The notion that the development of society is not the result of rational, goal-oriented action is a well-known theme in the history of political thought. Instead, a common belief is that the history of humanity not only is caused by unintended consequences, but also that it ought to be. This idea has been advocated by such various theorists as Hegel, Marx, Hume, Burke, Smith, Merton, Popper and Hayek (Vernon 1979). In this view the side effects of human action are thought to be positive and advantageous. The lack of goal-oriented rationality in politics does not prevent history from being rational. On the contrary, it is seen as a prerequisite for history to be rational.

Thus, evolutionary optimists have usually placed reason outside of the human intellect - in God, History, Nature or Tradition. Reason is something is supposed to come into being through a process or a system where individual human intellect is an integral part, and this supra-individual is thought to be disturbed by human attempts to influence history by rational, goal-oriented action. Various terms for this supra-

Resumé

The theme addressed is the criticism of constructivism and rationalism in politics as presented in the political theory of F. A. Hayek. Hayek's thesis is that goal-directed rationality in politics is counter-final. The main argument of this article is that on closer examination Hayek appears to be a constructivist himself. The purpose of Hayek's anti-rationalism is primarily ideological and instrumental. Hayek wants to induce certain anti-rationalistic beliefs because he believes rationalism has bad consequences. Yet this very project can be considered a case of goaldirected Furthermore, the argument is that Hayek mixes abstract philosophical doctrine with empirical theory. In his critique of constructivism, Hayek confuses two notions of 'rationality'. On the one hand he argues against epistemological rationalism associated with Descartes, and on the other he argues against the subjectivistic, action-oriented notion of 'rationality' associated with the idea of 'economic man'. Arguments against Cartesian rationalism not, however, imply the impossibility of goal-directed rationalism in politics. The so-called information problem cannot be solved at an epistemological level. It is a practical problem, which Hayek has certain ideas how to solve. The argument here, therefore, is that Hayek's thesis of 'tacit knowledge' is not an expression of skepticism as is often believed.

---

*I thank anonymous reviewers for helpful comments on this article.

individual reason have been coined in the history of political ideas, terms such as 'the wisdom of history', 'the cunning of reason' and 'the invisible hand'. These expressions have not only been used metaphorically; in the teleological tradition, reason has had an independent existence as an Essence (Aristotle), a God (Leibniz), or a World Spirit (Hegel) (Woodfield 1976).

Among secular evolutionary optimists, by comparison, these metaphysical have been replaced by natural selection, functionalism, and the invisible-hand. Reason in history has become synonymous with adaptation and fitness (Simon 1983, 37 ff.). This secularization, however, does not change the fact that the good society is presumed to lie outside of human control. The responsibility for evolution does not fall upon humans as political actors, but rather on external forces. The foremost representatives of this secularized evolutionary optimism belong to the tradition of so-called spontaneous order which includes Hume, Mandeville, Smith and Ferguson (Barry 1982; Hamowy 1987). Among these theorists, optimism on humanity's long-term development is coupled with an equally strong pessimism about humanity's ability to guide social processes rationally. A. Hayek is the social scientist in modern times who has most enthusiastically supported the thesis that attempts to plan and control a society lead to negative consequences - i.e. serfdom (Hayek 1944). Progress its very nature', writes Hayek (1960, 41), 'cannot be planned'. In contrast, refraining from attempts to control society induces positive consequences.

The objective of this article is to examine the arguments behind such claims more closely. It can, in my view, be shown that Hayek's criticism of constructivism is based on a theory which in itself is constructivistic. For this purpose, Hayek's normative theory and his theory of spontaneous orders are briefly considered. In particular, Hayek's thesis of tacit knowledge, springs from his theory of spontaneous orders, is examined in greater detail. It is subsequently shown that Hayek, in arguing for this thesis, mixes abstract philosophical doctrine with empirical theory. My conclusion, therefore, is that Hayek's philosophical arguments, which concern questions, are ideological and instrumental. The thesis of tacit knowledge is basically empirical and part of a constructivist project.

Hayek's Struggle Against Constructivism

The struggle against constructivism' is a continuous theme in Hayek's works. He gives this label to many political theories which share a common belief in the possibility of constructing social institutions with the aim of reaching certain goals (Hayek 1978, 3). The quintessence of constructivism

is best expressed by Voltaire: 'if you want good laws, burn those you have and make new ones' (Hayek 1973, 25). Among constructivists, Hayek includes contract theorists such as Hobbes, the French enlightenment philosophers such as Descartes and Voltaire, and the English utilitarians such as Bentham and J. S. Mill. Even the French so-called laissez-faire economists should be classified as constructivists according to Hayek (1960, 60). A market cannot be created through goal-oriented measures; instead, Hayek claims that it must grow spontaneously.¹

In Hayek's view (1973, 34), constructivism is irrational in a deeper sense, since it recommends an incorrect path to a certain goal. It is, furthermore, 'counterfinal', because it leads to results which the actors do not desire (Hayek 1973, 62 f.). The opposite of constructivism, on the other hand, is anti-rational evolutionism. This tradition stems from the Scottish enlightenment Hume as the forerunner (Hayek 1967, 106-121). Antirationalism, to Hayek, represents a deeper form of wisdom, because it recognizes the limits of human rationality (Hayek 1967, 82-95).

At this point it is reasonable to ask just what practical conclusions emerge from a claim about the limits of human rationality? One might think that the quintessence of anti-rationalism, somewhat simplified, can be formulated as the reversal of Voltaire's view on laws: 'If you want good laws, keep those which exist'. This advice, however, is an expression for the very same goal-oriented rationality as in Voltaire's recommendation. Instead, a genuine anti-rationalist would presumably exclaim: 'Do not worry if the laws are good or not - follow them anyway!' But is perhaps this recommendation in fact rational? To answer this question one must of course first decide what is meant by 'rational'. According to a 'thin' theory of rationality, an action is rational if the actor believes that the action leads to goal-fulfillment (Elster 1985). This notion of 'rationality' is subjectivistic and action oriented. One instance of such a subjectivistic theory of rationality the neo-classical idea of 'economic man', who seeks to 'maximize expected utility'.

Hayek, however, is an outspoken opponent of the idea of 'economic man', a concept which he calls a 'celebrated figment introduced by J. S. Mill' (Hayek 1960, 71). He thinks this conception of 'rationality' is wrong both normatively and empirically. But his denial of a goal-oriented model of rationality does not imply that he himself, as a political theorist, does not act according to this model. His own recommendation to political actors (to follow an 'anti-rational', rule-governed way of choosing measures) may well be interpreted as a case of rationality in accordance with the idea of 'economic man' in its own right; i.e. as a case of maximization of values defined in his normative theory.

In his polemic against rationalism in politics, Hayek also makes use
of another notion of 'rationality'. He argues against an epistemological

rationalism associated with Descartes. This kind of rationalism concerns the epistemic foundation of knowledge, not the choice of action from subjective values and beliefs. Yet there is no self-evident logical connection between these two different notions of 'rationality'. One can be an antirationalist the epistemological level, and at the same time believe in the possibility of rational politics. There is, however, a political and ideological connection between epistemological rationalism and political rationalism. Therefore, and this is the main theme of this article, it is important for Hayek to rejeetit. But this rejection is part of a normative project which is a case of goal-directed rationality in its own right, i.e. of constructivism.

For the sake of further argument, 'rationality' will hereafter be taken to mean (1) that an actor chooses the action he or she believes will lead to goal-fulfillment, and (2) that the actor in question has well-founded reasons for the belief that an action will lead to goal-fulfillment. The second condition implies, in the case of Hayek, that metaphysical reasons are considered non-rational, whereas empirical or scientific reasons are considered ²

Hayek's Normative Theory

Hayek is not a neutral scientist who only has the ambition of explaining the development of society. In common with teleological philosophers, he has a conception of purpose and a goal of development. He also shares the teleologician's ambivalence and antipathy for a purely normative discussion. and 'Ought' are combined in an evolutionary imperative.

But although Hayek sometimes slips into a teleological form of expression (Hayek 1979a, 160, 163), his normative theory is based on a value-relativist tradition (Gray 1984, 8 f.). There are no objective aims for the development of civilization (Hayek 1988, 27). What is good and bad is determined by our subjective values. Hayek presumes that we want to satisfy as many of our preferences as possible. A good society is characterized by a high chance for an unknown individual being able to fulfill her life's goal. Or as he puts it in a somewhat contractarian formulation: '. . . the best society would be that in which we would prefer to place our children if we knew that their position in it would be determined by lot' (Hayek 1976, 132). Hayek's normative theory, in short, can be described as a form of indirect, probabilistic utilitarianism (Gray 1984, 104).³

In his last work, The Fatal Conceit (1988), however, Hayek emphasizes the survival of a social system more than its capability of satisfying preferences. value of a social institution is determined by its effects on the group's reproductive ability. Naturally, as many critics have pointed out, a conflict can arise between the survival of a social system and individual

freedom (Barry 1982; Vaughn 1984). There is no guarantee that liberal institutions are those with the greatest prospects of survival. It is possible that a number of different social systems have the same value of survival. As James Buchanan (1982, 7) has appropriately observed,'. .. there exists a very large "cushion" between where we are and where we might be pushed to before species survival might be threatened'.

This inconsistency in Hayek's normative theory disappears somewhat if one takes into consideration that he is a pure consequentialist. It is the total, long-term probability of satisfying preferences that is to be maximized. This implies that all future consequences are relevant for a moral judgement of an institution. If an institution actually decreases a group's reproductive capability, the likelihood of satisfying preferences in the future will also decrease. Hayek has an aggregative, consequentialist perspective of individual in other words, which allows him to legitimize even illiberal institutions.⁴ This is clearly expressed in a speech given to libertarian where he warns against general tolerance for culturally deviating groups in a society (Hayek 1987). According to Hayek, there is no right to freedom. In this respect Hayek has the same skeptical and consequentialist view on rights as Hume (Hayek 1967, 106 ff.).

Hayek's normative theory implies that he looks at institutions from an instrumental perspective. The rationality of an institution is goal-oriented. But the goal is not metaphysically given. Rather, it is determined by what Hayek considers to be desirable (or what he thinks others consider desirable).⁵ If Hayek formulates concrete policy recommendations - which he obviously does - then they are an expression of goal-oriented rationality within politics. Yet such a viewpoint, it may be noted, contradicts, prima facie, Hayek's own anti-rationalism.

The Theory of Spontaneous Orders

In his theory of spontaneous order, Hayek attempts to reconstruct an antirational
of liberalism. The theory consists of three theses (Gray
1984):

(1) The thesis of tacit knowledge.

(2) The thesis of natural selection of cultural rules.
(3) The thesis of the 'invisible hand'.

In the theory of spontaneous order, Hayek (1973, 37) sets a Cosmos (grown order) against a Taxis (created order). A Cosmos is governed by Nomos - i.e. universal rules - independent of individual intentions. In Taxis, by contrast, Thesis predominates - i.e. constructed rules that aim toward reaching specific goals.

Rules have a central role in Hayek's political philosophy - both theor-

etically and normatively. The following citation expresses the core of his
entire political philosophy:

Man is as much a rule-following animal as a purpose-seeking one. And he is successful not because he knows why he ought to observe the rules which he does observe, or is even capable of stating all these rules in words, but because his thinking and acting are governed by rules which have by a process of selection been evolved in the society in which he lives, and which are thus the product of the experience of generations (1973, 11).

The three theses on which Hayek bases his theory must be understood in a normative perspective. Rule-governed action is a prerequisite for all moral values. His idea of 'rules of just conduct' integrates the notions of efficiency, justice and freedom (Hayek 1976,133-152). Spontaneous orders operate on all levels of human activity - physiological, psychological, social, economic, political and moral. On all these levels rules have the same optimizing effect.

Two types of rules are important for Hayek (1967, 66): biologically inherited and culturally transmitted. Spontaneous orders can emerge as a consequence of both. On the other hand, discretionary decrees or constructed (Theses) cannot lead to a spontaneous order. Hayek's anticonstructivism the question of how one should react toward culturally transmitted rules. Hayek argues that they should not be the object of rational calculation. The idea of tacit knowledge prohibits deliberate construction. two other mechanisms of change are suggested: natural selection and the invisible hand.

It is the thesis of tacit knowledge that leads to an anti-rationalist stance toward political and social institutions. Anti-rationalism as such, however, implies a basic paradox: We cannot know how we can reach a goal, but in admitting our lack of knowledge as to how we will reach the goal, we - according to anti-rationalism - can reach the goal. This paradox expresses a crucial dilemma in Hayek's entire political philosophy. If it has practical implications, it is basically rationalistic and constructivistic. If it doesn't, it is metaphysical and fatalistic.

The Thesis of Tacit Knowledge - Epistemological or Empirical?

The thesis of tacit knowledge is often taken as an argument for epistemological (Gray 1984, 34; Rowland 1987, 11 ff.). But the question is whether the thesis of tacit knowledge actually deals with epistemological Even if it is interpreted as an epistemological thesis, however, it lacks practical implications. Epistemological theories are, by most philosophers, considered to be non-empirical (Chisholm 1977). They express some kind of meta-theoretical presumptions and thus are not true

or false in the same sense as empirical theories. One cannot derive a
statement about the effects of certain actions, from an epistemological
theory.

The fundamental mistake of constructivism, according to Hayek, is that it regards human reason as an exogenous variable in the evolution of civilization. This way of thinking falls prey to a 'synoptic delusion' - the belief that society can be observed from an Archimedean point and understood as it actually is (Hayek 1973, 14). In Hayek's evolutionary epistemology, human reason has the same evolutionary relativity as other social institutions. Actually reason is a social institution (Hayek 1973, 13) and like other institutions, it is a sort of spontaneous order, part of a greater 'cosmos'. Therefore, it is impossible to remove human reason, place it beside society and analyze society as it really is. It is just as impossible as lifting oneself by the hair.

Hayek's epistemology has important similarities with Kant's (Gray 1984, 21; Hayek 1978, 45). Both emphasize the subjectivity of human knowledge and its dependence on pre-conscious conceptual categories. To be able to observe society as it is, is just as impossible for Hayek, as it is impossible for Kant to see das Ding an Sich. The difference is that Hayek has a dynamic and evolutionary perspective on cognitive structures and conceptual (Hayek 1952).

In Hayek's theory, 'tacit knowledge' denotes different types of cultural rules or behavioral dispositions that lead to optimal action. The principles behind these rules are inaccessible for conscious reflection. But the expression knowledge' suggests a contradiction. How can something which is tacit be known? Knowledge implies awareness of the objective of knowledge. This contradiction expresses another paradox in Hayek's antirationalism: reasons do we have for believing that tacit knowledge can lead to optimal behavior?

As a consequence of the thesis of tacit knowledge, an 'information problem' emerges. This problem forms the theoretical basis of Hayek's criticism against all forms of state intervention - from monetary policy to economic planning. The problem arises because, in Hayek's view, the knowledge which is necessary for rationally guiding social processes is inaccessible to conscious reflection.

In essence, the problem consists of two parts - the problem of articulation and the problem of dispersion. The first implies that some knowledge is practical and impossible to articulate. On this count Hayek draws heavily from Michael Polanyi's theory of 'personal knowledge' (Polanyi 1958). He also refers to Gilbert Ryle's distinction between 'knowing how', and 'knowing what' (Hayek 1978, 38). One can know how one rides a bicycle without being able to formulate the principles behind these movements. One can also speak a language without knowing how to articulate its syntax

and grammar (Hayek 1978, 39). Rules of just conduct, according to Hayek function in the same manner. One knows what is right and wrong, without being able to give explicit reasons (Hayek 1973, 76). Even actors on the market such as an entrepreneur can be guided by a sort of tacit knowledge in her search for new investments (Gray 1984, 37).

The problem of dispersion, on the other hand, implies that knowledge is spread out among many actors and is impossible to incorporate within one particular actor. In the market, for example, knowledge is spread among an infinite number of actors. In fact, the problems of dispersion and articulation coincide in the market (Gray 1984, 39). Knowledge that leads to efficient resource allocation is both spread and tacit.

The solution to the problem of information, according to Hayek, lies in an evolutionary growth of institutions which coordinate and utilize tacit and dispersed knowledge. In Hayek's view, the paradigmatic example of such an information generating institution is the price system in a market economy (Hayek 1945).

A fundamental problem with the thesis of tacit knowledge is that it is based on arguments from very different theories - theories that are logically independent. Hayek cites theses from epistemology, cognitive psychology, economics and even logic. Godel's impossibility theorem is taken as an argument for the impossibility of finding an Archimedean point from which a consistent theory of society can be formulated. Cantor's theorem in set theory, which states that no system of classes can be elements in themselves, is interpreted by Hayek (1967, 60-62) as support for his thesis that the human intellect cannot formulate the logical basis for its own operations (Gray 1984, 22).

But Hayek's most important source of inspiration is Kant's doctrine on categories and forms of Anschauen (Gray 1984, 21 ff.). Hayek's epistemology as is Kant's and all epistemological theories, aprioristic and not based on empirical investigations. When Hayek criticizes Descartes' rationalist idea of an Archimedean point', his discussion is purely epistemological. problem concerns the objective foundation of knowledge - not its contents. Descartes maintains that reason exists independently of its object and that it is therefore possible to formulate absolute and objective knowledge about society. Against this rationalism, Hayek argues for a subjectivistic theory that denies the independence of human reason from its object (Hayek 1979b). But strictly speaking, Hayek's polemic against Descartes doesn't have any consequences for deciding which concrete theses on society are reasonable in scientific terms. Hayek's discussion on Cartesian rationalism is logically distinct from empirical arguments. Descartes' idea on pure reason lies on the same level as Kant's critique of pure reason.

Cartesian rationalism has had great historical influence on ideas regarding

the possibility of constructing a society based on rational principles (e.g. Saint-Simon, Comte,etc). But, neither Descartes' idea of an Archimedean point, nor Kant's idea of Das Ding an Sich has any substantial scientific consequences, because they both lack empirical content; they cannot be falsified. They deal with how one understands the epistemological basis for what we call science. Statements about the possibility of rationally controlling social processes are empirical and not epistemological. A declaration certain measures are ineffective is either true or false.

In The Fatal Conceit (1988), Hayek emphasizes that his arguments in the
debate on 'Socialist Calculations' are basically empirical:

That socialists are wrong about the facts is crucial to my argument... As a question of fact. this conflict must be settled by scientific study. Such study shows that, by following these spontaneously generated moral traditions underlying the competitive market order ... we generate and garner greater knowledge and wealth than could be ever obtained or utilised in a centrally-directed economy whose adherents claim to proceed strictly in accordance with 'reason"" (1988. 6-7).

Even so, Hayek's critique of the French sociologists, the forerunners of socialist economists, in The Counter-Revolution of Science (Hayek 1979b) deals primarily with what he believes to be the epistemological roots of their empirically incorrect conceptions of society and the space for goaloriented political measures. Hayek maintains that the origin of their scientistic 'hubris' is Cartesian rationalism. But this connection is not logical. If one argues that there are limits to human rationality, this can be interpreted either epistemologically or empirically. But an epistemological limitation does not imply an empirical limitation. It is one thing not to be able to see das Ding an Sich and quite another to not be able to predict the effects of a particular economic policy.

Hayek's Confusion of Theoretical Levels

The logical gap between empirical theory and epistemology makes it difficult to identify the practical implications of Hayek's arguments. It is doubtful if Hayek's evolutionary epistemology can be falsified. This is a general problem in interpreting Hayek's political theory. He mixes together different theoretical levels. He makes use of both abstract philosophy and concrete empirical theory, without a clear conception of the theoretical links between them.

It is important in this regard to note that Hayek's thesis of tacit knowledge is a general theory about the inaccessibility of certain principles of action for conscious reflection. Hayek does not state exactly where the limits of conscious reflection are. The fact that there will always be phenomena that one will never understand, does not lead to any practical conclusion as long

as one cannot set the limit to the unknown. Attempts at specifying exactly where the borderline for tacit knowledge should be drawn would be contradictory, since it would be a statement about something which is unknown. Arguing that the basis of a certain regulative principle will always be inaccessible for human knowledge, is to claim something about this regulative principle, which is presumed to be impossible to do. The contradiction however, if 'basis' implies an epistemological rather than empirical category.

If the thesis of tacit knowledge is a variation on Kant's epistemology - as John Gray claims - then it must be of an abstract nature without any practical implications. The statement that there is a Ding an sich, impossible to see, does not imply that people should refrain from trying to widen the pool of conscious, empirical knowledge. Neither does Hayek's thesis that there is no Archimedean point from which one can see society as it is, imply any concrete recommendations for giving up attempts at understanding The lack of an Archimedean point implies only that the knowledge we obtain does not have universal validity. Our knowledge about society is logically dependent on premisses that we do not know anything about. Nonetheless, knowledge can still have an instrumental value, regardless of its epistemological foundation and thus can make rational politics possible.

If Hayek maintains that his evolutionary epistemology is true, this implies that he himself has a synoptic view. But it is hard, to see how Hayek's own thesis can stand above the evolutionary relativism that he claims characterizes all human knowledge. It suffers from the same paradox of self-reference as the so-called sociology of knowledge.⁶ This problem can be illustrated with the famous liar's paradox: 'All Cretans lie'. Is the Cretan telling the truth? Bertrand Russell has formulated an analogous paradox, which he solves with his so-called type theory (Wedberg 1966, 144). The solution is to forbid self-referring propositions. The liar's paradox dissolves into a nonsensical statement. In the same manner, the thesis of tacit knowledge becomes a nonsensical statement, if it is allowed to refer to itself.

Hayek's theory - as do all forms of relativism - risks being self-defeating. If the judgement that Hayek makes on the limits of human rationality is a general empirical proposition, then it must be restricted by the same relativity and incompleteness of all empirical judgements. Consequently, it would be more reasonable to interpret Hayek's thesis as an epistemological rather than empirical thesis, i.e. a statement about empirical propositions. But then Hayek's thesis loses its practical political relevance.

If, on the other hand, it is empirical, it needs in order not to be selfdefeating
imply a statement of certain facts - a statement that optimal
actions cannot be chosen through conscious, rational calculation, but rather

they are reached through blindly following evolutionary grown cultural rules. In this case, however, the content of the thesis of tacit knowledge is not tacit, but rather articulated and conscious. If so, the thesis can form the basis of a goal-oriented rationality at the political level - i.e. for constructivism.

Hayek argues zealously for a free market economy (katallaxi). But his reasons are not epistemological, rather they are empirical and practical. An institution cannot solve an epistemological problem. The price system is an institution that solves a practical coordination problem. Consequently, Gray (1984, 40) is wrong when he states that a socialist, planned economy, according to Hayek, is an 'epistemological impossibility'. By definition, epistemological problems cannot have empirical consequences. It is more reasonable to interpret Hayek as maintaining that the rational planning of a society is empirically impossible. This implies that he himself thinks he possesses information showing that a planned economy leads to results that nobody wants (counterfinality). He believes thai the measures he suggests - i.e. a free market - bring about better outcomes. But if Hayek proposes such a policy recommendation, he himself becomes a constructivist.

The argument against constructivism is therefore primarily empirical rather than epistemological. Constructivism leads, according to Hayek, to undesirable effects. The constructivist theory of society is incorrect. But if their mistake has political implications, it must be because they are wrong in the empirical sense.

Gray (1984, 40) contends that market institutions, in Hayek's view, serve 'an epistemological function'. Giving empirically existing phenomena an epistemological function is to make oneself guilty of a category mistake. Kukahtas (1989, 53) seems to be guilty of the same fallacy when he writes that, 'social rules are "used" by the individuals of society to solve the epistemological problems posed by social life; problems involving knowing how to act (and so, to co-ordinate and distribute)'. Yet epistemological problems are abstract and analytical, not empirical. Gray's and Kukahtas's conflation of epistemological and practical problems mirrors a fundamental ambiguity in Hayek's political philosophy - an ambiguity that hides the fact that he himself is a constructivist. The thesis of tacit knowledge is the exact opposite of epistemological skepticism. Claiming that something is tacit, but at the same time effective, is to say something substantially empirical.

One way to save Hayek from constructivism is to refer to a metaphysical thesis on the direction of evolution (Rowland 1989). This thesis claims that the effect of spontaneously emerging institutions is always better than the effect of constructed institutions. But such a thesis can hardly be rational (in the sense defined above). If it is rational, it should be supported by scientific reasoning.

It should be noted that Hayek's views on the problem of information have both a pessimistic and an optimistic side. The pessimistic side concerns the possibility of discretionary goal fulfillment. The optimistic side concerns the possibility of reaching an efficient social order through a universal system of rules. Both pessimism and optimism can be given metaphysical as well as empirical reasons. If the reasons are metaphysical, Hayek's anti-rationalism is non-rational. If the reasons are empirical, which they apparently are, Hayek is not an anti-rationalist, but rather a rationalist. His policy recommendations are made from an empirically based picture of how society functions. In this case, he represents a form of constructivism, fundamentally different from the kind of rationalism he criticizes.

It is important to underline that the limits of constructivism are not a priori valid; they can be continuously tested with the help of new empirical hypotheses on the effects of political measures. There are no reasons, therefore, for drawing conservative conclusions from Hayek's political theory. The fact that there is always going to be an unknown area of social reality does not imply that we know the limit of our conscious knowledge and thereby of our ability to act from rational calculations.

Conclusion

The paradox of anti-rationalism is that it advises the actors not to have any goals for their actions in order to reach certain goals. But if anti-rationalism is an attitude or belief which is chosen in order to reach certain goals, the paradox can be eliminated. Anti-rationalism then can be a second order choice - a form of indirect rationality. This indirect rationality is a prerequisite Hayek's indirect utilitarianism. Indirect utilitarianism assumes goal-oriented rationality. Hayek denies, however, the possibility of goaloriented at the political level. But this anti-rationalism at the political level can be interpreted as an exhortation to rational self-deception'. Hayek thinks he has well-founded empirical evidence that an antirationalistic leads to goal fulfillment, then, from his viewpoint, it is rational.⁷

There are two levels in Hayek's writings: an ideological level and a scientific level. His epistemological polemic against constructivism should be interpreted as an ideological critique of a particular view of society. Hayek wants to imprint an anti-rationalistic belief, but for rational reasons. It is the consequences of the constructivism he wants to fight - with the help of an anti-rationalistic ideology. But Hayek's project is basically goaloriented and constructivist. This conclusion is aptly formulated by Oakeshott (1981, 21), who dismisses Hayek as a true anti-rationalist: 'A

plan to resist all planning may be better than its opposite, but it belongs to
the same style of polities'.

The so-called information problem, which is so fundamental in Hayek's critique of constructivism, must be empirical and scientific if it is to have any practical consequences. If the problem is epistemological, it lacks practical consequences by definition. An epistemological problem cannot be solved by a particular institutional framework. Therefore, it is misleading to maintain that market institutions, in a way, transcend epistemological limitations. When Hayek asserts that an institution solves an information problem, he is dealing with coordination of human aspirations. Language, like market institutions, fills the function of coordinating human activity. But the coordination problem is practical, not epistemological. In this respect, Hayek is not a skepticist. On the contrary, he assumes to have scientific evidence for how to solve it.

---

NOTES

1. It is not unusual to interpret Hayek as an advocate of laissez-faire policy. See Crespigny 1975, 55n.

2. The conception of 'rationality' used in this article is close to the definition offered by Oppenheim (1981. 125): \ . . an action X done by actor A in situation S will qualify as rational if, in the light of the information available to A in S. it is an optimal means to the attainment of some ultimate goal of his'.

3. Hardin (1988. 15) claims that Hayek is an act-utilitarian, contrary to Hayek's own strong criticism of utilitarianism. Hayek's theory suggests, according to Hardin. a way of finding the optimal actions in a complex and uncertain decision-making situation.

4. The liberalism of Hayek is fundamentally different from the liberalism advocated by Nozick. before his change of opinion (No'zick 1989). Nozick (1974, 28 f.) claims that a moral value, i.e. freedom, can be seen either as a goal or as a restriction for human actions. A deontological moral theory views moral values as restrictions, whereas a consequentialist moral theory views them as goals. Contrary to Hayek, Nozick is a pure deontologist; for Nozick the freedom of an individual is a 'moral side-constraint'.

5. Hayek (1978, 292) claims that there 'are no absolutevalues. Not even human life itself.

6. See (Carlsnaes 1981, 217) for an analysis of this problem in Karl Mannheim's sociology of knowledge. See also Vedung (1982, 147-51).

7. For a discussion on self-deception and rationality, see Pears (1984,1986) and Oksenberg Rorty (1986).

REFERENCES

Barry, N. P. 1982. 'The Tradition of Spontaneous Order". Literature of Liberty 5. 7-58.

Buchanan, J. 1982. 'Cultural Evolution and Institutional Reform'. Paper presented to Liberty
Fund Conference, March 1982.

Carlsnaes. W. 1981. The Concept of Ideology and Political Analysis. New York: Greenwood
Press.

Chisholm, R. M. 1977. Theorv of Knowledge. New York: Engelwood Cliff.

Crespigny, A. 1975. 'F. A. Hayek: Freedom for Progress'. In Crespigny. A. & Minogue. K...
Contemporary Political Philosophers. London: Methuen & Co.

Elster, J. 1985. Sour Grapes. Cambridee: Cambridge University Press.

Hamowy. R. 1987. The Scottish Enlightenment and the Theory of Spontaneous Order. Carbondale:
Illinois University Press.

Hardin, R. 1988. Morality within the Limits of Reason. Chicago; The University of Chicago
Press.

Hayek. F. A. 1944. The Road To Serfdom. London: Routledge & Kegan Paul.

Hayek. F. A. 1945. 'The Use of Knowiedge in Society'. American Economic Review 35. 519-530.

Hayek. F. A. 1949. lndividualism and Economic Order. London: Routledge & Kegan Paul.

Hayek. F. A. 1967. Studies in Philosophy, Politics, and Economics. New York: Simon &
Schuster.

Hayek. F. A. 1973. Law, Legislation and Libertv. Vol. 1. Chicago: The University of Chicago
Press.

Hayek. F. A. 1976. Law, Legislalwn and Liberty. Vol. 2. Chicago: The University of Chicago
Press.

Hayek. F. A. 1978. New Studies in Philosophy. Politics. Economics and the History ofldeas.
London: Routledge & Kegan Paul.

Hayek. F. A. 1979a. Law, Legislation and Liberty. Vol. 3. Chicago: The University of Chicago
Press.

Hayek. F. A. 1979b. The Counter-Revolution of Science. Indianapolis: Liberty Press.

Hayek. F. A. 1987. Tndividual and Collective Aims". in Mendus. S. & Edwards. K., eds.,
On Toleration. Oxford: Clarendon Press.

Hayek. F. A. 1988. The Fatal Conceil. London: Routledge & Kegan Paul.

Pears. D. 1986. The Goals and Strategies of Self-deception", in Elster, J.. ed., The Multiple
Self. Cambridge: Cambridge University Press.

Polanyi, M. 1958. Personal Knowiedge. Chicago: The University of Chicago Press.

Rorty Oksenberg, A. 1986. 'Self-deception, Akrasia and Irrationality'. in Elster, J., ed.. The
Multiple Self. Cambridge: Cambridge University Press.

Rowland. B. M. 1987. Ordered Liberty and the Constitutional Framework. The Political
Thought of Friedrich A. Havek. New York: Greenwood Press.

Rowland. B. M. 1989. 'Beyond Hayek's Pessimism: Reason, Tradition and Bounded Con-

Simon. H. 1983. Reason in Human Affairs. Oxford: Basil Blackwell.

Vaughn K.I. 1984. 'The Constitution of Liberty from an Evolutionary Perspective". in Seldon,
A.. ed.. Hayek's Serfdom' Revisiled. London: The Institute of Economic Affairs.

Vedung, E. 1982. Political Reasoning. London: Sage.

Vernon, R. 1979. 'Unintended Consequences", Political Theory 7, 57-73.

Wedberg. A. 1966. Filosofins historia. Del 3. Från Bolzano Wittgenstein. Stockholm: Bonniers.

Woodfield, A. 1976. Teleology. Cambridge: Cambridge University Press.