This is a slightly ill-organised essay which I submitted for the final element of a part-time MA in philosophy. It's a bit dense and probably none too clear, and was returned ungraded for rewriting. I did not really know what to do with it, and got no useful guidance, so it never got rewritten.
The Earnings Thesis: moderating libertarian rights
Timothy David Wilkinson
A dissertation submitted to the University of Bristol in accordance with the requirements of the degree of Master of Arts in the Faculty of Arts, Department of Philosophy in November 2003
13,260 words
In this essay I argue that Nozickean entitlement theory does not incorporate a right to surplus value. Part of a constructive interpretation of Nozick’s theory is presented, which retains its basic commitments and discards what cannot be reconciled with them. Section I contains an analysis of some salient features of Nozick’s theory of justice. It is concluded that Nozick’s position should be seen as resting on the notion of compossible rights. Section II presents a position which satisfies the desiderata identified in section I, but according to which rules of entitlement are less expansive than in Nozick’s theory – specifically, according to which neither party to an exchange thereby gains title to more than the minimal reward necessary to induce him or her to enter into such an exchange. Moreover, this minimally-motivating reward is to be calculated in isolation from the possibility of other exchanges. From this it is concluded that Nozickean property-rules confer no automatic entitlement to any surplus - measured against the standard of foregone use-value. Some considerations from Locke and Nozick are adduced to support the thesis. Section III is an examination and attempted rebuttal of some Nozickean objections to the earnings thesis – based on arguments from self-ownership and rights of transfer. In section IV, an analysis of the notion of unproductivity presented in Anarchy, State and Utopia is offered. Some ambiguities and indeterminacies are resolved. Section V gives some reason for concluding that unproductivity should be regarded as central to Nozickean theory. In particular, a much-needed reconstruction of Nozick’s justification of the minimal state is presented, making essential reference to unproductivity. Section VI returns to constructive analysis, introducing and motivating some subsidiary principles. From these premises, it is argued that a consistent Nozickian conception of unproductivity implies the Earnings Thesis.
I declare that the work in this dissertation was carried out in accordance with the Regulations of the University of Bristol. The work is original except where indicated by special reference in the text and no part of the dissertation has been submitted for any other degree.
Any views expressed in the dissertation are those of the author and in no way represent those of the University of Bristol.
The dissertation has not been presented to any other University for examination either in the United Kingdom or overseas.
SIGNED:
Contents
I. Nozick’s theory
Nozick presents a number of interrelated arguments against rights to ‘assistance’ – where this means either rules of entitlement which would upset life plans in order to redistribute property, or unchosen duties to perform particular actions. I propose to focus on an argument which appears to generate many of Nozick’s desired consequences by appeal to a formal feature of rights. The approach is introduced as a response to Sen’s results with respect to the so-called ‘impossibility’ of a Paretian Liberal[1]. I will present a paraphrase of Sen’s proof of the incompatibility of a Pareto criterion and a Liberal rights criterion.
Sen conceives a Liberal (or libertarian) right to consist in decisiveness between two possible states of society. That is, A has a right iff there is some pair (x and y) of alternative states of society between which A’s preference is decisive: if A prefers x to y then x is to be preferred to y by society as a whole. The Pareto principle holds that where alternative x is preferred unanimously to alternative y, x is to be preferred to y by society as a whole. Sen then introduces two further principles whose conjunction with the Pareto principle is, he claims, impossible to satisfy:
Minimal Liberalism: at least two persons exist and each has at least one Liberal right (in the sense introduced above).
Unrestricted domain: the social decision procedure must work for any logically possible set of individual preferences.
Sen then demonstrates that this conjunction can give rise to inconsistent results. We suppose a society of two persons, whose preferences between the four alternatives w, x, y and z are:
Person A: w > x > y > z
Person B: y > z > w > x
We also assume that A has a liberal right (decisive preference) as between x and y, while B has a liberal right as between w and z.
The Pareto principle operating on these preference orderings give two partial orderings: w > x, and y > z. Both these are unanimous preferences. However, adding A’s right to this partial ordering means that since the preference x > y is to be decisive, the social ordering of alternatives must be w > x > y > z, whereas applying B’s decisive preference (z > w) instead gives us y > z > w > x. These are incompatible, and the impossibility result is arrived at.
It should be noted that ‘impossibility’ here does not mean that it is impossible for the Pareto principle and Liberal rights to be simultaneously satisfied by a given social decision. The condition of unlimited domain requires that the array of social orderings we arrive at must cover every possible configuration of preferences, and given that for any pair of Liberal rights, there will be some set of preferences which has the structure given above (i.e. each person has identical preference-orderings between the members of each of two pairs, but their decisive preferences place the two pairs in opposing orders), this mean that any attempt to produce a comprehensive array of social orderings for all possible preferences will fail.
The notion we are dealing with then, is not so much one of impossibility (it is not impossible to reconcile Liberal rights with the Pareto Principle), but one of possible impossibility: it is possible to devise (or locate among possible worlds) a situation in which it becomes (nomologically) impossible, given the prevailing preference-rankings in that situation, to satisfy both the Pareto principle and the requirements of Liberal rights.
Nozick responds to this finding:
The trouble stems from treating an individual’s right to choose among alternatives as the right to determine the relative ordering of these alternatives within a social ordering…A more appropriate view of individual rights is as follows. Individual rights are co-possible; each person may exercise his rights as he chooses. The exercise of these rights fixes some features of the world. Within the constraints of these fixed features, a choice may be made by a social mechanism based upon a social ordering; if there are any choices left to make!…If entitlements to holdings are rights to dispose of them, then social choice must take place within the constraints of how people choose to exercise these rights. If any patterning is legitimate, it falls within the domain of social choice, and hence is constrained by people’s rights. How else can one cope with Sen’s result?[2]
So the requirement of compossibility[3] provides for Nozick a reason to reject any principle which could conflict with the outcome of a procedural, rights-based system of political morality. The notion of compossibility is, as we have noted in connection with Sen’s result, expressible informally as a requirement that there be no (nomologically) possible situation in which rights could conflict. However, we have seen that the Nozickian rejects the conception of individual rights as decisive preferences between alternatives, feeding into a social decision function (subsequent work by Gibbard[4] has shown that even without the Pareto principle such a conception of rights still gives rise to impossibility results). It may therefore no longer be evident what exactly is meant by conflict of rights. There is no longer any social decision function which might be rendered unstable by some possible set of preferences.
To explicate the concept of compossibility, then, we introduce the intuitive notion of rights-satisfaction. A right is satisfied when it does its job: if it is a right to non–interference of a certain kind, it is satisfied if no interference of that kind occurs. If it is a right to prompt and expert medical treatment free at the point of delivery, it is satisfied if such treatment is provided. Not too much rests on a precise definition of this concept: it will be discarded once it has done its job of elucidating the nature of compossibility.
We have already noted that compossibility is a stronger requirement than the mere possibility that all rights are satisfied: it requires that in all possible situations, it is at least nomologically possible that all rights be satisfied. This immediately forestalls the possibility of rights which can go unsatisfied merely as a result of non-human circumstances: I cannot have a basic right to have the use of three goats, since there might not exist three goats. Equally, I cannot have the basic right not to be struck by lightning.
Within the realm of human action, then, what is required for rights to be satisfied? Simply that no person violates a right. This requires that there must be no intrapersonal conflict between the duties imposed by rights – no ‘jural dilemmas’. It also requires that in every possible situation every person must be able to carry out their duty simultaneously – there must be no incompatibility between the ground-level requirements (going to x place, picking up object y) which rights impose on different people.
More subtly, given Nozickean premises regarding compensation, it requires that where rights have previously been violated, it is possible to identify the rights-violator, so that the claimant’s right to compensation can be satisfied (though this may have more to do with principles of entitlement than rights proper – see below).
Nozick achieves compossibility by stipulating that all rights are negative rights to protection from interference with property (including property in one’s person). This means that it is always possible (unless one is being overpowered, etc: in which case the person who is using you as a tool will be the right-violator) to perform one’s duty – because it is always sufficient for doing one’s duty, that one simply do nothing.
Nozick’s main principle of rights then is that everyone has the right to be free of physical impingement on their property. Bearing in mind the requirement of compossibility, notice that this right does not protect any specific interest, except trivially the interest in one’s property being free from uncompensated and unpunished impingements by other persons – an interest which hardly seems to correspond with any intuitive notion of a basic interest prior to political organisation. Some principles are required to specify the extent of entitlement. These principles – those of justice in acquisition, transfer and rectification, along with the Lockean Proviso and possibly some others – should be subject also to a compossibility requirement – that it must be impossible that any two persons should ever simultaneously have title to a single object, and that any person should gain entitlement to something which does not exist, or is not available to be appropriated. Although I take it that these compossibility conditions are almost irresistibly plausible (to a Nozickian), it may be noticed that such an overlap would not entail the non-satisfaction of any rights: since the only rights arising from title to an object are Hohfeldian claims to non-interference, it is quite consistent with these rights being satisfied that two people should have title to the same object – they are in that case both excluded from the object – or that a person should be entitled to something which does not exist – and therefore their negative rights would be impossible to violate. However I shall assume that some Nozickian reason for this compossibility claim could be formulated, although I shall not do so here for reasons of space.
II. The earnings thesis
The Earnings Thesis states that neither party to an exchange thereby gains title to more than the minimal reward which would be necessary to induce him or her to enter into such an exchange, if no other exchange were possible.
This thesis satisfies the basic requirement of an unpatterned historical theory[5]: it does not involve any ‘emergent’ rights or entitlements, does not advert to considerations other than rights and entitlements, and is a principle of commutative justice, which judges each transaction making reference only to the intrinsic nature of that transaction (including the motivations and preferences of the parties).
x has a horse that pleases him and is for his turn ; this y would buy of him; x tells him he has no mind to sell; y presses him to set him a price, and thereupon x demands and takes £40 for his horse, which in a market or fair would not yield above twenty. But supposing y refusing to give £40, z comes the next day and desires to buy this horse, having such a necessity to have it that if he should fail of it, it would make him lose a business of much greater consequence, and this necessity x knows. If in this case he make z pay £50 for the horse, which he would have sold to y for £40, he oppresses him and is guilty of extortion whereby he robs him of £10, because he does not sell the horse to him, as he would to another, at his own market rate, which was £40, but makes use of z’s necessity to extort £10 from him above what in his own account was the just value, the one man’s money being as good as the other’s. But yet he had done no injury to y in taking his £40 for a horse which in the next market or fair would not have yielded above £20 because he sold it at the market rate of the place where the horse was sold, viz. his own house, where he would not have sold it to any other at a cheaper rate than he did to y.
This passage states a principle that (under certain conditions) one may not charge a person more than one would charge any other, from which we might derive the Earnings Thesis. Locke states that it is fair to charge the market price, even when one would accept less; but this is on the grounds: 1. that others may otherwise make up the excess profit by reselling (which would in turn contravene the Earnings Thesis) and 2. that any attempt to fix a maximum level of profit might deter trade altogether (which ex hypothesi cannot happen under the provisions of the Earnings Thesis).
The assertion that the ‘market rate’ is different in one’s home seems an odd way of putting things: the locale of the exchange seems irrelevant. It is introduced during discussion of price differentials between places, which are perhaps justified by transport and coordination costs (including effort, etc.)[7]
In his discussion of compensation and ‘boundary-crossings’ (p.58: Nozick uses the phrase ‘boundary-crossings’ when he wants to discuss the possibility that they may be allowed, ‘rights-violations’ in other cases.) , Nozick raises the question ‘Why ever prohibit?’. This question seems an odd one for Nozick[8]. There is an obvious reason to prohibit ‘boundary-crossings’, which is that they alter holdings in an unpredictable way .
I would suggest that this is the best way to read Nozick’s commitment to prohibition (punishability) of rights-violations. Although Nozick appeals to ‘apprehension’ felt by those who know that a deterrent punishment is not in place (ASU, p. 66), this seems misguided. It does not cover unfeared rights-violations; it would be undesirable, given Nozick’s association of personal projects with settled property-rights, for Nozick to regard unfeared theft as a ‘civil’ rather than ‘criminal’ matter - though he does suggest that inherent fearability might demarcate the boundary between public and private wrongs. More importantly, it does seem prima facie to cover fear-inducing non-rights-violations, including those due to mistake, superstition or nervous dispositions. in general. Nozick’s rights are intended to respect the fact that each person has their own life to lead, and Nozick emphasises the need to protect each person’s ability to make and pursue a ‘life-plan’.
But Nozick does not answer the question in this way. He addresses the issue of the ‘division of benefits of exchange’ and the effects of allowing unilateral expropriation with (merely) full compensation. Full compensation is the amount of compensation which is just sufficient to return the person compensated to the indifference curve they were on before the rights-infringement. This may for our purposes be considered equivalent to the minimally motivating price[9]. Nozick states that it is unfair to allow exchange to occur at one extremity of the contract curve[10]. The contract curve represents the feasible Pareto-efficient pairs of payoffs to the two parties to an exchange. This approach assumes that all possible benefits must be allotted to the parties: that is, the pair of payoffs must correspond to a point on the contract curve. If each party’s payoff corresponds to that represented by her least favoured point on the contract curve (as under the Earnings Thesis) this unfairness (asymmetry) between them is avoided. If Nozick wants to reject this alternative, it cannot be on the basis of unfairness as asymmetry between the parties.
Perhaps another kind of fairness might be appealed to: fairness to a single party without reference to an other persons. This seems better phrased in terms of a conception of desert: certain rewards (entitlements) accrue in virtue of certain acts or properties of the entitled person. The word ‘desert’ is perhaps misleading here. I propose a conception of earning: this captures the desert-like aspect of entitlements without bringing in other considerations of desert, which stem from undeserved entitlements or lack of deserved entitlements.
On the Nozickean view, the ‘desert’ that is expressed in entitlement is not defeated by the undeservedness of the capacity to work hard, etc. Nozick asserts that the bases of desert ‘need not themselves be deserved all the way down’. Thus objections from undeserved entitlement are ruled out.
The lack of deserved entitlement (in the sense of unrewarded sacrifice) might pose greater problems. But as Goodin[11] points out, it is plausible to see desert as negative: it’s not so much that everyone has a claim to what they deserve as that no one has a claim to what they don’t deserve. Nozick rejects positive claims (say, of need) as a basis for distribution partly on the grounds that ‘things come into the world already attached to people with entitlements over them’. Since ‘particular rights over things rights fill the space of rights’[12], no rights exist in conflict with them. This argument of Nozick’s is insufficient to show why we call these claims ‘positive’ and why they rather than his entitlements should be rejected. Claim-rights need not conflict with existing ownership rights to be unsatisfiable. ‘Filling the space of rights’ is not the same as filling space: not everything need be owned (hence initial acquisition). But there still might not be enough to satisfy all positive claim-rights. The relevant difference between positive claims and Nozick’s entitlements is that the latter and not the former, are compossible in the sense alluded to in section I.
Nozick’s argument against these claims need not depend on the question-begging assertion of conflict with existing rights, though. In section I, the notion of compossibility was introduced for just this purpose. The view of how compossibility is achieved we may call ‘the primacy of production’.
In section I it was suggested as a desideratum of any rights-theory that an entitlement or claim cannot arise unless the object concerned is available to be claimed. So there is a one-to- (no more than) one correlation between claim-begetting acts and rights of ownership. This can be achieved by conditionalising claim-rights along the lines of ‘where available’, and providing adjudication principles to decide on whom the unconditional right to a scarce resource should fall. This approach ‘renders property rights conditional on need, and gives them entirely the wrong type of character for Nozick.’[13] Alternatively compossibility can be achieved through a direct linkage between acts of production and the gaining of rights of ownership by some specifiable person. The latter is Nozick’s idea, and he makes the salient choice of the producer as owner.
The negative conception of desert ensures that rights are still self-guaranteeing, because the notion of earning limits entitlement; it cannot extend it. I do not, merely by doing some ‘deserving’ thing, earn a reward. I must make something or find a partner in exchange. This parallels Marx’s use of ‘socially necessary labour-time’ as the measure of value, which takes into account both the subjective preferences which determine value, and the objective basis of the value in labour.
Just by spending time labouring, I don’t create value. To do so, it is necessary both to expend effort, and to do so in an efficient (productive) way. This gives a clue as to how to go about characterising ‘earning’: I shall take working for something as the basic type of earning.
In considering the part labour plays in earning, we should note that the dual necessary conditions for earning outlined above are reflected in Locke’s conception of labour. On the one hand, it is viewed as making, or value-creation. On the other, work is viewed the giving up of time, comfort, energy; the taking of pains. Locke frequently uses terms cognate with ‘painstaking’ in chapter 5 of the Second Treatise[14], notably at §37:
he that so employed his pains about any of the spontaneous products of Nature as any way to alter them from the state Nature put them in, by placing any of his labour on them, did thereby acquire a propriety in them
On some persuasive interpretations[15], Locke states that labour is required for acquisition of property (where labouring is an option - I leave aside Locke’s variously interpreted right to ‘charity’[16]). I shall characterise earning as requiring a sacrifice, as well as presupposing the availability of the reward (achieved by the ‘making’ aspect of labour).
Labour is the salient choice for the most basic sort of sacrifice, and hence of earning. Even if there were such a thing as effortless labour, my time is sacrificed. And time, or life-time, is characterised (in this world and epoch) by equal ‘income’ distribution. This might provide an egalitarian basis for a theory of the type I am gesturing towards. This can explain the central place of labour in an agent-centred political morality. The theory of entitlement based on the notion of earning states that persons are entitled to the product of their labour, on the basis that they have earned it, and that they are subsequently entitled to the minimal motivating reward for making the sacrifice of this property in exchange (iff they can negotiate some such exchange). Compare Sreenivasan[17], interpreting Locke:
for able bodied individuals...the only legitimate title is labour by that individual...Thus, the possibility of an able-bodied potential owner’s having a right of acquisition by transfer must turn somehow on her labouring...It is possible, perhaps even plausible, to construe the labour that earned the money as being in some sense equivalent to the labour that made the cheesecake and as thereby entitling the present owner of the cheesecake to acquire the money, despite the fact that she didn’t labour on it.
My non-Locke-interpreting account expands on this by stating that both labour and exchange are characterised as sacrifices. And it is not equivalence of (actual or necessary) labour that allows acquisition, but the sacrifice of giving up property which was gained through the sacrifice represented by labour. On such an account, all entitlement (beyond entitlement to one’s person) flows from labour.
Two immediate problems arise for this view. First, the well-established problem of the resources used in production through labour. This is a separate issue. Of more immediate concern is the question of naturally occurring producer surplus. A producer surplus occurs where the price one can command for a resource is less than the minimum price one would accept. To take an example of Cohen’s[18], imagine that I shake trees and collect the coconuts. Even if the return in coconuts per hour’s shaking were halved, I would continue to engage in my shaking activities. Given the earnings thesis, if I would be willing to sacrifice the same time and toil to produce less, shouldn’t my earned entitlement extend only to that lesser amount?
To answer this question, I shall sketch a distinction between exchange and labour. A trade, at least if we assume the existence of money, is indefinitely divisible. Rather than exchanging two apples for two oranges (and ‘losing’ one fruit each), two parties might have their accountants record the transaction as two exchanges of one apple for one orange, for the purposes of tax-avoidance. But this does not, given the divisibility of exchange, get round the provisions of the Earnings Thesis. For if we alter our originally envisaged exchange so that I exchange your original minimally motivating payment for mine, each becomes more-than minimally motivating, since each of us has given up less than the originally envisaged amount. Money enables these measurements to be made, and where entitlement in goods is exceeded, it may (if entitlement to the surplus is not gained under some other provision) be forfeited in money. I can be regarded as having given up two apples for your one. The giving up of apples has no essential connection with the gaining of oranges. (In fact, I do not envisage the Earnings thesis affecting these kinds of outcome: as I suggest later, the Thesis leaves leeway for other principles to apportion any surplus. One such principle might state that barter (on a small scale) should be left alone.)
Work in the state of nature has an internal aim, and is not arbitrarily divisible. Consider the limiting case of working for survival. I would if necessary work full-time and full-pelt to produce a subsistence. If I need not (say I pick fruit, as above), doesn’t my earned entitlement extend only to a lesser amount? I think not. For the prospect of starving is not an alternative for me. I would not accept less than subsistence for any work (Here the strain is showing on the rational conception of humans: I probably would work for less than subsistence, even if I knew this would only slow down my awful death. But I shall ignore this problem, in the fine tradition of Nozickian argument, adding only that one certainly wouldn’t adopt a life-plan involving starvation.). The only effect of this restriction of entitlement (assuming that the unearned resources might go elsewhere) would be to force me to do more, and possibly useless, work. The example of subsistence extends to other aims. If I set out to build a hut, I will not be satisfied with half a hut. The minimal motivation for doing the making-of-a-hut is the ending-up-with-a-hut. To see this we need only consider the individual procedures I go through to make the hut. At each stage, I would not voluntarily do that procedure (putting that log there) except to achieve the internal aim of the procedure. The effort I (usefully) expend is at every stage and every level of detail directly correlated with the result I want.
This approach also highlights the plight of alienated labour. The essence of such labour is that it does not have an internal aim, merely an instrumental one. It is an exchange of time and effort spent under the direction of another for some (usually financial) reward. This asymmetry between ‘free’ labour in the state of nature and wage-labour under capitalism raises further questions, which may even be susceptible of an equitable (by my lights) answer within my extended Nozickian theory. The reduced return to labour due to the unavailability of autonomous labour might constitute a worsening of the proletarian’s situation; but I cannot here pursue this possibility.
Assuming then that earning by labour does occur by a juridically basic process (on individualist, Crusoe premises, labour is clearly more basic in some sense than exchange), and that the ‘earning’ model is established as the basis of entitlement, reward gained in exchange must be seen as earned. How is it earned unless by giving something up? And how would more be earned than is given up? I can see no reason from this standpoint why that should be so.
So the Earnings thesis is coherent and (I suggest) plausible, and would not demand that all of the benefits of exchange be shared between traders.
III. Objections to the Earnings Thesis
Perhaps the main (principled) objection to this view is based on the idea of maximal self-ownership (See, e.g., Cohen (1995) pp. 213 ff. Cohen regards self-ownership as more or less equivalent to Gauthier’s ‘right to one’s natural endowment’.) The idea is that each person is entitled to herself, her talents, and whatever flows from them, to the full extent compatible with other’s like entitlement. Cohen[19] forcefully suggests that this is coherent, contra Waldron[20] who want talents to be regarded as akin to functions taking social structures as arguments. The point, I think, may be forcefully put thus: where two parties ‘cooperate’ in exchange, why should each only receive their minimal cooperative payoff? Whose self-ownership interests are damaged by, say, allotting title along the lines of a bargaining solution like Nash’s[21] or Gauthier’s [22]?
This suggests that the Earnings Thesis has lost the original point of unproductivity theory: protecting the interests of a victim of exploitation, which reflects the right-based nature of Nozick's theory. But as Waldron[23] points out, Nozick’s theory is better seen as duty-based: side-constraints don’t seem based only on the interests of right-holders, but on agent-centred considerations. One example of this is the permission to override the interests of ‘innocent threats and shields of threats’[24] on the basis that I am not constrained in certain circumstances to respect (what would otherwise be) the rights of others. Similarly, Nozick rejects natural procedural rights in favour of an (agent-centred) ‘epistemic principle of border-crossing’[25]. ‘Does a person who did violate another’s rights’, he asks, ‘have a right that this fact be determined by a fair and reliable procedure?’ After introducing the EPBC, Nozick (apparently amnesically) remarks, ‘I am not sure that this is the proper focus. Perhaps the person...does have such procedural rights.’ The objection reasserts itself: ‘But what is a guilty person’s complaint against an unreliable procedure? That it is too likely to mispunish him?’[26] .
It is far from clear that all considerations of transactional justice must be concerned solely with furthering the interests (I shall include the ‘interest’ represented by having a choice) of those party to the transaction. Indeed Nozick’s ‘Lockean Proviso’[27], on any interpretation, requires that the interests of others have some effect on the amount of property that can be successfully acquired or transferred. Since the situation of others may not be worsened (in comparison to some baseline) by an appropriation or transfer, some ‘leakage’ from the entitlements-machine is clearly mandated. The Earnings Thesis is in this respect less un-Nozickian than Nozick, since unlike the Lockean Proviso, it does not require that the situation of third parties be an input into the entitlement-determining process. This mean that unlike the Proviso, it does not open the Nozickian theory to the charge of having regard to the pattern of entitlements (where might that lead to?)
To consider self-ownership more closely, I think it should be remarked that Nozick’s (imputed) position is that of upholding maximal self-ownership, not maximal ownership. Not everything need be owned: there must be some linkage between the ‘self’ and the ‘ownership’. The earning thesis supplies this linkage, and seeks to show that the full benefits of exchange do not in fact ‘flow from’ or ‘arise out of’ my (admittedly owned) talents, abilities or effort. So the mere appeal to self-ownership does not rule out the Earnings Thesis.
An alternative way of construing the relevance of self-ownership to this question sees the rights (entitlements) it confers from a choice- rather than benefit- orientated view: ownership comprehends the right (more precisely, the power[28]) of transfer, so on this view, the entitlement of one party to his share of the total surplus arises from the rights of the other party to the transaction.
To achieve this result, the power of transfer must be distinct from the power to alienate property, or to allow others to use it: it must involve the power of vesting full property-rights in another person.
Property transfer has not been regarded as unproblematic. Gaus (1994) points out that no combination of waivers and permissions of rights is sufficient to replicate the effects of transfer, and that a power to alter the status of others (including third parties) is required. Waldron (1994) points out that if the reason for respecting entitlements acquired through transfer is that one thereby respects the wishes of the original earner (maybe long-dead) then the respect for these derived entitlements must eventually dwindle.
We might compare Nozick’s mention of ‘rights with hooks’[29]:
We should note an interesting feature of the structure of rights to engage in relationships with others, including voluntary exchanges…Rights to engage in relationships or transactions have hooks in them, which must attach to the corresponding hook of another’s right that comes out to meet theirs.
To acquire something through exchange, one must have the power to gain full ownership rights over it; if this power is absent, the intention of the other party is not sufficient to transfer ownership to me. And this is especially relevant when the other party doesn’t (actively) intend that I receive anything. The bargaining process which Nozick believes ought to be allowed to take its course, subject to his principles of justice) is emphatically not about persons volunteering property to each other. It is about ceding the minimum of ones own property required to extract goods or services from another.
This has some bearing on the Wilt Chamberlain argument,[30] in which it is claimed that since each spectator of Wilt’s basketball-playing voluntarily gives up a hypothecated 25¢ toward Wilt’s share of the box-office takings, Wilt’s enrichment could only be countered by measures which obstruct the will of property-owners to use their property as they choose. (Wilt is a plausible candidate for receiving more than his minimal motivating reward). For the spectators don’t actually care if the money goes to Wilt: they are merely willing to give it up to see him play. It is no part of their intention to transfer the money to him: merely to alienate it in order to get something else. This kind of ‘voluntary’ transfer seems to have lost any direct connection with autonomous action and the furthering of life-plans which form Nozick’s central concern.
I therefore make a more radical assertion than that of Quest[31] that not ‘any consequence of any conjunction of voluntary acts is something people voluntarily bring about’. My assertion is that even if we do not allow that the aggregated consequences of voluntary transactions should also be assessed for voluntariness, we should distinguish between acquiescence and choice in ‘voluntary’ transactions. The spectators choose to give up their money: they merely acquiesce in that money’s going to Wilt. My focus differs from Quest’s: he asserts that the spectators may have cause for complaint if Wilt gets rich (they don’t acquiesce in his accumulation). I assert that they don’t have cause for complaint if he doesn’t (they don’t choose that he be paid).
This view of the matter might enable gifts to be made, (since they are unearned, they might cause problems for this view) even if there weren’t other considerations which could fill the gap left in entitlement by the Earnings Thesis.).
Sreenivasan regards transfer as requiring something very like acquisition on the part of the transferee:
the possibility of there being a right of the present owner to transfer her property turns on the possibility of construing transmission as use. This construal is manifestly possible. [One uses the resource to get hold of something else.] ... Some may argue that B’s right of acquisition is (conceptually or otherwise) essentially tied to A’s right of transmission. But even if that argument is correct (which is uncertain), it would only show that A’s right presupposed B’s right, and not that B’s right exists. ... A’s prior right of transmission does not entitle her to give B a gift to which B has no right.’ (Sreenivasan (1995), pp.107-8.
Kant[32] feels the force of this, and (by an unwarrantedly conservative transcendental deduction) concludes that
Transfer by contract of what is mine takes place in accordance with the law of continuity (lex continui), that is, possession of the object is not interrupted for a moment during this act; for otherwise I would acquire, in this condition, an object as something that has no possessor (res vacua), hence would acquire it originally, and this contradicts the concept of contract. Because of this continuity, however, that which transfers what is mine to the other is not one of the two separate wills (promittensis et acceptantis), but their united will. So the transfer does not take place in such a way that the promisor first abandons (derelinquat) his possession for the other’s advantage, or renounces (renunciat) his right, and the other immediately takes it up, or the reverse. Transfer is therefore an act in which an object belongs, for a moment, to both together, just when a stone that has been thrown reaches the apex of its parabolic path, it can be regarded as, for just a moment, simultaneously rising and falling, and so first passing from its rising motion to its falling.
I would challenge Kant’s analogy: it is surely far more plausible to say that the projectile is neither rising nor falling during that instant. The conception of contract as a ‘meeting of minds’ is a civilian, not a common-law conception. The Anglophone tradition might better be represented by Hampshire (1999, p.86) :
In any exchange and whatever the commodity exchanged, the buyers and the vendors have conflicting interests in setting the price. This remains true when the commodity in exchange is labour...The resulting conflict over labour costs is expressed at a political level in adversary arguments circling around justice and fairness of reward.’
The conflict is expressed at a legal level in adversary arguments about what liabilities each party has incurred under contract- not in an attempt to discover the truth about an agreement.
On this view, we can regard the right of transfer as the right to alienate property in another’s favour. Their right to keep it depends on whether they have earned it.
Having introduced the theory and established that it cannot be rejected out of hand by the Nozickian, I shall introduce a derivation of the principle from one of Nozick’s own concepts: unproductive exchange.
IV. Unproductive Exchange
Unproductivity is a property possession of which by a putative exchange requires that the exchange be blocked. I will assume that this blocking is accomplished by a principle of entitlement which states that an unproductive exchange does not (insofar as it is unproductive) confer entitlement on the unproductive party. The notion of unproductive exchange (along with that of unproductive activity) is described by Nozick as exchange in which one party is made no better off. An example is blackmail:
Though people value a blackmailer’s silence, and pay for it, his being silent is not a productive activity. His victims would be as well off if the blackmailer didn’t exist at all, and so wasn’t threatening them. And they would be no worse off if the exchange were known to be absolutely impossible.[33]
However, this is only one necessary condition for unproductivity, as Nozick points out, illustrating this fact with another example: one’s payment to a neighbour in exchange for his[34] not building a monstrous edifice which would ruin one’s view.
If your next-door neighbour plans to erect a certain structure on his land, which he has a right to do, you might be better off if he didn’t didn’t exist at all (no one else would choose to erect that monstrosity.) Yet purchasing his abstention from proceeding with his plans will be a productive exchange. Suppose, however, that the neighbour has no desire to erect the structure on the land; he formulates his plan and informs you of it solely in order to sell you his abstention from it. Such an exchange would not be a productive one: it merely gives relief from some thing that would not threaten were it not for the possibility of an exchange… . that such exchanges are not productive ones, and do not benefit each party, is shown by the fact that if they were impossible or forceably [sic] prohibited so that everyone knew they couldn’t be done, one of the parties to the potential exchange would be no worse off.[35]
There is, then, a second necessary condition, which jointly with the first is sufficient for an exchange’s being unproductive. This condition involves the motivations of the ‘unproductive’ party: it is central to the account that a threatened consequence is not independently desired; the threat is merely a bargaining tool.
A preliminary formulation of the conditions for the existence of an unproductive exchange E between exploiter[36] ER, who provides R (an unproductive act or its relevant consequence – N.B.: R will normally be a forbearance from action), and exploited ED who provides D is:
‘Holding Out’:
(HO) If E were impossible or forcibly prohibited, ER would still (be willing to) provide R.
‘Paying Off’:
Given standard (over-) simplifying assumptions about the relations between preferences and behaviour (which are in any case presupposed by Nozick in formulating HO and PO) it is possible to paraphrase the subjunctives embodied in these two conditions as indicative statements about the (inherently subjunctive) preference rankings of the parties. In the case of HO; ER (weakly[40]) prefers supplying R to not supplying R in those situations (ceteris paribus) where E is not an available option. In the case of PO; ED prefers that E should be (known to be) unavailable, but given that E becomes available, ED prefers to give up D in exchange for R than go without R.
Abstracting from strategic considerations concerning subsequent interactions, as until further notice I shall, we may observe that the possibility of unproductive exchange is dependent on imperfect information. If we allow ourselves the assumption that the carrying out of threats will have some cost (however minimal), we may conclude that rational actors with perfect information about each others’ preferences (and knowledge, strength of will?) would not engage in unproductive exchange. For if I know that you have no independent motive for withholding R, and that there is some reason for you not to do so, I will call your bluff, withhold D and receive R without unproductive exchange. Nozick’s economistic treatment of the difficult questions of justification depends on an idealised rationality on the part of his hypothetical contractors, so in this interpretative context, I feel justified in ignoring irrational threat-fulfilment[41].
The assumption that imperfect information has a part to play in defeating the justice of transactions by vitiating consent is shared by the rejection of fraud[42], and unproductivity might be seen as a kind of fraud: misrepresentation as to one’s own preferences – and in the circumstances, therefore, as to the ‘market value’ of the commodity in question.
V. Unproductivity and the state
This preliminary formulation of the mini-theory of unproductivity raises many questions, but before dealing with them, I feel it sensible to answer a conceptually posterior but pragmatically prior question: Why is unproductivity of interest? For reasons of brevity, I will discuss only one area of Nozickean theory in which unproductivity has an important role[43]. This is the context in which unproductivity is introduced, and is the central moment of ASU: the justification of the minimal state.
Nozick’s central problem is that whatever justification is made for prohibition, by the majority who buy into the state, of the enforcement of private (purported) determinations of justice (PJ), it must at the same time impose an obligation on that majority to compensate for this prohibition by providing protection for those who can’t afford to pay for it. (I assume that a general prohibition will induce those who can to pay, if unproductive ‘holding out’ may not take place. Not to do so would be to reject the assumptions about rational self-interest and the importance of personal security on which Nozick’s method of ‘fundamental potential explanation’ of the state depends[44].)
The idea that the obligation to protect rights is a form of compensation implies that the prohibition on PJ is imposed by the party who compensates, and is not a general negative duty arising from natural justice (in fact there is a general natural right to administer outstanding punishments[45].) The fact that it seems difficult to justify the imposition of a prohibition[46] under a system of natural rights (either the conduct is ‘naturally’ prohibited or not) gives rise to Nozick’s real though blithely dismissed dilemma ‘either you have a right to forbid it so you needn’t compensate, or you don’t have a right to forbid it so you should stop’.[47]
This problem is avoided if we allow that no new fundamental[48] duties/prohibitions can be imposed by anyone, but that the determinate content of a naturally subsisting prohibition can depend on circumstances, and the requirement it imposes on one party can therefore be rendered more stringent by the actions of another party, giving rise to new determinate prohibitions.
An example of a duty such as this is Nozick’s ‘epistemic principle of border crossing’(EPBC). My interpretation of it is a duty not to take any action without having used the best existing method of judging that one is not thereby violating others’ rights. Wolff[49] suggests that the EPBC may be unavoidably over-stringent; that it might require flag-bearers before motorcars. Detailed discussion of the merits of this objection is unnecessary here, since my version refers to violating rights, not running the risk of violating rights in the future. On the Nozickean assumption that the content of rights is well known, there are relatively few situations in which I need to know facts extraneous to the immediate situation to know whether I am (now) infringing rights. These situations include taking not-obviously-owned things, making potentially libellous statements, and imprisoning, killing or fining people[50]. Only in cases such as these does my version of the EPBC apply, so this objection is met.
There are three reasons why the EPBC is the best choice for the principle underlying the state monopoly on legitimate body- and property- invasions. First, it provides a reason why the prohibition of PJ should coincide with the duty to respect ‘procedural rights’. Second, it explains how procedural rights need not have the same determinate content in a primitive state of nature as they do in a more complex state. And third, it explains why the state has a duty to respect the procedural rights of all.
On the view I advocate, the state is required to offer its services to all in exchange for a (feasibly) affordable payment, in order to make the state method of determination of guilt available and thereby render PJ unjust (prohibited). The assumption is that the state (members of the protective association) would consider it worth their resources to offer its services for free were this the only feasible payment it could demand; perhaps on the grounds that one never knows who might prove dangerous if allowed to self-protect. If the state makes its services available only at a prohibitively high (though feasible) price, it violates HO. It would offer protection even if it were impossible to receive payment in exchange. And the independent would rather the exchange were impossible, so that he could at least have the right to protect himself.
The consequences of all this are:
1. The state and its clients are obliged to observe proper procedures in determining any person’s guilt and more importantly in applying punishment. (By EPBC)
2. In order to extend this obligation to all and thereby safeguard its client’s procedural rights, it must make its procedures (‘protection’) available to all by offering to make them clients on terms which they may feasibly meet. (By EPBC)
3. It may not threaten to charge persons more than they can reasonably be expected to pay, with the consequence that they do not receive protection. (By unproductivity)
I have presented a sketch of arguments whereby both the Lockean proviso’s consent-bypassing power and the legitimacy of the Nozickean state can be explained, in both cases making essential reference to unproductivity theory. In this way, I hope to have given antecedent reason for a Nozickean to accept the importance of unproductivity.
Having given some reason to think that unproductivity is of interest in the context of a Nozickean theory, I shall consider some subsidiary principles which give wider application to the theory.
VI. Refinement of the theory of Unproductive exchange
In section II, three formulations of the PO necessary condition for unproductivity were presented. Nozick generally uses the conjunction of PO1 and PO2 most of the time, but uses PO3 once, and the conjunction of PO3 and PO1 once. The task of deciding which (if any) of these formulations is the best one falls to his interpreters. I assume that a complex condition composed of more than one of them would be unmotivated and counterintuitive.
Nozick[51] states as a possible objection to PO1, which applies to PO2 as well, that ER might pre-empt ER2 who would have demanded even more for supplying U (or U2, a good of the same type). The overdetermination of the unproductive outcome should not, he feels, impugn the conclusion that the exchange itself is unproductive. We may note that for this purpose ER2 need only demand the same amount, since the principle states that ED would be no worse off , not better off without ER’s interaction.
In addition, Nozick tells us to ‘ignore the complication that one may sell a bona fide good to [one] he generally harms’[52] Conversely, one might engage in an unproductive (coercive, exploitative) exchange with one he generally benefits. Such a situation is envisaged by Mack[53] in his example of shoppers exploitatively boycotting a shopkeeper. The implications of these examples will be investigated further in what follows.
Consider the situation in which a positive externality eventuates on ED as a result of ER’s activities. A (bizarre) example is the enjoyed throwing of books into houses. Nozick rejects a ‘principle of fairness’[54] whereby acceptance of benefits entails a duty of reciprocation: ‘you certainly may not [charge and collect (without prior agreement?)] for benefits whose bestowal costs you nothing’[55]. Nozick’s conception of coercion[56] allows that one may place reliance on, say, one’s drug-dealer[57] so that the expected course of events includes the dealer selling one drugs.
I suggest that these considerations make PO3[58] the only viable formulation. As well as avoiding the complications mentioned, and those arising from irrelevant relationships like parenthood[59], PO3 adds some symmetry with HO: both are now defined relative to a putative prohibition. There seems no reason to reject PO3 (henceforth, ‘PO’) in favour of the obviously inadequate alternatives.
An important consequence of this interpretation is that the distinction between HO and PO is blurred. The facts that ED is no worse off under the prohibition, and that ER is still motivated to provide R despite it, are two sides of the same state of affairs. This feature of the account does not mean that something important has been left out: if unproductivity is a unitary notion, one necessary condition is better than two. (Compare Descartes on the ‘dual essence’ of a human.)
The principle of partition with respect to HO (henceforth PHO) specifies that HO may be satisfied in a partial way. Partial satisfaction of HO occurs when ER has some other independent motivation for withholding R, but if compensated for this foregone opportunity and if E (the unproductive exchange) were not available, would not be motivated to perform U. In other words, his asking price exceeds his total opportunity cost[60]. This principle is asserted by Nozick[61] (Mack points out that it is very commonly true of market transactions)[62].
An example is given[63]: that of a blackmailer who is motivated to reveal my secrets for profit elsewhere. I may compel her abstinence from publishing, but only on payment of full compensation (cost-covering compensation: this includes opportunity cost.)
This principle removes some counterintuitive decisions which might be arrived at in its absence. To take an extreme example a kind of ‘tacking paradox’ could arise, whereby I offer not {to shoot you or publish my memoirs} (R), in exchange for your money (D). Since I could receive some small fee by publishing my memoirs, I have an independent motive for {shooting you or publishing my memoirs}. This possibility parallels the problem which arose in connection with the rejected formulations of PO: that if interactions (exchanges) between parties are to be taken as a whole, the injustice of specific (sub-) transactions may be masked.
In the case under consideration, PHO entails that the ‘not publishing my memoirs’ part of R is the part for which I should be compensated. The obvious way to compensate for this is by (re-) granting permission to publish. In this way, the memoirs can be cut out of the picture when arriving at the (intuitively) correct result. (This limiting case of compensation is the most clearly unobjectionable form of compensation possible. Compensation for taking of x by instantly restoring x (or not taking x at all) is compensation in a formal sense only. The problems Nozick identifies with schemes of posterior compensation are in such a case entirely sidestepped!
So PHO states: the HO necessary condition may be satisfied with respect to some part of a transaction; specifically, when the amount demanded (D) is more than the minimum consideration which would motivate ER to give up R (perform U). Under PHO, D[64] is partitioned into a part which is necessary for motivating ER’s provision of R, and a part which is not.
Given the fact that partition of exchanges has been admitted in the case of PHO, and a transaction is no longer being regarded as a unitary ‘given’, the way would appear to be open for the introduction of an equivalent principle of partial satisfaction of PO: the principle of partition w.r.t. Paying Off (PPO).
The proposed principle PPO holds that D may be partitioned into the part (D1) whose transfer could not be blocked[65] if ED is to be made better off, and the rest (D2). Given the prospective exchange of R for D1, ED would be no worse off if the transfer of D2 were blocked: thus PO is satisfied in a partial way.
To reuse the example of a positive externality imposed in the course of an enjoyed activity like book-throwing, imagine that the thrower (ER) finds a source of fake books, use of which would preserve his enjoyment at a (slightly) lower cost. The erstwhile recipient of the books (ED) offers to pay D - the difference in costs, plus transaction costs, plus a small sweetener - if the thrower will continue to use real books (provide R). The thrower, sensing a profit, demands a higher price. Without PPO, the thrower may press his exploitative advantage. Under the provisions of PPO, however, any payment above D would be blocked, since given ER’s motivation to provide R at any price of D or more, the blocking of higher payments would not worsen ED’s position.
We may note that the counterfactual equivalence observed between the nonpartitioned forms of PO and HO still holds when each principle is subject to its own principle of partition. The part of D whose transfer may be blocked without rendering ED worse off is necessarily (given the standing assumptions about our homines oeconomici) the very same part whose transfer may be blocked without discouraging ER from providing R – since it would be R’s unwillingness to provide ER that would (relatively speaking) impoverish ED.
Pleasing symmetry does not in general, however, justify principles of entitlement (and I have no illusions as to just how pleasing this specimen of symmetry really is, in the grand scheme of things). I believe, however, that support for PPO can be derived from Nozick’s earlier account of coercion[66], already alluded to in support of the PO3 interpretation of PO.
Nozick’s discussion centres on the concept of a threat: I take it that the relevance of this concept to the central cases of unproductivity, e.g. blackmail, is evident. Threats are distinguished from two other phenomena; the two distinctions exactly parallel the two conditions (or the two faces of the single condition) for unproductivity.
The first distinction is between threats and offers. Each, on Nozick’s account, is characterised by a change in the (perceived) consequences of one or both of a pair of alternative courses of action, A1 and A2. The change is intentionally imposed by a person P in order to reverse the preference ordering over these alternatives on the part of another person Q[67].
We stipulate that pre-intervention, Q prefers A1 to A2. The reversal of this ordering may be done in one or both of two ways: by worsening the (perceived) consequences of A1, or improving those of A2:
This resultant change predominantly involves a threat to Q if he does A1 if Q prefers doing the old A1 (without the worsened consequences) to doing the new A2 (with the improved consequences).
This resultant change predominantly involves an offer to Q to do A2 if Q prefers doing the new A2 (with the improved consequences) to doing the old A1 (without the worsened consequences).[68]
This distinction between willingness and unwillingness to make the change mirrors the PO condition, while the fact that the formulations are in terms of predominance underlines the fact that the situation may contain two distinct elements of threat/offer, which may therefore in principle be separated, as in PPO.
Nozick’s discussion centres on how to classify well-specified situations into one of two exclusive categories. In such a situation, a major problem is that of specifying the baseline[69] against which changes in consequences are to be measured. Nozick suggests that the appropriate baseline is the ‘normal and expected course of events’, where ‘the term expected is meant to shift between or straddle predicted and morally required’.[70]
Either or both of these may, in Nozick’s view, allow that benefits flow from P to Q in the expected course of events. The test for threateningness is applied in incremental fashion once this baseline is established. This goes some way toward providing a rationale for PPO.
The consequences of the introduction of Partition w.r.t. PO are extensive. The limitation on receiving a more than-minimal profit or surplus would apply in both directions, meaning that neither party to a transaction could ask more than their own minimally-motivating price.
This is the crux and also the potential sticking-point of my interpretation. The consequence I have drawn appears to undermine the role which Nozick intended unproductivity to play: a method of apportioning the benefits of exchange, in circumstances where ordinary negotiation is not appropriate (a condition, moreover, which I imply is standardly satisfied).
The point may be put thus: where two parties ‘cooperate’ in exchange, why should each only receive their minimal cooperative payoff? Whose self-ownership interests are damaged by, say, allotting title along the lines of some bargaining solution[71]?
The original point of unproductivity theory appeared to be protecting the interests of a victim of exploitation, which reflects the right-based nature of Nozick's theory. However, as Waldron[72] points out, Nozick’s theory is better seen as duty-based: side-constraints don’t seem based only on the interests of right-holders, but on agent-centred considerations. As we have seen, the inadequacies of a strictly right-based account leads Nozick to reject natural procedural rights in favour of the (agent-centred) EPBC[73]. ‘Does a person who did violate another’s rights’, he asks, ‘have a right that this fact be determined by a fair and reliable procedure?’[74].
To clarify this issue, we consider Nozick’s second distinction in ‘Coercion’: that between threats and warnings. To simplify, the distinction rests on the reasons for the expression of conditional consequence. Nozick’s example is that of an employer who announces that he will go out of business if his employers vote in favour of a union[75]. This is a mere warning because the employers does not make the conditional decision to go out of business in order to worsen the union alternative. He is independently motivated to do so: the union will drive up costs, his is a marginal firm, etc. The distinction rests on the alternatives seen from the point of view of the employer, and the presence or absence of merely strategic, manipulative motivations for adopting (feigning adoption of) the threatened/warned course of action. Depending on what the baseline thus specified involves, the announcement is either a threat or a warning. The effects on the other party are of secondary importance. As we have had reason to note earlier, the PO and HO principles are counterfactually extensionally equivalent. At this stage we may add that the causal or constitutive relation between them seems to move from HO to PO: that is, it is the motivations and preferences of ER which determine the consequences of putative ‘blockings’ of exchange for ED.
In the preceding discussion, I have limited myself to consideration of exchange in a two-person world. The addition of other persons complicates matters further. One such complication is mentioned above: the possibility that a potential exploiter (in Nozick’s example, a blackmailer) might pre-empt another, who would have made a similar threat. The overdetermination of the ‘exploitative’ outcome would appear to vitiate the claim of unproductivity we might make for the exchange. If my only alternative is to deal with a second ‘exploiter’, then I am not worse off for dealing with the first. But intuitively, each is an exploiter, and the fact that the exploitation is overdetermined does not alter the fact[77].
This view is endorsed by Nozick in his treatment of the LP; he states that an ‘agglomeration’ may act in such a way as to worsen the situation of others. Mack also adverts to such a principle en passant, when he states ‘If someone else would hold the land if S didn’t, then B’s purchase from that party would be unproductive.[78]’
An iterative principle of agglomeration w.r.t. PO (APO) would rule that where a prospective exchange does not count as unproductive only because ED’s alternative to dealing with ER is dealing on at least as unfavourable terms with another party (ER2), then ER and ER2 may be regarded as a single agent for the purposes of the PO condition for determining unproductivity. The explanation of why ER and ER2 do not bid each other down need not concern us, except to note that implicit collusion or the formation of a cartel would only underline the need for such a principle.
It should be borne in mind that the expression of the principle in terms of treating several agents as one is purely heuristic. No ineliminable ontological or methodological commitment is made to an aggregative entity. The principle is concerned only with ED’s situation with respect to a proposed exchange. In any case, we should note that APO may be rendered redundant by the introduction of PPO, since the effect of that principle is to render unproductivity an entirely agent-centred notion: and consideration of ED’s position drops out of the argument.
A second possibility is that of multiple victims - Nozick mentions, of the hideous building example, that it ‘generalises to the case where the neighbour’s intention does not focus solely on [ED]’ - he peddles his scheme around several neighbours. ‘Whoever purchases the abstention is ‘served’ unproductively’[79]. The point of mentioning this is clearly to affirm a principle of agglomeration w.r.t. HO; the idea is to avoid allowing the would-be exploiter to play one potential victim off against another, by claiming that the potential gain from each constitutes a benefit foregone when contracting with the other[80].
A principle of this kind would state that in deciding the minimally-motivating payment, the alternative possibilities of exchanging the same good with some other party may be disregarded. The adoption of PUE (or its variant NEU) as the juridical principle accompanying the definition of unproductive exchange lends plausibility to this approach. Since a prohibition is of a type of action, we would expect its effects to be global: so in our subjunctive situation whereby an exchange at a higher price is impossible or prohibited, the impossibility should cover all exchanges at a higher price, not just the particular exchange envisaged. This approach is reminiscent of the distinction between use-value and exchange-value. As Mack points out, the theory of unproductivity seems to embody a ‘Marx-like aversion to persons perceiving and valuing the goods which they provide for others...as commodities.’[81]
Nozick states that ‘complicated combinations of subjunctive conditionals and [sic] counterfactuals might perhaps succeed in disentangling an owner’s preferences from his knowledge of [what others] are willing to pay. But no one yet has actually provided the requisite combinations.’[82] A first step towards the requisite combination is, unless I have missed something, incorporated in AHO: a party’s minimally-motivating reward for a given voluntary sacrifice of goods is to be determined with reference to the least he would accept if no greater reward could be gained from any party in exchange for that sacrifice.
Principles of agglomeration appear to move in the direction of a ‘class’ analysis of relations between contractors. There is a great deal more to be said on the subject of this kind of principle, which aims to prevent responsibility’s dispersal between individuals, but no space. I will simply add that these principles will, I think, allow whatever results may be generated by consideration of two-person worlds to be extended directly to many-person worlds, while still restricting entitlement to foregone use-values; not fetishised exchange-values.
Conclusion
"In our earlier discussion of border crossing we noted the absence of any compelling theory of just price or compelling reason why all the benefits of voluntary exchange should go to one of the parties."[83]
In this essay, I have presented and given an internal justification for a salient solution to the ‘bargaining’ problem: if any division is arbitrary, and since not all the benefits can go to both parties, the only remaining solution is that none (or almost none) of the benefits of exchange go to either party.
The position I have canvassed does not entail a prohibition on the parties keeping whatever share of the surplus they could get from bargaining. All it entails is that entitlement does not extend to the surplus; I suppose that this fact might be expressed by stating that the surplus reverts to the state of unownedness. Since we might assume (at least for state-of-nature style transactions) that the exchangers will be in possession at the time this reversion occurs, they will be in the best position to reappropriate ‘their’ goods. So some reason will have to be given why their appropriation may be blocked. One way in which this might occur is that others could use the surplus; the Lockean proviso would demand that they either consent or not be made worse off by the reappropriation.
Bibliography
Alexander, Lawrence A., ‘Zimmerman on Coercive Wage Offers’ Philosophy and Public Affairs 12, 1983
Appley, M.H. (ed.), Adaptation-Level Theory, London, 1971
Arneson, Richard J., ‘Lockean Self-Ownership: Towards a Demolition’ Political Studies 39, 1991
Arnold, N. Scott, ‘Why Profits are Deserved’ Ethics 97, 1987
Ashworth, Andrew., Principles of Criminal Law, Oxford, 1991
Avineri, Shlomo and de-Shalit, Avner (eds.), Communitarianism and Individualism, Oxford, 1992
Barry, Brian, Theories of Justice, Hemel Hempstead, 1989
Barry, Norman P., On Classical Liberalism and Libertarianism, Basingstoke, 1986
Beiner, Ronald and Booth, William James (eds.), Kant and Political Philosophy, New Haven, 1993
Booth, William James , ‘The Limits of Autonomy: Karl Marx’s Kant Critique’ in Beiner and Booth (eds.), 1993
Braybrook, David., ‘Social Contract Theory’s Fanciest Flight’ Ethics 97, 1987
Brazier, Margaret, The Law of Torts (9th ed.), London, 1993
Brickman, Philip and Campbell, Donald T., ‘Hedonic Relativism and Planning the Good Society’ in Appley (ed.), 1971
Carver, Terrell and Thomas, Paul (eds.) Rational Choice Marxism, University Park (PA), 1995
Child, James W., ‘Can Libertarianism Sustain a Fraud Standard?’ Ethics 104 , 1994
Chomsky, Noam., Class Warfare, London, 1996
Cohen, G. A., Karl Marx’s Theory of History, Oxford, 1978
Davis, Michael, ‘Nozick’s argument for the Legitimacy of the Welfare State’ Ethics 97 , 1987
Dick, James C., ‘How to Justify a Distribution of Earnings’ Philosophy and Public Affairs 4 , 1975
Dworkin, Ronald, Taking Rights Seriously, London, 1977
Elster, John and Moene, Karl Ove (eds.), Alternatives to Capitalism, Cambridge, 1989
Exdell, John, ‘Distibutive Justice: Nozick on Property Rights’ Ethics 87, 1977
Fried, Barbara, ‘Wilt Chamberlain Revisited: Nozick’s “Justice in Transfer” and the problem of Market-Based Distribution’ Ethics 97 , 1987
Fischer, John Martin and Ravizza, Mark (eds.), Perspectives on Moral Responsibility, Ithaca (NY), 1993
Forgacs, David (ed.), A Gramsci Reader, London, 1988
Gaus, Gerald. F. , ‘Property, Rights and Freedom’ in Paul et al. (eds.), 1994
Gauthier, David, Morals by Agreement, Oxford, 1986
Gibbard, A., ‘A Pareto-consistent Libertarian Claim’, Journal of Economic Theory 7, 1973
Gilbert, Margaret, ‘Agreements, Coercion and Obligation', Ethics 103, 1993
Goodin, Robert, ‘Negating Positive Desert Claims', in Pojman and McLeod, 1999
Gray, Kevin, Elements of Land Law, London, 1993
Gregor, Mary , ‘Kant on “Natural Rights” ’in Beiner and Booth (eds.), 1993
Guest, A. G.(ed.), Oxford Essays in Jurisprudence, Oxford, 1961
Hart, H.L.A., Punishment and Responsibility, Oxford, 1968
Hampshire, Stuart, Justice is Conflict, London, 1999
Hohfeld, Wesley N., Fundamental Legal Conceptions as Applied in Judicial Reasoning, New Haven, 1923
Honoré, A. M., ‘Ownership’, in Guest, 1961
Hutton, Will, The State We’re In, London, 1996
Kagan, Shelley, ‘Baier on Killing and Letting Die’, in Schneewind (ed.), 1996
Kant, Immanuel, The Science of Right (trans. Hastie, W.),
http://eserver.org/philosophy/kant/, 1790
Kavka, Gregory S., ‘An Internal Critique of Nozick’s Entitlement Theory’ Pacific Philosophical Quarterly 63, 1982
Kirzner, Israel M., ‘Entrepreneurship, Entitlement, and Economic Justice’ in Paul (ed.), 1981
Kleinig, John , ‘The Concept of Desert’, in Pojman and McLeod, 1999
Kymlicka, Will, Contemporary Political Philosophy, Oxford, 1990
Lamont, Julian , ‘The Concept of Desert in Distributive Justice’, in Pojman and McLeod, 1999
Lasker, I.B. (ed.), Karl Marx: Selected Works Vol.1, Moscow, 1946
Lee, R.W., The Elements of Roman Law (4th ed.), London, 1956
Lessnoff, Michael, Social Contract, London, 1986
Lewis, David, Counterfactuals, Oxford, 1973
Lomasky, Loren, ‘Personal Projects as the Foundation for Basic Rights’ in Paul et al. (eds.), 1984
Locke, John, The Second Treatise of Government in Wooton (ed.), 1993 (a)
Lyons, David, ‘The New Indian Claims and Original Rights to Land’ in Paul (ed.), 1981
Mack, Eric, ‘Nozick on Unproductivity: The Unintended Consequences’ in Paul (ed.), 1981
Mackie, J.L., Ethics, Harmondsworth, 1977
MacPherson, C. B., The Political Theory of Possessive Individualism, Oxford, 1962
Marx, Karl, Grundrisse (trans. Nicolaus, Martin), Harmondsworth, 1973
Mayer, Martin, Markets, New York, 1988
MILL, John Stuart, Principles of Political Economy, London, 1929
Nagel, Thomas, ‘Libertarianism without Foundations’ in Paul (ed.), 1981
Narveson, Jan, ‘Libertarianism vs. Marxism: Reflections on G.A.Cohen’s “Self Ownership, Freedom and Equality” ’, Journal of Ethics 2, 1998
Nash, J.F, ‘The Bargaining Problem’, Econometrica 18, 1950
Norton, David L., ‘Individualism and Productive Justice’ Ethics 87, 1977
Nozick, Robert, Anarchy, State, and Utopia, Oxford, 1974
O’Neill, Onora , ‘The Most Extensive Liberty’, Proceedings of the Aristotelian Society 80, 1980
Otsuka, Michael , ‘Killing the Innocent in Self-Defence’, Philosophy and Public Affairs 23, 1994
Paul, Ellen Frankel; Miller, Fred D.; Paul, Jeffrey (eds.), Property Rights, Cambridge, 1994
Paul, Jeffrey (ed.), Reading Nozick, Oxford, 1981
Pojman, Louis P. and McLeod, Owen (eds.), What Do We Deserve?, Oxford, 1999
Quest, Edward, ‘Whatever Arises from a Just Distribution by Just Steps is Itself Just’, Analysis 37, 1977
Rawls, John, A Theory of Justice, Oxford, 1971
Raz, Joseph, The Morality of Freedom, Oxford, 1986
Reiman, Jeffrey, Justice and Modern Moral Philosophy, New Haven, 1990
Roemer, John, Free to Lose, London, 1988
Rothbard, Murray N., Power and Market, Menlo Park (Ca.), 1970
Rowe, William L., ‘Causing and being responsible for what is inevitable’ in Fischer and Ravizza, 1993
Ryan, Alan, Property and Political Theory, Oxford, 1984
Schneewind, J.B. (ed.), Reason, Ethics, and Society:Themes from Kurt Baier, with His Responses, Chicago, 1996
Sen, Amartya, ‘The Impossibility of a Peretian Liberal’ Journal of Political Economy 78, 1970
Shiller, Robert J., Macro Markets: Creating Institutions for Managing Society’s Largest Economic Risks, Oxford,1999
Slote, Michael A., ‘Desert, Consent and Justice’, in Pojman and McLeod (eds.), 1999
Sreenivasan, Gopal, The Limits of Lockean Rights to Property, Oxford, 1995
Steiner, Hillel, ‘The Structure of a Set of Compossible Rights’ Journal of Philosophy 74, 1977
Tully, James, A Discourse on Property, Cambridge, 1946
Waldron, Jeremy, ‘Enough and As Good Left for Others’, Philosophical Quarterly 29, 1979
Wessel, Robert H., ‘A Note on Economic Rent’, The American Economic Review 57, 1967
Williams, Bernard ‘The Minimal State’ in Paul (ed), 1981
Wittgenstein, Ludwig, Culture and Value, Oxford, 1980
Wolff, Jonathan, Robert Nozick, Cambridge, 1991
Wooton, David (ed.), John Locke: Political Writings, Harmondsworth, 1993
Zimmermann, Michael J., The Concept of Moral Obligation, Cambridge, 1996
[3] The term is first used in the context of rights by Steiner (1977) and is now more widespread than Nozick’s ‘co-possible’.
[5] Wolff (1991, p.148, n.5), cites R. van der Veen and P Van Parijs, ‘Entitlement Theories of Justice’, Economics and Philosophy, 1 (1985), for ‘the case that a wide variety of theories may all have a claim’ to this status.
[8] cf. Mack, 1982, p. 186: ‘The shift to this outcome-orientation is a product of making the wrong of boundary-crossing rest upon the inadequacies of schemes of posterior compensation for boundary-crossing.’, Lomasky, 1984, p. 41: ‘For beings who undertake commitments, not everything is open to trade-offs.’
[9] Mack points out that the minimally-motivating price ought to be greater (though only by say, a penny) than full compensation
[11] Goodin, 1999, pp. 236-7. Goodin also suggests an analysis of undesert: what is undeserved is what would not occur in the expected course of events. This, as he points out, corresponds to Nozick’s (1997) use of the ‘expected’ course of events in his analysis of coercion. This puts great weight on our choice of baseline; in the case of the Earnings Thesis, the baseline is the least I would accept. I am not sure how (if at all) this could be regarded as the ‘expected’ course. Cf. note 5 supra
[16] See Waldron (1986) for a conception of charity consistent with all rights and duties being negative.
[18] Cohen (1996, pp.217-220) mentions this in connection with Gauthier’s (1986, pp. 272 ff. inapt use of the concept of economic rent.
[28] For the canonical source of the distinctions between different kinds of ‘right’, see Hohfeld (1923).
[36] For convenience, ‘exploit’ and its cognates are used freely in a non-technical sense throughout the essay.
[40]I weakly prefer A to B if I do not unequivocally prefer B to A: i.e. I either unequivocally prefer A to B or am indifferent between A and B.
[41]Perfect information is an idealised assumption common to many economic models, e.g. that of perfect competition. This discussion raises some questions which might reward further investigation: what are the limits of perfect information? Do economic models require godlike knowledge? Does perfect information imply perfect intelligence?
[43] Unproductivity also, I believe, provides justification for the limitation of the prima facie ‘voluntary’ exchange in cases of blackmail, and for Nozick’s Lockean Proviso.
[44]This point reflects a certain tension in Nozick’s position between what Lessnoff (1986) calls the voluntaristic and rationalistic readings of the social contract: the latter appeals to real interests (in a paternalistic way), without requiring actual consent. In the case of hypothetical contracts with determinate consequences, the former seems out of place.
[45]We should notice that the ‘right to punish’ is more properly described as a liberty to punish corresponding to a no-right on the part of an offender not to suffer deserved punishment. A person may not be punished twice, and no one has the right to be the one who does the punishing. Positive rights in punishment are at best a right that punishment occur- which seems incompatible with Nozick’s rejection of positive claim-rights (since the means to satisfy them may not exist, and Nozick does not countenance the possibility that rights might be impossible to satisfy - cf. the right to sustenance). However, the whole area of prohibition, punishment and enforceability of rights is a complex one which I hope to examine elsewhere.
[48]A ‘fundamental’ duty being one which is not derived from another duty + circumstances, as my duty not to kill my boss derives from my duty not to kill any human being + the facts that my boss exists and is human.
[50]Duties exist at common law, for example, to take steps to trace any owner of found objects when taking possession of them (see Brazier, 1993, p44). Tort law, however, is concerned only with cases in which some injury is alleged. The EPBC as it governs punishment of the guilty might however be recognisable within the context of common-law principles regarding criminal attempts, if an (impossible) reckless attempt can be so recognised. An attempt can occur when no actual injury is done, and reckless ignorance of antecedent fact (e.g. consent to sexual intercourse) can take the place of intention. Since the punished person is guilty, the ‘attempt’ to violate her rights is impossible - but an attempt nonetheless. See Ashworth ,1991, pp. 399-406; Hart, 1968, pp. 130-1.
[51]ibid., p. 85, note: ‘But if he didn’t exist, mightn’t another have stumbled upon the unique piece of information and asked a higher [or the same] price for silence? If this would have occurred, isn’t the victim better off [or no worse off] because his actual blackmailer exists? To state the point exactly in order to exclude such complications is not worth the effort it would require.’
[52] ibid., p. 84. I assume that we are told to ignore the complication because it can be dealt with, and the overall harm will not impugn the ‘good’ status of the particular good sold.
[53]Mack, 1981, pp180-1. Mack envisages a situation in which a group of customers C boycott their shopkeeper S in order to elicit a change in his employment policy. C would not give up shopping there should S prove intransigent, so HO is clearly satisfied. Mack seems to say that PO is satisfied because (1) if C had never existed S would have set up shop elsewhere, or (2) if C had had nothing to do with S, this would be because the shopkeeper had set up shop elsewhere. I agree with Mack that this boycott is a case of unproductive exchange, but cannot accept his reasoning regarding the satisfaction of PO. Consideration of what might have happened had C never existed, as in 1, seems to leave open far too many questions to constitute even an idealised decision procedure. Would the shopkeeper have already opened up shop before the birth of (some members of) C? Would other customers have developed ‘heightened racial sensibilities?’ (p181) Would an otherwise non-existent hurricane have occurred because one of the customers was not there to crush a butterfly, etc.? Once we extend the ambit of our counterfactuals in the way suggested, these questions do arise; the fact that they are obviously irrelevant does not impugn my objection; it strengthens it. The alternative, 2, is equally unacceptable: it violates Lewis’s (1973) canonical formulation of how counterfactuals should be formed: roughly, that the world in which the events represented in the consequent are envisaged as occurring should be the closest to the actual world of those worlds in which the antecedent is true. In particular, the two worlds should have identical histories up to the point at which the envisaged difference occurs. In short, the aim of deciding whether S benefits from C’s custom is to be decided relative to the relevant variable: the occurrence of the boycott. If the boycott were not possible, C would (‘regretfully’) return to the shop. it is this fact which makes the putative exchange pro tanto unproductive: the shoppers do not care enough about the S’s employment policy to take their custom elsewhere. The counterfactuals invoked should not extend back before the relevant time: the time at which the boycott was to start. In the case of prospective (therefore non-counterfactual?) subjunctives, this point is clearer still.
[55]Nozick, op. cit., pp93-5. Nozick makes important use of this principle in arguing against ‘ideal’ contract theories of political obligation, though it is unclear that his theory is not in this category: his story about the emergence of a state is a hypothetical history about contracts.
[57]Does this rely on the assumption that addiction is present? If so, does it also rely on that addiction’s having been intentionally engendered by the dealer? And in this case, what price Nozick’s assumption that one’s personal projects are a ‘given’: the unassailable Archimedean point of justice?
[58]Whether we should adopt impossibility or prohibition as the relevant property in these formulations seems a matter of little importance. It would not be, were the effects of a failure of deterrence to be allowed for in our idealised subjunctives: if X is impossible, it will not be done; if prohibited, it may still be done. That this possibility is not relevant is apparent from the fact that the subjunctives are really about the motivations of the parties, as I suggest below.
[59]If my parents had not existed (or had nothing to do with me?), I would not exist. Would I be worse off? Similar problems to these are raised by Kavka (1982).
[61] Nozick, 1974, p.85: ‘On the view we take here, the seller of such silence could legitimately charge only for what he foregoes by silence.’
[64] D is selected as the dependent variable here. Since D has in most examples been a sum of money, it is the more convenient choice for a divisible commodity. R could, however, take this role instead, as in the ‘tacking paradox’ mentioned earlier.
[65] I use ‘blocked’ henceforth rather than ‘prohibited’: as specified in section IV it is a limitation on entitlement, rather than on action itself.
[66] Nozick, 1997, p.16: ‘This study of coercion is intended as a preliminary to a longer study of liberty…’.
[74]ibid, p.103. After introducing the EPBC, Nozick (apparently amnesically) remarks, ‘I am not sure that this is the proper focus. Perhaps the person...does have such procedural rights.’ The objection reasserts itself: ‘But what is a guilty person’s complaint against an unreliable procedure? That it is too likely to mispunish him?’ (p.107)
[76]Nozick’s use of the term ‘agglomeration’ (noun) in connection with group violations of the Lockean proviso (1974, p. 180) is felicitous. As well as being a good word, it is etymologically distinct from ‘conglomerate’ (noun), and ‘conglomeration’ - which has the effect of removing the connotation of concerted cooperation.
[77]There may be a parallel here with an account of the persistence of moral responsibility in the face of overdetermination of the outcome: see Rowe, 1993.
[80]This might have beneficial consequences for investment: see for example Hutton, 1996, p.153; where the lack of any state-assisted long-term investment bank in Britain (and to a lesser extent the USA) is held to be a major cause of a ‘bias to disengagement and short-termism. the banks have to make high returns to satisfy their own investors...The demand for quick profits is fundamental to their relationship with industry...The freer and more dominant the stock-market, the higher the returns required from the banks.’ There might be an ‘invisible-hand’ explanation of how such an effect can come about. Hutton cites the power of shareholders without any commitment to the company to sell control to the highest bidder in a contested takeover. ‘Takeovers, far from being an economic boon, were frightening companies into making absurdly high dividend payments and forcing them into focusing only on immediate profit growth...’ (p.156). The use of individual valuations rather than the market rate might also tend to avoid speculative ‘bubbles’. Chomsky (1996, pp.39-40) suggests that the growth of financial capital speculating against currencies leads to pressures for low growth, low employment and low wages. A ‘tax on speculative capital’ has been proposed, ‘simply to try to shift the balance towards productive investment instead of speculative and destructive interchanges.’ To properly assess the consequences of the principles of unproductivity in these areas would require much more work.
[82]Nozick, 1974, p. 64. The footnote appended to his remark refers to the explanation of mutually beneficial exchange in terms of opposed preferences, rather than equality of something or other. The assumption is that each party prefers having the parcel of goods they hold after the exchange to that which they held before. Nozick remarks that this assumption falls to the example of a three-way trade, where each party acquires a ‘dispreferred’ good in order to exchange it for a preferred one, unless we adopt this principle: that if one knows a certain good is exchangeable for another, then one is indifferent between them. But this principle ‘has the consequence that a person [engaging in exchange] does not prefer having another’s good to having his own. For his own can be [exchanged for] the other.’ This ignores the fact that the other good can’t subsequently be exchanged by him for his own. The relevant consideration is the distinction between use-value and exchange-value. One may prefer having (i.e. keeping, using) the other’s good, but still recognise the goods’ equivalence in (market) exchange-value, using a convenient if arbitrary measure, say, money. The existence of money, of course, renders a three-way trade unnecessary, except for extraneous (say strategic) reasons. The bundles of subjunctives Nozick refers to again in this footnote seem less complicated than he claims.
Posts
