CAN LIBERTARIANS GET AWAY WITH FRAUD?

Abstract:

In this paper I argue that libertarianism neither prohibits exchanges in which consent is gained through deceit, nor does it entail that such exchanges are morally invalid. However, contra James Child’s (1994) similar claim, that it is incapable of delivering these verdicts, I argue that libertarians can claim that exchanges involving deceitfully obtained consent are morally invalid by appealing to an external theory of moral permissibility.

1. INTRODUCTION

Consider a transaction between two persons, A and B, in which A sells his car – which he knows is a ‘lemon’ – to B for $1,000. Suppose B agrees to purchase the car on the condition that it is in good working order. However, A lies and assures B the car is fine. Further, suppose that had B known the truth about the car’s condition, she would not have consented to the terms of the transaction.

Unlike outright theft, which involves a complete lack of consent, transactions like the above involve ‘imperfect consent’. In cases like the car sale, the imperfect consent is generated through deceit: although B ostensibly tokens consent to the transaction, she would not have agreed to the terms if she had known the truth about the car’s condition. Nevertheless, there are possible transactions that are ‘close’ to the above transaction to which B would have consented. Transactions in which B’s consent is imperfect because A deceives B about a property of the goods being exchanged are ‘fraudulent transactions’.¹

Fraudulent transactions are not the only kind of transaction involving imperfect consent. Exploitation, extortion, blackmail and various forms of duress all involve tokened consent that is rendered imperfect in various ways.² However, one feature that distinguishes fraud from these forms of interaction is that in fraud it is deceit that undermines consent. Call consent that is imperfect because it is predicated on deceit ‘deceitfully obtained consent’. The scope of deceitfully obtained consent is broader than the common usage of ‘fraud’. It applies not only to market based transactions for goods, but also to any consensual agreement in which the consent of one of the parties is secured through deceit. Thus, it also applies, for example, to consent to sex, labour and other uses of the body. In this paper I shall consider a problem that deceitfully obtained consent presents for libertarianism.

2. THE PROBLEM OF DECEITFULLY OBTAINED CONSENT

Because fraud and related acts involve deceitfully obtained consent, their outcomes are, prima facie, unjust. I take it that this ordinary moral intuition is based on three claims.

1. 1. Deceit is morally impermissible.

2. 2. If deceit is morally impermissible, then consent based on deceit is morally invalid.

3. 3. Holdings resulting from consensual exchanges of moral rights that lack morally valid consent are unjust.

4. C. Holdings resulting from consensual exchanges of moral rights that are based on deceitfully obtained consent are unjust.

Libertarians can accept (3) and I shall assume in what follows that it is true.³ However, it is not clear that libertarians can sustain (C) via (1) and (2). The problem for libertarians is that their theory does not appear to prohibit deceit. If deceit is not prohibited within the libertarian framework, then it is unclear how exchanges that involve deceitfully obtained consent can be unjust. One way libertarians can respond to the problem is to show that their framework does (or can) imply the truth of (1). Alternatively, they could provide a different argument for (C) that bases the invalidity of deceitfully obtained consent on considerations independent of (1) and (2).

The primary focus of this paper will be on whether libertarianism can sustain (C). However, I suspect many people think that A’s sale of the car is not only unjust, but that it should also be illegal. That is, they think

1. 4. Exchanges based on deceitfully obtained consent should be prohibited by the state.

Although the injustice of such exchanges is, for libertarians, a necessary condition of (4), it is not always a sufficient condition, since many libertarians place more stringent conditions on the use of coercive power by the state. In what follows I shall discuss (4), but my primary focus will be on whether libertarians can deliver the weaker conclusion (C).

The problem of deceitfully obtained consent is more pressing than, I suspect, most libertarians realize. First, it creates problems beyond the domain of market based transactions that are typical of fraud: it also affects the domain of self ownership. If libertarians cannot distinguish deceitfully obtained consent from morally transformative consent, then they cannot condemn cases where the consent to the use of one’s body is based on deceit. Consent to sexual acts, medical trials and labour contracts predicated on deceitfully obtained consent will also not be proscribed by the framework.⁴ Second, as many economists have pointed out, if not prohibited, widespread fraud would also severely undermine the existence of markets, an important libertarian institution.⁵ Not only do libertarians value markets intrinsically (Narveson 1988), they are also instrumentally important for the determination of market values that are used to calculate just distributions in certain libertarian theories (Steiner 1994). Thus, the presence of deceitfully obtained consent poses a serious problem that threatens to both erode the market mechanism and undermine rights of bodily integrity.

It is important to note that virtually all⁶ libertarians believe their framework not only supports (C), but that it also entails the stronger claim (4). Existing libertarian theories explicitly include state enforced prohibitions on fraudulent transactions. For example, in Anarchy, State, and Utopia, Robert Nozick includes a prohibition on fraud and argues that ‘a minimal state, limited to the narrow functions of protection against force, theft, fraud, enforcement of contracts, and so on, is justified’ (Nozick 1974: 5). Similarly, Jan Narveson argues that ‘restraints on force and fraud . . .simply [are] the market’ that libertarians champion (1988: 212). Finally, Murray Rothbard claims that fraud is an offence ‘actionable by law’ (1998: 143).

3. THE LIBERTARIAN FRAMEWORK

Libertarianism recognizes three increasingly stringent moral categories. First, acts are morally prohibited – impermissible – when they violate either the right of self ownership, the right of property ownership, or second order rights derived from these fundamental rights. Second, in certain cases the coercive use of force is permitted to protect against the violation of these rights by others, or to rectify their prior violation. Third, certain morally prohibited acts are also subject to enforceable prohibition by the state.⁷

Although libertarians agree that threats to, or violations of, these rights are necessary for the use of coercive force, not all libertarians believe all such violations or threats are sufficient for the coercive use of force. For example, some libertarians argue that the use of force for the defence of these rights against innocent threats is subject to stringent limits (Otsuka 1994). For these libertarians threats or violations of rights must be accompanied by other factors (say, the non-innocence of the violator) before the use of force is justified. This is why establishing the injustice of fraud in a libertarian framework is insufficient to establish claim (4).

The libertarian theory of justice is entitlement based. For libertarians, a distribution of legal or social rights over goods is just iff the goods are acquired in accordance with a principle of just initial acquisition, or are justly transferred from another person who had a prior just entitlement to those goods (Nozick 1974). Because the issue of deceitfully obtained consent arises in the context of exchange and transfer, it falls under the category of justice in transfer. Thus, if the holdings resulting from an exchange involving deceitfully obtained consent are unjust because of the deceitfully obtained consent, then deceitfully obtained consent must violate principles of just transfer. Principles of just transfer may be violated in one of two ways. First, if transfers involve morally prohibited acts, that is, if they violate self ownership or rights, property ownership rights, or second order rights. Second, if they involve morally invalid acts; that is, acts that, though not directly prohibited, come about in a procedurally flawed way. So, in summary, (C) is true within a libertarian framework when exchanges involving deceitfully obtained consent are either impermissible because they violate self ownership or property ownership or when they are morally invalid. I shall consider each of these possibilities below.

4. THE IMPERMISSIBILITY OF DECEITFULLY OBTAINED CONSENT

In a somewhat overlooked article in Ethics, James Child considered the narrower question of whether libertarianism entails a prohibition on fraud (rather than the broader category of deceitfully obtained consent). Child concluded that basic libertarian principles ‘do not support a prohibition of fraud’ (1994: 722). He argues that rights to self-ownership and property ownership prohibit transfers using force, threats of force, and theft by stealth, but because fraud falls under none⁸ of these forms of transfer, it is not prohibited by the fundamental libertarian rights. If fraud is not prohibited by the fundamental rights, then it is not prohibited by any other second-order rights that may be derived from the fundamental rights. Child’s analysis is case-based. He enumerates what he takes to be an exhaustive list of the kinds of acts that violate libertarian rights. He concludes that, since fraud falls into none of these categories, it does not violate these rights.

Although Child’s case-based argument is fairly convincing, it leaves open the possibility that fraud may fall under a different category of prohibited action Child fails to consider. However, there is also a principled argument that deceit is not prohibited. As Hillel Steiner points out, for libertarians who base their theory on a Kantian foundation, there is a distinction between those factors that make an action virtuous (permissible) and those which make it just. He writes, “The virtue of an action depends, for Kant, on the intention with which it’s done . . . its justness has to do with whether and how far it restricts someone else’s freedom” (Steiner 1994: 212). He continues,

whether an act is just doesn’t depend on the intention with which it’s done. It doesn’t depend on its having a certain intensional description. It depends, rather, on its having a certain extensional description: that is, on whether it’s compatible . . . with a certain set of rights. (Steiner 1994: 212)

Deceit is an intentional act. Whether a false statement is also a deceitful statement depends on the intentions of the speaker. Thus, whether a statement is deceitful depends on its having a certain intensional description. But according to Kant, whether an action is just depends – extensionally – on whether it respects a certain set of rights. The same is true of libertarian theories. Whether an action violates self ownership or property ownership rights depends similarly on its extension. Property and self ownership rights imply, inter alia Hohfeldian claims against others (Hohfeld 1919) performing certain actions and these claims do not depend on the intentions or motivations behind these actions. Of course, given that an action does violate rights, then what counts as an appropriate response to this violation may depend on the violator’s intention. Murder, for example, is distinguished from manslaughter by malice aforethought, the intention to kill. But crucially, manslaughter is already a violation of the victim’s self ownership rights; a verdict that is independent of the killer’s motives.

In the case of deceit, it is far from clear that the underlying action independent of intention – the conveyance of a falsehood – is something that persons have libertarian rights against. Neither self ownership nor property ownership imply rights to not receive false information. One exception occurs in special cases where a party has explicitly contracted to receive information. It is plausible that provided they contracted to justly acquire information that was previously justly held, they have a valid property right to that information.

This exception has been used to motivate a potential solution for libertarians.⁹ In the purchase of a car from A, B might contract not only to receive the car from A, but also to receive only true information from A, thus securing a right to the truth and precluding A’s providing false information about the car’s condition. Though promising, this solution cannot succeed. It is based on the idea that information is a good like any other that can be transferred between persons. However, if this is so, then exchanges of information are also subject to deceitfully obtained consent. Just as A might lie about the car’s condition in order to obtain B’s consent to a transaction she would not otherwise have consented to, A might lie about the properties of the information he is providing in the no-fraud clause that is intended to solve the fraud problem. Of course B could then try to contract against A’s selling him a fraudulent contract, but by now it should be clear that this strategy is subject to a vicious regress. Deceit may be used to obtain consent at any stage of the process and, unless deceit itself can be condemned, the problem re-emerges.

It appears that libertarianism is incapable of sustaining the claim that deceit is impermissible. Recall the argument behind the intuition that the outcomes of agreements predicated on deceitfully obtained consent are, prima facie, unjust: (1) deceit morally is impermissible; (2) if deceit is morally impermissible, then consent based on deceit is morally invalid; (3) since holdings resulting from consensual exchanges of moral rights that lack valid consent are unjust, it follows that (C), holdings resulting from exchanges based on deceitfully obtained consent are unjust.

I have argued that for libertarians, (1) is false. Libertarianism entails acts are impermissible only when they violate basic libertarian rights. Since deceit does not violate basic libertarian rights, it is not impermissible. Thus, if libertarians wish to defend (C) they must do so via an alternative argument.

5. THE INVALIDITY OF DECEITFULLY OBTAINED CONSENT

In Section 3 I noted that principles of just transfer may be violated not only through the performance of impermissible acts that violate basic rights, but also when an exchange is morally invalid. The distinction between impermissibility and invalidity is subtle, but important. While certain transactions, such as theft, are impermissible – universally proscribed – other forms of transaction may be invalid because they are procedurally flawed. Consider the following example from the context of law. A contract to ‘buy’ my neighbour is, in nearly all jurisdictions, prohibited. It is never legally permissible to purchase another human being. A contract to buy my neighbour’s boat, however, is not directly prohibited by law. Such a contract may or may not be legally binding, depending on the conditions under which the contract was made, what conditions it specifies, and so on. Though not impermissible, the purchase contract for the boat can still be invalid if its establishment is procedurally flawed. Similarly, libertarians might accept that their basic rights do not entail an outright prohibition on exchanges involving deceitfully obtained consent while nevertheless maintaining that the way in which these exchanges arise suffices to make them morally invalid and thus, not morally transformative. This approach seeks to find an alternative reason that deceit invalidates consent. The moral invalidity of deceitfully obtained consent would then follow not from the fact that deceit is impermissible, but other considerations. Here I consider two ways in which deceit, though permissible, might render exchanges involving deceitfully obtained consent morally invalid.

5.1. Domain Conditions

Alongside the basic rights of self ownership and property ownership, the libertarian framework also contains domain conditions that limit these principles to the domain of morally responsible persons. As Child notes, morally responsible agents must possess the deliberative capacity to ‘acquire, understand and appraise information, which includes considering its probability of truth or falsity and its relevance’ as well as the capacities to ‘entertain a stable set of preferences by which choice among various options with various payoffs can be made’ and ‘in light of this information and these values, to consider choices and weigh possible risks, costs, and benefits of these choices. This includes the risk and cost of acting on false information’ (Child 1994: 730).

Exchanges with persons who lack these capacities are generally invalid. For example, suppose A convinces B, a toddler and heiress to a large fortune, to trade her inheritance for candy. Libertarians can claim the consent of the toddler does not validate the transaction, not because self-ownership or property ownership are violated, but because B lacks the aforementioned capacities, she is not a morally responsible agent (and, therefore, not a self-owner). Although exchanges of food for annuities are not prohibited by libertarian principles, in this example the violation of domain conditions renders the exchange morally invalid. If A were to take possession of B’s fortune in such a case, he would act unjustly.

However, none of these domain conditions apply to deceitfully obtained consent. B may possess the capacities necessary for competent transaction while still being deceived by A. If agents retain the capacities necessary to qualify as moral persons, then they must evaluate the quality of the information they receive (as in capacity (3) above). Caveat emptor applies very strongly within libertarianism. In the car sale case, although B did not actually verify A’s claims about the functionality of the car, she retained the capacity to do so. B must decide whether to believe A – whether to give his claims any weight and whether to act on his information. In general, A merely expands B’s information set, he does not force her to act, or to consent to the transaction. Thus, because deceit does not undermine the domain conditions that characterize moral personhood, an appeal to these conditions cannot be employed as an alternative to the claim that deceit is impermissible in an argument for the moral invalidity of deceitfully obtained consent.

5.2. Influencing Consent

A second route to moral invalidity references the fact that deceit has an influencing effect on consent. I think it is widely accepted that deceit does influence consent and it might be argued that it is this influence, rather than its impermissibility, that renders the outcome of exchanges involving deceitfully obtained consent morally invalid.

Though it is widely accepted that deceit influences consent, it is not the only factor that does so. For example, the availability of alternative options influences consent. B’s consent to a transaction that offers her unfavourable terms is ‘less perfect’ than her consent to a transaction with better terms. If the terms are unfavourable enough, she may not consent to the transaction at all. B will be less willing to purchase a good for $10 from A if she knows it is available elsewhere for $8 and she may refuse to purchase it from A at all if it is available elsewhere for, say, $3. Furthermore, this may be so even if A’s price is a price B would otherwise have been willing to pay; that is, even if A’s price is lower than B’s reservation price. Thus, the presence of other options can have an effect on B’s consent to transact with A for $10. The effect of other options on consent is most pronounced in cases of duress. If B has a pressing and immediate need for the good A is selling and no alternatives are available, she may consent to pay a price far beyond her ordinary reservation price.

Despite the fact that the availability of alternative options can influence consent, libertarians have denied that considerations such as mere need or the availability of alternatives should be taken into account in determining the validity of exchanges or the justice of holdings. An argument used against the inclusion of these considerations is that since need and alternative options do not play a normative role in either the framework’s basic principles or domain conditions, these factors do not undermine the moral validity of consent, even when they influence consent.¹⁰ The mere fact that a consideration can influence consent does not suffice to undermine the moral validity of consent. The factor must also be, in some way, morally condemned.

If this argument applies to the availability of alternative options, then it seems, it must also apply to deceit. And again, because deceit does not have a normative role in libertarianism, libertarians cannot claim that merely in virtue of its effect on consent deceit invalidates transactions. If deceit carries no moral weight in the framework, then why should it have any influence on the moral validity of exchanges? Although deceit influences consent, since it does not have a negative normative status within the framework it is difficult to see how its mere influence on consent should be treated differently from other (non-normative) factors that also influence consent.

I have argued deceit does not render transactions morally invalid because it violates libertarian conditions of moral personhood. Nor does the fact that deceit has an influencing effect on consent undermine the moral validity of deceitfully obtained consent. Consequently, neither alternative is convincing.

Therefore, libertarians cannot support (C) via an argument based on the impermissibility of deceit, nor via alternative arguments appealing to domain conditions or deceit’s effects. Libertarian principles neither prohibit, nor render morally invalid, exchanges predicated on deceitfully obtained consent. Of course, this does not mean libertarians cannot amend their framework to deliver (C). The following section considers the prospects for this strategy.

6. THE POSSIBILITY OF ADDITIONAL CONDITIONS

In his discussion of fraud, Child has argued that not only do libertarians not prohibit fraud, they cannot prohibit fraud, since any additional condition that might be added would be inconsistent with the rights to self ownership and property ownership. He argues that these rights ‘cannot be sacrificed without giving up the very nature of the system’ (1994: 736). He grants that appending a prohibition on fraud is ‘not, of itself, necessarily inconsistent’ with basic principles, but he claims that because any addition would be ad hoc, libertarians ‘have no nonarbitrary criteria’ for keeping out other potential prohibitions (1994: 736). Fraud, according to Child, cannot be meaningfully morally distinguished from many other moral problems. He asks,

What about innocent mistakes by either party resulting in a ‘bad bargain’ for one? What about transactions which transpire at exorbitant and unfair prices? What about transactions stemming from unequal bargaining power, including those entered under economic duress? Why not rescind transactions the consequences of which do not work out as planned by one party? . . . How can you ban fraud without banning these transactions?” (1994: 737)

The conjunction of prohibitions on these cases will, he argues, undermine the scope of self ownership and property ownership to a degree that is inconsistent with the spirit of libertarianism. He concludes that not only do libertarians not prohibit fraud, their basic commitments entail that they cannot do so.

Child’s argument is not entirely convincing. These cases can be morally distinguished from fraud. Although independent arguments might be made for prohibiting or invalidating some of these other cases, unlike fraud or deceitfully obtained consent, none involve deceit. However, Child’s remarks do hint at a problem for the additional conditions solution.

As I have suggested, the issue is not the narrow problem of fraud that occupied Child, but rather the broader and more fundamental problem of deceitfully obtained consent. To reiterate, libertarianism does not prohibit deceit. If there is nothing morally problematic about deceit from a libertarian perspective, then it is difficult to see how there is anything morally problematic about consent predicated on deceit. And if deceitfully obtained consent is not morally problematic, then libertarians cannot sustain (C), the claim that holdings that result from exchanges based on deceitfully obtained consent are unjust.

The problem libertarians face is not a slippery slope comprising various unattractive forms of exchange, as Child suggests. The problem is that any prohibition on exchanges involving deceitfully obtained consent (including fraud) also entails one overly demanding prohibition, namely, a prohibition on deceit. Yet, as Child himself notes, “a general duty of veracity enforceable by the state . . . is surely a libertarian nightmare, completely inconsistent with self-ownership” (1994: 737). Libertarians must find a way to prohibit acts based on deceitfully obtained consent that does not also entail a prohibition on deceit tout court.

Furthermore, any proposed solution must not be ad hoc. Prohibiting deceit simply to solve the problem of deceitfully obtained consent would be ad hoc and arbitrary. Libertarians must offer a solution that explains why other prohibitions on unpalatable outcomes could not be added as well, for absent such a justification, Child’s slippery slope argument would indeed be plausible.

Libertarians can avoid producing an ad hoc solution by tying the justification for the prohibition of deceit to the justification for enforceable prohibitions on violations of property rights and self-ownership rights. If a non-ad hoc solution to the problem is to be found, it must connect with one of two forms of justification already present in the libertarian tradition: instrumentally motivated justifications or natural rights based justifications.

7. THE INSTRUMENTAL LIBERTARIAN SOLUTION

According to instrumental libertarians working within consequentialist and contractarian traditions,¹¹ rights of self-ownership and property ownership are justified by the important and beneficial social consequences they bring. While contractarian and rule consequentialists differ with respect to their particular justifications for rights, both claim that the establishment of these rights allows us to avoid socially undesirable states and to enjoy greater benefits than we would without the rights. This appeal to valuable social consequences can also be used to justify the prohibition of certain exchanges involving deceitfully obtained consent.

George Akerlof’s economic models show that in markets with goods of varying quality and in which there is no enforcement mechanism against fraud ‘the presence of people . . . who are willing to offer inferior goods tends to drive the market out of existence’ (1970: 495). This creates practical problems for some libertarian theories. For example, Georgist libertarians hold that ‘agents may appropriate unappropriated natural resources as long as they pay for the competitive value of the rights they claim’ (Vallentyne 2000: 8).¹² Because Georgist provisos depend upon competitive market prices and fraud undermines markets, such provisos may be rendered indeterminate in the presence of widespread fraud. In addition, for many libertarians markets are also intrinsically important. As Jan Narveson writes, ‘the defense of the market is clearly prominent on the libertarian agenda. In a sense, it is the only thing on that agenda’ (Narveson 1988: 212). If forms of deceitfully obtained consent such as fraud threaten markets, and markets provide intrinsically and instrumentally valuable social benefits, then libertarians can appeal to the preservation of these benefits to justify the following solution:

The Instrumental Solution. The protection of benefits provided by social goods, such as markets, justifies the prohibition of exchanges involving deceitfully obtained consent, when the prohibition of these exchanges results in a net benefit.

The mere fact that a prohibition protects social goods does not suffice to include it alongside preexisting basic rights. The values of adding the prohibition must outweigh the cost it imposes on other values, such as personal liberty. Where this line should be drawn is a tricky question, but if deceitfully obtained consent is as threatening a problem for libertarians as I have claimed, it seems likely that at least the loss of benefits in the most egregious cases would be sufficient to justify a prohibition on deceitfully obtained consent.

Note that if the justification of a prohibition is to secure social goods, then the prohibition must be stronger than a mere moral prohibition. If A sells B a lemon and the transaction cannot be enforceable rectified (presumably by the state), then there will be no (or no strong) incentive for A to refrain from engaging in this behaviour. Although he may face some social pressure to refrain from such transactions, if a merely moral prohibition is enshrined as a norm, it is not clear that social pressure will be sufficient to curtail fraud’s averse market effects. If it is to be effective, then the instrumental solution must be committed to the stronger claim (4): exchanges based on deceitful consent should be prohibited by the state.

One disadvantage of the instrumental libertarian solution is that it is a non-starter for natural rights libertarians since they do not base the justification of basic principles on instrumental considerations. A more universal solution that applied to both natural rights and instrumental libertarians would be preferable.

A more pressing problem is that the solution does not do justice to the intuition that the acquisition of goods via an exchange involving deceitfully obtained consent is always unjust. According to the instrumental solution, whether such an exchange is unjust is contingent on whether it sufficiently threatens social goods. It is difficult to see how permitting deceitfully obtained consent in cases such as non-repeated transactions or in private acts, such as sex, can produce significant negative effects on social goods. These occurrences are either too rare, or too private to lead to negative market effects. Nevertheless, the intuition that an individual is unjustly treated when consent to sex is predicated on deceit or when a unique item for which there is no market is fraudulently acquired is no less strong, despite the diminished social effects in these cases. What is upsetting about such cases is not the effect these acts have on society, but rather the undermining impact of deceit on individual consent.

Furthermore, any attempt to salvage the instrumental solution by setting the threshold of social impact very low threatens to significantly undermine individual liberties. And, by setting a low bar for the amount of social benefit required to justify an additional prohibition, the door is opened to alternative prohibitions that go beyond the domain of deceitfully obtained consent. If low levels of social benefit are required to justify prohibiting deceitfully obtained consent, then why not prohibit other acts that also have (slightly) net negative social effects? Instrumental libertarians face an uncomfortable dilemma. On the one hand, it seems their appeal to social benefits cannot prohibit all instances of deceitfully obtained consent, since certain instances will not produce significant social harms. In this case they cannot condemn some intuitively unjust acts. Or, alternatively, if they set the threshold low enough, they can capture all cases, but in setting a low bar their justification does not apply only to these acts and thus, the prohibition of the target cases would require draconian limits on individual liberty.

The instrumental solution attempts to associate the wrong of deceitfully obtained consent with its net negative social effects rather than directly with deceit. Yet, despite the fact that many exchanges involving deceitfully obtained consent do considerable social harm, many do not. Libertarians appear to be in a bind. They cannot seem to get away from the straightforward fact that what makes deceitfully obtained consent morally problematic is deceit; yet, their framework does not recognize deceit as morally problematic. Consequently, they appear unable to condemn transactions predicated on deceitfully obtained consent, since according to their theory there is nothing wrong with deceitfully obtained consent.

8. THE EXTERNAL SOLUTION

Despite this fact, none of the libertarians I know are profligate liars. And I suspect their virtue is not only caused by a fear of the social consequences of lying. They also refrain from lying because they believe lying is morally wrong. If libertarianism does not condemn deceit, then these libertarians must form this belief by appealing to external considerations. So long as they do not take libertarianism to be a complete moral theory, there is nothing inconsistent in their doing so. Although some libertarians may view the theory as a full theory of moral permissibility, for most it functions as a theory of justice, distinct from, but closely related to a theory of morally permissible behaviour (Nozick 1974; Steiner 1994). Still others see libertarianism as (primarily) a theory about the role and scope of the state, and in particular, its right to the coercive use of force. For this group, libertarianism does not apply to non-enforceable duties, such as duties of beneficence or, perhaps, even certain duties of justice (Vallentyne and van der Vossen 2014). The latter two groups of libertarians who do not view the theory as a full theory of moral permissibility may adopt the following solution:

The External Solution. Consent is morally invalid when it is obtained through acts that are impermissible according to a full theory of moral permissibility.

The external solution allows that certain forms of ostensive consent are morally invalid when they involve the violation of libertarian principles, since these principles form part of a full theory of moral permissibility. For example, consent gained through the torture of an innocent person is morally invalid because it involves the violation of self ownership rights. But importantly, the solution also implies that consent gained via deceit is morally invalid since deceit, despite not being prohibited by libertarian principles, is (pro tanto) prohibited by any plausible full theory of moral permissibility. In this way, libertarians can prohibit acts based on deceitfully obtained consent without claiming that deceit itself is unjust or subject to state-enforced prohibitions. In what follows, I will refer to that part of a full theory of moral permissibility that falls outside the scope of libertarian principles as general morality; a full theory of moral permissibility is the conjunction of libertarian principles and general morality.

Consider again the fraudulent car sale case from the first section. By appealing to the external solution, libertarians can consistently maintain A’s lie about the condition of the car is not a violation of libertarian principles while nevertheless claiming that consent gained through deceit is morally invalid because it is impermissible according to general morality. And, because deceitfully obtained consent is morally invalid, then – assuming the truth of premise (3) – libertarians can endorse claim (C): holdings that result from exchanges based on deceitfully obtained consent are morally unjust.

Above I claimed that the problem for libertarians wishing to add an additional condition to prohibit such cases is not, as Child claimed, the presence of a slippery slope comprising many small prohibitions that eventually undermines libertarian commitments to liberty. Rather, it is that a single prohibition on deceit would have this effect. However, the external solution does not employ a libertarian prohibition (and certainly not a state enforceable prohibition) on deceit. It claims that exchanges that satisfy two conditions are morally invalid: first that they involve acts prohibited by general morality, and second that these acts have an effect on consent. Both conditions are necessary since the absence of either would lead to intolerable restrictions on liberty.

In addition, the solution avoids the charge that it is ad hoc since libertarians already place a great deal of emphasis on consent. Finally, the solution is one to which both instrumental and natural rights libertarians can appeal. Thus, by adopting the external solution, libertarians can sustain claim (C), that exchanges involving deceitfully obtained consent are unjust, and they can do so without appending ad hoc conditions or placing intolerable limits on individual liberty.

9. THE NOVELTY OBJECTION

There are a few objections to the external solution. The first is that it seems an obvious solution. After all, apart from the appeal to external prohibitions in the domain of general morality, the argument leading to (C) is identical to the intuitive argument supplied in Section 2. However, as I shall show, the solution has four rather novel and unexpected implications for libertarianism.

First, in conversation, many libertarians have agreed that the theory does not prohibit fraud or other forms of deceitfully obtained consent, but they have insisted that it is clear that libertarians claim exchanges involving deceitfully obtained consent are morally invalid. While I share this conclusion, the justification of this claim in the form of the external solution is not one that can be reached using the resources of libertarianism alone. As I argued, libertarianism lacks the resources to condemn deceit, and thus, apart from a general intuition that deceit is wrong, it remains mysterious how deceitfully obtained consent can therefore be morally invalid. I suspect that many libertarians’ endorsement of the moral invalidity approach implicitly relies on the intuition that deceit is wrong. However, the external solution makes explicit that a solution to the problem must draw a distinction between libertarian prohibitions and general prohibitions and must locate the impermissibility of deceit in the latter domain. Furthermore, libertarians who wish to adopt the external solution to the problem cannot simultaneously maintain that libertarianism provides a full theory of moral permissibility. Consequently, the external solution has implications for the scope of the libertarian theory of moral permissibility.

Second, the problem of deceitfully obtained consent also challenges the common libertarian claim that among competent agents, all rights respecting transactions are justice preserving.¹³ Yet, if libertarians think it is true as claim (C) asserts, that holdings resulting from consensual exchanges of moral rights that are based on deceitful consent are unjust, then, since deceit does not involve the violation of any libertarian rights, it follows that it is not the case that all rights respecting transactions are justice preserving. Libertarians who wish to retain the claim that all rights respecting transactions are justice preserving must either abandon (C) or show libertarian principles do prohibit deceit. And, since I have argued libertarianism does not prohibit deceit, libertarians must choose between the justice preservation claim and claim (C). Because the external solution entails an acceptance of (C), it also implies that the justice preservation claim is false: not all rights respecting transactions are justice preserving, an implication that has significant ramifications for some libertarian theories.

Third, the solution blurs the Kantian distinction between an intensional theory of permissible action based motives, maxims, or intentions and an extensional theory of justice. An implication of the external solution is that the transfer of A’s car to B is a just transfer only if A does not deceive B. Because the juridical status of the transfer depends on whether A deceives B, and whether A deceives B or makes an honest mistake depends on A’s intentions, the juridical status of the transfer depends on an intensional description of A’s action. I say ‘blurs’ rather than ‘destroys’, since criteria of right action may still fully rely on intensional descriptions. The external solution impacts only the claim that an act’s justice depends only on its having a certain extensional description.

Finally, note that the external solution can be applied to wrongs other than deceit that affect consent. Perhaps the most controversial case involves what might be called duressed consent. As I noted above, the availability of alternatives can alter consent. According to many moral theories, persons are obliged to provide aid to those who are in dire need and without sufficiently attractive alternatives, especially when the cost of doing so is low. Libertarians – and their critics – have agreed that libertarianism does not condemn agreements made under duress as morally invalid. Yet, even if duress does not entail any obligations according to libertarian principles, if it does trigger obligations under a complete theory of moral permissibility, and duress has an effect on the provision of consent, then, it appears the external solution can be employed to argue that exchanges involving duressed consent are also not morally transformative. This result would contradict many libertarian claims.

While it is hard to deny that duress can impact consent, libertarians can resist this conclusion by denying the claim that there exist obligations to aid in general morality. Note, however, that if libertarians accept the external solution, their strategy for rejecting obligations of aid under duress changes. They cannot merely claim that libertarian principles do not entail such obligations, they must also show that a full theory of permissible action does not entail such obligations. Of course, this is no problem for libertarians who think libertarianism already provides a full theory. But some libertarians maintain that since libertarianism does not recognize such obligations they cannot be enforceable obligations related to justice, even though withholding aid may be morally wrong according to a full theory of moral permissibility. The external solution undermines this line of argument by claiming that when they influence consent, general moral obligations external to libertarianism can undermine the moral validity of exchanges (and the justice of holding resulting from such exchanges).

There is an undeniable sense in which the external solution is an obvious solution, but this fact follows from its close connection to ordinary intuitions about the wrong of fraud. It does not, as I have argued, entail only minimal modifications to libertarianism. It challenges the completeness of libertarianism as a moral theory by drawing a distinction between libertarian principles and external principles, it shows that the common libertarian claim that all rights respecting transactions are justice preserving is false, it blurs one way of distinguishing between morality and justice, and it opens the door to a much broader form of libertarianism in which the availability of alternative options plays a role.

10. THE LIMITED LIBERTY OBJECTION

Despite the novelty of the solution, this last implication gives rise to a second concern: does the external solution entail limits on liberty that are at odds with the spirit of libertarianism?

First, it is important to distinguish what the external solution claims from what it allows. The solution claims that if an act is impermissible under a full theory of permissibility and that act also effects consent, then the act renders that consent imperfect. Any exchange resulting from this imperfect consent is morally invalid and holdings resulting from such exchanges are unjust. The solution makes no claim about the content of a full theory of moral permissibility. Nor does it make any claim about the role of either the state or other individuals in rectifying unjust holdings. It therefore does not entail the stronger claim (4), that exchanges based on deceitfully obtained consent should be prohibited by the state.

Only if one accepts the external solution and one thinks a full theory of moral permissibility entails obligations to help those in duress does the external solution imply the invalidity of duressed consent within libertarianism. And only if one thinks that all unjust holdings are subject to state enforced rectification and that any exchanges that would lead to unjust holdings are subject to state enforced prohibitions does it follow that the state has a role in enforceably prohibiting exchanges based on deceitfully obtained consent. The restriction of individual liberties and expansion of state powers thus depend respectively on one’s theories of permissibility and the use of force. So, while the external solution is consistent with an expanded form of libertarianism, it does not entail these expansions.

As an illustration, it is perhaps helpful to consider a pedestrian case of deceitfully obtained consent involving a surprise birthday party. Suppose B’s friend A knows B does not like surprises, but because A wants to throw her a surprise party, A lies and invite her to what he describes as a ‘quiet dinner’. Sadly, the surprise ruins B’s birthday. If she had known the truth, then she would have stayed home. The case meets the conditions of the external solution. B’s consent is gained through deceit (which, let us assume, is impermissible according to external principles of moral permissibility). The consent A secures from B is, consequently, morally invalid.

Does this result make sense – has B a moral complaint against her friend? Note that the case meets the definition of fraud given in the introduction, and since existing libertarian theories explicitly (but incorrectly) claim fraud is prohibited, libertarians should be willing to accept the moral invalidity result. Those who might be troubled by the birthday party case are, I suspect, influenced by one of two mistaken beliefs.

The first mistaken belief involves a conflation of claim (C), that consent predicated on deceit unjust with claim (4), that exchanges based on deceitfully obtained consent should be prohibited by the state. The external solution only entails that (a) B has a moral complaint against her friend. She may, depending on one’s commitments, also have (b) a legal (tortious) claim against her friend. More strongly, one might think (c) her friend has also committed a crime that – stronger yet – (d) could be enforceably prevented by the state. The external solution is committed only to (a) but consistent with (a) through (d).

The second mistaken belief concerns an appeal to objective harms. Since it is ‘merely’ a surprise birthday party, B should ‘let it go’, since, after all, she is not ‘really’ harmed. Outside a libertarian framework, these considerations may indeed play a role in morally evaluating the case. However, an important feature of libertarianism is that it is agnostic about the content of individual preferences. It is not committed to any objective values (Steiner 1984: 225). For example, non-consensual sex is rape and non-consensual transfers are theft regardless of the reasons the victims withhold consent. This reasoning also extends to B’s birthday party. If A secures her attendance through deceit, B has a moral claim against A. It does not matter that we happen to enjoy birthday parties and think B should too, or that there are not good reasons for thinking B has been objectively harmed. If the deception concerns facts material to B’s attendance, then these facts are sufficiently important to B. A’s behaviour cannot be excused by appealing to objective harms.

I hope to have shown that, although the external solution opens a door to a very different kind of libertarianism, it does not mandate a radical departure from existing theories. Nor does it entail intolerable restrictions on the domain of personal liberty or a broad expansion of enforceable state powers.

11. CONCLUSION

I have argued, following Child, that libertarianism does not prohibit exchanges involving deceitfully obtained consent. Nor does it entail that these exchanges are morally invalid. However, contra Child, I have claimed that it can deliver this second verdict by appealing to an external theory of moral permissibility. Consequently, libertarians can sustain the intuitive claim (C): holdings resulting from exchanges based on deceitfully obtained consent are unjust. This external solution not only avoids contingency problems that plague a possible instrumental libertarian solution, it is also, in principle, open to all libertarians who do not think libertarianism delivers a full theory of moral permissibility.

In the final portion of the article I considered objections to, and implications of, the external solution. I noted that while it is consistent with a state prohibition on exchanges involving deceitfully obtained consent, it does not entail such a prohibition. Nor does it entail a broader form of libertarianism in which duress plays a greater role. But it does allow for such a theory, which rather than representing a hurdle to their theories, provides libertarians with a principled form of libertarianism that avoids many of the theory’s harshest objections.