The parsimony of libertarianism.

• Author: Fleming, James E.

I want to begin by congratulating Randy Barnett on writing The Structure of Liberty,(1) one of the most radical and provocative works of political and legal theory that I have ever read. I consider myself to be a liberal who prizes liberty. Barnett claims to provide an account of the structure of liberty along with "[t]he liberal conception of justice" and the rule of law.(2) His is a radical libertarian account centrally concerned with protecting the fundamental natural rights of property, first possession, freedom of contract, and self-defense. In Barnett's world, the fabled libertarian night-watchman state has been downsized and privatized: It is a world of private courts, private police, and private prisons where inmates work to earn enough money to pay restitution to their victims.

If this is liberalism, it is enough to make me an anti-liberal, at least of the civic republican, progressive, and radical feminist strains. Indeed, for years I have been defending liberalism against criticisms by such anti-liberals, charging that they were attacking a caricature of liberalism. Well, Barnett's book embodies the caricature of liberalism that they attack. I mean this as a compliment: His book provides a caricature of liberalism by boldly exaggerating some of its characteristic features, in particular, its libertarianism and its fear of state power. Indeed, Barnett's The Structure of Liberty should join (if not replace) Robert Nozick's Anarchy, State, and Utopia(3) in the standard, obligatory footnote references to libertarian political and legal theory.

There is much to praise and much to criticize in Barnett's provocative book, and it deserves vigorous and thoroughgoing engagement. I shall focus on three points. The first relates to his account of "the" liberal conception of justice. The second concerns his natural law method of reasoning, in particular, his ambitious expansion of H.L.A. Hart's famous notion of the minimum content of natural law. The third relates to his account of the rule of law, in particular, his aggressive elaboration of Lon Fuller's well-known conception of the formal principles of legality to include a requirement of "compossibility" and the parsimony of rights. Thus, my essay is titled "The Parsimony of Libertarianism."

1. "THE" LIBERAL CONCEPTION OF JUSTICE

   John Rawls gave his classic work of political theory the modest title A Theory of Justice.(4) Barnett labors under no such modesty. He claims to develop "[t]he liberal conception of justice."(5) To be sure, he sometimes calls his conception that of "the classical liberal approach."(6) But he does not systematically distinguish, or articulate the connection, between his conception of liberalism and other, non-classical versions of it. If both Rawls and Barnett are liberals, we must ask, "How capacious is the tradition of liberalism?"(7)

   As I stated previously, I have spent years defending liberalism against anti-liberal critics, including civic republicans, progressives, and radical feminists. My move is generally to argue that (a broadly Rawlsian) liberalism is more capacious than such anti-liberals have recognized and to argue that liberalism, properly conceived or reconstructed, can be synthesized with civic republicanism, progressivism, and feminism.(8) The resulting liberal republicanism or liberal feminism yields a conclusion that liberalism can sponsor a limited formative project. Under such a liberalism, government should secure the basic liberties that are preconditions for self-government in two senses: not only deliberative democracy but also deliberative autonomy. Securing these two sorts of preconditions, I argue, would afford everyone the common and guaranteed status of free and equal citizenship in our morally pluralistic constitutional democracy.

   In making such moves, I sometimes explicitly and sometimes implicitly distance liberalism from classical liberalism or libertarianism.(9) Barnett's book prompts the question: Should liberalism be capacious enough to include libertarian theories like his? Or should it exclude such theories, notwithstanding Barnett's claim to develop "the" liberal conception of justice? In analyzing the development of the tradition of liberalism from classical liberalism to contemporary liberalism, Stephen Holmes has argued that it has been characterized by a shifting understanding of insecurity, ranging from fear of state power to fear of "private" power.(10) Barnett's liberalism reflects no such shift in the understanding of the sources of insecurity: The only ground of insecurity in his account is the fear of state power.(11) Or, with respect to other sources of insecurity that might seem to call for the intervention of state power, he argues that the medicine is "far more troublesome and dangerous than the disease."(12)

   What is at stake here is whether liberalism, properly so called, can justify a limited formative project of government to secure the preconditions for self-government through legislation. Such a formative project would share (some) common ground with civic republicanism, progressivism, and radical feminism. Barnett's version of liberalism radically questions the justice and the legitimacy of such legislation (and, for that matter, practically all legislation). Indeed, if there were a contest as to what work of libertarian theory runs the most pages before acknowledging the existence of legislatures and of legislation, Barnett's book might win: legislation is not mentioned until page 124. And even there, he does not acknowledge the legitimacy of legislation. Instead, in the course of waxing eloquent about the characteristics of common-law adjudication, he expresses doubts about the original idea that errors in the common law "were supposed to be corrected by occasional acts of legislation."(13) He goes on to complain: "Today, legislation is hardly extraordinary and is hardly confined to correcting doctrinal errors of courts."(14) He continues:

   Indeed, for some time now the legislative process has tended to overshadow and even to supplant common-law processes as the principal engine of legal discovery and change. This has meant that legal evolution has sometimes been replaced by legal revolution--and the disruption and hubris that typically accompanies revolutions--as the dominant approach to legal change.(15) Even more remarkably, in a section on "[l]aw and [l]egislation" in his fable imagining a polycentric constitutional order, he does not refer to legislation enacted by legislatures.(16) Instead, he states that "judicial opinions are commonly supplemented by reference to `codes' or legislation written by authoritative outside institutions" such as the American Law Institute's Restatements of the Laws or the legal experts who write treatises.(17)

   I am grateful to Barnett for sharpening and clarifying the differences between classical liberalism and the liberalism that I wish to develop and defend (by synthesizing it with civic republicanism, progressivism, and feminism). And I am grateful to him for providing me with a citation to use in deflecting anti-liberal critiques of liberalism. I no longer have to object that they attack a caricature of liberalism. Instead, I can employ a strategy of confession and avoidance: I can confess that their criticisms are well-taken against versions of liberalism like Barnett's, but I can avoid those criticisms by arguing that they do not apply to versions of liberalism like Rawls's (or for that matter, Ronald Dworkin's,(18) Bruce Ackerman's,(19) or my own).

   Here we should ask Barnett, just how capacious is liberalism? Does he mean, by claiming to elaborate "the" liberal conception of justice, to write Rawls, Dworkin, and others out of the canon of liberalism? How would he conceive the core, and the boundaries, of liberalism? How does he conceive the differences, and the connections, between his conception of liberalism and those of Rawls, Dworkin, and others?

2. THE NATURAL LAW METHOD OF ANALYSIS

   A

   Barnett opens his book with a perspicacious...