James M. Buchanan’s contractarianism and modern liberalism

Abstract

This paper contrasts Buchanan’s contractarian–constitutional liberalism with Hayek’s evolutionary liberalism and Rothbards free-market liberalism as representative branches of the classical liberal tradition. While Hayek and Rothbard focus on individual liberty as private autonomy, Buchanan posits that individual sovereignty should be recognized as the fundamental normative premise of liberalism. He insists that a consistent application of this premise requires liberals to respect individuals as sovereigns not only in their capacity as private law subjects but also at the constitutional level of choice where, as sovereign citizens, they choose, jointly with their fellow citizens, the rules under which they wish to live. It is argued that by supplementing the notion of individual liberty as private autonomy with the concept of individual sovereignty in constitutional matters Buchanan lays the theoretical foundation for complementing the well-developed liberal theory of the market with a consistent liberal theory of democracy.

The title I have been asked to comment on requires me, I suppose, to answer two questions. First, what distinguishes Buchanan’s contractarian liberalism from other branches of the classical liberal tradition? And, second, in what sense is the distinctiveness of Buchanan’s contractarianism relevant for a ‘modern’ liberalism?

Before I address these questions directly, a few introductory remarks are in order to prepare the ground for the arguments I want to develop.

• I presume all varieties of liberalism that may be legitimately counted within the classical liberal tradition to share as their common core

  • the emphasis on individual liberty as the foundational principle of a desirable social order, and

  • a prima facie preference for markets as institutional arrangements within which voluntary contracts are the principal means by which individuals coordinate their activities.

• In contrasting Buchanan’s contractarianism with other branches within the classical liberal tradition I need to be selective. I shall concentrate my comparison on Murray Rothbard’s free-market libertarianism as the extreme counterpart to Buchanan’s contractarian liberalism and on F.A. Hayek’s evolutionary liberalism as an intermediate position. In so doing, I presume that what I argue with regard to them is, more or less, relevant for other branches of the classical liberal tradition as well.

• As F.A. Hayek has emphasized, when we talk about social order we must distinguish between the order of rules and the order of actions, i.e. between the rules of the game that govern the behavior of the individuals involved and the patterns of actions that result from the individuals’ choices within a given framework of rules. This corresponds to Buchanan’s distinction between two levels of choice, choice within rules (the sub-constitutional level) and choice of rules (the constitutional level).

• The order of rules, the constitutional framework, is essential in determining the nature of the order of actions. While the latter is the immediate object of interest when the ‘desirability’ of a social order is at issue, the order of rules is the indirect object of evaluation because of its instrumental role in shaping the order of actions. Rules are not ends in themselves but instruments for shaping the resulting patterns of actions.

• Any conception of a desirable social order must include two components, namely a normative criterion against which the ‘desirability’ of different social orders is to be compared, and factual assumptions about what are suitable measures (in particular: rules) to create a ‘desirable’ social order. Accordingly, disagreements on what qualifies as a desirable social order may be due either to disagreements on which normative criteria should be applied or to different factual assumptions about what are suitable means for bringing about what is considered desirable (or both).

• In comparing Buchanan’s, Rothbard’s and Hayek’s respective versions of liberalism my primary interest will be in identifying the nature of the differences that separate them. If their disagreement is about the normative criterion that a classical liberal outlook implies it is concerned with what I propose to call matters of principle. If their disagreement is about what are adequate means for achieving a liberal order it is concerned with what I propose to call matters of prudence.

• At the constitutional level matters of principle are intimately linked to the issue of legitimacy, the issue of who is authorized to decide what counts as a ‘desirable’ constitutional order (or order of rules). Matters of prudence are about the issue of instrumental adequacy, about what provisions should be prudently included in a constitutional order as suitable instruments for achieving a ‘desirable’ order of actions.

1 Individual liberty, private autonomy and constitutional choice

As noted above, I presume that all branches within the classical liberal tradition share as their common core an emphasis on individual liberty as the foundational principle of a desirable social order. From this it would seem natural to conclude that Buchanan, Rothbard and Hayek equally base their liberal concepts on a normative individualism in the sense that the evaluations of the individuals themselves are the only source from which legitimacy in social matters can ultimately be derived. Accordingly, one might be inclined to assume that if any differences exist between their respective approaches they can only be about matters of prudence but not about matters of principle. And, in fact, when Buchanan (1999 [1986]: 461) speaks of “the individualistic value norm on which a liberal social order is grounded,” Rothbard and Hayek appear to invoke the same “value norm” when the former insists that “only individuals have ends … only individuals can desire and act” (Rothbard 1970: 2), and when the latter notes that “it is the recognition of the individual as the ultimate judge of his ends … that forms the essence of the individualist position” (Hayek 1972: 59). Yet, even though Buchanan (2001 [1977]: 27) posits that the “libertarian anarchist and the contractarian share the individualistic value premise,” on closer inspection one cannot fail to notice that there are subtle differences between the ways in which the three authors interpret “the individualistic value norm.”

It is the emphasis on individual liberty as private autonomy that constitutes the unifying core of the classical liberal tradition. And it is this understanding of individual liberty that Hayek implies when he describes individual liberty as “freedom under the law” (1960: 153), as a condition in which liberty is “limited only by the same abstract rules that apply equally to all” (ibid.: 155).¹ The chief aim of what he refers to as “liberal constitutionalism” is, as he puts it, “to provide institutional safeguards of individual freedom” (1973: 1) and to protect an “assured private sphere” (1960: 13), “a recognizable private domain” (1967: 162).² Liberty as private autonomy means freedom of choice within rules—the private law order and public regulations—that define individuals’ (property) rights.³ Accordingly, what individual liberty means in substance depends on how these rights are defined, and to the extent that their definition changes over time and differs across different communities individuals’ private domains will be differently demarcated. Hayek (1960: 229) explicitly points to this fact when he notes that “the recognition of the law of private property does not determine what exactly should be the content of this right.”⁴ In other words, Hayek recognizes that, in establishing a liberal order, we face a problem of constitutional choice in the sense that the rules of the game are, explicitly or tacitly, selected from a set of potential alternatives. This raises of course the question of how different rule-systems can themselves be comparatively evaluated in terms of the fundamental “individualistic value norm” which assigns to the individuals themselves the role as “ultimate judges.”

Hayek does not directly address this question, a question that can clearly not be answered by invoking individual liberty qua private autonomy as the normative standard since it is itself defined in terms of that (the “abstract rules”) which it were to judge. He clearly invokes, though, a normative criterion when he speaks of “appropriate rules”⁵ and when he warns against the error of presuming that “the law of property and contract were given once and for all in its final and most appropriate form” (Hayek 1948: 111). He also offers sporadic hints at what he means by “appropriate rules,” e.g. when he describes them as rules that “will make the market economy work at its best” (ibid.: 111), when he states that our aim in “altering or developing them should be to improve as much as possible the chances of anyone selected at random” (1976: 129f.), or when he notes that “we should regard as the most desirable order of society one which we should choose if we knew that our initial position in it would be decided purely by chance” (ibid.: 132). Yet, he does not systematically discuss how these standards of “appropriateness” are related to the fundamental “individualistic value norm.” Instead, he emphasizes the need to rely on experience in constitutional matters and draws attention to the process of cultural evolution as an experimental discovery procedure that helps men to find out, by trial-and-error, which rules are and which are not “appropriate.”⁶

It is no more than a conclusion from what has been said above when Hayek (1960: 158) states: “What exactly is to be included in that bundle of rights that we call ‘property,’ … what contracts the state is to enforce, are all issues in which only experience will show what is the most expedient arrangement.” He thereby expresses a view, however, that is in stark contrast to a central tenet of Murray Rothbard’s free-market liberalism. While Rothbard fully shares Hayek’s emphasis on individual liberty as private autonomy, he strongly disagrees with Hayek’s cautious view on the issue of how we can know what rules are “appropriate” for demarcating the “assured private sphere” that defines the substance of “individual liberty.”⁷ According to Rothbard (2002, [1973]): 26, it is “the natural rights basis for the libertarian creed” that provides the definite answer to this issue. In other words, from Rothbard’s natural rights perspective the law of property and contract is indeed “given once and for all in its final form.” There exists no problem of constitutional choice that would need to be addressed in terms of the “individualistic value norm.” The question of what rules should be adopted becomes, instead, a cognitive problem for which there can be only right or wrong answers, not one of subjective evaluation to be decided by “the individual as the ultimate judge.”

Rothbard’s ideal libertarian society is a “society formed solely by … an unhampered market, or a free market.” (1970: 77). It is “a market society unhampered by the use of violence or theft against any man’s person or property” (ibid.: 152),⁸ or, as Rothbard’s follower H.-H. Hoppe calls it, a pure “private law society.”⁹ For Rothbard the question of “whether or not a certain practice or law is or is not consonant with the free market” (ibid.: 654) is to be judged in terms of whether or not it involves “implicit or explicit theft” (ibid.), for which, in turn, the “institution of private property” (ibid.: 156) provides the required standard. And, as noted above, the question of how we are to determine the appropriate legal specification of the “institution of private property” Rothbard (1998 [1982]: 17) answers by invoking the “natural law discoverable by reason.” As he asserts: “In fact, the legal principles of any society can be established in three alternative ways: (a) by following the traditional custom of the tribe or the community; (b) by obeying the arbitrary, ad hoc will of those who rule the State apparatus; or (c) by the use of man’s reason in discovering the natural law … Here we may simply affirm that the latter method is at once the most appropriate for man” (ibid.).¹⁰

Once the ground for further argument is provided in this fashion it is only a matter of logic for Rothbard to conclude that, since all legitimate within-market transactions are based on voluntary contracts among the parties involved (market exchange being the paradigm example of a voluntary contract),¹¹ the market society or private law society is a contractual society,¹² a “genuinely co-operative society” (Rothbard 1970: 84) with “benefits for all participating individuals” (ibid.: 78). As Rothbard (1956: 250) reasons: “The free market is the name for the array of all the voluntary exchanges that take place in the world. Since every exchange demonstrates an unanimity of benefit for both parties concerned, we must conclude that the free market benefits all its participants.” The problem with such reasoning is, though, that it disregards the fundamental difference between choices within rules and choices of rules, the very distinction that is at the heart of Buchanan’s contractarianism. To be sure, from the fact that the participants voluntarily agree to the transactions they carry out within the rules of the “market game” we may justly conclude that they all expect to benefit from these transactions. Yet, it does not allow us to conclude that they all expect to benefit from the “market game” as such, with its particular rules, compared to “games” played under different rules. The question whether players agree to moves within a game with given rules must surely be distinguished from the question of whether they agree to the rules themselves.

What sets Buchanan’s contractarian liberalism apart from Rothbard’s as well as from Hayek’s (and also other) interpretations of the classical liberal tradition is that he not only explicitly distinguishes—as Hayek also does—between two levels of choice, the constitutional and the sub-constitutional level, but insists that “the individualistic value norm upon which a liberal social order is grounded” requires us to respect individuals’ freedom of choice at both levels. In Buchanan’s (1999 [1991]: 288) own words:

“The justificatory foundation for a liberal social order lies, in my understanding, in the normative premise that individuals are the ultimate sovereigns in matters of social organization, that individuals are the beings who are entitled to choose the organizational-institutional structures under which they will live. In accordance with this premise, the legitimacy of social-organizational structures is to be judged against the voluntary agreement of those who are to live or are living under the arrangements that are judged. The central premise of individuals as sovereigns does allow for delegation of decision-making authority to agents, so long as it remains understood that individuals remain as principals. The premise denies legitimacy to all social-organizational arrangements that negate the role of individuals as either sovereigns or principals. On the other hand, the normative premise of individuals as sovereigns does not provide exclusive normative legitimacy to organizational structures that—as, in particular, market institutions—allow internally for the most extensive range of separate individual choice. Legitimacy must also be extended to ‘choice-restricting’ institutions so long as the participating individuals voluntarily choose to live under such regimes.”

It is the very question of “the ultimate justification for regimes of social interaction” that, as Buchanan (ibid.: 281) charges, has often been neglected by “advocates of a liberal or free society embodying the maximal exercise of individual liberties.” And it is indeed his singular merit not only to have persistently put this question on the liberal agenda but to have shown how it can be consistently answered from a classical liberal perspective.

The emphasis on individual liberty as private autonomy is for Buchanan no less than for Hayek at the heart of the classical liberal tradition,¹³ and no less than Rothbard he praises the market as an arena for voluntary cooperation, as “the institutional embodiment of the voluntary exchange processes that are entered into by individuals in their several capacities” (Buchanan 1979: 31). What distinguishes his contractarian liberalism is the insistence that classical liberals, if they consistently adhere to their normative individualism, are required to respect individuals as “ultimate judges” not only when they act as private law subjects within the market arena but also in matters of constitutional choice. Buchanan does not dispute Hayek’s arguments on the beneficial role that cultural evolution may play as an experimental discovery process, but he insists that such arguments cannot per se answer the question of how alternative institutional arrangements—or systems of rules—are to be judged in terms of the classical liberal “individualistic value premise.”¹⁴ And as far as the “natural rights’” are concerned that Rothbard invokes as the standard for how the boundaries of individuals’ private domains should be drawn, Buchanan insists that, if they are not meant outright to deny individuals’ freedom to choose the rules under which they wish to live, such “natural boundaries” can have legitimizing force only to the extent that the individuals involved agree on where they lie. The “ultimate test for the existence of natural boundaries must,” so Buchanan (2001 [1977]: 25) argues, “lie in the observed attitudes of individuals themselves.”

From a contractarian perspective à la Buchanan, a free-market libertarian who “finds it necessary to presume that there are definite and well-understood ‘natural boundaries’ to individuals’ rights” (ibid.: 23f.) would seem to face the following choice: If his “natural rights” are supposed to command authority independently of agreement among the individuals concerned, he must claim that those who correctly “read” these rights are entitled to override the judgment of erring dissenters, denying them the status as “ultimate judges” in constitutional matters. Alternatively, if he acknowledges that “natural rights” cannot be determined independently of “the observed attitudes of individuals themselves,” he must address the issue of how, according to the individualistic value premise, such rights are supposed to be derived from individuals’ “observed attitudes.”¹⁵ In other words, the free-market libertarian would have to deal with the very same problem for which Buchanan seeks to provide a consistently individualistic solution.¹⁶

The solution that Buchanan proposes starts from the diagnosis that, since “rules of the game,” if they are to serve their coordinating function, must be shared among the parties involved, their definition is necessarily a matter of (explicit or implicit) collective choice,¹⁷ and that, if such choice is to comply with the individualistic value premise, it can derive its (ultimate) legitimacy from no other source than agreement among the individuals involved, i.e. from the same source that provides legitimacy to voluntary market exchanges. As far as the manifold private collective associations, such as business firms or clubs, are concerned which exist within the market arena, free-market libertarians appear to have no difficulty conceding that, what legitimizes their day-to-day operation is the participants’ voluntary agreement to their respective constitutions, even if these constitutions impose significant limits on the individuals’ within-period freedom of choice. Buchanan does no more than extend the very same criterion that the libertarian applies to constitutional choice within the market to the constitutional choices that define the institutional framework of the market, choices that, by their very nature, cannot be made other than through the political process. This is the principal tenet of what he describes as “exchange conceptualization of politics” (Buchanan 1999a [1986]: 461), a conceptualization that generalizes “the model of voluntary exchange among individuals” (ibid.) from simple bilateral exchange-contracts to the inclusive constitutional contracts among individuals as citizens of political communities, comparable to the inclusive constitutional contracts they enter into as members of private associations. As Buchanan (1989: 179) argues: “If politics … is modeled as a cooperative effort of individuals to further or advance their own interests and values … it is evident that all persons must be brought into agreement. … The complex exchange that describes a change in the constitution (in the rules) is no different in this fundamental respect from simple exchange between two traders.”¹⁸

Bringing persons into agreement poses, of course, additional problems when we move from exchange transactions, whether bilateral or multilateral, to organizational-collective arrangements. Exchange transactions take place only if and when all trading parties agree. Absent agreement there is no transaction. By contrast, organizational-collective arrangements are based on continuing contractual relations the purpose of which is to allow a group of persons to carry out joint actions over a (definitive or open-ended) period of time. For such joint enterprises the agreement-issue arises at two levels, the constitutional level at which the rules that define the terms of operation are to be chosen, and the sub-constitutional level at which the day-to-day operational decisions are to be made. At both levels collective choices have to be made and, strictly speaking, bringing all participants into agreement would require that these choices be made unanimously. As Buchanan (1999 [1986]: 463) puts it: “The political analogue to decentralized trading among individuals must be that feature common over all exchanges, which is agreement among the individuals who participate. The unanimity rule for collective choice is the political analogue to freedom of exchange of partionable goods in markets.”

If the unanimity rule is to serve for collective choices the same role that agreement among the trading parties plays in legitimizing market transactions, obvious questions of practicability arise, especially in large-number settings. As far as collective choices at the sub-constitutional level are concerned, Buchanan’s and Tullock’s classic contribution The Calculus of Consent (1962) specifies the arguments why rational individuals have prudential reasons to agree to forgo the veto-right that the unanimity rule would grant them. Yet, for constitutional choices the practicability of the unanimity rule remains an issue. For private organized collectives operating within a market arena there is a straightforward answer to this issue. Just as exchange contracts concluded within the rules of the market can be claimed to be legitimized by the voluntary agreement among the trading parties, constitutional contracts on which private organized collectives are based can equally be claimed to be legitimized by the participants’ voluntary choices to join and to remain within the joint enterprise.¹⁹ The situation is, however, obviously different as we move to the level of organized polities, the level where the very rights individuals enjoy as private law subjects are defined. Admittedly, the task to specify how individual sovereignty can be effectively exercised, and the legitimizing force of agreement be secured, at the constitutional level of politics poses a serious challenge to Buchanan’s contractarian liberalism. Yet, this is a challenge that every classical liberal must face as soon as he seeks to consistently apply the individualistic value premise not only at the level of choices made within a framework of pre-defined individual rights but also at the level at which these very rights are defined.

2 Matters of principle and matters of prudence in classical liberalism

The specific differences between the three branches of the classical liberal tradition that I have contrasted above come into sharper relief when one compares them in light of the following question: Who is the intended addressee of the arguments that they advance in support of their respective concept of the “liberal order”? It is in answering this question that the essential achievement of Buchanan’s contractarian liberalism can be shown most clearly. With his insistence that a classical liberal outlook, if it is to consistently adhere to its individualistic value premise, must respect individuals as “ultimate judges” at the constitutional level of choice no less than at the level of ordinary market choices, Buchanan forces his fellow liberals to be more conscious about whom they want to convince of the merits, or the attractiveness, of their “liberal ideal.” As I will argue below, clarity on this issue helps to distinguish more convincingly between what should be treated as matters of principle and what as matters of prudence within the classical liberal tradition.

If, as Buchanan’s contractarianism asserts, the individuals themselves must be respected as ultimate judges in constitutional matters, if it is their judgment that ultimately decides what counts as a “good society,” then they must surely be seen as the ultimate addressees of whatever proposals in matters of social organization are made. To be sure, assigning to the individual group-members or citizens the status as “ultimate addressees” cannot mean that advocates of liberal constitutional proposals must always cast their arguments in a language that is suitable for direct communication with “the common man.” As intellectuals and scholars contractarian liberals, no less than their fellow liberals, will naturally address their writings directly at their academic peers and it is to their critical examination that they submit their arguments. The requirement to respect individuals as the “ultimate judges” means, however, that in their academic discourse liberal scholars impose a disciplining constraint on their arguments, namely to support whatever constitutional proposals they advance with arguments that show why—paraphrasing what Hayek has said about the rules of the market as the “game of catallaxy”²⁰—the individuals concerned should “have reasons to agree to the proposed constitutional regime.” It is in this sense, i.e. with regard to their own judgment on what they regard as a “desirable” constitutional order, that the individuals themselves must be regarded as the ultimate addressees. Even if they are not the direct addressees of the arguments exchanged in academic discourse, these arguments must specify reasons for why the individual constituents of the group whose constitutional regime is under consideration can be expected to be in favor of what liberals advocate as “appropriate” rules. Or, stated in yet another way, while the respect for individuals as sovereigns, in matters of constitutional choice no less than in matters of private autonomy, must be treated within the liberal doctrine as a matter of principle, particular liberal recipes for how people should organize their social life, including the liberal preference for markets, must be argued for on prudential grounds. Their advocates must provide arguments for why the individuals concerned would serve their own interest when adopting the recommended recipes.

In the case of Rothbard it is quite obvious that in advocating his free-market liberalism he does not have the individual citizens in mind as the ultimate judges. Nor does he argue in terms of reasons why the individuals concerned can be expected to consider his liberal ideal more desirable than potential alternative constitutional regimes. If, as Rothbard claims, the rules of the (market) game are unequivocally given by “natural law discoverable by reason” then what the “rules of the game” should be is pre-ordained and cannot be a matter of individual preferences. In Rothbard’s free-market liberalism individual liberty is fully realized within the market arena. At the constitutional level individuals are simply not “free to choose” or, more precisely, they would be simply in error if they were to choose rules of the game that are not consonant with the natural law. The question of how property rights should be specified and which contracts should be enforced is not a matter of evaluative judgment but of logical reasoning.

When Rothbard (1970: 653) characterizes the “purely free market” as an arena “where the individual person and property are not subject to molestation,”²¹ and when he argues that so-called “external diseconomies” are not “a defect of the free market … (but) the result of invasions of property, invasions which are ruled out of the free market by definition” (Rohbard 1956: 259), he makes it appear as if the question of how the line between what individuals, in exercising their freedom of choice, are allowed and not allowed to do is a matter of logical deduction from apodictically true first principles, not something that sovereign individuals decide among themselves, weighing the advantages and disadvantages of differently drawn demarcations. Just as in the tradition of Misesean apriorism, Rothbard (1970: xi) claims to be able to deduce “the entire corpus of economics from a few simple and apodictically true axioms,”²² his “ethics of liberty” is praised for being based on “axiomatic-deductive arguments and proofs” (Hoppe 1998: xxii).²³

On the basis of such axiomatically deduced ethical standards Rothbard proclaims, for instance, that, by contrast to copyright, the “patent is incompatible with the free market” (1970: 654). “The crucial difference,” he reasons, “is that copyright is a logical attribute of property right on the free market, while patent is a monopoly invasion of that right” (ibid.: 655). Where such “logic” reins there is no room whatsoever for individuals to consider among each other as sovereign citizens how they would want to define copyright- and patent-rules in light of the advantages and disadvantages that they expect to result from potential alternative rules. Similarly we can, according to Rothbard, decide on apriori grounds whether cartel contracts should or should not be permissible in a “free market.” Against “theorists who attack cartels” because they involve collusion he argues that the “whole concept of ‘restricting production’ is a fallacy when applied to the free market” (Rothbard 1970: 573,568). Since in both cases, cartels and business firms, individuals voluntarily pool assets “according to rules agreed upon by all from the beginning” there is in his view “no essential difference between a cartel and an ordinary corporation or partnership” (ibid.),²⁴ “nothing distinctively immoral about such action” (ibid: 564).²⁵

It is because of his natural rights-based axiomatic-deductive ethics that Rothbard (1970: 562) feels entitled to reject as “inconsistent” William Hutt’s arguments on “’consumer’s sovereignty’ as an ethical ideal against which the activities of the free market are to be judged.” He does not even consider the possibility that individuals may have reasons, as sovereign citizens, to prefer a constitution that seeks to implement this “ethical ideal,” and that, if he wants to dissuade them from such constitutional choice, he would need to provide arguments why they can expect their common interests to be better served by what he suggests. To provide reasons why individuals may, as a matter of prudence if not of axiomatic logic, want to adopt “consumer’s sovereignty” as a constitutional ideal is, however, exactly the point of Hutt’s arguments²⁶ who, in this regard, simply restated in more elaborate ways what Adam Smith (1981 [1976]: 660) had in mind when he said in critique of the mercantile system: “Consumption is the sole end and purpose of all production; and the interest of the producer ought to be attended to only so far as it may be necessary for promoting that of the consumer. The maxim is so perfectly self-evident, that it would be absurd to attempt to prove it.” Hutt as well as Smith provided arguments why, in comparing their merits as constitutional ideals, citizens are well advised to prefer the principle of consumer sovereignty over Rothbard’s (1970: 657) “principle of maximum service to consumers and producers alike.” In other words, Hutt was aware of the fact that as Buchanan (2001 [1990]: 40) puts it, at the “constitutional stage decisions must weigh the predicted costs and benefits of the alternative institutional arrangements” (Buchanan 2001 [1990]: 40). He did, as Buchanan (1991: 119) notes, “not commit the libertarian blunder of extending the defense of the liberties of individuals to enter into ordinary voluntary exchanges to a defense of the liberties of individuals to enter into voluntary agreements in restraint of trade.”

Whatever their disagreements on issues such as patents or cartels may be, my main point in comparing Buchanan’s contractarian liberalism and Rothbard’s free-market libertarianism is not that they differ in what they advocate but in how they advocate it. Because he insists that liberalism’s individualistic value premise requires one to respect individuals as sovereigns at the constitutional level of choice no less than in their capacity as private law subjects, it is a logical conclusion for Buchanan that liberals must view their fellow citizens as the ultimate addressees of their arguments in favor of the liberal ideal they advocate. That individuals must be respected as sovereigns in constitutional matters is, from such perspective, a matter of principle, while the particular institutional features of the liberal ideal are considered a matter of prudence, to be discussed in terms of reasons that appeal to citizens’ common interests. It is in this sense that Buchanan (1995/96: 267f.) argues against a too narrow interpretation of the ideal of individual liberty:

“What is the ultimate maximand when the individual considers the organization of the political structure? … (T)his maximand cannot be summarized as the maximization of (equal) individual liberty from political-collective action. … A more meaningful maximand is summarized as the maximization of (equal) individual sovereignty. This objective allows for the establishment of political-collective institutions, but implies that these institutions be organized so as to minimize political coercion of the individual. … So long as one’s agreement to such political action is voluntary, the individual’s sovereignty is protected even though liberty is restricted.”

It is the narrow interpretation of the ideal of individual liberty from which Buchanan distances himself that is at the very heart of Rothbard’s free-market libertarianism. From its natural rights-based perspective the “purely free market” is seen as the order that is mandated, as a matter of principle, by Rothbard’s “axiomatic-deductive” ethics. It is definitely not seen as a constitutional ideal that liberals must advocate to their fellow citizens on prudential grounds, i.e. in terms of reasons that appeal to their common interests.

When, in terms of their ways of advocating the ideal of individual liberty as private autonomy, Buchanan and Rothbard represent the polar ends of the liberal spectrum, Hayek’s evolutionary liberalism must be placed somewhere in between. Hayek’s argument for the liberal order is obviously not cast in Buchanan-type contractarian terms, nor is it grounded in a Rothbardian natural-rights approach. There are good reasons, though, to locate Hayek’s thoughts on the virtues of the spontaneous market order and the discovery process of cultural evolution closer to the Buchanan-end of the spectrum than to Rothbard’s side. A careful reader of Hayek’s arguments cannot fail to see that he advocates his liberal ideal by providing reasons why individuals may, on prudential grounds, want to live within a constitutional framework that provides room for private autonomy and experimental discovery.²⁷ He does not, like natural-rights libertarians, portray the liberal order as the imperative implication of unquestionable first principles, but emphasizes that the “defense of the free society must … show … that the members of such a free society have a good chance successfully to use their individual knowledge for the achievement of their individual purposes” (Hayek 1967: 164f.). For him the market order is not a matter of logical necessity—or of principle—but a matter of prudential constitutional choice. This is clearly reflected in his already mentioned statement on the market as the “game of catallaxy”: “The individuals have reasons to agree to play this game because it makes the pool from which the individual shares are drawn larger than it can be made by any other method” (Hayek 1978: 137).

Hayek’s arguments on the “use of knowledge in society”²⁸ and on competition as a discovery procedure which are at the core of his entire work can, in this sense, be understood as providing prudential reasons for why individuals should prefer a market order and a political constitution that leaves room for institutional variation and for cultural evolution to play their knowledge-creating role.²⁹ As the great advantage of a spontaneous, self-generating order he emphasizes that it makes “possible the utilization of widely dispersed knowledge” (Hayek 1978: 136), and on the advantage of competition he notes that it “is the most effective discovery procedure which will lead to the finding of better ways for the pursuit of human aims” (ibid.: 149).

Since for both, Hayek and Rothbard, Ludwig von Mises has been a most important influence, it is worth mentioning that their divergent views reflect an ambiguity in Mises’ own writings. While Rothbard may cite for his support Mises’ (2005: 61) statement that “liberalism is derived from the pure science of economics and sociology which make no value judgements,”³⁰ Hayek’s view may find support in Mises’ assertion that “liberalism has always in view the good of the whole” (ibid.: xxii),³¹ and that liberals, if “they considered the abolition of the institution of private property in the general interest, they would advocate that it be abolished” (ibid.: 11).

3 Buchanan’s contractarian constitutionalism and modern liberalism

Once classical liberals recognize that consistency requires them to respect individuals as constitutional sovereigns, and that it is to them that their arguments in support of the liberal order must ultimately be addressed, it should be obvious that the question of how the process of collective constitutional choice can be organized in such fashion that individuals can meaningfully exercise their sovereignty at that level must be part of the liberal research agenda. It is Buchanan’s singular merit to have devoted much of his research effort to addressing this very question and to have, thereby, developed a classical liberal outlook at democracy that complements, as a consistent counterpart, the traditional theory of the spontaneous market order and the private law society. In fact Buchanan deserves to be credited for having shown how the two liberal traditions, the “British” and the “Continental,” that Hayek contrasts as if they were estranged relatives, can be integrated into one coherent theoretical framework. About the two traditions Hayek (1978: 120) writes: “While to the older British tradition the freedom of the individual in the sense of a protection by law against all arbitrary coercion was the chief value, in the Continental tradition the demand for self-determination of each group concerning its form of government occupied the highest plain. This lead to an early association and almost identification of the Continental movement with the movement for democracy which is concerned with a different problem from that which was the chief concern of the liberal tradition of the British type.”

By characterizing the British tradition as the “evolutionary type of liberalism” (ibid.: 132) and the Continental tradition as the “constructivist type” (ibid.) Hayek made it appear as if there were a fundamental conceptual divide between the two traditions, due to differences in their respective views on how the rules of the liberal order came about or may be usefully modified,—the British tradition emphasizing the role of evolutionary forces,³² the Continental tradition emphasizing the role of deliberate legislation. Focusing thereby attention on what should be treated as matters of prudence Hayek distracted from the much more important fact that, due to their shared individualistic value premise, the two traditions are united in matters of principle, namely equally respecting individuals as sovereigns, a unity that calls for their theoretical integration.

The need to achieve a theoretical integration of the liberal outlook at markets and the individualistic approach to politics is exactly what Buchanan has emphasized in his work, along with important suggestions for how such integration may be achieved. It is, as he has stressed once and again, the generalization of the notion of “mutual gains from voluntary exchange” from the market arena to the arena of political collective action that is the guiding theme of his research program. As he puts it:

“If we adhere strictly to the individualist benchmark, there can be no fundamental distinction between economics and politics, or more generally between economy and polity. The state, as any other collective organization, is created by individuals, and the state acts on behalf of individuals. Politics, in this individualistic framework, becomes a complex exchange process, in which individuals seek to accomplish purposes collectively that they cannot accomplish noncollectively or privately in any tolerably efficient manner. The catallactic perspective on simple exchange of economic goods merges into the contractarian perspective on politics and political order” (Buchanan 2001[1988a]: 62]).

This is what Buchanan has in mind when he speaks of an “individualist-democratic methodology” (1975: 5), when he describes himself as, “in basic values, an individualist, a constitutionalist, a contractarian, a democrat—terms that mean essentially the same thing to me” (ibid.: 11), or when he characterizes his approach as “democratic, which in this sense is merely a variant of the definitional norm of individualism” (ibid.: 4).

Even if he has not as expressly as Buchanan emphasized the fact that their common individualistic value premise unites the liberal ideal of individual liberty and the democratic ideal of citizen sovereignty, Hayek (1948: 29) asserts: “True individualism not only believes in democracy but can claim that democratic ideals spring from the basic principles of individualism.”³³ He has repeatedly pointed out that the legitimizing principle in democratic politics must be seen in “the consent of the people” (Hayek 1979: 3 and 4) and that the “ultimate justification” for conferring the power to coerce to a democratic government is “that all have … an interest in the existence of such power” (ibid.: 6).³⁴

By contrast to natural-rights libertarians—whose failure to adequately separate matters of principle from matters of prudence in the liberal doctrine impedes their capacity to productively discuss policy issues with their fellow citizens—Hayek explicitly recognizes the need for liberals to convince their fellow citizens of the advantages of the principles they advocate.³⁵ While they may well, so he argues, disagree with what the majority decides, they can nevertheless agree to “majority rule as a method of deciding.”³⁶ The principles that the liberal advocates, so Hayek (1960: 115) argues, are “not proved wrong if democracy disregards them, nor is democracy proved undesirable if it often makes what the liberal must regard as wrong decisions.”³⁷

Before concluding this paper it is worth mentioning an aspect of Buchanan’s constitutional economics that even some of his most sympathetic readers have found puzzling, which is, however, as I submit a straightforward implication of his contractarian outlook. As an economic approach Buchanan’s constitutionalism starts naturally from the Smithean view that it is not in counting on human “benevolence” that we can hope to build a “good society” but in establishing and enforcing a framework of “rules of the game” that channel self-interested human behavior in socially beneficial directions. In the founding treatise of constitutional economics, The Calculus of Consent, Buchanan and Tullock (1962: 23) note that “the theory of markets or the competitive organization of economic activity is based” on the Smithean insight that “insofar as possible, institutions and legal constraints should be developed which will order the pursuit of private gain in such a way as to make it consistent with” the common interest of the group as a whole. The aim they pursue with The Calculus is, as they put it, to extend this very insight to the realm of politics, “pointing the way toward those rules for collective choice-making, the constitution, under which the activities of political tradesmen can be … reconciled with the interests of all members of the group” (ibid.: 23) in similar ways in which Smith had shown how “the self-seeking activities of the merchant and the moneylender tend to further the general interests of everyone in the community” (ibid.).

It is, however, one thing to explain how under “appropriate rules” self-interested individuals can come to play mutually beneficial games, and it is quite another thing to explain how self-interested individuals come to establish among themselves “appropriate rules,” rules which serve their common interests. Common interests are a necessary but not a sufficient requirement to achieve effective agreement in constitutional matters. To be sure, to the extent that individuals are free to move between communities with different constitutional regimes they may be able individually and separately to realize their constitutional preferences by joining those regimes that best suit their preferences. Yet, in order for individuals to be able to choose between different regimes these regimes must, in the first place, be established and maintained within the communities between which they may move. And for this internal problem of constitutional choice individual migration does not provide a solution. Within each community rules that serve the common interest of all its members are obviously a paradigm example of a collective good, and in large number settings there may, as Buchanan (1999 [1989]: 370) points out, “exist little or no incentive for any single player to participate actively in any serious evaluation of the rules,” such that “the fully rational player will refrain from participating in the choice among regimes.”

What may have irritated some readers is the conclusion Buchanan (ibid.: 371) draws in regard to the collective-good problem in constitutional choice, namely “that becoming informed about, and participating in the discussion of, constitutional rules may require the presence of some ethical precept that transcends rational interest for the individual.” Such conclusion, they suspect, looks as if Buchanan’s economic approach admits defeat at the very level of analysis that is its special domain, abandoning, when it comes to matters of constitutional choice, the very behavioral assumption that is at the core of the economic tradition in the social sciences. Yet, in my reading, Buchanan is here not calling for a shift in behavioral assumptions but, instead, he diagnoses a plain fact, namely that in a community the prospects of “playing a better game” will critically depend on its members willingness to constructively participate in a constitutional discourse with the aim of identifying and implementing rule-changes that promise mutual benefits. A community of narrowly self-interested individuals who are unwilling to invest in such constructive participation will simply fail to realize mutual gains that could be had. In the face of such dismal prospects rational individuals should, in their capacity as member of communities, be able to recognize that there are prudential grounds for them to encourage in each other an attitude of democratic citizenship, a sense of shared responsibility for their common affairs. This is, I submit, the essential message when Buchanan (ibid.; 369) states: “I want to suggest here that each one of us, as a citizen, has an ethical obligation to enter directly and/or indirectly into an ongoing and continuing constitutional dialogue.”

The constitutional interests that members of a community may have in common do not become effective by themselves. They have to be politically implemented. This insight leads to the recognition that the willingness to constructively participate in the project of constitutional maintenance—or, in short, democratic citizenship—is an attitude that should be deliberately cultivated. Calling for such an attitude does not mean to ask individuals to sacrifice their own interests for the “common good.” It means, instead, asking them to do—in their own interest—their share in maintaining the very prerequisites that allow them to successfully pursue their own interests.

4 Conclusion

Hayek has been rightly praised for his most important role in the modern revival of classical liberalism. I want to submit, though, that a truly “modern” liberalism must fill a void in the classical liberal tradition that Hayek only started to address, namely to complement the well-developed liberal theory of the market by a consistent liberal theory of democracy.³⁸ As I have argued above, to have shown how this void may be filled is the specific contribution of James Buchanan to a modern liberalism.

An anonymous referee has censured my argument for not recognizing that, by invoking Buchanan’s level of constitutional choice, it “opens the possibility that utilitarian considerations can compromise liberty.” The “weighing of costs against benefits” at the constitutional level, the referee objects, “opens the door to a move away from the classical liberalism” Buchanan is claimed to defend, because “people might trade off some liberty for some security,” or even choose “at the constitutional level a socialist society.”

This objection pertains in fact to the central claim that this paper seeks to defend. If the supreme normative principle of “classical liberalism” is individual liberty as private autonomy then utilitarian considerations at the constitutional level that compromise such liberty must, indeed, be ruled out. My point is, however, that such understanding of individual liberty represents a too narrow interpretation of the normative premise on which classical liberalism is based. Its fundamental normative principle must be seen, so I submit, in the principle of individual sovereignty, the norm that individuals should be free to choose how they wish to live, individually and collectively, and that their voluntary mutual agreement is the only source from which social arrangements can derive their legitimacy. The constitutional emphasis of Buchanan’s liberalism emphasizes the fact that the principle of individual sovereignty includes individuals’ freedom of constitutional choice, their liberty jointly to submit to rules that they expect to allow for a more attractive life than would otherwise be possible.

Recognizing individuals’ freedom to choose at the constitutional as well as at the sub-constitutional level implies that we ought to distinguish between two different kinds of claims in classical liberalism. The claim that individuals’ sovereignty in constitutional matters must be respected no less than their liberty conceived as private autonomy is, so I argue, for classical liberalism a matter of principle which must be unconditionally respected. This is not meant to diminish the importance of the rich arsenal of forceful arguments in favor of a constitution of liberty that the classical liberal tradition commands. It is, instead, meant to recognize that these arguments concern matters of prudence. They point out prudential reasons why, in light of the benefits it promises, people should want to embrace a constitutional order that provides broad scope for private autonomy.

With his constitutional liberalism Buchanan challenges his fellow liberals to acknowledge more explicitly than they usually do that with their arguments in favor of the liberal order they cannot bypass their fellow-citizens’ sovereignty in constitutional matters, but must seek to convince them why there are good reasons to prefer such an order. In other words, Buchanan admonishes his fellow liberals to acknowledge that the fundamental normative premise of classical liberalism requires them, as a matter of principle, to respect their fellow citizens’ constitutional choices, even if they go against the prudential arguments offered.

In giving it the label “modern liberalism” I am surely not suggesting that Buchanan’s contractarianism is meant to revoke in any way the classical liberal belief that a constitution of liberty provides for a more humane, productive and overall attractive social order for people to live in than any known alternative. Buchanan’s liberalism is modern not in moving away from its classical predecessors’ belief in the value of individual liberty as private autonomy but in insisting that liberals must address their arguments for this belief not to some “external judge” but to their fellow citizens, in respect of their sovereignty in matters of constitutional choice. Buchanan’s liberalism is modern in requiring liberals to adopt this attitude in advocating their cause, the only attitude appropriate in a democratic society in which equally free persons with equal rights have to come to terms with each other in such matters.