HomeMust Politics Rest on Morality?EducationAtlas University
No items found.
Must Politics Rest on Morality?

Must Politics Rest on Morality?

10 Mins
June 22, 2010

BOOK REVIEW: The Structure of Liberty: Justice and the Rule of Law, by Randy E. Barnett (New York: Oxford University Press, 1998. 347 pp.)

In The Structure of Liberty, Randy Barnett takes on a complex and ambitious project. His main goal is to provide a systematic argument defending the liberal (that is, the classical liberal) conceptions of justice and the rule of law. For the liberal conception of justice, he also provides and defends his own formulation, in the form of a list of rights. His basic argument is that his list of rights, and the rule of law, are necessary for "handling the serious and pervasive social problems of knowledge, interest and power" (Randy Barnett, The Structure of Liberty, p. 3. [Hereafter, Barnett]).

Fortunately, in undertaking this project, Barnett combines a clear, readable style with serious logical reasoning and a clear structure. He also includes chapter summaries, with numbered lists of the chapters' chief points. And on the book's main theme-the definition of the liberal conception of justice-Barnett builds his formulation gradually, providing eight formulations at key points in the book; each additional formulation contains added material based on new points that have been discussed; the added material is printed in boldface for easy comparison. Lastly, a meticulous index makes it easy for the reader to go back and refresh his memory about points made previously.

But proceeding carefully is no guarantee of truth, and The Structure of Liberty, I believe, unintentionally provides a demonstration of the need to proceed in proper order as well. Specifically, it demonstrates how important ethics-and specifically an individualist morality-is in defending liberal politics. As I will argue below, Barnett's failure to consider fundamental ethical issues (and to rely on an Objectivist ethical base) ultimately causes his argument to fail, and also leads him to wrong conclusions on some important issues.


In his first chapter, "Introduction: Liberty vs. License," Barnett describes his views on the meaning of rights and on how they are to be defended through the concept of natural law. He declares that natural-law reasoning about ethics and politics is analogous to reasoning in engineering or architecture. To be exact, natural law is of the form "given-if-then": Given certain facts; if you want to achieve certain ends, then you must act in certain ways. In natural-law ethics, principles are of the form "Given certain facts that we can identify about human nature; if you want to live a good life, then you should act in certain ways."

The Objectivist ethics clearly fits within this definition of natural-law ethics. The major contribution of Objectivism -and its major difference from neo-Aristotelian views of natural-law ethics (such as those of Barnett's philosophical mentor, Henry Veatch)-is that Objectivism replaces "if you want to live a good life" with "if you want to live." That is, it demonstrates the connection of its ethical standards to the requirements of man's survival rather than to a conception of "the good life" over and above survival.

At this point, Barnett introduces a distinction between natural-law ethics and natural rights.

While a natural law analysis could be applied to a variety of questions, including the question of how human beings ought to act (i.e., vice and virtue), the question of how society ought to be structured is a separate and quite distinct inquiry. Given the various problems that arise when humans live and act in society with others, the classical liberal answer to the latter question was that each person needed a "space" over which he or she has sole jurisdiction or liberty to act and within which no one else may rightfully interfere. The concepts defining this "liberty" or moral space came to be known as natural rights. . . . Whereas natural law ethics provides guidance for our actions, natural rights define a moral space or liberty-as opposed to license-in which we may act free from the interference of other persons (Barnett, pp. 13-15).

Again, Objectivists would agree with the idea that rights are the application of natural-law reasoning (reasoning of the "given-if-then" form); with the distinction between the question of how the individual should act and the question of how society should be organized; and with the idea that the concept of rights applies only to the latter. They would also agree with Barnett that much confusion has been caused by the failure to distinguish between the subjects of ethics and of rights. (On this distinction, see Eyal Mozes, "Deriving Rights from Egoism: Machan vs. Rand," Reason Papers, No. 17 [1992].)

But granted ethics and rights are two distinct issues, what is the relation between them? Objectivist value theory approaches the question of rights by establishing the importance of rationality, independence, and productivity. Specifically, it demonstrates that to live man must guide his actions by his independent, rational thinking, using it to produce the resources he needs to survive. The basic social requirement of man's survival, therefore, is that other people not prevent him from acting rationally, independently, and productively. Fundamentally, the only danger to man's ability to act in this way is the possibility that some other person will initiate physical force against him. Rights are therefore justified as the principles on which society must be organized to fulfill the basic social requirement: to preserve man's ability to act rationally, independently, and productively by protecting him from the initiation of force. Thus, in the logical structure of Objectivism , rights come later than ethics and are based on it. In the words of Ayn Rand , rights are "the concept that preserves and protects individual morality in a social context-the link between the moral code of a man and the legal code of a society, between ethics and politics." ( Ayn Rand , "Man's Rights," The Virtue of Selfishness, paperback edition, p. 92.)

Barnett's view is different:

In short, natural law ethics purports to instruct us on how to exercise the liberty that is defined and protected by natural rights [italics in the original]. . . . Thus, one can reject a natural law approach to proscribing [Sic] the ethics or propriety of human conduct, and still accept the usefulness of a natural rights approach to specify the appropriate principles of justice that comprise a social structure in which people can pursue happiness, peace, and prosperity. (Barnett, p. 15.)

The italicized sentence implies a logical hierarchy that is the reverse of Objectivism's; it suggests that natural rights are more fundamental than natural-law ethics. (To that extent, the approach seems to be similar to the one taken by Douglas Rasmussen and Douglas Den Uyl, when they characterize rights as "meta-normative principles." Liberty and Nature: An Aristotelian Defense of Liberal Order, La Salle, Illinois: Open Court, 1991, pp. 96-115.) In any event, Barnett clearly believes natural rights are not based on natural-law ethics and can be defended independently of it. And it is just such a defense, independent of any particular view of ethics, that Barnett tries to provide in the bulk of his book. So, let us turn to his argument and see how well it succeeds.


Barnett argues for the liberal conception of justice by trying to show that it is necessary for "handling the serious and pervasive social problems of knowledge, interest and power." Under these three headings, Barnett lists eight problems (See Figure 1.)

Barnett then defends natural rights as needed for dealing with several of the problems on the list, and for our purposes his first formulation of the liberal conception of justice will be sufficient to show how he argues.

Formulation 1.

Justice is respect for the rights of individuals and associations.

(1) The right of several property specifies a right to acquire, possess, use, and dispose of scarce physical resources-including their own bodies. Resources may be used in any way that does not physically interfere with other persons' use and enjoyment of their resources. While most property rights are freely alienable, the right to one's person is inalienable. (2) The right of first possession specifies that property rights to unowned resources are acquired by being the first to establish control over them. (3) The right of freedom of contract specifies that a right-holder's consent is both necessary (freedom from contract) and sufficient (freedom to contract) to transfer alienable property rights (Barnett, p. 83).

How do these rights solve the problems on Barnett's list?

The first-order problem of knowledge. This is the problem made famous by Friedrich Hayek: How can we wisely coordinate the use of resources, given that the relevant knowledge is dispersed among all members of society and is ever changing? The liberal answer is that the right of several property decentralizes decision making, causing each person's knowledge to be taken into account in the use of some resources, by allowing him ownership of resources about which he can decide based on his own knowledge. The right of first possession provides a way for individuals to gain control over resources without disrupting others' plans for the use of resources; when someone claims control over a previously unowned resource, he is the only person so far who has demonstrated knowledge of how this resource is to be used; if anyone else then tries to claim control over the resource without the first claimant's consent, he will not be taking all of the first claimant's knowledge into account, and this is addressed by the right of first possession's favoring the first claimant. The right of freedom of contract allows people to transfer control over resources while taking into account the knowledge of both the former and the new owner, and this is necessary for creating market prices that reflect one's knowledge.

The partiality problem. (The word "partial" is used here as the opposite of "impartial," rather than as the opposite of "complete.") Barnett defines the partiality problem as "the need to (1) allow persons to pursue their own partial interests. . . , (2) while somehow taking into account the partial interests of others whose interests are more remote to them" (Barnett, p. 137). The right of several property, by limiting people's jurisdiction, limits the impact on others of their self-interested actions. The right of freedom of contract is needed to make people take the interests of others into account, by requiring others' consent for obtaining control over the resources they own.

The incentive problem. This is the problem of "ensuring that persons have an adequate incentive to make choices reflecting the knowledge to which they have access and to discover new information" (Barnett, p. 153). The right of first possession is needed to give individuals the incentive to invest scarce resources in discovering and improving new resources. The right of freedom of contract is needed to ensure that the transfer of resources is in the interests of both parties, and to create incentives to investigate and discover opportunities for mutually beneficial exchanges.

The problem of enforcement abuse. After we have established the need for enforcing people's rights, in order to deal with the enumerated problems, there arises the problem of preventing those responsible for this enforcement from using the power they are given in a manner that violates rights. The right of several property is needed to give individuals resources that they can use to resist such abuse if it occurs.


Barnett's discussion of his list of problems, and of how liberal rights help to solve them, is frequently instructive and provides some highly original insights. The best parts are Barnett's discussion of the rule of law; "the problem of knowledge" and what is needed to solve it; and, in connection with "the incentive problem," his refutation of arguments alleging that some "public goods" need to be provided by coercive means to address "the free-rider problem" (Barnett, pp. 161-167).

However, the fundamental problem with Barnett's book is its social perspective. As Barnett states clearly in the beginning, his project assumes a social goal shared by the members of society. Objectivists may agree with the goal Barnett posits-achieving "a social structure in which people can pursue happiness, peace, and prosperity" (Barnett, p. 15). But this goal can't simply be taken for granted; for example, economic equality, ensuring that the will of god is done, or preventing social changes that would disrupt people's routines, are all goals that some people would regard as more important than allowing people to pursue happiness, peace, and prosperity. Barnett's goal for a social structure must be based on more fundamental principles in ethics and must be discussed from the perspective of the individual moral agent.

Barnett's discussion of the "problem of interest" is the part that Objectivists would find the most to disagree with. He writes that "'the partiality problem' arises from the fact that people tend to make judgments that are partial to their own interests or the interests of those who are close to them at the expense of others" (Barnett, p. 136). Thus, he does not distinguish between acting on irrational desires and acting on rational self-interest. Barnett gives no examples to concretize "the partiality problem" or explain why people acting on "partial self-interest" are in general a problem; he simply takes it as self-evident that this is the case. However, people acting on their rational, "partial" self-interest are not a problem, since men's rational interests are not in conflict. (For more on this, see Rand, "The 'Conflicts' of Men's Interests," in The Virtue of Selfishness ); problems arise only when people act on irrational desires, harming others as well as themselves. The valid way of analyzing Barnett's "partiality problem" is, therefore: the problem of how to protect individuals from suffering the consequences of others' irrationality.

When he comes to the "incentive problem," Barnett's formulation sounds bizarre. He describes that problem as "ensuring that persons have an adequate incentive to make choices reflecting the knowledge to which they have access and to discover new information" (Barnett, p. 153). But given an individual's need to survive, and the fact that reason is man's means of survival, what greater incentive could he possibly have to make choices reflecting the knowledge to which he has access and to discover new information? The only problem is to ensure he is able to make those choices and discover that information. And to be fair, that is probably what Barnett means. But it is not what his formulation says.

It is, then, a useful exercise to take the problems on Barnett's list and reformulate them in individualistic terms. As noted above, there is really one basic social problem: How to make it possible for people in a social context to use their rational, independent judgment in taking the actions, and producing the resources, they need in order to live. Barnett's problems can all be seen as aspects of this basic problem. Making sure there are no social forces preventing the individual from producing and then using the resources he needs is Barnett's "incentive problem." Making sure that others' irrationality does not get in the way gives rise to two of his problems. The irrationality of other people in general gives rise to Barnett's "partiality problem." The irrationality of those charged with protecting rights gives rise to Barnett's "problem of enforcement abuse." Allowing the individual to make use of other people's knowledge of which he is ignorant, and which may be relevant to his goals, is Barnett's "first-order problem of knowledge."

Had Barnett focused his discussion on this basic problem-which we may call "the problem of independent action" -it would have made a significant difference to some points in his list of rights. For example, Rand's argument for property rights focuses on man's right to the product of his own efforts:

"Without property rights, no other rights are possible. Since man has to sustain his life by his own effort, the man who has no right to the product of his effort has no means to sustain his life." (Ayn Rand, "Man's Rights," The Virtue of Selfishness , paperback edition, p. 94.)

Barnett, in contrast, never explicitly discusses an individual's right to the product of his own efforts (whether the right of someone working by himself, e.g., a farmer on his own land, to the product of his work, or the right of an organizer of a business concern to the product of the production process he has organized). There is a special right on Barnett's list for resources appropriated from nature, "the right of first possession." And while it seems clear that Barnett believes the individual should have property in what he has produced, this is left to implication. In any society more advanced than a hunter-gatherer tribe, the vast majority of resources used in sustaining human life are produced by human action rather than appropriated from nature; so the issue of the right to the product of one's own efforts is of much greater importance than the "right of first possession," but Barnett apparently does not regard this issue as important enough to discuss explicitly.

Could Barnett defend such a right based on his list of social problems, without referring to the more basic "problem of independent action"? I submit the answer is: no. If we start, not from the individual's need for a social order that protects his ability to act independently and produce, but from a social need to solve "the first-order problem of knowledge" or "the partiality problem," then there is no argument to prove that allowing each person to keep the product of his own effort is the best rule for solving these problems. Opponents of liberalism could offer other ways of distributing "several property," and-with no less validity than Barnett's arguments-claim that distributing property their way would solve Barnett's list of social problems better.

For example, some might claim that to handle Barnett's problems, it is important to prevent great concentrations of property in the hands of a single individual. To answer such claims, it is necessary to distinguish between political and economic power, a distinction based on the Objectivist ethics; Barnett's argument for rights, through his list of social problems, does not leave room for that distinction. And without clearly distinguishing political from economic power, great concentrations of property would be regarded as no less harmful than the power given to a central political authority.

Barnett reveals this implication of his approach through his preference for the term "several property" over "private property." Barnett says the term "several property" is better because it makes it clearer that jurisdiction to use resources is dispersed among the "several"-meaning "diverse, many, numerous, distinct, particular, or separate"-persons and associations that comprise a society. (Barnett, p. 65.) Opponents of liberalism could argue that allowing each person to keep the product of his own effort, since it allows great concentrations of property to form, endangers the existence of "several property" and its ability to solve Barnett's list of problems. If to address "the first-order problem of knowledge" we need to decentralize decision making about the use of resources, would not an extremely wealthy individual present a danger to such decentralization, since that individual, with his limited knowledge, will be making decisions about the use of a vast amount of resources? And shouldn't we also guarantee to each individual ownership of some minimum amount of property, in order to make sure his personal knowledge will be taken into account in the use of resources? If to address "the partiality problem" we need to limit the impact of people's self-interested actions by limiting their jurisdictions, doesn't this require limiting the extent of any single individual's jurisdiction by limiting the potential for accumulation of property?

Barnett would no doubt respond to such arguments by stating that any attempt to let a central authority enforce limits on the accumulation of property would create its own problems of knowledge, as well as problems of incentive and of enforcement abuse. But this would shift the discussion to a cost-benefit analysis of the problems created by great concentrations of property versus the problems solved by adherence to the liberal conception of justice. It is not obvious Barnett can demonstrate that the latter solves more problems than it creates. To win this sort of argument, advocates of liberty must demonstrate that the alleged problems created by great wealth are different, not only in degree but in kind, from the problems created by political power; and this is precisely what Barnett's analysis, based on his list of social problems, does not make possible.


Barnett presents the rule of law as a separate principle from justice, introduced to handle "the second-order problem of knowledge," which is "the need to communicate knowledge of justice in a manner that makes the actions it requires accessible to everyone" (Barnett, p. 85). Barnett discusses the formal requirements for legal rules that can guide people's actions ex ante; that is, by allowing people to know in advance, before any dispute occurs, whether an action they are considering would be just or not. The rule of law requires the existence of specific rules, even in areas in which the principles of justice in themselves leave a wide range of options; even if several possible optional rules are consistent with justice, people need to know in advance to which rule their actions have to conform.

Barnett then discusses the "third-order problem of knowledge," which is, "the need to determine specific action-guiding precepts that are consistent with both the requirements of justice and the rule of law" (Barnett, p. 108). He then describes the common-law process of adjudication as a solution to that problem.

Barnett also discusses the role of the rule of law in handling two of the "problems of interest": The incentive problem: a stable framework of rules is necessary for creating an environment in which people can be secure in making long-range plans. The partiality problem: publicly accessible rules, by which the administration of justice is expected to operate, creates a "warning sensor" that helps detect any hidden irrational motives on the part of people charged with administering justice, since such motives will often cause them to depart from the rules.

Barnett also provides an intriguing discussion of the role of lawyers in the legal system and why, contrary to the way they are often perceived, they help preserve impartiality in the legal system. Lawyers, unlike most of their clients, deal with the legal system regularly rather than on a onetime basis; this gives them a direct stake in the just operation of the legal system, and in preserving their reputation among judges and other lawyers. This counteracts the motivation of parties to a trial to use illicit tactics to win by any means.

Overall, Barnett's discussion of the rule of law is the strongest part of his book, containing the most valid and original insights about the requirements for preserving freedom and the role of legal institutions.

Eyal Mozes
About the author:
Eyal Mozes