The ethical system of John Rawls, properly understood, justifies libertarian political institutions.
John Rawls (1921–2001) was arguably the most important political philosopher of the 20th century. His systematic works, A Theory of Justice and Political Liberalism, set the stage for myriad debates within political philosophy. 1 The books also staked out a form of liberal egalitarianism. Although Rawlsians and libertarians agree on the basic priority of certain core liberal rights, such as freedom of speech and freedom of religion, they disagree dramatically on the scope of economic liberties. Rawlsians argue that only the freedom to choose one’s occupation and the right to own personal property (not capital goods) are fundamental liberties that states must protect.
Libertarians and classical liberals adopt a much broader scheme of rights, including the right to set one’s own prices for goods and services, the right against government regulation, the right to freedom of contract, and the right to own and operate private capital.
So what is a chapter on Rawls doing in a book on arguments for libertarianism? The answer is that Rawlsian arguments can be used to defend classical liberalism. In fact, there are two distinct, contrasting, and somewhat incompatible methods of grounding classical liberal institutions in Rawlsian political philosophy. The two methods strongly correlate with the lines of reasoning in Rawls’s two books, A Theory of Justice and Political Liberalism. John Tomasi uses the framework of A Theory of Justice to defend what he calls “free‐market fairness,” which differs from Rawls’s theory, “justice as fairness,” by expanding the list of economic liberties to include those celebrated by classical liberals. In contrast, Gerald Gaus has pursued the line of argument Rawls set out in Political Liberalism. Gaus offers a strong defense of markets within the Rawls‐inspired “political liberal” or “public reason liberal” framework.
In this chapter, I will focus on explaining both the Tomasian and Gausian defenses of classical liberalism by comparing them with Rawls’s approaches in A Theory of Justice and Political Liberalism, respectively. Tomasi more directly follows Rawls, whereas Gaus’s early versions of public reason liberalism predate Political Liberalism but show strong similarities with the Rawlsian approach. Toward the end, I will explore their compatibility. Importantly, I will not review libertarian criticisms of Rawls’s position. The goal of this chapter is strictly limited to using Rawlsian tools to justify libertarian institutions.
I proceed in five parts. The first reviews the basic features of Rawls’s project in A Theory of Justice, and the second explains Tomasi’s defense of classical liberalism that revises this project. The third section reviews the basic features of Rawls’s transition to Political Liberalism and the idea of public reason. The following section outlines the more complex Gausian defense of classical liberalism based in public reason liberalism. In the last section, I offer an argument favoring the Gausian defense, but I also argue that Gaus’s theory permits Tomasi’s free‐market fairness to be the correct theory of justice and to form a basis for political activism in a publicly justified polity.
Rawls’s Project in A Theory of Justice
Rawls is well known for his prominent role in reviving social contract theory in the United States in the late 20th century. His work followed—and helped displace—nearly a century of political thought, which included, prominently, both utilitarian and Marxist views.
Rawls argues that a society is just when its basic structure is regulated by principles that would be selected through a thought experiment. He imagines people in an “original position,” tasked with choosing between different principles of justice. Their choice is constrained by the “veil of ignorance,” which denies them information that could bias their selection of principles in ways that we would normally regard as inappropriate, such as drawing on one’s race or class. The principles of justice selected will form the rules for distributing what Rawls calls “primary goods” or goods that any person with a rational plan of life would want, such as rights, liberties, income, and wealth. These goods include (a) basic rights and liberties, (b) freedom of movement and free choice among a wide range of occupations, (c) the powers of offices and positions of responsibility, and (d) income and wealth. More elusively, Rawls includes among primary goods (e) the social bases of self‐respect, which he understands as “the recognition by social institutions that gives citizens a sense of self‐worth and the confidence to carry out their plans.” 2
Critically, Rawls’s veil of ignorance prohibits people from choosing conceptions of justice based on deep features of their identities that do not ordinarily seem inappropriate bases for determining what justice requires. For instance, Rawls denies that persons could appeal to their conceptions of the good, such as one’s worldview or religion, in selecting principles of justice.
More radical still, Rawls denied that people in a just society could appeal to their natural talents, such as a person’s mathematical or musical ability, to claim a greater share of primary goods. For instance, Jane may not claim a larger share of primary goods even if she produced a surplus of goods by using her natural talents. Imagine that Jane invents a computer program that substantially reduces her firm’s costs, and that invention leads to her securing a higher income from her job. In Rawls’s view, although a society’s particular constitution or laws may allow her to keep her income, she has no basic right to it in virtue of using her natural talents. The reason is that natural talents are undeserved. Rawls claims that the distribution of natural talents is arbitrary from a moral point of view, meaning that natural talents are distributed by nature at random, and so not in accord with justice or equity. This latter claim has raised libertarian hackles, and rightly so, but the two versions of Rawls‐inspired classical liberalism show that the Rawlsian framework can progress even if Rawls’s position on natural talents is mistaken.
Central to Rawls’s theory is that the parties in the original position are not real‐world persons. Instead, their choice is a model of the process of reflective equilibrium, where persons theorize together in order to harmonize their considered judgments about the requirements of morality and justice. What the parties choose is therefore the best way, or one of the best ways, to identify the conception of justice to which real‐world persons are rationally committed. Rawls’s is not a hypothetical consent theory of politics, where the justification of political order is rooted in what persons would agree to under certain conditions but that they have not in fact agreed upon. For Rawls, no important normative claim is made true by the consent of hypothetical persons. Instead, hypothetical consent is a heuristic for identifying principles of justice that render the broadest, most coherent explanation of the moral and political judgments we share.
When parties choose, their choice is both rational and reasonable in the sense that their choice both follows the canons of rational choice and is suitably impartial or unbiased. Also note that the parties do not choose for us. Instead, they choose principles to govern a well‐ordered society, which models persons like us living under favorable conditions. Once the parties select principles of justice, Rawls argues that the principles must be tested against a psychologically and sociologically realistic model of society to see whether the rules can self‐stabilize among persons disposed to be just.
Rawls wants to ensure that normally functioning persons can come to comply with the principles of justice because they will recognize doing so as good for them. Only then will principles of justice comport with our pretheoretical expectation that the true principles of justice be ones that can survive public scrutiny. Just institutions, that is, need not hide their normative basis. People should be able to access the foundations of their institutions and abide by them on the basis of approving those foundations, given their conception of justice and the good.
The basic idea is that the correct principles of justice should promote social stability in a public fashion, in contrast to views, like utilitarianism, that may require that governments hide their utilitarian principles so as not to discourage citizens from complying with governmental dictates. So again, Rawls’s construction does not involve parties choosing principles for us. Instead, the construction attempts to locate principles of justice that satisfy our considered judgments about justice, both our substantive judgments about what is and is not just, and our procedural judgments about the nature of justice, such as whether institutionalizing justice will be stable for normally functioning human beings.
Libertarians sometimes mistakenly construe Rawls as a hypothetical consent theorist. This has led to much confusion. As we can see, Rawls’s real view is different, richer, and more plausible.
Rawls then argues that parties will choose two principles of justice, known together as “justice as fairness”:
First Principle: Each person has the same indefeasible claim to a fully adequate scheme of equal basic liberties, which scheme is compatible with the same scheme of liberties for all.
Second Principle: Social and economic inequalities are to satisfy two conditions:
They are to be attached to offices and positions open to all under conditions of fair equality of opportunity.
They are to be to the greatest benefit of the least advantaged members of society (the difference principle). 3
The parties choose these principles because the principles will protect and promote the parties’ capacity to exercise their two moral powers: their power to be rational and their power to be reasonable. The first moral power is exercised when people form a rational plan of life, whereas the second power is exercised when people formulate and live out a conception of justice.
The first principle ensures that people will be able to claim and exercise the basic liberties required to exercise their two moral powers, by protecting freedom of speech, religion, and press, among others, and by protecting procedural rights, like the right to a fair trial. Citizens must also have the means to realize the “worth” of those liberties, that is, to be able to exercise them in meaningful ways on a regular basis. This principle has lexical priority over the second principle; the second principle never overrides the first.
The second principle ensures that people will be able to enjoy the worth of those liberties on equal terms. Fair equality of opportunity protects against the dominance of any one social group, whereas the difference principle ensures that even the poorest have access to the primary goods necessary to exercise their two moral powers.
Importantly for the entire project, Rawls claims that choice based on our two moral powers derives from our shared conception of the person. Reasonable people have different views about what the human person is, but they agree on at least some common features, and those common features form the basis of the Rawlsian project.
Notice how egalitarian Rawls’s theory of justice is. Yes, Rawls protects a great many liberties as strongly as he can. But his list of liberties includes only two economic liberties: the right to freely choose one’s occupation and the right to own personal property. That means that the state may not order persons to perform jobs; neither may it expropriate the personal—that is, noncapital—goods of persons. But no other economic liberties are protected. Therefore, in principle, a Rawlsian can support vast amounts of government redistribution, regulation, and even socialist economic production.
Regarding the second principle, libertarian discussions of Rawls’s view have tended to focus on the difference principle. But the fair equality of opportunity is in many ways more egalitarian, and it is lexically prior to the difference principle in that conflicts between the two principles must always resolve in favor of fair equality of opportunity.
Fair equality of opportunity, for Rawls, means that social inequalities can be justified only if they promote fair equality of opportunity. The only reason the state may allow you to become wealthy, influential or, powerful is if you do so under a system of rules that promotes equal opportunity for everyone. Your unequal wealth, influence, and power—no matter how nobly and virtuously achieved or deserved—can be eliminated, even if it arises within a social system that provides great but unequal opportunities to everyone. Unless the opportunities are equal, they are unjust. Rawls allows but one exception: opportunities used to make the least well‐off better off.
The difference principle is, in my view, a bit milder. It requires that inequalities of primary goods maximize the position of the least advantaged members of society. Again, even if you acquire more wealth through productive activities that benefit everyone and exploit no one, the state is not morally required to protect that wealth. Your efforts leading to inequality must occur within a system of rules that maximize the position of the least advantaged members of society. Few theories of distributive justice are more egalitarian.
The institutional implications of Rawls’s position are similarly egalitarian. Rawls reviews five regime types: (a) laissez faire capitalism with a social minimum (such as a guaranteed minimum income), (b) a capitalist welfare state (with more extensive social insurance, regulation, and government power), (c) property‐owning democracy (where the ownership of capital is constantly redistributed away from large capital holders), (d) liberal socialism (basic liberal rights plus government ownership of the means of production), and (e) command‐economy socialism. Laissez faire fails to realize the principles of justice because it does not guarantee the ability of persons to enjoy the “worth” of their liberties, especially their political liberties, as the rich and powerful will dominate politics. Although welfare states do better on this score, they are also vulnerable to great political inequalities and so will be dominated by the rich. Command‐economy socialism violates basic liberties, such as freedom of occupation. That leaves only property‐owning democracy and liberal socialism, both of which involve enormous amounts of government intervention in the economy.
Rawls acknowledges economic problems for capitalism but ignores problems for property‐owning democracy and liberal socialism. Capitalism falls victim to inequality, insufficiency, and allows the wealthy to rule. Property‐owning democracy and liberal socialism can be assumed to work largely as they are intended. That is because Rawls’s work is situated within ideal theory, where we choose a conception of justice for persons living under favorable conditions who are prepared to comply with their political institutions so long as others do likewise. Why Rawls didn’t apply the same assumptions to his model of capitalism is an important question, one Tomasi tries to remedy.
In sum, justice as fairness appears to be both highly egalitarian and deeply statist.
Tomasi’s Free Market Fairness
After all this, you must be wondering how Rawls’s apparatus could be used to support libertarian positions. It’s a good question. I will set aside attempts to show that capitalism satisfies the difference principle better than socialism. Some have pursued this line of thought, arguing that the free market in fact maximizes, for instance, the wealth and income of the least advantaged. This is a shallow attempt to use Rawlsian ideas to justify libertarian conclusions. The fix is too easy. Instead, we need to consider again the basic motivations of Rawls’s theory of justice and see whether they lead where he believes they do.
Enter John Tomasi’s recent book, Free Market Fairness. 4 Tomasi takes on many features of Rawls’s methodology of justice, such as original position reasoning and the veil of ignorance. He embraces ideal theory, arguing that identifying conceptions of justice requires finding a set of general principles to govern a well‐ordered society. This is to say that a well‐ordered society contains persons living under favorable conditions who are prepared to comply with a conception of justice. Tomasi also takes on many aspects of Rawls’s conception of the person, though he describes that conception as a “responsible self‐author,” which is not the same. However, because Tomasi spends little time contrasting the two ideas, I will not explore the differences here. 5
Tomasi also argues that parties will select basic liberties that are based on their shared conception of the person as a responsible self‐author. Those basic liberties have a similar special priority over other considerations of justice. Tomasi’s major departure from Rawls comes from the list of economic liberties he endorses. Tomasi accuses Rawlsian liberals of “economic exceptionalism” that singles out a few economic liberties for protection as basic rights. Tomasi’s research program, market democracy, of which free‐market fairness is a Rawlsian variant, does not endorse “absolute” property rights. Instead, it “affirms a thick conception of economic liberty as part of a broader scheme of rights and liberties designed to enable citizens to exercise and develop their moral powers.” 6
Tomasi’s critique draws our attention to Rawlsian arguments for protecting a short list of economic liberties and then attempts to show that these arguments also support expanding the list. In other words, Tomasi claims that thick, libertarian‐like economic liberties should be treated on a par with freedom of occupation and the right to own personal property. Tomasi argues, for instance, that if the right of occupational choice is a basic right, “it becomes unclear how the other liberties of working can be excluded.” 7 Thick liberties of working include the rights to sell, trade, or donate one’s labors. That is, persons have basic rights to determine the conditions of their working lives in general. The right to own productive property can also be justified on the basis of an analogy with the right to own personal property: both ownership rights “can provide a person with personal security” and can serve to express the owner’s identity.
One of Tomasi’s illustrations of that latter point is “Amy’s Pup‐in‐the‐Tub,” a small business owned and operated by a woman in the Rhode Island area, whose values are expressed in her work. Amy loves animals, especially dogs, and she derives great meaning from making her living caring for them. Tomasi also argues that people should be free to engage in “long‐term financial planning” to avoid diminishing “the capacity of citizens to become fully responsible and independent agents.” 8 So even persons whose values and commitments are not tied to their work should have the basic right to enjoy the fruit of that work, namely, their income.
Tomasi’s second line of argument in defense of libertarian‐like conclusions is that respect for responsible self‐authorship requires a “distributional adequacy condition,” where
a defense of any version of liberalism is adequate only if it includes the claim that the institutions being endorsed are deemed likely to bring about some desired distribution of material and social goods. 9
This condition could be egalitarian, if the distribution must satisfy some standard governing the “relative holdings of citizens” where some equalization is required by justice. But it could also be a sufficientarian view where the distribution must ensure that all persons have access to some minimum amount of resources. 10 Or the condition could require the maximization of available resources. Nonetheless, a distributional adequacy condition is required.
Lest you think Tomasi is rejecting libertarianism, Tomasi claims that nearly all historical classical liberals and libertarians accept a distributional adequacy condition and argue that free‐market regimes satisfy the condition. 11
Tomasi’s third basis for classical liberal conclusions is his alteration of the traditional Rawlsian list of primary goods. Rawls argues that primary goods are all and only those essential for persons to develop and exercise the two moral powers that anyone would want no matter his or her rational plan of life. Recall that primary goods include rights and liberties, along with the social bases of self‐respect. Thus, Rawls’s second principle, which distributes primary goods, has a role in distributing basic rights, liberties, powers, and positions rather than, say, mere money.
However, Tomasi argues that the Rawlsian commitment to maximizing the share of primary goods enjoyed by the least advantaged is ambiguous between different lists of primary goods. Some lists will include major improvements to rights and liberties and minor improvements to the social bases of self‐respect. Other indexes will choose the reverse. Some difference principle goods (DP goods, as Tomasi terms them), then, are rivalrous, and it is not clear how to trade them off.
Tomasi argues that, in some cases, reasonable people will prefer more wealth to more workplace democracy. That is, reasonable people might prefer to have bosses but higher income, whereas others might prefer the reverse. If so, we must attend to levels of social wealth in maximizing the bundle of primary goods. And because, as Tomasi claims, capitalism is such an effective wealth generator, we have an argument that the difference principle requires capitalism (which is not the same as the easy‐fix approach mentioned earlier). Tomasi does not mean to claim, however, that free‐market fairness requires trading off rights and liberties. Instead, maximizing wealth helps the least advantaged enjoy their rights and liberties.
The alteration and development Tomasi proposes for the list of primary goods will alter the recommendations of the difference principle and fair equality of opportunity. 12 Rawlsian “high” liberals give pride of place to workplace democratic liberties among these goods, but Tomasi argues that free‐market fairness focuses much more on the maximization of wealth, combined, perhaps, with a social minimum of some sort. 13 He claims that free‐market fairness affirms traditional liberal ideals of formal equality of opportunity and, in doing so, holds that “every citizen, regardless of birth status or economic class, is owed high‐quality health care and education as a matter of justice.” 14
But Tomasi denies that justice must be carried out by the “direct” mode of social construction—the form of direct, deliberate government intervention that attempts to design social outcomes. Instead, classical liberals rely on an “indirect” mode of social construction that “emphasize[s] market mechanisms in pursuit of a superior system of education and health care for all.” 15 The “market democratic strategy is to create systems with the maximum number of decision points.” 16
Finally, Tomasi uses Rawlsian ideal theory to show that free‐market fairness, much like justice as fairness, implies ideal regime types, but ones very different from property‐owning democracy and liberal socialism. The two regime types are “democratic laissez faire” and “democratic limited government.” The former is more radically libertarian; Tomasi associates it with Murray Rothbard–esque opposition to state power. 17 Democratic laissez faire will involve quite limited government and seeks to use market mechanisms as often as it can to satisfy the abstract demands of free market fairness. It “marks the high point of optimism about market mechanisms (and skepticism about political decision making).” 18 Democratic limited government is weaker: although “enthusiastically capitalistic, this regime type allows a greater degree of direct government intervention in economic affairs.” 19 Tomasi associates this view with F. A. Hayek and Milton Friedman.
The point of ideal theorizing is twofold. First, it allows us to steer clear of the common failings of human beings that should not pollute our conception of justice, even if these failings should alter our institutional recommendations. Second, it allows us to see how our preferred institutional proposals link up with our commitment to justice under favorable conditions. Satisfying these two commitments allows ideal theorizing to give us inspiring ideals that can establish the moral superiority of some conceptions of justice over others. Many libertarians are wary of ideal theory, seeing it as an escape strategy for socialists. But Tomasi argues that ideal theory favors libertarianism 20 —an important twist.
The Turn to Political Liberalism
After the publication of A Theory of Justice (TJ), Rawls became increasingly convinced that the well‐ordered society he described was “unrealistic.” 21 In Part 3 of TJ, Rawls had assumed that members of a well‐ordered society would agree on a partial conception of the good. Consequently, he could demonstrate that TJ’s notion of a well‐ordered society is stable by showing that this partial conception of the good was congruent with justice as fairness. Such a society would be inherently stable because each person believes that compliance with institutions that manifest justice as fairness is compatible with his or her good and the good of others, and so compliance is the best response to the actions of others.
Rawls came to believe that citizens would share in the affirmation of those goods only if they had a shared conception of the person as a free and equal moral agent. But reasonable and rational persons might reject this conception of the person and so reject justice as fairness as an appropriate framework for regulating their behavior. Justice as fairness, therefore, would no longer be stable in the right way, as some agents would be practically rational to defect from the institutions that manifest justice as fairness. True, the state could forcibly impose justice as fairness on an unwilling populace, but Rawls thought that a conception of justice should receive the free support of its members, or else that conception had a significant flaw.
The problem faced by TJ is that its notion of a well‐ordered society, one regulated by a conception of justice, allows for the free use of practical reason. People can reason freely about what is best and most important in life. But doing so leads people to affirm a variety of incompatible worldviews, philosophies, and religious beliefs. Some of the doctrines will reject justice as fairness and the conception of the person on which justice as fairness is based. And the people who hold those doctrines will begin to dissent from the social order that realizes justice as fairness, and their behavior will follow suit. Stability, therefore, will break down.
The breakdown of stability led Rawls to recast his theory. First, Rawls converted justice as fairness into a “political conception of justice” 22 whose conception of the citizen is “free‐standing” from the details of the reasonable comprehensive doctrines prevalent in a well‐ordered society. A political conception of justice must be compatible with each reasonable comprehensive doctrine so that an “overlapping consensus” can form around the political conception. An overlapping consensus occurs when all reasonable comprehensive doctrines in a society converge on or overlap on a political conception of justice. 23
Rawls also relaxes the congruence relation. A citizen need not endorse a political conception of justice as part of his or her personal good. Reasonable positions need only be “congruent with, or supportive of, or else not in conflict with” political values, or the values that are part of a shared, political conception of justice. 24 In Political Liberalism (PL), stability, now called “stability for the right reasons” can occur even if diverse comprehensive doctrines merely fail to provide citizens with sufficient reasons to reject justice as fairness. In fact, justice as fairness need only “normally outweigh” each citizen’s privately held doctrines and values.
To demonstrate that an overlapping consensus is possible, Rawls tries to show that persons share conceptions of citizenship and of political society regardless of their reasonable comprehensive doctrines. Reasonable people conceive of citizens as free and equal and of society as a cooperative venture for mutual gain. As a result of sharing those ideals, citizens will want to abide by principles that each reasonable person can accept. The implication of this motivation is that political justification takes the formdescribed by Rawls’s “liberal principle of legitimacy,” which he defined as follows:
Our exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in light of principles and ideals acceptable to their common human reason. 25
By sharing political conceptions of the citizen and society, citizens of a well‐ordered society should all be committed to the liberal principle of legitimacy, which says that state coercion can be justified only by reasons that draw on shared political values. In other words, reasons must be public. By using public reasons in discussions about justice and constitutional matters, citizens can assure one another that their political activities are compatible with their shared political conception of justice. Citizens are obliged to cooperate only when others do likewise, so they must have stable expectations that others are disposed to cooperate. The use of public reasons, by virtue of drawing on shared normative resources, becomes not only a sign of mutual respect but also a signal of mutual assurance. Rawls then specifies an ethics of citizenship that is based around his “duty of civility,” which requires that citizens, when discussing constitutional essentials or matters of basic justice, offer public reasons for their positions.
The notions of a freestanding conception of the person, a political conception of justice, shared conceptions of citizenship and society, the idea of an overlapping consensus, the importance of publicity and public reasons, and the duty of civility all flow from Rawls’s attempt to outline the structure of a well‐ordered society that is stable for the right reasons. The main ideas in PL and their motivation are now in view.
In Rawls’s first versions of political liberalism, he assumed that persons endorse the same conception of justice, namely, justice as fairness. But the fact of reasonable pluralism, he would later admit, would also lead to reasonable disagreement about justice. Paul Weithman argues that this admission came long after he began to “recast justice as fairness.” 26 Pluralism about justice, then, was not an important part of Rawls’s political turn. But we shall see that justice pluralism is an important part of determining whether the framework of political liberalism is compatible with or even supportive of classical liberalism.
The reason for this is that, for Rawls’s last formulations of political liberalism, he recognized that different liberal political conceptions of justice could be legitimate for a well‐ordered society. Those conceptions must share a “criterion of reciprocity” in social cooperation and recognize the burdens of judgment, or the features of political and moral reasoning that lead rational and informed persons of goodwill to disagree on matters of fundamental import. The conceptions must also have three common features:
First, a specification of certain rights, liberties, and opportunities (of a kind familiar from democratic regimes); second, a special priority for these freedoms; and third, measures assuring all citizens, whatever their social position, adequate all‐purpose means to make intelligent and effective use of their liberties and opportunities. 27
Notice how broad that range is, at least in principle: Tomasi’s free‐market fairness fits within the range. Tomasi’s view specifies the relevant rights, liberties, and opportunities and gives them lexical priority; and his distributional adequacy condition is meant to ensure that all citizens have access to means to make effective use of their rights and liberties. I suspect Rawlsians will find a way to deny that Tomasi’s free‐market fairness is a reasonable liberal political conception. But from what Rawls says, there are no grounds to rule it out. Not even PL’s special focus on political liberty can do so, as Tomasi draws on the value of political liberty to ground his case for classical liberal institutions. Perhaps Tomasi has the wrong conception of justice, but his view, on Rawlsian terms, is surely reasonable.
Rawlsians will object that Rawls rejected libertarianism in PL because it has “no special role for the basic structure” of society, where a basic structure is “a society’s main political, social, and economic institutions, and how they fit together into one unified system of social cooperation from one generation to the next.” 28 The problem with libertarianism is that it views the state as just one more “private association,” and so there is “no uniform public law that applies equally to all persons, but rather a network of private agreements.” 29 In this way, libertarianism “rejects the fundamental ideas of the contract theory.” 30
I will not take on Rawls’s criticism of Nozickian libertarianism here, 31 but it should be plain that some versions of libertarianism advocate a uniform public law applied equally to all persons and that makes a special place for the basic structure in its theory. Tomasi’s view is one example, but so are most forms of constitutionalist classical liberalism, such as those advanced by Hayek and Friedman. So the Rawls of the later versions of PL is likely committed to the position that some forms of libertarianism are reasonable political conceptions of justice. That is an interesting and surprising result.
Gaus’s The Order of Public Reason
Classical liberals can use Rawlsian tools to bolster the case for classical liberalism in light of Rawls’s developments in PL. In fact, the line of argument flowing from PL concerning the implications of diversity and diverse reasoning may have strong classical liberal implications. Though, as the reader will discover, the reasoning can become subtle and complex.
In The Order of Public Reason (OPR), Gerald Gaus argues that impartial Kantian reasoning, such as Rawls’s original position abstraction, cannot, by itself, generate a uniquely justified set of rules or principles, because of reasonable (what he calls “evaluative”) pluralism. Gaus claims that we should “give up on the hope that we can construct a compelling description of members of the ‘realm of ends’ that will lead them to agree on the same rule.” 32 Instead, we must allow extensive disagreement in personal judgments by acknowledging the existence of multiple “optimal eligible proposals.” Impartial reasoning of the Rawlsian variety can establish only the boundaries of reasonable disagreement; it cannot yield a determinate choice of principles, like justice as fairness.
The bulk of OPR tries to explain how free and equal members of the public can converge on a determinate set of social rules. Gaus proceeds by drawing on the tradition of moral thought that stretches “from Hobbes through Ferguson, Hume, and Smith to F. A. Hayek and contemporary game theorists,” all of whom claim, in one way or another, that “moralities are social facts with histories.” 33 Morality, in this view, is the path‐dependent output of cultural and moral evolution. Rawlsian impartial reasoning is indeterminate, so Gaus argues that social evolution is required to lead members of the public to converge on a member of the eligible set of proposals, those proposals for which no reasonable member of the public has sufficient reason to reject.
Importantly, Gaus uses original position reasoning because it helps evaluate whether the rules and principles we’ve converged on are justified. If we throw out a Rawlsian test of public justification, “the evolutionary view cannot distinguish authoritarian from non‐authoritarian positive moralities.” 34 The solution to the indeterminacy of Rawlsian political reasoning, then, is to “evaluate our evolved rules from the perspective of public reason: we must seek to determine whether the outcome of the social evolutionary process is within the optimal eligible set” of rules, principles, and so forth. 35 We can use original position–style reasoning to rule out many proposals for justified laws and policies, and even principles of justice, but such reasoning will not yield a uniquely best choice. Consequently, we must allow nonrational processes, such as social evolution, to converge on one of the undefeated, as in unrefuted, outputs of original position–style reasoning.
Gaus’s work drives public reason to classical liberal conclusions in virtue of six features of the account: (a) the focus on justifying moral conventions rather than law alone; (b) a moderate account of idealized reasoning by contracting parties, which will generate enormous diversity among the doctrines and reasons persons endorse; (c) the justification of a right to private property; (d) the refutation of socialism; (e) the right against legal coercion; and (f) an emphasis on nonideal theory.
Focus on Justifying Moral Conventions
Unique among public reason liberals, Gaus claims that social morality must be justified, in addition to laws. A social morality is “the set of social‐moral rules that require or prohibit action, and so grounds moral imperatives that we direct to each other to engage in, or refrain from, certain lines of conduct.” 36 Social morality has several distinguishing characteristics. 37 It structures social interaction and so helps persons coordinate their behaviors and cooperate for mutual gain. Although social morality restrains our aims via social punishment and ostracism, it ultimately extends our ability to achieve our goals. Persons do not comply with the rules that compose social morality—“social-moral” rules—for merely instrumental reasons, but rather because the rules are seen as genuine moral imperatives. Finally, social‐moral rules require an individual to defer to the judgment of others who enforce the rules when his or her private judgment conflicts with theirs.
I believe Gaus focuses on social morality for two reasons. The first is that laws and moral norms share critical features and so raise similar justificatory problems. 38 Both law and morality impose demands on individuals that they may well be better off without. The problems of moral and political justification, then, arise for the same reason: norms might restrict individual liberty without cause.
The second reason for focusing on social morality is new to Gaus’s work. Gaus notes that the problem with specifically Hobbesian and Lockean solutions to the problem of moral justification is that they are “inherently political” and so “politicize the resolutions of all moral disputes.” 39 This theory implies an unattractive form of statism where state authority invades all domains where the content of social morality is disputed. 40 So the focus on social morality is critical to avoid politicizing all moral conflicts—and so to avoid a strong form of statism.
The focus on social morality is critical for Gaus’s defense of classical liberalism because it assumes that the ultimate source of social order is not the state but rather the local moral conventions that evolve, as a spontaneous order, from our normal moral agency and routine interaction. Social morality even has a special priority over law because it can resolve social problems without resort to coercion. The purpose of politics is to perform functions that noncoercive social morality cannot. Thus, there is a presumption on behalf of social solutions to social conflict, rather than political solutions.
Moderate Account of Idealized Reasoning
The next feature of OPR’s main line of argument concerns the sorts of reasons that can block the justification of coercion and the forms of coercion our objections undermine. Gaus uses the language of “defeat” here, such that refuted justifications are “defeated” and some reasons serve as “defeaters” for coercion. 41 Rawls’s account of public justification appeals to coarse‐grained concepts, like a political conception of justice and an overlapping consensus of reasonable comprehensive doctrines. 42 Political conceptions and comprehensive doctrines are sprawling complexes of norms and rules. Gaus fractures these two ideas into much smaller ones: sets of social‐moral rules, as we have seen, and sets of diverse, intrapersonally related reasons.
Gaus fragments political conceptions into rules in part because he thinks that only rules can solve the problem of indeterminacy. Rawlsian political conceptions of justice are much more general in scope, such that evaluations of social practices depend on appealing to principles that can be justified only by appealing to sophisticated philosophical judgment. Rule‐based evaluation does not require such complexities. Consequently, principles are not the primary basis of moral evaluation. 43 Gaus follows Hayek in holding that human beings are as much rule followers as goal seekers. Therefore, a great many of our actions are driven by following rules, far more than are driven by reasoning from principles. Accordingly, our ordinary forms of moral evaluation concern rules rather than principles. In this way rules, not principles, provide “mutually understood guides for social life” available to all normal agents, not merely moral experts. 44
The move from doctrines to reasons is similar. First, Rawlsian comprehensive doctrines are systematic views of the world, God, right and wrong, good and bad, just and unjust. As such, the doctrines are complex relations of a great many reasons for action. Accordingly, using them to evaluate social‐moral rules will prove overwhelming. Further, most citizens lack comprehensive doctrines because of the cognitive demands of accepting and living by them. Public reason liberals require a more local unit of justification for rules—reasons.
In both cases, Gaus appeals to attractive models of human cognition to set the standards of judgment and evaluation involved in public reason. We evaluate rules on the basis of reasons. An additional implication is that any account of our sufficient reasons must involve moderate idealization, where we judge what reasons persons have by appealing to what they would endorse if they had engaged in a respectable amount of information collection and processing. 45 To solve our social problems, we must model agents’ reasoning in ways that are realistic, given the bounded nature of human rationality, and that preserve our real‐world commitments, which more radical forms of idealization may destroy.
The implication of evaluating rules based on the reasons that persons would affirm under conditions of moderate idealization is that justificatory reasons will vary dramatically between persons given their different points of view, histories, and rational commitments. Further, Gaus declines to restrict justificatory reasons to the set of shared or accessible reasons, another departure from most public reason liberals. 46 Far more reasons, then, can be used to object to coercion and to successfully defeat the justification for it.
For instance, it is much easier to defeat a law that forces people to take jobs mandated by the state, given all the different reasons that can plausibly override the mandate, and the few reasons that would prove sufficient to justify the mandate to a wide variety of people. Moral rules and coercive laws cannot be justifiably imposed on persons if those individuals have sufficient reason to reject those rules and laws. Given moderate, diversity‐preserving idealization, then, different citizens will have sufficient reason to reject a great many laws and policies, so much so that we might end up with no justified laws or policies at all (though Gaus thinks that our need to live together is sufficiently strong that concerns about moral anarchy can be answered). That means that an enormous range of laws and policies that Rawlsians propose will be defeated by the diverse reasons of members of the public.
The point of the foregoing, at times technical, discussion is to explain in detail why Gausian public reason tilts in a classical liberal direction. Gausian public reason acknowledges far more sources of diversity among citizens, and that diversity brings orders of magnitude more successful objections to coercive laws. Combining accounts of what is to be justified (rules), the group to whom coercion is justified (moderately idealized agents), and their objections to coercion (objections as diverse as the persons who offer them) shows a dramatic shift in the attitude a publicreason liberal should have toward coercion. We are now in a position of skepticism of the justification of coercive laws. This will tend to manifest itself in forms of moderate libertarianism.
Justification of a Right to Private Property
So on Gausian public reason, diverse, rational reasoners affirm distinct and conflicting moral rules, and they will not agree on which rules are best to govern their common social life. But Gaus insists that those reasoners do not face a “null” set of justifiable rules. Instead, the reasoners often face the problem of too many eligible (undefeated) proposals.
In Part 2 of OPR, Gaus uses three conceptual devices to narrow indeterminacy to the optimal eligible set (OES): (a) the abstract deliberative model that maps out the order of public justification, (b) an account of social evolution that will generate social convergence on a member of the OES, and (c) a “testing conception” of public justification that limits the determination of whether a presently practiced rule is in the OES to cases where the rule has become morally controversial.
Let us begin our account of Gaus’s right to private property by explaining the idea of the abstract deliberative model, which is similar to Rawls’s original position. Gaus’s abstract deliberative model holds that diverse members of the public determine what is publicly justified in a series of steps or in an order that proceeds from more abstract determinations to more concrete ones. 47 He then argues that members of the public will agree on certain basic rights by taking the “perspective of agency.” 48 That is, they will endorse a number of rights by reasoning as bare agents, stripped of some of their individuating characteristics. In doing so, members will adopt a presumption in favor of liberty, respect for autarchy (self‐directed action, but not full‐blown autonomy), rights not to be coerced, freedom of thought, rights against harm, and rights to assistance. 49
Next in the order of justification are jurisdictional rights, rights that devolve collective choices to individual members of the public. Privacy rights and private property rights are the quintessential jurisdictional rights. 50 In sum, constraints on the optimal eligible set follow an order from the more abstract (formal features of rules) through the relatively abstract (the abstract rights of agency) to the more concrete (jurisdictional rights).
The abstract rights of agency are just that—abstract. They involve rights that all liberals, and indeed most nonliberals, endorse. Among these are rights to assistance, which are a kind of positive right. Libertarians are bound to recoil. But it is important to note that Gaus allows the commitments of libertarian and conservative members of the public to defeat more extensive enforceable duties of aid because of their belief that some persons are undeserving. Even so, here enters an element of Gausian public reason that is not fully libertarian. For Gaus, this is a good thing, for his theory should not be coextensive with the comprehensive commitments of a small, sectarian group within liberal orders.
For our purposes, the most important feature of the order of justification is the notion of a jurisdictional right, which assigns individuals and groups rights to govern particular, specific social domains in accord with their own wishes and choices. Public reason derives from more monistic forms of moral reasoning, such as those characteristic of Rousseau, Kant, and the early Rawls. But once we introduce diversity into public reason, it is clear that we will seldom be able to agree when making collective decisions. As a result, we have reason to “partition” moral space so that people can make their own choices when it matters most to them. Jurisdictional rights are moral partitions, “individualized spheres of moral authority in which the rightholder’s judgment about what is to be done provides others with moral reasons to act.” 51 Thus, jurisdictional individual rights are a kind of public justification, a set of institutional methods that we can use to settle disputes that we cannot resolve collectively.
The most fundamental jurisdictional right is the right to private property, as private property rights “economize on collective justification.” 52 To own property, Gaus argues, just is to have a legitimate social space in which an agent’s evaluative standards not only have free reign but also impose duties on others not to interfere. Gaus denies that the jurisdictional right to private property is fully extensive or absolute; nonlibertarian members of the public will have diverse reasons to defeat such radical authority claims. 53
Nonetheless, Gaus argues that “members of the public will endorse a system of property rights” that, in general, “are not easily overridden and that are extensive, including private property in capital assets.” 54 Gaus denies that public reason requires worker ownership of capital because public reason is not meant to promote a controversial ideal of autonomy but rather to preserve the moral agency of diverse persons under conditions of disagreement. Consequently, those who insist upon worker ownership as a condition of autonomy are simply insisting on their own sectarian doctrines.
Rights against Legal Coercion
We now turn to Gaus’s case for a right against legal coercion. For Gaus, rights of agency are among our most fundamental rights, and they include a right not to be coerced in the absence of a compelling justification. Libertarians might rejoice here, but not so fast: Gaus claims that reasonable people disagree about which forms of social order are most coercive, so some nonlibertarian views about the coerciveness of the market may present problems for libertarian property rights claims. Nonetheless, there are certain obvious core cases of coercion, such as the forms of coercion used by police, that all recognize as coercive. Legislation is also typically coercive.
Gaus then argues that, insofar as we have a settled scheme of basic liberties (which include a right of private property), these rights set a baseline against which coercion must be justified. If the government proposes to use coercion to force you to wear a motorcycle helmet or evict you from your home and if you have a publicly justified right to make your own decisions with regard to those issues, then the right against legal coercion prohibits state action.
Gaus also notes that publicly justified legislation, despite justifying the use of legal coercion, must be evaluated by counting its coerciveness as one of the reasons to oppose the law (even if those reasons are ultimately overridden). States are not exempt from this standard. 55 Gaus stresses that the right against legal coercion is a right that coercion be prohibited in the absence of a public justification. Further, a law’s coerciveness continues to count against it even when the factors favoring the law are stronger.
It should be plain that a right against legal coercion pushes public reason in an even more classically liberal direction. Not only is there a strong right against the state to not be coerced in the absence of a public justification, but also the cost of coercion must still figure into the justification of a law we think is, all things considered, publicly justified. Given the diverse, moderately idealized reasons of the public, the right against legal coercion will be a powerful one, including when it comes to the protection of private property rights.
Refutation of Socialism
Gaus also argues that socialism and strongly egalitarian states are almost always ineligible as systems of political and economic governance. Let’s start by considering state socialism. Even if substantial redistribution of wealth is justified, “socialist systems would still be outside the socially eligible set,” 56 because socialism necessarily requires collective decisionmaking and so requires consensus about how various parts of the economy should be run. The right of private property is publicly justified because it is a solution to the inability of members of the public to agree on common standards. Socialism, understood as government ownership of the means of production, requires agreement where none can be reasonably expected.
A second argument against socialism is empirical in nature, which is that “extensive private ownership—including capital goods and finance—is for all practical purposes a requisite for a social and political order that protects civil liberties.” 57 Rawlsians, Gaus argues, fail to appreciate how difficult it is to divorce the good aspects of markets from private ownership. Gaus states, “There has never been a political order characterized by deep respect for personal freedom that was not based on a market order with widespread private ownership in the means of production.” 58
Rawlsians are apt to complain that such an order allows sufficient economic inequality to undermine democratic freedom. Anticipating this reply, Gaus next appeals to empirical evidence that political freedom and economic freedom are correlated. Gaus argues that little evidence exists to support the claim that the economic inequalities found in liberal democracies today endanger democratic institutions.
Now, against the libertarian, I should stress that Gaus rejects both market anarchist and minimal or small‐state libertarianism, because many suitably idealized members of the public will oppose libertarianism and so oppose coercive laws that institutionalize more radical forms of libertarianism. A pure capitalist order would allow for no redistribution of wealth, no government production of public goods, or any regulations other than the regulatory effect of torts and contracts. But many reasonable people, if not the vast majority of them, believe, with epistemic justification, that a less‐than‐minimal state cannot be publicly justified, because it would coercively prohibit people from collectively deciding to extend state power. As a result, they have respectable objections to the sorts of coercion required to limit the state as much as libertarians would like.
Public reason also cannot prohibit redistribution, since the argument against a more egalitarian state “succeeds only if the justification of property is not dependent on questions about justified redistribution and/or conceptions of social justice within the evaluative standards of Members of the Public.” 59 But that is implausible, as these issues are arguably not “even remotely independent issues.” 60
Turning now to egalitarian redistributive liberalism, Gaus argues that, despite rejecting libertarianism, public reason liberals must nonetheless hold that in many contexts, taxation is coercive, and the coerciveness of taxation increases as the tax rate increases. Accordingly, the right against legal coercion will provide stronger and stronger reason to oppose taxation as the tax rate goes up. 61 Further, in the production of public goods, the state is bound by a Paretian standard of public finance and public policy. Public programs must be reasonably expected to advance the evaluative standards of all, or they are seldom justified; that is, programs must be Pareto improvements.
Further, Gaus argues that evaluating policy requires a clear sense of their benefits and costs, which may require bills to contain their own funding directives (a recommendation Gaus adopts following Swedish economist Knut Wicksell), which in practice would substantially limit the public policy process. Gaus then adds supermajority voting rules into the mix, limiting government even more. All of those restrictions are barriers to the public justification of a strongly redistributive state.
In consequence, Gaus concludes that public reason liberalism
Leads not to socialism, or a thoroughgoing egalitarian liberalism, or to libertarianism [understood as minimal state or anarchist libertarianism—KV], but to the more nuanced approach to legislation we find in the fifth book of Mill’s Principles, allowing that there are a number of tasks that government justifiably performs, but having a strong overall inclination toward less rather than more “authoritative” (i.e., coercive) government. 62
The argument for this position, Gaus claims, rests on five core claims:
Individuals are free and equal.
Free and equal persons have a moral right not to be forced or coerced without justification.
A member of the public will hold that the greater his or her estimates of coercion, the stronger must be the justification.
Free and equal members of the public reasonably disagree on many matters involving degrees of coercion, but many reasonable people believe that large states with high rates of taxation and redistributive institutions are more coercive.
Only laws that can be justified to all members of the public can reconcile coercion with respect for everyone’s freedom and equality. 63
Consequently, public reason liberalism must “tilt” against coercive states, and public reason liberals are responsible for showing that their favored proposals do not use coercion, at least not in excess of what can be justified given the legal right against coercion. 64
Emphasis on Nonideal Theory
The final feature of Gaus’s view that pushes public reason in a more libertarian direction is Gaus’s emphasis on nonideal theory. Gausian public justification is identified with a “testing conception” of evaluation, where the aim of a model of public reason is not to reconstruct social institutions from the ground up, but rather to help us evaluate our real‐world practices in a way that accords with our real‐world normative standards and commitments.
Moral evaluation for Gaus, following Hayek, must begin from our actual social practices and then “test” those practices via the deliberative model. 65 In other words, public justification does not begin by asking whether our entire set of coercive institutions embodies a commitment to general principles of justice. Rather, we test real rules via a test of public justification only when challenges or concerns about the rules arise.
Further, Gaus does not assume that citizens and political officials will fully comply with the law; he limited original position reasoning and allows for far more economic realism than other public reason liberals. Readers of this volume are likely already sympathetic to the claim that formal economic models and empirical data favor libertarianism or alternatives, so by allowing nonideal behavior and real‐world data into public justification, public reason liberalism is pushed further still in a classical liberal direction. Libertarians have frequently understood the case for libertarianism as resting in a commitment to nonideal theorizing, where the failures of government are impossible to ignore. In this respect, then, Gaus’s emphasis should be welcome.
Which Rawlsian Libertarianism?
The reader will notice that I have spent much more time developing the Gausian position than the Tomasian position. I did so in part because Tomasi is not reconstructing public reason from the ground up but is instead modifying justice as fairness to accommodate classical liberal economic liberties and a broader classical liberal outlook on political life.
But I have also focused more on Gaus because I find the Gausian view more plausible. Rawls was right to move from A Theory of Justice to Political Liberalism because of the problems of reasonable pluralism about the good and justice. 66 Gaus plausibly develops public reason in a diversity‐accommodating direction, which I regard as a natural extension of a broadly Rawlsian project. Tomasi’s free‐market fairness, therefore, will fall prey to the same concerns as justice as fairness. Reasonable people disagree about what justice requires; thus, we cannot use Rawlsian reasoning to vindicate justice as fairness or free‐market fairness. Ultimately, Gaus has set the stage for the next version of the Rawlsian project, a Rawls 3.0 that accommodates reasonable pluralism about the good and justice, and other forms of diversity as well. Tomasi’s project is associated with Rawls 1.0, the Rawls of A Theory of Justice.
All the same, the Gausian model does not rule out appeals to one’s own conception of justice in shaping a publicly justified polity. It is a great good in Gausian public reason for people to appeal to their sectarian conceptions of the good and justice, and not merely to defeat laws that cannot be publicly justified. Appeal to diverse values, Gaus now argues, helps society discover new and better ways of living together. 67 So the advocates of free‐market fairness need not give up their view, but they must accept that free‐market fairness cannot claim a special status in governing a society’s basic structure. Free‐market fairness might embrace the correct principles of justice, but not the correct principles in the fully Rawlsian sense, as they cannot survive public justification due to deep disagreement about justice, and so cannot stabilize a diverse social order in the right way. Thus, these two Rawlsian libertarianisms are in tension. Free‐market fairness cannot form the foundation of an order of public reason. But it can form part of that foundation.
John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1999); Political Liberalism, expanded ed. (New York: Columbia University Press, 2005). ↩
John Rawls, Justice as Fairness: A Restatement, ed. E. Kelly (Cambridge, MA: Harvard University Press, 2001), pp. 58–59. ↩
Thus, “sufficientarianism” refers to a principle of justice that everyone be given enough resources to live minimally decent lives. Contrast this with egalitarianism, which requires equalizing resources even among persons who already have enough. ↩
See Tomasi, Free Market Fairness (pp. 127–42) for a description of his “hit parade” review of major figures, such as Herbert Spencer, Milton Friedman, and F. A. Hayek, and contemporary libertarian political philosophers, such as Loren Lomasky and David Schmidtz. ↩
Tomasi also argues that Rawlsian emphasis on protecting political liberty can be turned to classical liberal ends, but for reasons of space, I omit that discussion. ↩
Samuel Freeman uses the term high liberals to describe social democratic and egalitarian liberals whose conception of liberalism is the most evolved and inclusive of the insights of the liberal tradition as a whole. See Freeman, “Capitalism in the Classical and High Liberal Traditions,” Social Philosophy and Policy 28(2), pp. 19–55, 2011. Tomasi takes on Freeman’s term in his work. ↩
Though in later work, such as the introduction to the paperback edition of Political Liberalism, Rawls allows an overlapping consensus to form around a set of reasonable liberal political conceptions. See p. xxvi. ↩
Before The Order of Public Reason, Value and Justification details his most extensive effort. See Gerald F. Gaus, “Part II. A Theory of Moral Justification,” in Value and Justification: The Foundations of Liberal Theory (New York and Cambridge, UK: Cambridge University Press, 1990), pp. 251–378. ↩
Gaus’s The Order of Public Reason contains his critique of standard conceptions of idealization (pp. 235–44). For Gaus, an agent has a sufficient reason R to endorse a rule or law if and only if a “respectable amount” of good reasoning would lead the agent to affirm R as undefeated. See ibid., p. 250. A respectable amount of reasoning is determined by the degree of reasoning that members of the public expect of one another, a modest standard often determined contextually. ↩
See Gaus’s critique of the shared reasons requirement. Ibid., pp. 283–87. ↩
The extensive detailing of the deliberative model is developed in ibid., chap. 5. ↩
Importantly, Rawls’s concerns regarding pluralism about justice postdate the development of Political Liberalism, but the recognition of justice pluralism becomes critical in the introduction to the paperback edition and his later article “The Idea of Public Reason Revisited,” University of Chicago Law Review 64 (Summer 1997): 765–807. ↩
See Gerald Gaus, The Tyranny of the Ideal (Princeton, NJ: Princeton University Press, 2016). ↩