential arguments and problems of the cognitivistic and the noncognitivistic view, respectively. To these views and in particular to the problems that still plague them I will now turn first. In a second step, I will then take sides with non-cognitivism and argue that the fundamental concerns that seem to motivate the cognitivists' rejection of non-cognitivism are built on doubtful empirical assumptions.
In his recent book on ethics and economics (1987), Sen distinguishes two roots of economics. There is an engineering approach to economics and an ethics-related tradition. The first starts from given aims, ends, or values and scrutinizes optimal ways to reach them. It is 11 concerned with primarily logistic issues rather than with ultimate ends and such questions as what may foster 'the good of man' or 'how should one live"' (4). The second, ethics-related origin of economics can be traced back to philosophical efforts to deal with preciscly these latter questions.
Sen claims that the traditional behavioral assumptions of economics and the self-interest based economic view of human motivation suffer from severe deficiencies. As far as descriptive or explanatory issues are concerned his argument evidently has point. It should be quite clear that the desire for their own pleasure or well-being hardly can explain all actions of human individuals. Of course, one could make a definitional truth out of this. But, as is well known since Bishop Butler's criticism of definitional egoism this would lead us nowhere. In analogy to the argument that in coping with social reality definitionally egoistic behavior has to be classified still as "egoistically egoistic" and "egoistically altruistic" behavior we would for instance have to introduce "pleasurable pleasure" and "unpleasurable pleasure", or, a notion of higher order wellbeing that could stem from a direct or lower order reduction of well-being.
Such ways of concept formation would be quite inconvenient. People commit actions that are reducing their own well-being all the time. Human motives are so diverse and so complex that no single dominant motive can account for behavior under all circumstances. Many modem economists would agree. Still they would argue that their modem concept of utility is neutral with respect to the selfishness or unselfishness of human motivation, and thus covers any motive whatsoever. Utility, they say, may be used as a neutral term naming whatever it is that individuals maximize.
This concept of utility, though not the term "representative utility", became popular among economists almost immediately after the publication of v. Neumann-Morgenstern's book on game theory and economic behavior (1944). Not only then some theoreticians were mislead by the use of the term utility. They thought that the Ramsey -v. Neumann-Morgenstern usage of the term is related to the qualitative usage of utility in former time. But this is not true. In particular no individual acts because he wants to gain representative utility. Representative utility has no desirable qualities. It is not directly related to well-being and no reason for action at all. Individuals showing a certain degree of consistency in their choices merely behave as if they are maximizing something. But this "something" is a motivational "nothing". Maximizing utility is not a motive of acts but a side-effect of behavior motivated by some different end or other.
The utility function and the notion of utility maximization is only a convenient shorthand (cf. Elster 1979). As Sen is well aware maximization of representative utility as such does not explain behavior. It merely represents consistent orderings that have to be explained otherwise. As far as empirical problems of explaining and predicting social reality are concerned, one cannot avoid to specify the arguments of the utility function. The empirical content of the economic theory of human behavior, if there is any, rests on this specification. However, self-interest as the standard economic assumption will not, at least not under all circumstances, provide a convincing specification of utility.
Sen argues that economists have unduly neglected the ethics-related origin of economics. Staying firmly within this tradition they could have avoided the extreme and mistaken views of either taking self-interest or unspecified utility as universal explanation of human behavior. A moral science, unbiased by preconceived ideas, will acknowledge that all sorts of motives may affect human behavior including moral considerations and even ethical Or economic theories. Opinions, convictions, accepted rules etc. may as a matter of fact exert their influence on human behavior. The idealized model of pure self-interest will be approximately valid only under circumstances which more or less enforce that human beings behave as if motivated exclusively by selfish motives.
As far as the phenomenology of ethics is concerned, Sen is undoubtedly right. Ethical motives do play some role in the determination of behavior. However, Sen is not sufficiently aware of the fact that his criticism only applies to the economic approach to descriptive and explanatory questions. It has merely indirect bearing on the epistemological issue of the limits of justificatory normative argument. Therefore the Hobbesian limits to normative argument, which go back to the engineering-related origins of economics, remain untouched by Sen's criticism.
Of course, Sen could raise the cognitivistic claim here that on the justificatory level reason is not merely instrumental, too. But, this claim cannot be directly supported by empirical evidence. Besides stating the fact that individuals believe that they can ultimately or categorically know right from wrong in matters practical Sen does not offer much in favor of cognitivism as a meta-theoretical position. Referring to the prevalent beliefs or intuitions of people is not very compelling if we can develop a convincing view of normative argument without assuming that the cognitivistic intuitions are true. For, if such an account is possible, the non-cognitivist theory has less epistemological liabilities of doubtful credentials than a cognitivist one.
Such an alternative explanation is exactly what noncognitivistic moral scientists from David Hume (cf. 1948, 1985) to John Mackie (cf. 1977, 1980) tried to provide. At least to my mind, these efforts have been quite successful. One can very well subscribe to subjectivism and non-cognitivism as far as rational justification is concerned and at the same time acknowledge that as a matter of fact there are many human motives that go beyond self-interest. Human beings may even observe rules provided by ethical theories simply because they accept these rules. In their choices they can go beyond considerations of expediency. But, the fact that man can follow rules without considering the consequences in every instance of choice does, of course, in no way entail that the rules themselves have to be justified independently of consequences. It leaves as an open question whether or not a convincing subjectivistic or non-cognitivistic theory of the legitimation of rules starting from given individual interests is possible.
The fact that human behavioral technology contains the faculty of rule following as such does not theoretically limit the scope of admissible de facto interests of the addressees of normative arguments. The addressee of normative argument is the sovereign setting the goals and the theoretician is the engineer telling the sovereign how to reach them. However, this engineering-related subjectivistic view of ethical argument faces its own problems. To these I will now turn discussing the theories of James Buchanan as one of the leading subjectivistic philosopher economists of our time.
James Buchanan has been a long time critic of taking the model of rational economic man to its Hobbesian extreme in explanatory contexts. Not all actions can be explained by the Hobbesian motive of individual selfinterest. However, in normative as opposed to descriptive and explanatory analysis Buchanan starts from clearly Hobbesian premises. Within justifications of normative proposals the aims, ends or values of individuals have to be taken as what they are. Ultimately they are there and that is all to it. We can rationally discuss how to pursue given interests but we cannot rationally criticize ultimate ends, aims or values. Therefore, from a meta-ethical point of view Buchanan clearly should be classified as a non-cognitivist.
Despite of this subjectivist premise, Buchanan takes strong norm ativ e -theoretical stances in favor of certain social institutions. The Macchiavellian enterprise of developing prudential recommendations that may prove useful in a struggle for power falls short of what Buchanan hopes to accomplish. However, non-cognitivism does not seem to provide clearance to go any further than Macchiavellism would allow for. Thus, within Buchanan's writings we observe the "classical" tension between a non-cognitivistic meta-ethical starting point and quite outreaching aims of norm justification.
If, for instance, different groups of people do have conflicting ultimate ends, aims, or values then a noncognitivistic ethical theory cannot resolve the conflict. Normative theory reduces to an analysis of optimal strategies to pursue particular interests of groups of individuals. It boils down to a bargaining theory analyzing the status quo of the game of life and the threat points involved.
Such a strictly particularized ethical theory is truly Hobbesian in spirit. Its recommendations have normative force insofar only as they contribute to the ability or "present means" of their addressees to "obtain some future apparent good" -- which incidentally is Hobbes' definition of "power" (cf. Hobbes 1985, chap. X opening clause). They thus serve the same functions as the Hobbesian laws of nature. Normative analysis itself ultimately finds its justification in that it enhances the potential to reach the ultimate aims, ends, or values. It is moored to strictly particularized, contingent interests and it is itself subservient to "power" in the Hobbesian sense of the term.
Of course, Hobbes himself already wanted to go beyond this. He tried to provide some form of universalization superceding particular and merely contingent individual inclinations. To that purpose he introduced the notion of agreement in an original contract. The agreement of all individuals and their promise of obedience is meant to do the trick. However, since the days of Hobbes it seems doubtful how contractarian notions of universal agreement can be made compatible with the non-cognitivistic meta-theoretical premise of the Hobbesian enterprise itself.
Hobbes tried to overcome particularism by stipulating that every individual in the state of nature could be killed by any other one in that state. It is in no way implausible that from this a common interest of all in the foundation of a civil society might be derived. Nevertheless, even then the joint efforts of some subgroup might and presumably will be sufficient to bring about a civil society while simply forcing the rest into it. Thus, the agreement of all, though a conceivable possibility under conditions as those prevailing in the state of nature, will as a matter of fact not be necessary even then.
The state of nature of atomistically interacting individuals is purely counterfactual. It is not only true that it does not exist. It is also no real alternative that could be of any importance in determining the threat points of the individuals in the factual status quo. Therefore in the status quo of a non-atomistic real society it is completely implausible that the assent of virtually all would be of interest for powerful groups of individuals. If such groups can get their way anyway they do not have good reason to go beyond their particular interests.
It is exactly at this point that Buchanan's contractarian political ethics seems to run into severe trouble too. He repeatedly states the fundamental role of agreement within his theory. But, why is agreement important? From a purely non-cognitivistic point of view, a person would have to strive for the agreement of other individuals only insofar as this itself is helping him to reach his own goals. This may well be the case for some empirical reason or other. But it is far from obvious, even completely unlikely, that such reasons will regularly prevail in status quo.
Buchanan's theory is still plagued by the traditional problems of a Hobbesian "ethics of large numbers" or political ethics. As his predecessor he has not made much progress in defending his claim that universal agreement is to be taken as a source of justification. What could he say to those who, as a matter of fact, do not accept agreement as a source of justification, who simply neglect the opinion of other individuals unless its consideration is instrumental for reaching certain "given" goals of their own? Within a non-cognitivistic approach universal agreement cannot be taken as the ultimate source of justification. According to the very premise of non-cognitivism this role is reserved for purely particular or individual aims, ends, or values. To make it compatible with non-cognitivism, the role ascribed to universal agreement must therefore be reinterpreted.
There seem to be two principally distinct ways of reinterpretation. First, agreement may be an intermediate goal. To reach agreement can be instrumental in reaching some given ends, aims, or values of a higher order. Second, one might argue that public opinion formation already incorporates some requirement of universalizability as a side constraint imposed on any argument. Only those arguments can convince the general public that can at least conceivably gain universal acceptance. Both interpretations raise interesting questions to which I will turn now.
Following the first line of argument Buchanan could try to persuade people that it is in their personal interest to pursue the goal of institutionalizing the requirement of universal or some weaker form of agreement. Such an argument suggests that "we" should develop social institutions that require the agreement of everyone or almost everyone because such institutions would strengthen the position of individuals. The unanimity rule as a basic constitutional norm might serve that purpose. It incorporates universal agreement or the conceivability of universal agreement as a criterion of post- constitutional choices but it is itself not justified by agreement in constitutional choice.
The argument in favor of agreement as a goal is valid under non-cognitivistic premises only for those addressees who as a matter of fact do share appropriate ultimate aims, ends, or values. There is still no guarantee that all will agree on the ultimate level of argument and if they ultimately disagree there are no further rational arguments. There even seems to be no guarantee that efforts of persuasion will eventually induce convergent opinions rather than cementing disagreement. This leads to an important class of problems that are attached to the interpretation of universal agreement or some related requirement of universalizability as a side constraint.
This line of argument has been traditionally pursued by many scholars who wanted to do justice to the skeptical objections of subjectivist philosophers and at the same time sustain some form of practical argument transcending a mere technology of getting one's way in the pursuit of particular interests. Some have argued, for instance, that the meta-theorctical tenets of subjectivism logically entail the norm that all subjective values have to be respected equally. But, this is clearly mistaken. Without violating the principle that one cannot derive an ought from an is the metatheoretical tenets of subjectivism would have to be interpreted as meta-theoretical prescriptions rather than as descriptive epistemological claims. This is far from self-evident. Further, and more importantly, the argument that the subjectivist has good reason to exert an externality on individuals who do not share his own subjective aims, ends, or values applies again. His metatheoretical premises do not logically entail that he is bound to respect the spheres of other people.
Still, even if there is no logical relationship between non-cognitivism and some form of universal respect of the spheres of other individuals there may be some other relationship. Some, in particular some well known German scholars (e.g. Habermas and Apel), have argued that there is a kind of transcendental logic of argumentation that cannot be violated without getting into contradictions. Engaging in the very activity of arguing, they say, we have already acknowledged premises that allow for the derivation of some form of the principle of universalization.
To put it differently this amounts to the statement that engaging in a game we must accept the rules of the game. That this argument is not very convincing can be easily seen if one makes a distinction between taking an external and an internal point of view to the rules of a game. Individuals who take an internal point of view indeed have accepted the rules, for what reasons ever, and will subsequently apply them as standards of correct behavior. But, nobody "must" take an internal point of view. A participant can very well plan his moves in a game from an external point of view. He does not accept the rules of preexisting institutional structures but merely exploits his knowledge of the institutional facts in predicting what is going to happen within the game. The game of arguing is no exception to this. Arguments may be strategic moves that are planned from a purely instrumental external point of view like those of the attorney at court or the Sophist in politics. In particular there are no "logical" reasons why individuals must take an internal point of view and therefore there are no logical reasons why they should be honest participants in this sense (cf. on understanding as purposeful action also Baurmann 1985).
However, there is a more plausible interpretation of must" that is closer to empirical necessity than fullfledged transcendental pragmatics would suggest. The make up of the human mind may be such that we simply cannot look at other people as purely natural phenomena without a special effort to do so. Built into our view of the world there may be a notion of community with other human beings that we cannot evade and we cannot avoid to become aware of it. As P. F. Strawson (1974) has put it in his freedom and resentment (9): "We look with an objective eye on the compulsive behaviour of the neurotic or the tiresome behaviour of a very young child, thinking in terms of treatment or training. But we can sometimes look with something like the same eye on the behaviour of the normal and the mature. We have this resource and can sometimes use it: as a refuge, say, from the strains of involvement; or as an aid to policy; or simply out of intellectual curiosity. Being human, we cannot, in the normal case, do this for long, or altogether".
If this is interpreted as an empirical argument I would admit that it is a plausible one. The facts that are described by Strawson may indeed narrow the scope of potentially successful arguments. This becomes particularly clear if public ethical discussion is put to scrutiny. We are not talking about reality then but rather talking about talking about reality. There are conventions of normative "discourse" that seem to require universalizability. As long as these institutional facts do exist they are an important element in the status quo. If individuals may be expected to look at the institutional norms from an internal point of view, sometimes or even most of the time, then this is an important fact too. If the success of arguments depends on observing such factually existing and accepted rules then even the Sophist will have to frame his arguments in a universal manner.
It is a definite advantage of non-cognitivistic positions like that of Buchanan that they draw our attention to these empirical phenomena. From these phenomena the most interesting and troublesome problems of a non-cognitivistic ethical theory of moral institutions arise. These problems are basically problems of motivation. Theories of what is right as such do not solve the problem of motivation. We may know what is the good thing to do and still might not have any incentive to act accordingly. (This holds good, by the way, for categorical as well as for hypothetical norms.)
Some cognitivists at least tacitly seem to assume that only the belief in the categorical validity of a norm could socially support it. But, the assumption that the only motivational effective reason for entertaining certain opinions is their objective categorical validity is quite a strong one and far from obvious. We have good reason to praise and blame other individuals if we know that to do so will serve our long-term interests. This is true even if we should know that these activities cannot be based on a claim of objective validity. If the cognitivist assumes that only objective knowledge of right and wrong can provide a good reason to blame or praise other individuals and thus to support moral institutions then this comes dangerously close to a petitio principii.
In any case the cognitivist cannot avoid to look at the incentive structures that will govern individual behavior in the real world. In this context he will face roughly the same problems as the non-cognitivist and will have to use roughly the same economic and game theoretic tools. From this, interesting relationships between economic and ethical theory arise.
Economic and political theory tell us many sad stories about public goods' provision. This wisdom may be easily applied to the moral order as a public good. Even if it is in the interest of all that certain moral conventions exist -- and even if it is right in itself -- this does as such not provide an incentive for any individual actor so convinced to participate in the process of producing and enforcing it. Individuals who agree that the institutionalized existence of a public convention serves their own long term interests can at the same time try to take a free ride in the "production" process. From the individual point of view there is nothing irrational about such behavior though as a result of individually rational behavior every participant may end up worse than without it.
A cognitivistically minded believer in collective rationality could maintain here that the observed clash between the dominance and the Pareto principle of choice exhibits a contradiction between individual and collective rationality. The cognitivist may argue that it is simply right that collective results should be Pareto optimal -- and quite ironically many economists seem to share that view. But a strict Hobbesian individualist and subjectivist cannot do so coherently. For him there simply are neither collective rationality nor normative principles that are valid independently of individual interests. On the other hand, it is obviously true that every individual would be better off in terms of his own interests if everyone would act on behalf of a norm requiring participation in the enforcement process without exception.
Hobbes would have said that individuals may have an interest "in foro interno" that certain institutions exist and at the same time have no incentive "in foro externo" to act accordingly. This Hobbesian distinction is not merely of historical interest. On the contrary, recent discussions seem to show an enhanced awareness of the relevance of the distinction between fundamentally different forms of interests -- those related to opinion formation and those related to the determination of actions. Viktor Vanberg and James Buchanan, for instance, introduced the two concepts of "constitutional" and "action or operational interests" (c. f. 1988). Constitutional interests apply to our considerations of what we should hope to become true. They are basically opinions about what kinds of constitutional rules would serve best our given individual interests. Action or operational interests contrary to that are confined to the exigencies of the moment and to within rule choices. They take interests and rules or institutions as given.
On the constitutional level of the moral as well as of the legal order individuals may entertain the opinion that they are or would be better off if certain institutions exist. But, as a matter of fact and very trivially so a single isolated individual is almost never in a position to do anything significant about the existence of social institutions. Individual actions are insignificant for the provision of these fundamental public goods. This veil of insignificance (cf. on this Kliemt 1986 b) cuts off selective incentives that could affect action or operational interests. In particular the opinion that everybody would be better off if everybody behaved in a certain way or participated in the enforcement of certain rules requiring a certain way of behavior is not an operational interest warranting individual action.
Taking into account the last remarks it seems hardly conceivable that an interest based moral theory might get off the ground at all and exert some motivational influence on individual action and thus on social reality. But this problem vanishes if we carefully distinguish between the descriptive and the normative level of argument. If as a matter of fact people do act on behalf of their convictions sometimes then it should be expected that what they regard as their interest will influence behavior too. There can be -- as David Hume was already aware -- not only an opinion of right but also an opinion of interest (cf. Hume 1985, essay IV). Under certain circumstances especially if opportunity costs are not too high human individuals may act upon these opinions. Then their behavior is not motivated directly by the exigencies of choice situations anymore. It is motivated by their opinion on what they regard as their own long term interests.
Exactly here the discussion of constitutional interests enters the picture. This discussion starts from the personal interests of the addressee of the argument. But it does not presume anymore that the addressee has an operational or action interest. It merely shapes opinion in that it shows -- as Hobbes has said about his "laws of nature" in the Leviathan -- what the individual should "hope" to become true.
Contrary to the logic of the Hobbesian premises of self-interested behavior people may be expected to act sometimes simply on their opinion of interest even if the opinion does not echo an underlying action or operational interest. Sen is clearly right in stressing against some extreme forms of economics that people are motivated by their ethical convictions or opinions. If action or operational interests are not too strong opinion may overcome adverse incentives of a situation. This is of particular relevance as far as such activities like praising and blaming other individuals for their behavior are concerned. Because these activities involve merely low costs they may be expected to occur regularly even if they are opposed to operational or action interests (for the "low cost" condition in a more cognitivistic utilitarian setting cf. also Selten 1986).
It can be seen from the foregoing observations how an interest based moral theory if it is institutionalized as an ethical conviction system can at least in principle support a complementary institutional structure providing action interests. This will be the case if what we theoretically regard as our interests will fit in with the status quo. If for instance our public theory of what serves best individual constitutional interests is a contractarian one then the counterfactual argument that the status quo could conceivably meet general agreement is of factual force for those who entertain appropriate ethical convictions. It shapes our opinion and in certain situations at least some individuals will act on behalf of this opinion.
If, as David Hume said "it is on opinion only that government is founded" (Hume 1985, 32) it should be clear by now that moral and political institutions can be based on opinion of interest. Again, from the point of view of an interest based moral theory we may still ask whether or not they should be based on opinion of interest. Many cognitivistically minded theoreticians implicitly suggest that it would serve our interests better if moral and political institutions could be based on opinion of right and not merely on opinion of interest. This claim does not affect their own position but it could affect the acceptability of an interest based moral theory as far as such a theory must be based on opinion instead of direct interests. On this I would only allude again to the introductory remarks of the paper and stipulate that as far as historical evidence reaches all pernicious superstitions and enthusiasms of former and present times have been based on opinion of right as opposed to opinion of interest. Opinion of interest will in general support a more sober minded temper.
Even if the workings of a society cannot be understood completely in terms of direct or action interests of rational economic man, even if "it is only on opinion that all government is founded" opinion need not be necessarily, as cognitivists usually claim, opinion of right. Shifting attention from action to constitutional interests it should be clear that opinion of interest may well provide the foundation required.
In the latter case, the theory best adapted to shaping our ethical views of right and wrong would be economic theory in general and constitutional economics in particular. Within such an interest centered view of the world there would not be much need for the information created by traditional cognitivistic ethical theory. Further, if the conviction system of a society as a matter of fact were interest based then we would quite naturally end up with economic theory as the core element of the mores or "working ethics" of that society. Within an interest based every-day-view of the world economic theory is itself a central part of any adequate ethical theory or at least an adequate method of any ethical theory.
Of course, cognitivistically minded ethical theorists as well as adherents of religious truths raise objections against this world view. They usually pursue two lines of argument -- and most of the time simultaneously. On the one hand, they argue that cognitivism and the theses created within cognitivistic ethical theories are simply true. To this claim they add the further one that the public conviction system should be based on "the truth". For convenience the combination of these two tenets -- a descriptive and a prescriptive one -- may be called the "truth claim" for it is raised under the premise that cognitivism is true. On the other hand, cognitivists stipulate that even if cognitivism would not be true we would be better off should we accept cognitivism and some of its allegedly categorical ethical truths. This may be called the "pragmatic claim". It is raised for purely practical reasons independently of the truth or falsehood of cognitivism.
An analogous strategy is well known from discussions about religion. It is first argued that certain religious convictions are true and that we therefore should accept them. To this it is added then as a fall back position that even if religion were not true its institutionalized variants should be accepted and supported because of their usefulness. It is claimed that religious conviction systems serve important functions in society that could not be accomplished if people would simply mind their interests and shape their preferences accordingly.
Both cognitivists and adherents of religion seem to tru st that they are on firm ground with their claim to a superior social usefulness of their own as compared to non-cognitivistic systems of argument. To them as to many people it seems to be almost self-evident that all individuals would be better off should everybody share some strong views about ethical knowledge and accept some cognitivistic code or other as a basic conviction system. From this it seems to follow too that those who take an interest in the common good should feel morally obliged to fight against non-cognitivism.
The calculative mode of thinking, the purely instrumental reasoning in terms of means ends relationships itself seems to be morally objectionable because its probable consequences seem to be repelling. On the level of theoretical discussion such a defense of higher "ethical" values against the more mundane leads many ethical theorists to a rejection of economic modes of thinking. They seem to think that economic reasoning as such will tentatively erode a public capital of higher moral values. In particular they suggest at least implicitly that if for instance all or most individuals would merely pursue their interests and would answer questions about how and under what kind of institutions they should live according to economic reasoning then everyone including the non-cognitivists would be worse off.
It cannot be excluded on a priori grounds that cognitivists are right about the consequences of cognitivistic and non-cognitivistic conviction systems and about the probable effects of economic modes of thought. However, it should be noted carefully that their argument about consequences is not an epistemological one. Neither is it a typical cognitivistic claim. Instead of this, cognitivists are attacking here non-cognitivists on their own territory with arguments that are acceptable from the non-cognitivistic point of view too. They are appealing to certain aims, ends, or values or for short interests that human individuals as a matter of fact share. If evaluated in terms of these interests, they claim, non-cognitivism fares worse than cognitivism and thus should be rejected by the non-cognitivist himself.
Of course, the non-cognitivist has to admit that from the point of view of an interest based moral theory we may indeed ask whether or not moral, social, or political institutions should be based on opinion of interest. The non-cognitivist cannot state that he is dealing with questions of right and wrong that are beyond human decision. For, his meta-ethical tenets oblige him to concede that there is clearance for theory choice on practical grounds. He should ask in fact whether it would not be better if people should believe in the rightness or wrongness of certain acts in terms of their interests.
Would not in particular the proliferation of economic modes of thought indeed lead to an erosion of many public goods?
An affirmative answer to these questions could affect the acceptability of an interest based moral theory. However, the answer must be supported with empirical evidence. Regrettably our empirical theories of public opinion formation and the political effects of alternative forms of public conviction systems are in a rudimentary state still. Hard evidence could be hardly found. Still, we do have some clues from everyday experience. And it is to these clues that cognitivists in general allude. They seem to assume that the clues are telling a story favorable to their own position. At first sight this seems to be almost self-evident. Even if people are motivated not only by action interests but also by their convictions or opinions then socially useful behavior that goes beyond self interest can be expected only if they are convinced that such kinds of behavior should be performed.
Cognitivists can point to the acts of saints and heros here. Almost all of these individuals seem to hold strong views of what is right and what is wrong. This point may easily be extended to more pedestrian human beings too. If obstacles to collective action have to be overcome in bringing about certain desirable social states it seems to be clearly helpful if individuals are convinced of the rightness of participation, etc.
However, taking a closer look it can be easily seen that the evidence is not as clear as cognitivists seem to assume. First, it should be noted again that non-cognitivism does not logically exclude that human individuals as a matter of fact share some higher ideals. As the example of many non-cognitivists shows this is practically or psychologically viable too. Further it is far from clear that these attitudes of non-cognitivists could only flourish in a cognitivistic environment.
Taking into account that opinion of interest may motivate individuals as well as opinion of right it should be noted that opinion of interest may well serve the functions of an ethical code in education for instance. An analysis of constitutional interests in small group interaction would easily show that the development of certain dispositions that overrule opportunity taking behavior in expedient choices can be expected to be favorable not only to the group of other individuals but to the actor himself. Using some elementary facts of psychology it could be shown too how effects on large scale interaction may arise because individuals cannot switch off their disposition as seems fit.
Second, non-cognitivistic theories do not simply state that individuals should pursue their interests whatever they are. Instead of this the most convincing variants of non-cognitivism are, as has been indicated before, theories about moral institutions. Moral institutions are treated roughly in the same manner as legal ones and it is asked under what kinds of factual moral institutions rational individuals should want to live. This type of analysis is a natural extension of constitutional economics and is therefore subject to the objection of particularism extensively discussed before. In the last resort it can only hope to persuade as many people as possible. But, even in this respect cognitivism is not better off. If it comes down from the clouds of pure ethical theory its practical success as an institutionalized ethical code applied by earthly individuals depends on persuasion too -- though, admittedly, different persuasive arguments may be used.
Third, and most importantly, it should be observed that the cognitivistic evaluation of everyday evidence seems to blur the distinction between the meta-ethical position of cognitivism and material cognitivistic ethical theories (cf. also Hoerster 1982). There are many diverging cognitivistic ethical theories on the material level of ethical argument and there are many noncognitivistic ones. It may be true that some cognitivistic theories after transformation into ethical codes or moral institutions of a society will have better consequences than some non-cognitivistic ally founded codes. In other cases the comparison might lead to the opposite conclusion. But the question under discussion presently is whether non-cognitivism or cognitivism as general meta-theoretical premises (after transformation into public convictions) will lead to better consequences. To be more specific the question is: Is it better, in terms of consequences, if individuals in a society develop skeptical attitudes compatible with noncognitivism or is it better if they will accept cognitivistic reasons too?
In the first case individuals will accept only instrumental reasoning. They will look at social institutions from the point of view of social engineering and economic reasoning in the broad sense of the term will provide the logic of normative argument. All kinds of argument that go beyond means-ends relationships (on a variety of levels of arginnent) will be excluded on a priori grounds. In the second case individuals will be ready to go beyond instrumental reasoning. They will not exclude any cognitivistic claims on a priori grounds though they will prefer some theories to others.
Of course, the moral entrepreneur has any reason to support the latter attitude. And it comes as no surprise that most moral entrepreneurs -and, for that matter preachers -- regard skepticism as the worst of all vices. It is their "vested interest" that non-cognitivism is stuck with a bad reputation because their cognitivistic claims cannot reach the skeptic part of the audience anymore.
Due to the incentive structure of moral entrepreneurs skepticism perhaps might be a looser. Nevertheless, we might try to form our opinion of interest on whether we should want it to loose or not. On this I can only allude again to the introductory remarks of the paper and stipulate that as far as historical evidence reaches all pernicious superstitions and enthusiasms of former and present times have been based on opinion of right as opposed to opinion of interest. If it is really true that meta-theoretical non-cognitivism, as cognitivists are ready to admit, will as a matter of fact nourish a skeptical attitude in society then from the point of view of its practical consequences non-cognitivism should be given a chance. As far as the influence of public opinion systems is concerned things can hardly become worse and there seems to be a fair chance that skepticism if only more widely spread will prove clearly superior to cognitivism. Opinion of interest will in general support a more sober minded temper and thus provide the most valuable public good for a large pluralistic society.
The basic value premises ingrained in economic modes of thought not only seem to be ethically unobjectionable from this point of view. They may be regarded as conducive to the public good too. From a consequential ist and non-cognitivist ethical point of view the economic attitude towards ethical problems seems morally superior to the classical ethical attitude.
<< References >>
|
Kliemt, Hartmut: Papers on Buchanan and related Subjects. – Munich (Accedo Verlagsgesellschaft), 1990: p. 9 – 36. (Studies in Economics and Social Science (SESS), Vol. 1 (1990)) |
[Top]