114 Handbook of Political Theory
Scanton, Thomas (1998) What We Owe Each Other.
Cambridge, MA: Harvard University Press. Sen, Amartya IC. (1970) 'The impossibility of a Paretian
liberal'. Journal of Political Economy, 78 {Jan./Feb.}:
152-7.
Sen, Amartya K. (1990) 'Rights consequentialism'. In
Jonathan Glover, ed., Utilitarianism and its Critics.
London: Macmillan, 111-18. Shapiro, lan (2003) The Moral Foundations of Politics.
New Haven, CT: Yale University Press. Sher, George (1997) Beyond Neutrality: Perfectionism
and Politics. Cambridge: Cambridge University Press. Sidgwick, Henry (1962) The Methods of Ethics, 7th edn.
Chicago: University of Chicago Press. Simmons, A. John (1992) The Lockean Theory of Rights.
Princeton, NJ: Princeton University Press. Skyrms, Brian (1996) The Evolution of the Social
Contract. Cambridge: Cambridge University Press. Smith, Tara (1995) Moral Rights and Political Freedom.
Lanham, MD: Rowrnan and Littlefield. Spector, Horacio (1992) Autonomy and Rights: The Moral
Foundations of Liberalism. Oxford: Clarendon.
Spencer, William {1995) Social Statics. New York:
Robert Schaikenback Foundation. Steiner, Hillel (.1994) An Essay on Rights. Cambridge,
MA: Blackwell. Sumner, L. W. (1987) The Moral Foundations of Rights.
Oxford: Oxford University Press. Sunstein, Cass R, (1997) Free Markets and Social Justice.
Oxford: Oxford University Press. Ten, C. L. (1980) Mill on Libert}'. Oxford: Clarendon. Vallentyne, Peter, ed. (1991) Contractarianism and
Rational Choice. Cambridge: Cambridge University
Press.
Wall, Steven (1998) Liberalism, Perfectionism and Restraint. Cambridge: Cambridge University Press.
Wicksteed, Philip H. (1946) The Commonsense of Political Economy,, ed. Lionel Robbins. London: Routlcdge.
Williams, Bernard (1985) Ethics and the Limits of
Philosophy. London: Fontana/Coüins. Young, Robert (1986) Personal Autonomy: Beyond
Negative and Positive Freedom. London: Croom-Helm.
issical Liberalism and Libertarianisr The Liberty Tradition
ERIC MACK AND GERALD F. GAUS
THE LIBERTY TRADITION
AlasdairMacTntyre provides a helpful characterization of what constitutes a 'tradition' within moral or political thought. He says that such a tradition is:
an argument extended through time in which certain fundamental agreements are defined and redefined in terms of two kinds of conflict: those with critics and enemies external to the tradition who reject ail or at least key parts of those fundamental agreements, and those internal, interpretative debates through which the meaning and rationale of the fundamental agreements come to be expressed and by whose progress a tradition is constituted. (1988:12)
In Maclntyre's sense, libertarianism and classical liberalism constitute a tradition of political thought. Within a tradition the internal debates may be so important to its members that the criteria for genuine membership are tied to one's position on these issues, and so the criteria of membership in the tradition are themselves contested. Some in the tradition will seek to withhold the status of member to others who claim it. For example, some members of the socialist tradition (say Marxists) might not grant that label to other members of that tradition or might themselves insist upon some differentiation between themselves and others (e.g. 'utopian socialists') within the same basic tradition. In the case of the libertarian/classical-liberal tradition, the most radically anti-statist members of this tradition may claim the label 'libertarian' and deny that label to their less anti-statist fellow-travellers, while the least anti-statist members of the tradition may claim
the label 'classical liberal', which they deny to their most hard-core anti-statist comrades. Hence, the hyphenated designation of the tradition that concerns us here. However, despite this hyphenated designation, it is enlightening to understand classical liberalism and libertarianism as comprising a single tradition of political thought. All the positions that we shall place in that tradition share a significant family resemblance, which is acknowledged by most members of this tradition by their willingness to accept for themselves and for most other members both the designations 'libertarian' and 'classical liberal'. Rather than the awkward phrase 'libertarian/classical-liberal tradition' we shall refer to the more melodious 'liberty tradition'.
The family resemblance among members of the liberty tradition obtains at two related levels. Underlying the tradition is, first, a doctrinal resemblance, constituted by a substantial sharing of normative principles and more or less empirical generalizations about how the world works (or fails to work) — principles and generalizations that together yield conclusions about the normative constraints on legitimate states. Second, there is a consequent political resemblance, a substantial surularity of conclusions about the way these shared normative constraints are to be applied, and thus what sort of state, if any, is justified. As with all traditions in political thought, this commonality of outlook is conjoined with vigorous disagreements. Some members of the liberty tradition accept different versions of the characteristic doctrinal elements. Indeed, some members of the liberty tradition entirely reject some of its characteristic doctrines. Furthermore, members of the tradition accept a
Handbook of Political 1 heory
classical Liberalism and Libertarianferr
range of conclusions about what sort of state, if any. can be justified. This range constitutes a spectrum of political stances which, from the most anti-statist left to its least anti-statist right, encompasses what we shall call Market Anarchism,' Minimal Statism, Taxing Minimal Statism and Small Statism.
We begin by explicating the unity of the tradition in terms of the doctrinal family resemblance among its members. Following that explication, the diversity within the tradition is introduced in terms of its internal debate about what sort of state, if any, can be justified. This debate itself, however, reflects and motivates a complex discussion within the tradition about precisely which interpretations of which of the doctrinal elements associated with the liberty tradition ought to be affirmed, and which is central and which is peripheral. Thus, an examination of the diversity of political stances within the liberty tradition quickly brings us back to the doctrinal level of this tradition but, this tune, with a focus on internal doctrinal debate. We conclude by considering a recent attempt to extend the liberty tradition far toward statism - so-called 'Left Libertarianism'.
DOCTRINAL UNITY
Let us begin with formulations of a dozen doctrinal elements that unify the liberty tradition. For the sake of including all wings of the liberty tradition, each formulation allows for a range of interpretations. Not all members of the tradition endorse every doctrinal element, let alone all the same interpretations. And behind their endorsement of different combinations of different interpretations of these doctrines, members of the tradition differ hi the precedence attributed to various doctrines. Some depend most heavily upon bold versions of the normative doctrines articulated below, while others base then' case on bold claims about how the world works, conjoined with more modest versions of the normative doctrines. These dozen doctrinal elements are not independent axioms or theorems entailed by such axioms. Behind the doctrinal unity lies a diversity of deeper philosophical strategies -for example, deontological, contractariau, or conse-quentialist strategies - for vindicating some set of versions of these normative doctrinal elements. What provides unity at this philosophical level is each member's anticipation that his philosophical strategy best vindicates his interpretation of these doctrinal components which, in turn, supports a political position within the libertarian/classical-liberal spectrum.
(I) The liberty tradition is normatively individualist, affirming the separate value of each individual.
Each individual's life, well-being or preference satisfaction is thought of as having supreme importance in and of itself, not merely in so far as it contributes to social life, well-being, or preference satisfaction (Mack, 1999). This normative individualism underlies an insistence on the illegitimacy of actions and policies that impose losses on some individuals in the name of providing more extensive benefits to others (Nozick, 1974: 28-35). It might seem that utilitarian members of the tradition oppose this: utilitarians insist that only the greatest overall happiness is of value, and so it would seem that individuals are normatively important only as a means to aggregate satisfaction. While this might be the crux of utilitarianism in moral theory, it has not been in the liberty tradition. As Samuel Brittan has observed:
the traditional economist's case tor a form of market economy has been based on what might be called libera! utilitarianism. This is a belief that individual desires should normally be satisfied to the maximum degree possible without interfering with the desires of others. The utilitarianism involved is a highly qualified one. (1988: 43)
In addition, the liberty tradition is ontologically individualist in that it takes individuals, not classes, or races, or nations, to be in the final analysis the only sites of value, the only real agents, the only true bearers of rights and of responsibilities (sec Benfham, 1987: ch. 1, s. 4; Buchanan and Tullock, 1965: 11-12). Only individuals make choices; there is, literally, no such thing as 'social choice' (dc Jasay, 1991: 57-9).
(II) Normative individualism - the separate importance of each individual's life, well-being or preference satisfaction — is thought to endorse enforceable moral claims held by all individuals against interferences that diminish their lives, well-being or preference satisfaction. A moral claim against interference by others is basic to the liberty tradition (Nozick, 1974: 30ff; Machan, 1989: 7ff). The special importance that each individual's life, well-being or preference satisfaction has for her is an obstacle to justifying moral duties that require an individual to put aside her own concerns to work for the well-being or preference satisfaction of others (Lomasky, 1987: 94ff). At the very least, an additional burden of proof rests upon anyone who asserts that enforceable claims against being interfered with or harmed must be compromised in order to make room for enforceable claims to be assisted (Gaus, 1999: 117-19, 191^1).
(in) The liberty tradition takes individual liberty to be the core political or legal norm (Robbins, 1961: 104). Individual liberty is what each individual
may legitimately demand of each other individual. There may be many other things that are good in life as ends in themselves or as means to those ends, but - at least absent complicating special circumstances - these rarely can be demanded of others as a matter of right. Part of the reason that liberty is the only thing - or at least the primary thing - that may be demanded of others as a political right is that the demand for liberty is uniquely modest: to demand liberty is merely lo insist that one be left alone in one's solitary activities or in one's joint activities with other consenting individuals. Liberty as non-interference by others is thus a good that everyone with aims, goals or projects has an interest of demanding from all others, it can only be supplied by others, and it can be universally supplied at modest costs, unlike demands to be benefited or served at the expense of others (Lomasky, 1987: ch. 5).
(TV) According to the liberty tradition, respect for the individual and her liberty requires respect for that individual's control of extra-personal objects -tangible and non-tangible property - that she has acquired in ways that do not infringe upon others' equal liberty (Lomasky, 1987: ch. 6; Mack, 1990). Several related sub-themes are apt to be endorsed by members of the liberty tradition. First, seizing another's peacefully acquired holdings is itself a violation of her liberty. Second, seizing the fruits of another's labour or what a person has acquired through voluntary exchange of his labour or the fruits of his labour violates that person's entitlement or desert (Gaus, 1999: ch. 8). Third, a system that allows such seizures renders all other sorts of liberty insecure; secure private property is a background condition for a general regime of liberty (Gray, 1986: ch. 8). Fourth, secure private property is a background condition for economic prosperity. In general, the liberty tradition insists that freedom is only possible given the institutions of private property and the free market. Indeed, for some members of the tradition 'liberty is property' (Narveson, 1988: 66).
(V) In the liberty tradition, a desirable social order is an association of individuals (and sub-associations), each of whom has and pursues their own legitimate ends in life, but who themselves share no common goals (Oakeshott, 1975). Desirable social order emerges through the choices that individuals make when rights are secure. The liberty tradition denies that society is a collective enterprise, in which order is achieved through individual devotion (and subservience) to collective aims. According to the liberty tradition, individuals are able to enter into peaceful and mutually beneficial relations because of their general compliance with certain general rules - rules that are protective of the domains defined by individuals' rights to their
lives, liberties, and justly acquired estates. The ideal of the rule of law is that all are subject to the same general rules - rules that do not subordinate them to one another or lo some spurious societal end but rather protect and facilitate each person's pursuit of her own projects and ends (Hayek, 1973).
(VI) A social order that emerges out of the choices that individuals make when their rights are secure is more desirable than a centrally planned order because it allows and encourages individuals to bring their highly individualized and dispersed knowledge to bear on their decision-making (Barnett, 1998). When these rights are secure, individuals tend to benefit from their own productive decisions and bear the costs of their own unproductive choices; this provides an economically and socially desirable incentive structure. Moreover, institutions that reflect such decentralized decision-making, such as markets and non-coerced customs, themselves give rise to and convey important information for yet further individual planning and decision-making (Hayek, 1991). The liberty tradition emphasizes the extent to which political decision-makers are and must be ignorant of the knowledge they would have to have in order to engage effectively in the central planning to which they aspire (Hayek, 1944).
(VII) At least as a first approximation, we can say that in the liberty tradition the licit use of coercion ■ is limited to blocking or nullifying infringements upon the rightful claims of individuals. Any use of coercion that infringes upon an individual's control of her person or property (and, thereby, abrogates the basic condition for the emergence of mutually beneficial social interactions) is illicit. To seize another and enslave him is to engage in illegitimate force; to break the slave-catcher's ami to prevent one's capture or to escape from slavery is to engage in legitimate force. Within the liberty tradition, any proposal to expand the list of legitimate uses of force - either by expanding the list of rights for which people can demand coercive protection or by expanding the role of legitimate coercion beyond the protection of rights - bears a heavy burden of proof. And part of what must be proven within the liberty tradition is that the proposed expansion of legitimate coercion still leaves individual liberty, as originally conceived, as the central political norm.
(VnT) The distinctive feature of political institutions in the liberty tradition is that they authorize the use of force and can legitimately threaten and use force against citizens. Because the scope of legitimate force is so limited, political policies and institutions readily fail to be legitimate. As at least a first approximation, wc can say that political policies and institutions are legitimate if and only if they restrict their use of force to actions that block or
118
Hand bo
Doliti!;
riieory
Classical Liberalism and Libertarianism: "• h
nullify the violation of people's rights to liberty and property. Within the tradition, political institutions that use coercion more extensively must bear an especially heavy burden of justification, one that is too great to justify a state that extensively employs coercion against its citizens.
(IX) The liberty tradition rejects a basic distinction between the morality that applies to individuals generally and the morality that applies to public institutions and officials. To be sure, the special circumstances of public officials - including, perhaps, their possession of political authority - means that they may engage in particular actions, e.g. the punishment of wrongdoers, that is denied to citizens generally. But all sorts of special circumstances can render one individual, e.g. the executor of another's will, free to act in a way that is denied to citizens generally. The recognition of special rights and duties is consistent with the same principles of justice that apply to private individuals applying also to public officials. Illicit coercion is just as criminal when performed by public officials who pretend to be the instruments of justice as when it is engaged in by thieves and murderers. The liberty tradition certainly rejects the idea that there is a 'special distinction and dignity' attaching to the functions of government or the public (von Mises. 1985: 40).
(X) One familiar way of conveying the liberty tradition is to say that individual A may rightfully be subject to force by individual or institution C, if and only if that force will prevent A from harming B or will nullify some harm that A has already inflicted upon B. This formula, deriving from J. S. Mill (1991 [1859]), nicely highlights purposes that do not justify the use of force:
(XI) in contrast to the conventional left and right, members of the liberty tradition reject any fundamental distinction between 'personal' and 'economic' liberties (Machan, 1989: 98), and they certainly deny that there is a strong case only for so-called personal liberty. Because of the tradition's highly generalized endorsement of individual liberty, it is important for members of the tradition to hold that across all the various important dimensions of human life and interaction, desirable order tends to emerge from the exercise of individual liberty rather than from the imposition of some centrally determined structure or arrangement. For some members of the tradition, the fact that secure liberty is the fountainhead of beneficial order will be the generalization about how the world works that stands at the core of their doctrinal stance.
(XII) Members of the liberty tradition believe that most, if not all, political regimes have continuously and grievously infringed upon people's just liberties, and have continuously engaged in extensive acts of unjustified aggression, plunder, and meddlesomeness. The self-image of the liberty tradition is that it sees through the common demands and rationalizations for political power to understand the permanent tendency of political power to be oppressive, exploitative, and destructive of harmonious and mutually beneficial social arrangements. Members of this tradition hold that, even though some form of political authority is perhaps necessary and justified, citizens must always be jealous of such power, on their guard against it, be ready to condemn and resist its expansion and misuse.
POLITICAL AND DOCTRINAL DIVERSITY
(a) C's preventing A from harming himself does not justify C's forcible intervention.
(b) C's preventing A from acting in some sinful or ignoble way does not justify C's forcible intervention.
(c) C's preventing A from offending B (in a non-harming way) does not justify C's forcible intervention.
(d) C's causing A to bestow a benefit upon B does not justify C's forcible intervention.
As is well known, how such a Millian formulation works itself out depends upon such detailed matters as, for example, how one construes harm and what counts as an imposition of harm. The liberty tradition is highly sceptical of proposals that would question (a)-(d) or interpret them hi such a way that they provide no real barrier to state coercion. One can remain within the tradition if one chips away at the edges of some of these principles but not if one fractures any of them.
Let us proceed now to the internal debates within the liberty tradition. We can best articulate these debates by attending to four political positions along the liberty tradition spectrum - (1) Market Anarchism, (2) the Minimal State, (3) the Taxing Minimal State and (4) the Small State - and the intellectual disputes among them. Although one could simply catalogue the doctrinal debates among members of the tradition, the actual structure of and motivation for the tradition's internal doctrinal debate are better captured by seeing how certain key differences about doctrinal elements fit into the contentious dialogue among advocates of these distinct political positions.
Market Anarchism
The liberty tradition's doctrinal commitments easily endorse Market Anarchism (Friedman, 1973). Liberty requires private property and a market order
(IV), desirable order emerges ou[ of individual choices (V), the market uses the dispersed information of individuals (VI). the tradition is deeply sceptical of all coercion (VII) to the extent that most coercion is illegitimate (VIII), and, crucially, because the liberty tradition rejects an important distinction between public and private morality (IX), the grounds for justified coercion must lie in the rights of private individuals. Recall that according to John Locke, in a pre-political condition - a state of nature — individuals would have not only rights to life, liberty and property, but rights to 'punish the offender, and be Executioner of the Law of Nature' (1960: 290). If all rights, including rights to punish and enforce rights, reside in individuals, and if the market order is itself necessary for freedom, there is a strong case to be made that the protection of freedom itself can be left to the market.
Locke himself famously rejected life in anarchy (but see Simmons, 1993). Because each would judge for himself about the proper bounds of his and other rights, and about whether infractions had actually occurred, individuals would end up in conflict. The solution, Locke argued, would be that each would agree to a political society, 'all private judgement of every particular Member being excluded, the Community comes to be Umpire, by settled standing Rules, indifferent and the same to all Parties' and where only some have the authority to interpret and enforce these rules (1960: 342). Market Anarchists, however, do not concede the need for political authority to solve such disagreement. Although people employing their private judgements can come into conflict, Locke himself shows that they seek the good of shared judgement that would resolve the conflict. If so, there is no reason why such a good must be provided by that monopolistic provider we call the state. Against Locke, the market anarchist argues that a market regime of multiple, competing, protective agencies will not produce disorder and strife - so long as there is a strong demand for the orderly, peaceful, and just resolution of disputes. If we suppose that people desire the orderly, peaceful, and just resolution of disputes strongly enough that the powers of a minimal state would be confined to the provision of such resolutions, this very demand for orderly, peaceful, and just resolution of disputes would be strong enough to call forth their market provision.
Members of the liberty tradition attracted to anarchistic solutions thus endorse competitive providers of legal and police services. As in the market generally, competition between providers of judgements and enforcement will tend to produce high-quality goods; in this case, a high-quality good of impartial, efficient umpiring of conflicting rights claims. People will gravitate to impartial judging services for a variety of reasons. Verdicts from partial
or unreliable judges will be apt to be resisted by others, and private enforcement agents will not seek lo execute them, knowing that they are likely to be biased and, so, opposed by those who are found guilty or liable by them (Baniett, 1998). Market rivalry in the production and sale of protective services will motivate not merely price competition but also the discovery and production of new and better modes of law and rights protection. This is to be contrasted with a monopolistic provider's almost total disinterest in the cost-effective production of good law and good rights protection if it is allowed to maintain its monopoly by suppressing aspiring competitors. Desirable positive law will emerge as the articulated rights and rules that protective agencies will provide in order to satisfy consumer demand. As with other dimensions of desirable social order, claims the market anarchist, desirable positive law is more likely to be the product of market (or market-like) processes than of political processes.
In the world of the market anarchist, polities as we know it - including law that arises through legislation - completely withers away. The complete elimination of the political is a better yet alternative to eternal vigilance towards political power (see doctrine XII). The core claim of the market anarchist, then, is that anarchy better respects the very doctrinal elements that motivate the Lockean. All', members of the liberty tradition maintain that everyone ought lo enjoy the freedom of deploying one's property as one pleases - in any enterprise one pleases - as long as that deployment does not infringe upon anyone's rightful liberty. According to the market anarchist, an aspiring competitor who proposes to sell services very much like those offered by the minimal state is proposing to deploy his resources in ways which the champions of the minimal state cannot honestly claim are impermissible and subject to coercive suppression. The champions of the monopolistic provider called 'government' must themselves recognize that its suppression of these competitive endeavours would be criminal.
Minimal State Views
States as legitimate monopolies
A member of the liberty ttadition defending some form of government might reply to this anarchist case in two ways. The first claims that the enterprise of producing and delivering the protection of rightful claims is especially subject to natural monopolies or cartelization.
The argument from natural monopoly, representing a qualification of doctrinal element V, argues
Handbook of Political Theory
that judging and protection are characterized by increasing returns. If it is the case that over the full range of possible outputs, the (n + I )th unit costs less to produce than did the nth unit, then the larger a provider already is, the less its marginal and average costs. This may well be the case with protection services. Suppose a protection agency adjudicates all conflicts among its members peacefully, and typically protects with force the invasion of its members by non-members. If so, the larger an agency is in terms of members, the more conflicts it can adjudicate peacefully, and so its cost per member will decrease. If increasing returns hold, a monopolistic provider is apt to arise in a free market. In two ways this takes the sting out of the anarchist's condemnation of the government's monopoly (Nozick, 1974:52). First, if markets give rise to monopolistic providers, then the anarchist is wrong to think that his free market case is an anti-monopoly case. Second, if we arc stuck with a monopolistic provider, then there seems grounds for following Locke's lead and putting special restrictions on its behaviour. Hence, it might be argued, it must be subject to some other form of public regulation. Perhaps constraining constitutions ought to be thought of as the public regulation of this especially dangerous sort of natural monopoly.
A related argument concerns cartclization (Cowen, 1992). A protective agency will be able to compete effectively in the provision of desired protective services only if it can offer to its clients the enforcement of the rights articulations, rules, procedures, and appeal mechanisms that emerge from agreements among the competing protective agencies. And any particular agency will be able to offer this law enforcement to clients only if it itself is party to those inter-agency agreements. But, once such law-generating agreements are in place, it will be in the interest of the agencies already party to them to exclude further agencies from admission and, thereby, to preclude these agencies from becoming viable competitors to them. Moreover, argues the defender of the state, this exclusion involves no initiation of coercion and, hence, contravenes no liberty tradition norms. Thus, something like a confederation of rights-protecting agencies that enjoys something like a monopoly on the production and sale of defence services can be expected to arise by processes to which the individualist anarchist cannot morally object.
The second response to the anarchistic challenge docs involve some explicit refinement, if not weakening, of liberty tradition norms. According to this response, a protective agency or confederation of such agencies that aspires to the status of minimal state can more readily permissibly suppress the puta-tively rights-protecting activities of its competitors than may at first seem to be the case (Nozick, 1974).
For such an agency or confederation may permissibly suppress activities that pose even a moderate risk of violating rights (at least if it will not be feasible for the boundary crossers to compensate the victims of their violations). Or, to a similar effect, it may be held that such an agency or confederation may permissibly suppress activities of its competitors in the name of the procedural rights of its clients. Thus, considerations of risk or of procedural rights are invoked to refine liberty tradition doctrines in ways that seem to support the minimal statist against the anarchist critique.2
The minimal state and revenues
We thus arrive at the endorsement of the minimal state - a monopolistic agency legitimately employing force and the threat of force solely to protect people's lives, limbs, liberties, estates, and contractual rights against both internal and external threats. This minimal state achieves the protection of these rightful claims only in ways that are themselves respectful of people's rightful claims. The effective enforcement of these claims is taken to secure the background conditions out of which mutually beneficial and valued social and economic order is most likely and most extensively to emerge through individuals' own, well-motivated exercise of their protected liberties.
But how can the minimal state acquire the funds to provide protective services without itself violating people's rightful claims? According to its own champions, the minimal state is subject to the same moral strictures that apply to all of us. If it would be criminal for any one of us to seize funds from another, even if that first party proceeded to employ those funds to provide the second party with protection against third parties, then it will also be criminal for the minimal state to seize funds from any of us even if it proceeds to employ those funds to provide us with protection against (other) internal or external threats. One will at least draw this conclusion unless the advocate of the minimal state can bear the burden of showing that, contrary to appearances, the seizures conducted by the minimal state are really distinct from the deprivations conducted by ordinary thieves in morally significant ways (see element XI).
How, consistent with unreconstructed liberty norms, could the minimal state acquire the resources necessary to finance the services it supplies? The key to the minimal statist's answer is that individuals do not have original (pre-contractual) moral claims to the various forms of protection that the state proposes to provide. Whereas individuals have original moral claims not to be interfered with or harmed by others in certain ways, individuals do
All others
Agent
Pay my share
Do not pay my share
Enough others co-operate, so that defence is secured
Not enough others co-operate, so that defence is not secured
Agent gains the public good, but has to pay part of the cost for it	Agent pays the cost, but since others don't co-operate, she doesn't get the public good
Agent gets the public good for free	Agent doesn't get the public good, but at least she doesn't pay anything
Figure 9.1   A multi-person prisoner's dilemma
not have original moral claims that other agents protect them against those interferences or harms. If I am shipwrecked on that proverbial island with strangers A and B, I have a claim that A (and B) not attack me. But I do not have a claim against B that B protect mc against A"s attacks. B's protecting me would be a service and, if 1 want a claim to thai service, I must acquire that claim by paying B for it (in cash or kind). Similarly, say the friends of the minimal state, this agency is under no pre-contractual obligation to provide individuals with the protection in which it specializes. It is, therefore, free to offer its presumably highly valued protective services for sale - as is any aspiring supplier of valuable services. The minimal state's acquisition of the resources necessary to provide its services is vindicated as one side of a normal voluntary business transaction. (The minimal state may offer somewhat different packages of services at differing prices to its various potential customers.) Such a minimal state, of course, cannot require that people buy protection. As a monopoly it can charge consumers, and as a constitutionally unregulated monopoly it can charge consumers whatever the market will bear. But it cannot require anyone to purchase its services.
The Taxing Minimal State
Protection as a public good
The market anarchist and the minimal statist share a crucial premise, namely, that the value to individuals of their receipt of protective services will motivate almost everyone to pay for those services. Individuals eager to be protected in their lives, limbs, liberties, and estates will finance the production of protective institutions by patronizing either a range of competing protective agencies or a lmnimal
state that is the monopoly supplier of those services. Put somewhat differently, the shared premise is that the protection of rightful claims is a standard economic good which people will voluntarily pay for to the extent that they value it. Unfortunately, however, important parts or aspects of the protection of rightful claims are not like standard economic goods; important parts or aspects of the protection of rightful claims are public goods.
The crucial feature of a public good is that, if the good is produced, it will not be feasible to exclude1 individuals who have not paid for that good from benefiting from it. The usual example of a public good is the protective sendee of national defence. If a system of national defence is funded and produced, it will not be feasible to exclude occupants of the national territory from its benefits. The non-excludability of these goods provides people with an incentive not to purchase them. Rational individuals confront a multi-person case of the well-known prisoner's dilemma, depicted in Figure 9.1. Defection (not paying her share) is the agent's dominant strategy: no matter what the rest of society does, she does best by not paying her share. Thus public goods tend to be undcrsupplied even if everyone prefers paying their share for the public good to not having it. The parties thus end up at a Pareto-inferior result: each orders the north-west over the south-east cell, yet they all end up in the latter. If these special difficulties in soliciting voluntary market payments for public goods cannot cost-effectively be overcome, the public good will not be financed and produced, and every member of the public will be worse off than she would be had she paid her share of the cost of that good and it had been financed and produced. In the case of rights-protective goods, every member of the public will be worse off with respect to the protection of her rightful claims.
It is widely held that these special difficulties of marketing public goods cannot cost-effectively be
122 Handbook of Political "i heory
overcome by voluntary means (but see Schmidtz, 1991; Cowen, 2003), and it is also widely held that public goods can cost-effectively be financed by coercive means. The latter idea is that individuals can be coerced into paying their share of the cost of public goods and Ihis will result in each being a net beneficiary: the direct and indirect costs imposed upon each individual by requiring her to pay her share will be less than the benefits to her of having the relevant public good produced. These views amount to a qualification of the liberty tradition's general endorsement of markets and contractual relationships as the best devices for allocating resources to their most valuable uses (see elements V and VI). Government is justified largely on the grounds of market failure: although the market generally provides for both a free and a prosperous society, it is not perfect (Buchanan, 1975: ch. 3). Thus the classical liberal political economists of the nineteenth century - Adam Smith, J. R. McCulloch. Nassau William Senior, J. B. Say, David Ricardo, Robert Torrens - insisted that the market depended on a political framework that it could not itself provide; the market could not itself provide a coercive public apparatus for the enforcement of property rights and contracts (Robbins, 1961; Gaus, 1983).
Market anarchists and minimal statists may challenge these widely held views. They may argue, first, that coercive state provision of public goods tends to oversupply them, so that it has its own offsetting inefficiencies (Buchanan and Tullock, 1965). And, they may insist, market and contractual arrangements can be envisioned that will yield funding for public goods - especially rights-protective public goods - that is not significantly suboptimal (Buchanan, 1975; Narveson, 1988: 238). Advocates of the minimal state that depict it as a natural monopoly seem better positioned to make this argument than are market anarchists. Such a minimal state will, to a considerable degree, be able to tie its clients' purchase of non-public aspects of rights protection to their also paying for public aspects of rights protection. For instance, it will be able to say, 'We will sell you access to our courts for the settlement of criminal and civil disputes - which you need to purchase from us if you are to enjoy it -only if you also agree to buy national defence from us.' Of course the state's monopolist position poses its own problems; in so far as the state is a monopoly it tends to restrict supply and to make consumers pay more for its output than they would under market competition.
If crucial public goods would be significantly underproduced in the absence of individuals being required to contribute to their funding (and requiring such contributions would yield a satisfactory level of the production of those public goods), members of the liberty tradition are faced with a
hard choice. On the one hand, they may stick with unreconstructed versions of that tradition's basic norms at the cost of precluding the mutual benefits associated with those public goods (while no doubt insisting that the public good characteristics of law enforcement arc typically overestimated, and that most of what the state should do is to provide essentially privately consumed protection services). Of, on the other hand, they may legitimate the coercive takings that are, by hypothesis, needed to fund those valuable goods at the cost of weakening at least some of those central norms. The second alternative moves us to the right along the liberty spectrum to the taxing minimal state - a state which funds itself through taxation, but only in so far as is necessary to finance the production of protective services (or, perhaps, the production of these and other public goods).
Three justifications for public goods provision
How great will be the doctrinal cost of this weakening of liberty tradition norms? That depends upon how much independent reason - reason independent of the felt practical need to allow for the coercive funding of public goods - there is for the adoption of less restrictive versions of these norms. We can identify three approaches to justification: (1) that coercive public goods provision is fully consistent with the basic commitments of the liberty tradition; (2) that the goods at stake justify overriding liberty; and (3) that such provision is benign paternalism.
(1) Many members of the liberty tradition insist that, on the most plausible understanding of its basic norms, they arc not violated by coercive public goods provision. According to these defenders of the coercive funding of public goods, to recognize the 'separateness of persons' is to reject the idea that the gains that may accrue to some vindicate the infliction of losses upon others. So the core norm of the tradition is a prohibition on benefiting some individuals at the expense of others. This core, anti-redistributive norm is not offended by the coercive takings involved in financing public goods (which, by hypothesis, would not be voluntarily funded). For, we are assuming, those forced extractions leave everyone better off than they would be were people not subjected to them; these forced extractions, inducing Pareto-superior moves, are not redistributive.
This line of argument narrows the scope of the behaviour against which individuals have rightful claims; it says that further reflection upon the basis of liberty and immunity from injury narrows the
tradition's prohibitions to treatments by others thai, on net, are costly or harmful to the individual. Similar reasoning can be offered for a similar conclusion - but one that is cast in terms of the character, rather than the scope, of the individual's rightful claims. Here it is argued that an individual's rightful claims are justified by the service (hat those claims provide for her basic interests. People's rightful claims and the correlative constraints upon others' liberty of action ought to be as stringent as necessary to serve the relevant interests of those persons - and no more stringent than that. Rights can be divided by stringency into more stringent rights that straightforwardly prohibit certain interferences with the right holder, and less stringent rights that prohibit certain interferences if (but only if) those interferences leave their subjects injured in her basic interests. One standard way of characterizing these two levels of rights is to say that the more stringent rights are rights protected by a property rale (others simply must not trespass upon the right), while the less stringent rights are protected by a liability rule (others may trespass if and only if the intervention does not on net damage the interests protected by the right) (Calabresi and Melamed, 1972). If the rationale for rights is their service in protecting people's basic interests, the rationale for rights would point to rights of the less stringent sort. In the case of public goods, agents' basic interests are served by possessing rights of the less stringent variety. For this allows the coercive takings that are, by hypothesis, necessary for the public goods to be produced whereas more stringent rights would forbid those takings and leave the agents worse off in their basic interests.
(2) Members of the liberty" tradition whose primary philosophical orientation is contractarian or consequentialist will sympathize with more direct vindications of the coercive funding of public goods — vindications that do not focus specifically on the scope or stringency of rights. Contractarians will simply point to the mutual gains which, by hypothesis, the coercive financing of public goods will yield (Buchanan, 1975; Gaus, 1999: ch. 10). Because the consequentialists within the tradition adhere to the core normative commitment (doctrine I), they are highly sceptical of any measure of aggregate societal well-being that would justify harming some to benefit others. They are, instead, apt to hold that the only sound measure of societal improvement is the Pareto criterion: a social change can be said to be an improvement if and only if at least some individuals gain and no individuals lose. Absent special complications, promotions of public goods, if necessary even by coercive means, will be Pareto improvements. Hence, this generally sceptical consequentialism endorses such promotions.
(3) Finally, coercive Rinding for public goods may be defended within the liberty tradition as involving only a most benign form of paternalism and, hence, as involving little or no weakening of the tradition's strongly anti-paternalist stance (see element X). The problem that exists when individuals are faced with the choice of whether or not to contribute to the funding of a public good is that those individuals will tend to be too clever for their own good as shown by the prisoner's dilemma in Figure 9.1. Rational agents will cleverly refuse to volunteer to pay their share; regardless of what others do - whether enough others contribute or fail to - a rational agent does best by refusing to contribute. But rather than ending up with their most preferred outcome, namely receipt of the good with no payment, each will end up not receiving the good. The actual outcome will be worse, in terms of the actual values and preferences of the individuals involved, than the outcome of each individual being required to pay her share. Coercively requiring these payments from individuals merely helps them overcome their indulgence in all-too-clevcr strategic reasoning, which threatens to harm them.
This appeal to putatively benign paternalism points to a common feature of all these vindications of coercive takings - a feature to which many of the more anti-statist members of the tradition will take exception. From the perspective of these' more anti-statist members, these vindications share a failure to take choice or discretionary control seriously enough (Mack, 2000). These members of the tradition understand its crucial norms as protective of agents' authority (or jurisdiction or sovereignty) over themselves and their personal domains. The primary wrong involved in a coercive intervention consists in the intervener's impairment of the agent's choice and not in the setback to the agent's interests which normally accompanies such an impairment of the agent's discretionary control. Thus, coercive interventions remain wrongful even when they advance the interests of their subjects.
The Small State
If the arguments that support the Taxing Minimal State are extended to legitimize coercive takings for the production of other sorts of public goods (for example, the public good of mosquito abatement) or to correct other types of market failure (say, the regulation of natural monopolies), then we have gone beyond the Minimal State to the Small State. The more types of goods and sendees that are accepted as significantly public and, hence, as justifiably financed through taxation, the larger the Small State becomes.
Column
Contribute Don't contribute
Row
Contribute
Don't contribute
1
Figure 9.2 An assurance game
The nineteenth-century classical liberal political economists endorsed significant public activity on essentially market failure grounds. Public provision of roads, harbours and canals was generally endorsed, though Adam Smith at least was clear that these were not really public goods (Smith, 1976: 724; O'Brien, 1975: 275-6; Gaus, 1983). Widespread primary and secondary education are also often seen as, if not perfect public goods, goods that possess significant similarities to public goods, and so will be significantly under-supplied by free markets. However, while advocates of the Small State thus endorse taxation to fund general education, this does not justify public provision of that education. 'The strong case for government finance of at least general education,' says Hayek, 'does not however imply-that this education should also be managed by the government, and still less that government should acquire a monopoly of it' (1979: 61). Thus advocates of the Small State have endorsed vouchers, by which government compensates for the under-supply of education by additional funding, but leaves provision to the market (Friedman and Friedman, 1980: ch. 6).
The major movement towards the right of the liberty spectrum is driven by the belief that markets will feil to allocate resources efficiently to the production of public goods. Convictions about other sorts of market failure reinforce this right-ward movement. For example, a belief in the propensity of unregulated market processes to give rise to monopolies within certain industries will dispose members of the tradition to accept political regulation of those processes or those monopolies - unless it can be argued that the costs of these market failures will, in any case, be less than the costs of the governmental failures associated with this political regulation. Members of Ihe liberty tradition who accept the need and propriety for coercive taxation for the funding of an array of
public goods that go beyond protective services, and the need and propriety for some moderate level of governmental regulation in response to other market failures, are to be located at the Small State position on the liberty spectrum (but see Cowen, 1988).
The social minimum and the liberty tradition
A public goods argument can be advanced for general forced donation to the elimination of poverty. Hayek, for example, suggests that a scheme for assistance against severe deprivation is in the interest of all; indeed, he adds that 'it may be felt a duty of all to assist, within the organized community, those who cannot help themselves' (1976: 87). Now even if we all accept this duty, we may not freely contribute, as Figure 9.2 shows. The numbers in each cell refer to Column's (top right) and Row's (bottom left) preference orderings. In this 'assurance game', each gets his first choice if both contribute; they both prefer contributing to eliminate poverty over not contributing. But neither wants to be played for a 'sucker' in which one contributes but the other does not; not only might such unilateral contribution be less efficient, but the players may think it unfair that one bears the total burden. As can be seen from Figure 9.2, if Row and Column can be assured that the other will contribute, then each gets his best option by also contributing; but if one player thinks that the other will not contribute, then the rational thing to do is also not contribute (thus getting her third rather than fourth option).
It is not at all obvious, though, that charity is a good that must be achieved through joint action of this sort (Narveson, 1988: 258ff). Another difficulty with this argument is that any actual scheme of
forced contribution will ensnare individuals who do not much value Hie amelioration of poverty (who do not have the preference orderings of Row and Column); any actual scheme will, therefore, be a bad for some individuals. Furthermore, members of the tradition will emphasize that benevolent people commonly care not about the general elimination of poverty, but rather about they themselves making a contribution to amelioration of some particular instance or sort of poverty, and tins is something individuals can accomplish without the forced co-operation of others.
Nevertheless, the endorsement of state provision of an income floor or safety net remains the most likely and salient route to the Small State. Endorsement of this role for the state most prominently distinguishes advocates of the Small State from advocates of the Taxing Minimal state (and their more anti-statist brethren). Still, the endorsed redistributive role for the state must be modest, if it is not to carry its advocate outside of the liberty tradition. The state's redistributive function must be seen as something of an afterthought or supplement to the primary purpose of the state, namely, the protection of persons' rightful liberties. A further complication is that certain justifications for transfers from some individuals to others turn out not to be genuinely redistributive. Let us consider two doctrinal refinements that seek to vindicate required transfers yet are not genuinely redistributive before proceeding to genuinely redistributive proposals.
The first defends a minimal safety net on the grounds that it is a public good of the rights-protective sort. According to this argument, the presence of a safety net enhances the safety of those wrhose lives, limbs, liberties or estates would otherwise be threatened by those in free fall. If the benefits to non-free-falling individuals of the enhanced safety (in terms of the better protection of then rights) exceed the costs of their contributions to the safety net, there is as strong a justification for taxation to finance the net as there is for such taxation for national defence (Lomasky, 1987). However, such transfers are not genuinely redistributive, and for this reason this endorsement of required transfers does not amount to a move rightward from the Taxing Minimal State to the Small State.
The second non-redistributive doctrinal refinement supporting transfer involves the adoption of a version of the 'Lockean proviso' (a term coined by Nozick, 1974: 178). A Lockean proviso specifics some way in which people's otherwise unobjectionable acquisition, possession, or deployment of private property can have an objectionable net impact upon other individuals, for instance, by making those other individuals worse off than (hey would be were private property not to exist. An
advocate of such a proviso holds that acquisition, possession, or deployment of private property thai would have an objectionable impact will, nevertheless, be acceptable if the acquirers, possessors or deployers compensate those individuals so that the net impact is not objectionable (Gaus, 1999: ch. 9). Requiring property holders to make these compensatory payments will appear to be taxation for redistributive purposes. But, since the rationale for requiring these payments is much more like the rationale for requiring tort feasors to compensate those they have damaged, it is not actually taxation; requiring such payments is consistent with mere Minimal Statism.
In what ways, then, may a member of the liberty tradition seek to justify genuinely redistributive transfers'.' The ways are as many as the types of underlying philosophical strategies to be found among members oTthe tradition. Kantian members may argue that whereas the primary way in which we manifest respect for persons is non-interference with their persons and choices, respect for persons also requires that one not gratuitously fail to assist individuals who need assistance merely to sustain their personhood or agency. According to this argument, to fail to assist such persons when that assistance has no significant opportunity cost is to fail to recognize the separate importance or moral standing of these individuals. Contractarian members may argue that reasonable agents bargaining about the basic enforceable nouns that will morally govern their interaction will include a modest duty of assistance because, for each individual, the expected costs associated with this duty's enforcement will be less than the expected benefits (Morris, 1998: ch. 5). Consequentialist members may argue that the gains in social welfare generated by this modest level (and extent) of coerced assistance will exceed the losses in social welfare thereby generated.
Of course, advocates of each of these philosophical strategies who favour positions to the left of the small state will dispute their co-strategists' contentions. The more anti-statist Kantian will argue that any coercive talcing treats its subject as a means to others' ends, while no failure to assist an individual, no matter how necessitous she is, treats her as a means. The more anti-statist contractarian will argue that reasonable people who have an appropriate level of information about themselves and their prospects will not all sign on to a society-wide mandatory assistance programme. The more anti-statist consequentialist will argue for the greater social value of voluntary philanthropic and mutual aid institutions compared with the actual value of mandatory assistance programmes. Each of these more anti-statist members of the tradition, in response to pro-redistribution arguments offered by advocates of competing
philosophical strategies, will say that the appearance of those arguments proves that only adherents of her own philosophical strategy will be reliable enemies of the slate.
This structure is such that each person has the greatest possible opportunity to do whatever she might want to do (lexiwin opportunity) (1995:25).
'LEFT LIBERTARIANISM': EXTENDING THE LIBERTY TRADITION INTO STATISM?
We have argued that the liberty tradition, at its far right (i.e. statist) wing, accommodates a small state and perhaps some small genuinely redistrib-utive transfers. Others, however, have recently attempted to push the liberty tradition to embrace statist conclusions of an egalitarian nature. Although, on our account of the liberty tradition, any movement to more statist positions is a movement to the right, these more statist, egalitarian proposals are generally described as 'leftist' (Steiner and Vallentyne, 2000). We shall follow convention here and speak of this egalitarian push for a more extensive and activist state as 'left lib-ertarianism'. We can identify three strategies that have been pursued by left libertarians: (1) the endorsing of a more positive conception of liberty; (2) the supposition of equal claim to all social resources; and (3) an expansive interpretation of harming others.
Liberty as Effective Opportunity
The liberty tradition's devotion to individual freedom is, of course, a devotion to some version of negative freedom (Berlin, 1969). Now egalitarians have long criticized the liberty tradition for denying that freedom requires resources. Liberty, it has long been argued, is not simply about the absence of interferences with one's actions, but the ability to perform the actions a person desires. In short, a free person can do what she desires to do. As the British socialist R. H. Tawney put it, liberty implies 'the ability to act'(1931: 221).
Phillipe Van Parijs has recently advanced a far more sophisticated version of this so-called left-wing critique of negative liberty (for discussions, see Reeve and Williams, 2003). For Van Parijs, 'real freedom' involves three components: 'security, self-ownership and opportunity - in contrast to formal freedom, which only incorporates the first two' (1995: 22-3). A real libertarian society, Van Parijs argues, meets three conditions:
1 There is some well-enforced structure of rights (security).
2 This structure is such that each person owns herself (self-ownership).
As Van Parijs explains, this last condition requires that 'in a free society, the person with least opportunities has opportunities that are no smaller than those enjoyed by the person with the least opportunities under any other feasible arrangement' (1995: 25). And this in turn leads to the requirement that a society provides the highest sustainable basic income for all, including surfers who spend their days off Malibu (a position rejected by Rawls, 2002: 179).
Van Parijs argues - as he must, given his commitment to self-ownership - that the transfers of income required by such a scheme respect the self-ownership of the industrious and do not exploit them (1995: ch. 5). It is, he argues, the equalization of external endowments that drives redistribution: if someone produces without using resources in scarce supply she has the right to her full product, but because production always requires such resources, his basic income proposal does not lead to exploitation of the industrious.
Van Parijs's case that this proposal is not unfair to the industrious is long, sophisticated and complicated. It is so complex because he is trying to show what seems manifestly false: namely that a scheme in which the productive are required to provide the unproductive with the highest feasible income does not exploit the productive (using them as a mere resource for the surfing pleasure of others). If one has a claim to the fruits of one's labour, removing these fruits so that others can surf certainly appears unjust. Consider a version of an example of Van Parijs (1995: 160), in which the Greens and the Reds live in a society that only produces houses; each person receives a house of equal quality. In their production, the two groups work equally hard but the Reds, alas, are unlucky: their tools break, termites infest their structures, every time they paint it unexpectedly rains, and so on, such that they actually do not produce any houses. Van Parijs takes it as manifest that because the Reds and Greens work equally hard and get the same rewards no exploitation results; indeed he thinks it would be 'embarrassing' to see the Reds as exploiters. Yet for all their (wasted) effort the Reds have not managed to build houses to live in: they only live as well as the Greens because they have taken over some of the fruits of Green labour. Van Parijs's intuition is that luck must be irrelevant to justice (1995: 160): if some have managed to actually produce while others have tried and failed because of bad luck or natural adversities (see also Steiner, 2001), they have a claim on those who do produce. 'It's not our
fault." the Reds appear to claim, 'that we never actually build houses; those who succeed must give us houses as good as their own.' This is not an intuition shared by the liberty tradition (Rand, 1957). It violates doctrinal commitment IV of the liberty tradition against seizure of the fruits of another's labour, a comniitmcnt that, unsurprisingly, Van Parijs rejects (1995: 1451T). Moreover, Van Parijs's position seems to illustrate how violations of commitment IV also compromise persons' claims to self-sovereignty (doctrinal coinmitment II).
Equal Ownership of Original Property
In contrast to Van Parijs, the left libertarianism of Hillel Steiner advocates a radical version of negative liberty. Building his theory on a Hobbesian conception of negative liberty, Steiner holds that 'Broadly speaking, it suggests that a person is unfree to do an action if, and only if, his doing that action is rendered impossible by the action of another person' (1994: 8). If, am free to X if and only if I cannot be prevented by another from A'-ing, then it follows that I am free to A'if and only if none of the locations and objects necessary to A'-ing are controlled by others, or would be controlled by others should 1 attempt to X. Thus 'Freedom is the possession of things' (1994: 39). But to have a right to freedom requires more: it is to have a title to a domain of locations and things: it is to have prop-erty^rights (1994: 81). Thus all rights to freedom are property rights, and all property rights are rights to freedom, a claim made by many in the liberty tradition (see element IV).
Steiner claims that his account of liberty as property rights has a virtue lacking in competing theories of rights: compossibility. If rights are defined in terms of intentional actions - e.g. I have a right to see a film tomorrow and you have a right to wreck a building tomorrow — they can conflict:
Whether my seeing a film tomorrow afternoon, and your wrecking a building then, are or are not jointly per-formable actions depends inter alia on whether the building yon are to wreck is the cinema I'm to attend. 11 and only if we each have a duty to do these actions, those two duties are incompossible and so are the respective rights which they eorrelatively entail. (1994: 91-2)
Steiner thus argues that a system of rights can be guaranteed to be compossible - the performance of all the correlative duties are necessarily jointly possible - only if rights are defined in terms of property over a 'set of extensional elements' (control over objects, locations in time and so on).
Steiner's conception of liberty, rights and the compossibility requirement supports his entitlement account of justice. Justice involves a division of the
world into various domains, each person possessing rights over some of it, tins defining his sphere of freedom. Since. Steiner insists, rights protect our ability to choose, we are free (while alive) to transfer our holdings to others; the justice of a system of holdings will thus crucially depend on its history — whether it has been brought about by a series of legitimate, non-exploitative, choices.
Thus far all this looks like the liberty tradition, but we still need to know what constitutes a just initial distribution of the world's resources. Only if the initial distribution is just will the subsequent moves be just. Arid while Steiner insists that we own our bodies (though not our 'germ line genetic information'), he rejects the simple version of Lockean theory, according to which simply by mixing our labour with an unowned resource we appropriate the resource. But if we camiot justly appropriate nature that way, how do we generate just claims over natural resources?
Steiner is attracted to a quick route to egalitari-anism. This quick argument for equality requires two premises: (1) justice involves treating equally those who are in relevant respects equal, and treating unequally those who are in relevant respects unequal; and (2) in these matters there are no relevant differences; so (3) justice demands equality. This argument leads Steiner to the claim that everyone is entitled to equal freedom and so to some sort' of equal share of natural resources (1994: 216, 235). This equality of ownership may also be depicted as a version of the Lockean theory, in which wc originally hold the world in common (Otsuka, 1998). Steiner supports the crucial premise (2) by arguing that rules of justice are intended to adjudicate disputes among those with different moral codes; any attempt to specify the relevant differences between those with different moral codes is sure to draw on those very codes that are being disputed. Consequently it is question-begging to advance any account of relevant differences (1994: 215) and so, from the perspective of justifying a system of justice, we must be completely sceptical that there are any relevant differences. So, in essence, Ihe argument is that formal justice plus thoroughgoing scepticism about public relevant reasons equals equality: if 'no criterion for relevantly differentiating cases can be eligible to serve as a standard of distributive justice, the inference must be that no cases can be regarded as relevantly-different; that is, all cases are relevantly alike' (1994: 216).
But the inference is not pellucid. Consider: (1) if justice involves treating equally those who are in relevant respects equal; and (2') because of thoroughgoing scepticism about public reasons, it cannot be shown that people are equal in any relevant way; then (3') justice doesn't demand treating everyone
equally. As Steiner seems to realize, it is obscure how to apply a doctrine about relevant reasons (formal justice) given the complete absence of relevant reasons; formal justice makes sense in an account of the rule of law against a background of rights and relevant reasons. Although at some points in his Essay on Rights Steiner seems to distance himself from this argument, he comes back to its central claim in the same work: that in a world where all reasons are tendentious, only equality is impartial, and so only equality can be accepted by everyone as a 'lexically prior' rule of justice, overriding the rules of their own moral codes.
Steiner, and most left-libertarians, thus uphold some form of su1ffi^ti£0Jncpj^j'edistributionj>n the grounds of a claim by each to an equal share in natural resources. Unique to Van Parij's left liber-tarianism is a 'massive extension' of the scope of resources to be redistributed (taxed) by including job assets as an external resource (Van Parijs, 2003: 206). More specifically, Van Parijs seeks to identify and capture the employment 'rents' associated with jobs. (On the idea of rents, see Mack, 1992.) The core of Van Parijs' argument for treating jobs as assets is that, for a variety of reasons, workers are paid above the market-clearing wage; this constitutes a rent, and should be considered an additional clement of their endowment (Van Parijs, 1995: 108). In our economics a worker's endowment can have a value of A' (the income she would receive from her job given a market-clearing wage), but she may receive X+n; the value of n is a rent which Van Parijs counts as a social resource. Now suppose we are in a situation with a market clearing wage: the marginal entrant gets X and so collects no rent, but given uniform pricing all non-marginal entrants collect a surplus - they would work for X-m, but still receive X. Van Panjs, however, explicitly excludes Ihese as rents (1995: 264, n35). His complaint, then, is not that many people get more than they require to do tbeir job; it is that if Marge the marginal worker gets more than she requires, there exists some unemployed person Maggie who would take the job at a lower wage than Marge, but Marge has claimed this scare resource and so effectively denies to it Maggie. Van Parijs sees everyone who gets rent n as having claimed a scarce social resource, and so their rent may be taxed away. However jobs cannot be considered simply as unalloyed resources to be distributed, but as packages of rights and liabilities (Williams, 2003). Given these liabilities, many do not want a job even if it is offered to them: it does not count as a resource to them since they would not take it at any price because since they don't want to work. It looks as if Van Parijs is exploiting those who work to support those who do not want to: those uninterested in actually doing the job receive a sort of compensation for
being excluded from it. Moreover, this is especially odd in a 'libertarian' theory: the benefits of an uncoerced agreement between two free agents employing their labour and property - an employment contract -is said to generate a social resource to which all others have something like an equal claim.
the results of 'the global syc.!eiif. v\ [iich hk1 nlinosl solely the resuhs of other peopled Liclions and decisions.
CONCLUSION
to do wiih anarchism' ivon Mises, 1985: 37). The classical liberal, said von Mises. is no more an enemy of the stale than he is an enemy of sulphuric acid: bolh can be useful, and both are dangerous (1985: 38). At least about that last part, all members of the liberty tradition will agree.
Extending Harmful Action
If equal ownership of resources cannot be established as a default, the egalitarian interpretation of Locke must provide a positive argument as to why we all possess equal claims to external resources. If natural resources are owned in common, how did that come about? Far more plausible is the view thi.t natural resources are not originally owned at all -that we all equally lack original rights over natural resources - but that any act of acquisition must avoid banning others, which leads to something like the liberty h-adition's understanding of the Lockean proviso. Now a third 'leftward' (statist) push comes from attempts to widen the concept of harm, such that almost any economic activity constitutes a harm to others.
Thomas Pogge (2002) has recently argued along these lines. Pogge builds his case on what looks like the liberty tradition's understanding of negative rights, in particular the right not to be harmed (see element X). However, he argues that 'simple liber-tarianism' is flawed because it fails to appreciate how institutions create harm (2002: 172). In particular, Pogge insists that the imposition of the 'global economic order' causes harm: it 'engenders war, torture and starvation' (2002: 173), and so anyone who participates in that order is contributing to injustice (2002: 211). Indeed, all participants help starve the poor (2002: 214). Thus global redistribution is required to compensate for harms done, as well as to satisfy a version of the Lockean proviso (2002: ch. S; compare Steiner, 1994: ch. 8).
Note how this argument depends on the idea that there exists an overall global system which is to be Ihe object of our evaluation (rather than, say individual actions), that this system is coercivcly imposed on the poor, that as a consequence of this system the poor are harmed, and that each one of us is 'deeply implicated' in the harm perpetrated by the system (2002: 142). This is not the place to evaluate these controversial claims about international economics and politics [see further Chapters 17 and 22]. It is important here to note, though, the way that the argument seeks to avoid the normative individualism of the liberty tradition (element I). The idea that there exists an all-encompassing system in which all participate - it is almost impossible to opt out -and so we are all responsible for the results of the system, makes each individual responsible for all
We have characterized the liberty tradition both by its doctrinal commitments and by its internal controversies, and we have criticized attempts to extend it towards statism. The unity of the liberty tradition is comprised by its convergence - not, of course, perfect, but substantial - on the key doctrinal commitments. The tradition's core commitment is a normative individualism that grounds the primary political norm of individual liberty and responsibility; individual liberty is what each individual may legitimately demand of each other individual, and individuals arc responsible for their own actions and decisions. Respect for the individual and his liberty requires respect for that individual's properly rights. The liberty tradition refuses to divide up liberty into the personal and the economic, and it refuses to equate freedom with the power to do what one wants. The market, as well as private property, is fundamental to the tradition. In a market order, desirable social order emerges out of the choices that individuals make when their rights are secure. And when these rights are secure, individuals tend to benefit from their own productive decisions and bear the costs of their own activities. In the liberty tradition, coercion is first and foremost justified on the grounds that it is required to block or nullify infringements upon the rightful claims of individuals; use of coercion that infringes upon an individual's control of her person or property is illicit. Or, we might say, justified coercion prevents harms to others, but we must be careful not to stretch the concept of harm so that everything we do always harms others. As the liberty tradition sees it, the state - a mechanism of organized violence -ought to be bound by the morality that applies to all. But, of course, it has not been bound: the state regularly engages in unjustified aggression, plunder, and meddlesomeness.
The liberty tradition's internal disputes focus on the extent of the justified state, from the market anarchist who denies that any monopoly of violence can be justified, to the advocate of the Small State who is willing to countenance not only the state but coercive taxation to promote public goods and correct various 'market failures'. The market anarchist is apt to see the small statist as an ally of the modern expansive state, lending legitimacy to the idea that government's job is to 'fix the market'. The small statist responds that '[classical] liberalism is not anarchism, nor has it anything whatsoever
NOTES
1 Slightly more cumbersome labels for this stance would be 'private property anarchism1 and 'auarcho-capitalisra'. 'Individualist anarchism' would he fine except for the residue within actual individualist anarchists, such as Benjamin Tucker (1926) and Lysander Spouner (1971). of belief in the labour theory of value. See further Murray Rothbard (2000).
2 See Nozick (1974: chs 4, 5). ForNozick, this greater scope for suppression allows for a legitimate 'ultra-minimtil' state, but that legitimacy is only retained if die lultra-minimal' state compensates those who are disadvantaged, by these prohibitions and. thereby, transforms itself into the 'minimal' state.
REFERENCES
Barnctt, Randy E. (1998) The Structure of Liberty.'
Oxford: Clarendon. Bentham, Jeremy (1987) Introduction to the Principles
of Morals and Legislation.  In Alan Ryan, ed.,
Utilitarianism and Other Essays. Harmondsworth:
Penguin.
Berlin, Isaiah (1969) 'Two concepts of liberty'. In his Four Essays on Libert)'. Oxford: Oxford University Press.
Brittan, Samuel (1988) A Restatement of Economic Liberalism. Atlantic Highlands, NJ: Humanities.
Buchanan, James M. (1975) The Limits of Liberty: Between Anarchy and Leviathan. Chicago: University of Chicago Press.
Buchanan, .lames M. and Gordon Tullock (1965) The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor, MI: University of Michigan Press.
CalabresL Guido and A. Douglas Melamed (1972) 'Property rights, liability rules and. inalienability: one view of the cathedral'. Harvard Law Review, 8: 1089-128.
Cowen, Tyler (1988) The Theory of Market Failure. Lanham, MD: University Publishing Association.
Cowen, Tyler (1992) 'Law as a public good: the economics of anarchy". Economics and Philosophy, 8: 249-67.
Cowen, Tyler with Gregory Kavka (2003) 'The public goods rationale for government and the circularity problem'. Politics, Philosophy and Economics, 2 (June): 265-78.
Handbook of Political Theory
De Jasay. Anthony (1991) Choice. Cu/Uruci, Consent; A Restutement of Liberalism. London: Institute of Economic Affairs.
Friedman, David (1973) The Machinery of Freedom. New York: Harper and Row.
Friedman, Milton and Rose Friedman (1980) Free to Choose. London: Seeker and Warburg.
Gaus, Gerald F. (1983) 'Public and private interests in liberal political economy, old and new'. In S. I. Benn and G. F. Gaus, eds, Public and Private in Social Life. New York: St Martin's, 183-222.
Gaus, Gerald F. (1999) Social Philosophy. Armonk. NY: Sharpe.
Gray, John (1986) Liberalism. Milton Keynes: Open
University Press. Hayek, F. A. (1944) The Road to Serfdom. Chicago:
University of Chicago Press. Hayek, F. A. (1973) Law, Legislation and Liberty. Vol. I,
Rides and Order. London: Routlcdgc. Hayek, F. A. (1976) Law, Legislation and Liberty. Vol. II,
The Mirage of Social Justice. London: Routledge. Hayek, F. A. (1979) Law, Legislation and Liberty. Vol. Ill,
The Political Order of a Free People. London:
Routledge.
Hayek, F. A. (1991) 'The use of knowledge1. In RichEird
M. Ebeling, ed.. Austrian Economics. Hillsdale. Ml:
Hillsdale College Press, ch. 14. Locke, John (I960) Second Treatise of Government. In
Peter Laslett, ed.,  Two Treatises of Government.
Cambridge: Cambridge University Press. Lomasky, Loren E. (1987) Persons, Rights, and the Moral
Community. New York: Oxford University Press. Machan, Tibor (1989) Individuals and Their Rights.
La Salle, IL: Open Court. Maclntyre, Alasdair (1988) Whose Justice? Wliich
Rationality? Notre Dame, IN: University of Notre
Dame Press.
Mack, Eric (1990) 'Self-ownership and the right of
property3. The Monist, 73 (October): 519-43. Mack, Eric (1992) 'Gauthieron rights and economic rents'.
Social Philosophy and Policy, 9: 171-200. Mack, Eric (1999) 'In defense of individualism'. Ethical
Theoiy and Moral Practice, 2: 87-115. Mack, Eric (2000) 'In defense of the jurisdiction theory of
rights'. Journal of Ethics, 4 (January-March): 71-98. Mill, Jonn Stuart (1991) On Liberty. In John Gray, ed., On
Liberty and Other Essays. New York: Oxford
University Press. Morris, Christopher (1998) An Essay on the Modern State.
Cambridge: Cambridge University Press. Narveson, Jan (1988) The Libertarian Idea. Philadelphia:
Temple University Press. Nozick, Robert (1974) Anarchy, State and Utopia.
New York: Basic. Oakeshott, Michael (1975) On Human Conduct. Oxford:
Clarendon.
O'Brien, D. P. (19751 The Classical Political Economists.
Oxford: Clarendon. Otsuka. Michael (1998) 'Self-ownership and equality:
a Lockean reconciliation'. Philosophy and Public Affairs.
27 (Winter): 65-92. Pogge, Thomas (2002) World Poverty and Human Rights.
Oxford: Polity. Rand, Ayn (1957)-4rfos Shrugged. New York: Dutton. " Rawls, John (2002) Justice as Fairness: A Restatement.
ed. Erin Kelly Cambridge, MA: Belknap. Reeve, Andrew and Andrew Williams, eds (2003) Real
Libertarianism Assessed: Political Theoiy after Van
Parijs. New York: Palgrave Macmillan. Robbins, Lord (1961) The Tlteory of Economic Policy in _
Classical  English   Political  Economy. London:
Macmillan.
Rothbard, Murray (2000) 'The Spooner-Tucker doctrine: an economist's view'. In Egalitarianism as a Revolt against Nature, 2nd edn. Auburn, AL: Ludwig von Mises Institute.
Schmidtz, David (1991) The Limits of Government: An Essay on the Public Goods Argument. Boulder. CO: Westview.
Simmons, John A. (1993) On the Edge ofAnarchv: Locke,
Consent and the Limits of Society.  Princeton, NJ:
Princeton University Press. Smith, Adam (1976) An Inquiry into the Nature and
Causes of the Wealth of Nations, ed. W. B. Todd.
Oxford: Clarendon. Spooner, Lysander (1971) The Collected Works of
Lysander Spooner. Weston, MA: M. and S. Steiner, Hillel (1994) An Essay on Rights. Cambridge,
MA: Blackwell. Steiner, Hillel (2001) 'The ethics of redistribution', Acta
Philosophica Fennica, 68: 37^-5. Steiner, Hillel and Peter Vallentyne, eds (2000) Left-
Libertarianism and Its Critics: The Contemporary
Debate. Basingstoke: Palgrave. Tawney, R. II. (1931) Equality. New York: Harcourt,
Brace.
Tucker, Benjamin (1926) Individual Liberty. New York: Vanguard.
Van Parijs, Phillipe (1995) Real Freedom for All: What (If Anything) Can Justify Capitalism? Oxford: Clarendon.
Van Parijs, Philippe (2003) 'Hybrid justice, patriotism and democracy'. In Real Liberterianism Assessed: Political Theoiy after Van Parijs, ed. Andrew Reeve and Andrew Williams. Hoxmdsmill, Basingstoke: Palgrave: 201-214.
Von Mises, Ludwig (1985) Liberalism in the Classical Tradition. San Francisco: Cobden.
Williams. Andrew (2003) 'Resource egalitarianism and the limits to basic income'. In Real Libertarianism Assessed: Political Theoiy after Van Parijs, ed. Andrew Reeve and Andrew Williams. Huundsmill, Basingstoke: Palgrave: 111-35.
10
Conservative Theories
JOHN KEKES
Conservatism is a political morality. It is political because it aims at political arrangements that make a society good, and it is moral because it holds that a society is good if it enables people living in it to live good lives, that is. lives that are personally satisfying and beneficial for others. Conservatism, like liberalism [see Chapters 7-9] and socialism [see Chapters 6 and 30], has different versions, partly because conservatives often disagree with each other about the particular political arrangements that ought to be conserved.1 There is no disagreement among them, however, that the reasons for or against those arrangements are to be found in the history of the society whose arrangements they are. This commits conservatives to denying that the reasons are to be derived from a hypothetical contract, or from an imagined ideal order, or from what is supposed to be beneficial for the whole of humanity. In preference to these and other alternatives, conservatives look to the history of their own society because it exerts a formative influence on their present lives and on bow it is reasonable for them to want to live in the future. The conservative attitude, however, is not an unexamined prejudice in favour of the historical arrangements of the conservatives' society. They are in favour of conserving only those arrangements that their history has shown to be conducive to good lives.
Another reason for the disagreements among conservatives is that, although they agree in regarding certain questions as basic to political morality and in identifying the range of reasonable answers to them, they nevertheless give answers that fall at different points within that range. The combination of the questions that are thought to be basic and the answers to them that are thought to be reasonable defines different versions of conservatism,
explains their differences, and distinguishes between conservative, liberal, socialist, and other theories.
These questions are:
o   To what extent should political arrangements be based on history?
• How does the diversity of values affect political arrangements?
• What should be the relation between individual autonomy and social authority?
• How should political arrangements respond to the prevalence of evil?
The discussion will proceed by considering these questions and the different answers conservatives give to them. It will conclude by identifying a version of conservatism that appears to be the most reasonable.
TO WHAT EXTENT SHOULD POLITICAL ARRANGEMENTS BE BASED ON HISTORY?
Conservatives agree that history is the appropriate starting point, but some of them believe that it is not a contingent fact that certain political arrangements have historically fostered good lives, while others have been detrimental to them. Conservatives who believe this think that there is a deeper explanation for the historical success or failure of various arrangements. There is a moral orders in reality. Political arrangements that conform to this order foster good lives, those that conflict with it are bound to make lives worse. These conservatives are committed to a