1. Introduction
In the Doctrine of Right (DR) Kant argues that the reason we have a moral duty to exit the state of nature (exeundum e statu naturali) is that in the absence of a civil condition property rights are normatively ambiguous. In brief, Kant's argument is as follows: (1) Embodied rational agents possess equal freedom of choice and movement, i.e. external freedom (äussere Freiheit). (2) External freedom entitles agents to claim parts of the external world as their own. Yet (3) by claiming parts of the external world agents threaten the external freedom of others, until and unless (4) all agents collectively exit the state of nature and institute a civil condition.
For Kant, property (Eigentum) refers specifically to claims on the external world that involve discretionary use of an object not currently in one's grasp. Property entails the right to exclude others from use and to recover or retrieve an object from another (Reference Kant and Gregor[1797] 1996, 6:245; 270).Footnote 1 The result of the exit from the state of nature is ownership, a legal entitlement that holds vis-à-vis other free agents and is backed by the state's coercive power.Footnote 2 In this paper, I seek to understand why exactly Kant thinks that property rights are normatively ambiguous in the state of nature, an ambiguity he expresses by calling such rights ‘provisional.’
The concept of provisionality is first introduced in §9 of Kant's treatment of private or natural rights, i.e. rights concerning what is mine or yours that we have simply qua free, embodied individuals, and which constrain any legitimate political order (6:229, 6:242). Kant titles the section: ‘In a State of Nature Something External Can Actually Be Mine or Yours but only Provisionally’ (6:256; emphasis in original). By characterizing property rights as ‘actual’ (wirkliches), but only ‘provisional’ (provisorisches), Kant is claiming that there is some sense in which we have property rights in the state of nature and some sense in which we do not. §9 is a direct qualification of §8, the title of which asserts that property rights are impossible in the state of nature: “It is possible to have something external as one's own only in a rightful condition, under an authority giving laws publically, that is, in a civil condition” (6:255).
What I call Kant's provisionality claim (KPC) is §9's qualification of §8's apparent assertion that there are no property rights outside a regime of public law. Understanding KPC is key to appreciating Kant's account of how a political order is both authorized by and yet fundamentally transforms relations of natural right, as well as the kind of authority Kant thinks all rational agents possess to coerce others into forming a state. KPC thus illuminates Kant's account of the normative foundations of political life. Moreover, since KPC concerns the extent to which property rights are natural or conventional, it also pertains to whether the Kantian state may engage in property redistribution or must simply respect ex ante distributions of property. This, in turn, suggests larger questions about how to classify Kant's political philosophy: egalitarian or libertarian?
The plan of my argument is as follows: Section 1 describes two interpretations of KPC (weak and strong provisionality), briefly discusses their insufficiencies, and proposes an alternative (anticipatory provisionality). Section 2 explains the argument of DR leading up to KPC. Sections 3 and 4 criticize weak and strong provisionality in more detail. Section 5 develops anticipatory provisionality. Section 6 discusses the implications of KPC for property redistribution in the Kantian state.
1.1. Weak and strong provisionality
Given the obscurity of the idea that one could have something ‘actually … but only provisionally, ’ it is not surprising that there is little interpretive consensus about KPC. Alan Ryan gives voice to an understandable frustration with Kant's argument when he writes, ‘we find Kant both asserting … that men only have property in external things when a legal order gives them that property … and asserting that we have to assume a ‘natural right’ to appropriate unowned things and make them our property in a state of nature’ (Reference Ryan1984, 79–80).Footnote 3 Interpretations of KPC divide into what I call weak and strong provisionality.Footnote 4 Let me treat each in turn.
According to weak provisionality, KPC is that agents in the state of nature can make rightful claims to ownership, but such claims are not fully legitimate without the institutional context and enforcement apparatus necessary to render them secure. For example, Sharon Byrd and Joachim Hruschka write: ‘For Kant it is not in the first instance the state that introduces rights to external objects of our choice, because as Kant notes, property ownership must exist before one moves to civil society.’ Accordingly, Kant's view is not that ‘property ownership … needs to be approved by the state, ’ but rather that the state is required to ‘provide the institutions I need to be an owner of property’ (Reference Byrd and Hruschka2010, 101–102). Similarly, Paul Guyer argues that because the transition from the state of nature to the civil condition is one in which an antecedently rightful conception of property is made durable, the state's function is merely one of ‘introduc[ing] determinate boundaries between claims’ by ‘surveying … boundaries and recording … deeds’ (Reference Guyer and Timmons2002, 62).Footnote 5
Weak provisionality interpretations accept some or all of the following claims: (1) Agents in the state of nature are authorized to use force to protect their property, just as long as that force takes the form of coercing others to join a state with them. (2) State of nature property claims lack the fullest possible legitimacy because, in the absence of a state, boundary disputes cannot be settled in accordance with the freedom of all. As a matter of empirical fact, any resolution of these disputes will likely amount to the subjection of some wills to the private wills of others. (3) The role of the state is essentially to ratify pre-existing property claims, just so long as these claims did not involve interference with another's person or (provisional) possessions.
By contrast, strong provisionality interpretations accept some or all of the following claims: (1) In the absence of a state one cannot claim ownership of an object without wronging others. That is, the state of nature lacks the conceptual conditions under which a ‘claim’ to property-rights is anything other than an unjustified application of force. Using force to protect my ‘right’ to external objects is an unjustified (though perhaps prudentially rational) wrong that entrance into the public condition corrects. For example, Katrin Flikschuh writes, ‘Kant regards the commission of an act of injustice as a necessary condition of the possible establishment of relations of justice between persons’; KPC ‘provisionally counts as permissible an action that is strictly speaking prohibited’ (Reference Flikschuh2000, 136 and 140).Footnote 6 (2) The reason property is provisional in the state of nature is not primarily because boundary disputes cannot be settled in a way consonant with the freedom of all (though, as a matter of fact, such disputes may happen). Rather, the problem is that no agent could be rightfully authorized to unilaterally impose his will on another. (3) State of nature property norms are completely formal – specifying only that the state must make some consistent legal determination of ‘mine and thine’ – which implies that the state can specify any regime of property rights consistent with the equal freedom of all, even if this involves massive redistribution of individuals’ holdings.
In sum, for weak provisionality the Kantian state is an instrument for securing a right that already exists. For strong provisionality, the Kantian state institutes a right that is otherwise absent. The Kantian state of weak provisionality respects individuals’ ex ante claims to property, and is thus libertarian in spirit.Footnote 7 The Kantian state of strong provisionality treats property as a conventional artifact, which can be redistributed as needed so as to ensure equal freedom. It is thus liberal egalitarian in spirit.Footnote 8
1.2. Problems with weak and strong provisionality
Unfortunately, weak provisionality renders mysterious why Kant says that property rights are only provisionally rights – i.e. defective or not fully actualized rights – rather than complete rights whose empirical application is merely underspecified or indeterminate. Strong provisionality renders mysterious why Kant says they are rights, as opposed to wrongs we are temporarily authorized to commit.
With respect to Kant's position in the history of political philosophy, weak provisionality attributes to Kant an essentially Lockean view. It holds that there is a natural right to property in the state of nature, albeit one that cannot be securely implemented without public institutions.Footnote 9 For example, compare Guyer's description of the Kantian state as ‘surveying boundaries’ and ‘recording deeds’ with Locke's description of the state as providing the ‘guards and fences to the properties of all members of society’ (Reference Locke and Laslett[1689] 1988, §222).Footnote 10 By contrast, strong provisionality attributes to Kant an essentially Hobbesian view, on which ‘rights’ in the state of nature do not involve genuine moral authorizations to constrain another, and are thus mere placeholders for the granting of actual rights by a sovereign authority. For example, consider Flikschuh's description of the Kantian state of nature as one in which the act of claiming property is ‘strictly speaking unlawful, ’ but prudent given that ‘political circumstances leave open no other option’ (Reference Flikschuh2000, 138). This seems quite similar to Hobbes's claim that in the state of nature there is ‘no Mine and Thine distinct; but only that to be every mans, that he can get; and for so long, as he can keep it, ’ the result of which is that the Sovereign has ‘the whole power of prescribing the Rules, whereby every man may know, what Goods he may enjoy … without being molested by any of his fellow Subjects: And this is it men call Propriety’ (Reference Hobbes and Tuck[1651] 1996, 90 and 125; emphases in original).Footnote 11 For Hobbes, as for Flikschuh's Kant, to speak of natural property ‘rights’ is simply another way of describing prudential reasons for holding on to external objects by force. There is no necessary connection between those acts of force and the regime of property rights eventually instituted by the sovereign. Juxtaposing weak and strong provisionality, we see that by collapsing provisionality either into having a right or lacking one, the critical literature represents Kant either as a Lockean natural lawyer or a Hobbesian positivist.
Additionally, both weak and strong provisionality have trouble accounting for different aspects of DR’s argument. Weak provisionality cannot adequately account for Kant's argument that property rights are not rights until and unless they are institutionally embodied. Consider for example Kant's statement that a unilateral exercise of the will, such as claiming property in the state of nature:
cannot serve as a coercive law for everyone with regard to possession that is external and therefore contingent, since that would infringe upon freedom in accordance with universal laws. So it is only a will putting everyone under obligation, hence only a collective general (common) and powerful will, that can provide everyone this assurance. – But the condition of being under a general (i.e. public) lawgiving accompanied with power is the civil condition. So only in a civil condition can something external be mine or yours. (6:256; emphasis in original)
Here, Kant clearly states that property rights can only be realized collectively. In state of nature property claims one individual restricts the freedom of another by claiming an object as his own. But such restriction could in no way be consonant with the freedom of all unless it was based on a logically prior collective authorization through public institutions that speak in everyone's name: “only in a civil condition can something external be mine or yours.”
But strong provisionality runs up against the problem that Kant also states that claims to ownership in the state of nature impose obligations on other agents. For example, he discusses a ‘permissive law of practical reason, ’ which ‘gives us an authorization … to put all others under an obligation, which they would not otherwise have, to refrain from using certain objects of our choice because we have been the first to take them into our possession’ (6:247; emphasis mine).
The obscurity of KPC, and the inadequacy of the existing interpretations of it, stem from the fact that Kant appears to hold the following two theses:
Thesis 1: There is a natural right to property that is not itself generated by public authority.
Thesis 2: Entry into the civil condition does not simply render natural property rights more secure. Rather, there is no way for property rights to be fully rightful outside of an institutional and legal specification.
In other words, Kant appears to hold both that property rights are not just posited by the state, but also that property rights are incomplete without the state and its powers. Weak Provisionality accepts Thesis 1, but denies Thesis 2. If it reckons with the institutional concerns of Thesis 2 at all, it misconstrues the matter in terms of pragmatic rather than conceptual necessity. That is, the only role weak provisionality can afford to public institutions is the pragmatic one of resolving conflicts regarding property disputes. By contrast, strong provisionality accepts Thesis 2, and so accepts that for Kant rights must be institutionally specified in order to be genuine rights. But it denies Thesis 1 in that it treats state of nature property claims as a species of wrong.
If one reads weak provisionality as asserting the rationalist claim that reason apprehends an independently existing normative order (i.e. we have natural property rights), and strong provisionality as asserting the empiricist and skeptical claim that rights are merely the product of custom or convention (i.e. we don’t have natural property rights), reconciling Thesis 1 and 2 will be an instance of Kantian critical philosophy's general attempt to unite the truth in rationalism with the truth in empiricism.Footnote 12
1.3. Varieties of strong provisionality
Faced with the interpretive problems outlined above, some strong provisionality interpreters hold an intermediate position. They reject the weak provisionality view that the state is a mere instrument for giving greater stability to natural property rights, without adopting Flikschuh's view that claiming property in the state of nature is, strictly speaking, wrongdoing. For example, following from Kant's definition of right as the authorization to coerce (6:232), Arthur Ripstein argues that for Kant ‘all rights to external objects in a state of nature are merely provisional, because they are all titles to coerce that nobody is entitled to enforce coercively’ (Reference Ripstein2009, 165).Footnote 13 And Christine Korsgaard concludes that in the state of nature property rights are ‘provisional … in the sense that we have the right to defend them, but not in the sense that anyone else has a duty to respect them’ (Reference Korsgaardforthcoming).Footnote 14
While these formulations are not inaccurate, they do not fully elucidate the murky concept of provisional right. With respect to Ripstein one wants to ask, if right is defined as an entitlement to coerce (6:232), how could a provisional right be a right, absent its essential characteristic?Footnote 15 With respect to Korsgaard one wants to ask, if I have a right to defend something, but nobody else has a duty to respect my right, then in what sense have we progressed beyond a Hobbesian war of all against all in which no genuine claim-rights exist?Footnote 16
Focusing on passages from DR in isolation from its overall argumentative structure will not settle the debate between strong and weak provisionality.Footnote 17 Certain formulations suggest that the state simply secures rights whose content is already determined in the state of nature. For example, Kant writes, ‘The juridical state is that relation of human beings among one another that contains the conditions under which alone everyone is able to enjoy his rights’ (6:305–306; translation modified; emphasis in original). Byrd and Hruschka suggest that Kant's point here is that with the state ‘you become able to exercise or to enjoy the rights you have … the state simply confirms what is already yours and secures it’ (Reference Byrd and Hruschka2010, 25). But they downplay other passages in which Kant writes that in the public condition ‘it is determined [bestimmen] by law’ (6:312) what is mine or yours. ‘Determination’ implies that the content of ownership claims is formal – i.e. stating only that mine and thine must be possible – until determined, i.e. given content, by the state.
For an illustration of this interpretive tension consider the following passage. Kant describes our duty to enter a state as the command to “enter a condition in which what belongs to each can be secured to him against everyone else” (Tritt in einen Zustand, worin jedermann das Seine gegen jeden anderen gesichert sein kann) (6:237). ‘Secured’ (gesichert) could imply either that the state simply legitimates what each individual already has, or that the state, by conferring on the person the normative status of property owner, constructs the conditions under which one can rightfully speak of ‘having’ external possessions at all.
1.4. Anticipatory provisionality
I offer a different interpretation of KPC, anticipatory provisionality, which explains why property rights are both provisional and rights. According to anticipatory provisionality, state of nature property claims do not wrong other agents, but nor are they full-fledged rights claims that simply await institutional specification. Rather, they anticipate a condition in which the authority to make such claims can no longer be unilaterally determined. The meaning of ‘anticipate’ here is two-fold. State of nature property claims create the necessary background conditions for public authority. At the same time, they draw on a public authority they lack. Although they are defective as instances of right (for reasons to be explained below), they are still rights because they are opposed to the clear wrong of remaining in the state of nature.
Let me state the view in rough outline. For Kant we are justified in making property claims in the state of nature, as long as: (1) we are prepared to enter into a state with others, which means recognizing that such an entity could in principle decide that we are not in fact entitled to all that we have claimed, and (2) we only use unilateral force to defend our property claims as a last resort if and when other agents actively prevent the formation of a state. In sum, although the state does not create our authorization to make and defend property claims, in the state of nature property rights are provisional because what they essentially are is a right to exist under a public authority that specifies a determinate property regime. With respect to that specification, Kant's view is neither that the state must merely ratify pre-existing property distributions, nor that the state can implement any property distribution it wishes in the name of equal freedom. Rather, the Kantian state can redistribute, but it must respect the prima facie validity of pre-state property holdings, a concept I explain in the final section.
2. KPC: the background
This section briefly reconstructs Kant's arguments about property prior to the introduction of provisional right in §9:
(1) Ownership, i.e. the right to an object not currently in one's grasp (6:245), is an expression or material determination of the freedom of an embodied, rational being with the capacity to exercise choice (Willkür) by setting and pursuing ends. One cannot set ends without at least some discretionary control (even if only temporary) over the means to those ends.Footnote 18 Control over means distinguishes choice, i.e. ‘consciousness of the ability to bring about [desire’s] object by one's action, ’ from mere wish (6:213).
(2) But at the same time, property right (indeed, any form of right) depends on the possibility of the joint realization of each agent's power of choice. Rights are grounded in external freedom, i.e. freedom of action insofar as our actions affect one another. External freedom is governed by the Universal Principle of Right (UPR), which requires that ‘the freedom of choice of each … coexist with everyone's freedom in accordance with a universal law’ (6:230).Footnote 19 In other words, because rights form a system of reciprocal willing, one agent's rightful expression of external freedom cannot violate the rightful external freedom of others.
(3) By (2) ownership is only normatively legitimate if the discretionary use of means does not violate UPR. The external mine and thine constitutive of choice must be consonant with the freedom or choice-making power of all.
(4) Absent a public condition, agents cannot legitimately claim ownership without in some way violating the freedom of others. In the state of nature claims to ownership arbitrarily restrict the ends others can set in a way not consistent with the equal exercise of the choice-making capacities of each. Such claims thus represent unilateral restrictions of choice (e.g. 6:261).
By (1)–(4), external freedom looks to be in contradiction with itself. It seems both to require ownership (1) and prohibit it (4). According to the ‘antinomy of propositions concerning the possibility of [ownership], ’ it both is and is not possible to ‘have something external as mine even though I am not in possession of it’ (6:255).
(5) Ownership does not conflict with UPR under a public condition, ruled by the general will and structured through symmetrically binding laws. A law-governed state provides a shared or omnilateral authorization for what originally appeared to be unilateral declarations of ownership, such that claims to ownership no longer violate UPR.Footnote 20
In sum, Kant's argument establishes that property is both required by our status as free end-setters and yet morally intolerable, because at odds with reciprocal external freedom, until and unless free beings exeundum e statu naturali. From this he concludes that UPR licenses property acquisition in the state of nature only by also compelling each agent to coerce all others to form a state.
The crucial difference between weak and strong provisionality arises with respect to (4). What is under dispute is why exactly claiming ownership in the state of nature violates UPR. Is it because without mediating institutions boundary conflicts will arise – conflicts which, given our nature as equal, free beings, can only be rightfully settled by a shared, omnilateral will [weak provisionality]? Or is it because there is something about the very act of making a property claim in the state of nature that necessarily wrongs another, irrespective of the empirical possibility of disputes about borders and boundaries [strong provisionality]? Or is it because property claims in the state of nature draw on an authority they do not fully possess, even though they seek to engender that very same authority [anticipatory provisionality]?
3. Against weak provisionality: the state of nature and the problem of private judgment
Weak and strong provisionality might initially appear to be legitimate interpretations of different aspects of Kant's account of why the state of nature is conceptually unstable. DR offers three distinct versions of KPC. Kant argues that in the state of nature: (1) agents lack appropriate assurance that others will respect their property rights and so lack both rational motivation and obligation to do so themselves; (2) there is no definitive way to decisively delimit the boundaries between what is mine and what is yours; and (3) there is no way for one agent to place another under an obligation to respect his entitlement to external objects of choice, consonant with the equal liberty of all. Following Ripstein, I call these the problems of (1) assurance, (2) indeterminacy, and (3) unilateral obligation (Reference Ripstein2009, 145–147 and passim).Footnote 21 In this section I argue that a proper understanding of the relation between (1)–(3) shows the interpretive limitations of weak provisionality.
The problem of unilateral obligation pertains most obviously to the acquisition of property, assurance to the protection of what one has already acquired, and indeterminacy to both acquisition and protection. But the three instabilities are also clearly interrelated. After all, consider a situation in which there is no morally legitimate way of drawing a secure boundary between what is mine and what is yours because there are no shared legal norms which delimit where your property ends and mine begins. I may think that I own the fruit of a tree whose branches hang over into my yard, and you may disagree on the grounds that the roots of the tree are in yours. In the absence of a common standard that decides the case, I have no rational basis to expect that you will obey what I perceive to be your obligations to respect my property rights, and you have no assurance that I will respect yours. Here, problems of indeterminacy lead to problems of assurance.
Conversely, consider a situation that lacks shared legal norms concerning which actions constitute a legitimate defense of one's property rights and which are merely illegitimate acts of encroachment on another's turf. That is, imagine that there is no way for agents to know which kinds of defensive reactions are or are not sanctioned by their right to property, and so no way to know how to distinguish between aggression and defense. In this case, agent1 may constrain his actions with respect to agent2 in accordance with agent1's understanding of right, but agent1 has no assurance that agent2 will understand his obligations in the same way. Imagine that you think that climbing ‘your’ tree, the trunk of which is located in my yard, and picking all its fruit is simply defending what is yours from my acquisitive hands, while I think that this action is simply trespass. Obviously, neither of us can expect that what we do unto others will be done unto us in return. Lacking mutual assurance, neither of us knows where what we own ends and where what you own begins, since we cannot justify to one another claims concerning legitimate and illegitimate restriction on possible use. Here, problems of assurance lead to problems of indeterminacy.
Finally, imagine a condition in which mere declarations of property cannot generate rightful ownership, because there is no reason why an agent should think that your claim on an object constrains what he can or cannot do with it. In that case, drawing boundaries between mine and yours, and establishing schemes to protect such boundaries, are activities that lack moral sense. Securing assurance and resolving indeterminacy thus depends on rendering unilateral obligation rightful.
Instability (3) is obviously the most fundamental, in that it concerns the very possibility of making moral claims to ownership. Yet some commentators mention only (1) and (2).Footnote 22 But this gives the misleading impression that state of nature property claims are perfectly conceptually intelligible, albeit unstable. The result is a construal of the Kantian state of nature on which the practical life of pre-political agents already embodies an understanding of what it would mean to claim ownership of an external object of choice such as a tree or piece of fruit; it is just that such agents have no way to ensure that their claims are durable in the face of challenges.
By contrast, if one sees (3) as the main instability, one will favor strong provisionality. On this construal of Kant's state of nature, prior to a public condition agents have no genuine ground for their property claims, since they have no grounds to conclude that their particular acts of ‘obligating’ another to respect their property claims actually entitle them to anything more than the right to their own bodies, a point I will return to below. Insofar as such agents actively enforce their ‘rights’ to things not currently attached to their bodies, they are merely applying illegitimate coercive force. The situation marked by instability (3) is thus one in which agents in the state of nature make property claims, but absent the requisite justificatory grounds.
§8 shows why underpinning the problem of assurance is the more fundamental problem of unilateral obligation. Kant writes:
When I declare (by word or deed), I will that something external is to be mine, I thereby declare that everyone else is under obligation to refrain from using that object of my choice, an obligation no one would have were it not for this act of mine to establish a right. This claim involves, however, acknowledging that I in turn am under obligation to every other to refrain from using what is externally his; for the obligation here arises from a universal rule having to do with external rightful relations. I am therefore not under obligation to leave external objects belonging to others untouched unless everyone else provides me assurance that he will behave in accordance with the same principle with regard to mine. This assurance does not require a special act to establish a right, but is already contained in the concept of an obligation corresponding to external right, since the universality, and with it the reciprocity, of obligation arises from a universal rule. – Now, a unilateral will cannot serve as a coercive law for everyone with regard to possession that is external and therefore contingent, since that would infringe upon freedom in accordance with universal laws. So it is only a will putting everyone under obligation, hence only a collective general (common) and powerful will, that can provide everyone this assurance. – But the condition of being under a general external (i.e. public) lawgiving accompanied with power is the civil condition. So only in a civil condition can something external be mine or yours. (6:255–256)
Weak provisionality interpreters such as Guyer understand Kant's claim about assurance – ’I am therefore not under obligation to leave external objects belonging to others untouched unless everyone else provides me assurance that he will behave in accordance with the same principle with regard to what is mine’ – in terms of rational motivation. They think that Kant's point is that in the absence of mutual awareness of an enforcement agency, individuals are likely to defect from a cooperative scheme of property rights. Guyer writes:
[S]ince no one can reasonably expect to enjoy a claim to property unless others are also allowed to do so as well, but also that no one can reasonably be expected to confine his claims to his own property unless others can also be expected to do so, a system for the public enforcement of the boundaries of property claims is as necessary as a public system for defining them. Thus the office of the sheriff is as basic to the state as is that of the recorder of deeds. (Reference Guyer and Timmons2002, 62)
Guyer is certainly not wrong to think that without the assurance provided by the state's coercive laws, individuals have no rational basis to respect the property claims of others. What he does not see, however, is that this issue of rational motivation is a surface manifestation of a deeper conceptual problem. We encounter problems of assurance because, at bottom, without the state we can only accidentally share normative principles concerning property rights, and this returns us to the problem of unilateral obligation. Let me explain further.
As we have seen, Kant fundamentally understands a property right as a coercive authority vis-à-vis other free agents to defend what is one's own. The key question of §8 is how one externally free agent could assume such coercive authority over another by imposing obligations on her, consonant with the equal freedom of all.Footnote 23 Without an answer to this question, making a property claim would be an illegitimate arrogation of sovereign authority by an individual, a form of private domination. Kant's answer is that if an order of equal freedom is possible, declaring an object as mine must place me and all other agents under a reciprocal nexus of obligations. That is, I can only impose an obligation on another free agent if that very same act places me under symmetrical obligations with respect to her.
There is a temptation to think that one could constitute such a shared normative order by aggregating individual acts of obligation-imposition: in one act, I impose an obligation on you; in another act, you impose an obligation on me, and so on with respect to the entire network of dyadic interactions. But Kant's insight is to see that this will not solve the problem of unilateral obligation. Why not? Imagine that I impose an obligation on you to respect my property, and you impose an obligation on me to respect yours. If agents in such a condition do in fact respect each other's property, each is doing so merely out of his own good will, i.e. out of his own determination (and thus by his own capacity for practical reason) that he ought to fulfill his obligation to respect the property of others. This means that whether or not any agent is secure in his property depends on the good will of other agents, and vice-versa. But bi-lateral dependence on the good will of another is just a restatement of a condition of private domination. Moreover, what underlies this apparent relation of right is the merely accidental fact that our good-wills happen to converge on the same content.
Why exactly does the state of nature lack a shared normative order by which our rights can be realized together? I propose that the deepest instability of the Kantian state of nature, underlying even the problem of unilateral obligation, is the problem of private judgment.Footnote 24 Kant writes:
[H]owever well disposed and law-abiding human beings might be, it still lies a priori in the rational idea of such a condition (one that is not-rightful) that before a public lawful condition is established individual human beings, peoples, and states can never be secure against violence from one another, since each has his own right to do what seems right and good to it and not to be dependent on another's opinion about this. (6:312)
It is necessary to hear this passage in the properly a priori key. Pace Guyer, for Kant the reason we are not ‘secure against violence’ in the natural condition is not because human beings are very likely to disagree. It is because even if they agree and so think they share a moral order, this ‘order’ is simply a loose association of wills with overlapping content, propped up by an aggregation of individual judgments that such an order is worth sustaining.
The fragility of such an order comes to the fore in cases of conflicts about right. Kant writes, ‘when rights are in dispute (ius controversum), there would be no judge competent to render a verdict having rightful force’ (6:312). But this potential for dispute is the effect, rather than the cause, of the lack of non-accidentally shared norms. The necessity of an omnilateral enforcement agency follows from the necessity for principles to be grounded in a collective will; it involves no further thoughts about rational psychology. This is what Kant means when he writes that the reciprocity necessary for property rights ‘is already contained in the concept of an obligation corresponding to an external right’ (emphasis mine).
My reading of §8 shows why weak provisionality cannot be correct. Kant's aim in this paragraph is to show that property rights have no actuality without institutional embodiment. In the Kantian state of nature agents may make judgments concerning property norms and their application. But they lack the shared standards necessary to render such judgments objective determinations (e.g. ‘this is right’) rather than mere subjective preferences (e.g. ‘this seems right to me’). Without a shared normative order, any putatively symmetrical agreement is grounded only in independent and contingent acts of private will.
4. Against strong provisionality: the postulate and the permissive law
As we saw in Section 2, Kant argues that in the state of nature external freedom is in internal contradiction with itself. The finite, embodied nature of external freedom seems to require ownership of external objects of choice – or at the very least to license it – yet such ownership thwarts the equal external freedom of others. Kant's ‘Postulate of Practical Reason with Regard to Rights’ offers a preliminary resolution of the internal contradiction of freedom by generating a provisional entitlement to property. The ‘permissive law of practical reason’ authorizes the final resolution of the contradiction by allowing each agent to coerce all others to exit the state of nature, thereby rendering provisional entitlement conclusive.Footnote 25 In this section I explain both the contradiction and its resolution in order to show why strong provisionality, despite its initial plausibility as an interpretation of §8, is incorrect: KPC is not that claiming property in the state of nature wrongs other agents.
4.1. The postulate
Prior to the introduction of the postulate, Kant has argued that our right to external freedom implies a right to non-interference with our bodies (6:247). For Kant there is no ambiguity of right involved in my struggling to prevent you from wrenching an apple out of my hand. In taking the apple from my grasp you wrong my body, a wrong that Kant thinks is determinate and conclusive prior to any regime of public law. For the purposes of our concerns here, let us simply grant this assumption.Footnote 26
The postulate extends my normative entitlement to my body to objects in the world. It states, ‘It is possible for me to have any external object of my choice as mine, that is, a maxim by which, if it were to become a law, an object of my choice would in itself (objectively) have to belong to no one (res nullius) is contrary to right [rechtswidrig]’ (6:246; emphasis in original). In other words, the postulate establishes that there can be no absolute prohibition on property. Kant then argues that the possibility of property somehow generates an obligation on the part of all agents to respect each other's property claims: ‘It is … an a priori presupposition of practical reason to regard and treat any object of my choice as something which could objectively be mine or yours’ (6:246), and a corresponding duty ‘to act towards others so that what is external (usable) could also become someone’s’ (6:252). I treat the postulate in this section and the permissive law in the next.
To understand Kant's argument for the postulate, consider the conceptual structure of purposive activity. Many kinds of complex end-setting activity necessitate continual access to the material objects that are the means to that end.Footnote 27 For example, I could not execute my end of building a chair if I am continually preoccupied with protecting my dominion over the wood and hammer. But such reflections on the necessary role of possession in human end-setting activity are not sufficient to establish that enduring, unrestricted access to objects is necessarily rightful. The claim that unrestricted access to objects follows from the nature of human purposive activity only sets reason its task: to search for a possible legitimating ground for the rightfulness of such access.Footnote 28
So Kant's point cannot be that we have a right to exclusive use of external objects of choice because this is what we require in order to realize our embodied practical nature.Footnote 29 To argue in this manner would admit considerations of human need into the structure of right. But Kant clearly states that right ‘does not signify the relation of one's choice to the mere wish (hence also to the mere need) of the other’ (6:230). Indeed, the entirety of Kant's practical philosophy has ‘not nature but freedom of choice itself for its object’ (6:216). Accordingly, Kant's justification of the postulate must be grounded in UPR, which pertains only to the form of the relations between free agents and concerns the authorization to coerce. The fact that I need something to realize my own agency is an internal (monadic) property of the will rather than a relational (dyadic) property of its interactions, and so cannot generate a coercible duty for others (6:230).Footnote 30
By contrast, if UPR is the driving force of the argument, then, whether or not unrestricted access to property is necessary for the realization of purposive agency, such access is illegitimate unless it is compatible with the external freedom of others. The reason that there can be no absolute prohibition on property, i.e. ‘putting usable objects beyond any possibility of being used’ (6:251), is that a condition is possible under which ‘in the use of things choice [is] formally consistent with everyone's outer freedom in accordance with universal laws’ (6:251).
Ultimately, the postulate assumes a possibility. It says: Let us suppose that we can rightfully possess external objects of choice. It asks: What conceptual conditions render this assumption legitimate?Footnote 31 It answers: Such claims would be illegitimate if they are incompatible with UPR. But making property claims does not conflict with UPR under the condition that we enter into the civil state. As Ernst Weinrib helpfully writes, the postulate ‘allows us provisionally to hold the notion of external property in place until the thought of it can be completed in a further phase’ (Reference Weinrib2003, 277).
Although the postulate merely ‘holds’ the rightful possibility of property in place, it establishes two necessary features of any property regime that constrain the state's law-making activities. First, to have a right to property is to have a right to something that is not merely a right to one's body. Having a right to property entails that I can be wronged if you interfere with an object that is within my discretionary control, even if that object is not currently attached to my body (e.g. even if I am not currently holding my apple or wearing my sweater). Second, a property right is the right to enduring control over an object. Kant need not hold that the conceptual structure of property claims necessarily implies that in having an object now I must rightfully have permanent control over it. Nevertheless, it is part of the Kantian concept of property that I do not constantly have to defend my claim. Imagine, for instance, a regime of public law which declared that in order to have a right to land I must actually be on the land at all times. In this case, what is doing the normative work in prohibiting you from taking the land is simply my right not to have my body harmed, where it is currently located. Insofar as you enter my land without making contact with my body, you do me no wrong. By Kantian lights, such a regime would be illegitimate because it involves ‘putting usable objects beyond any possibility of being used.’
4.2. The permissive law
The postulate states that since rightful ownership may be possible, we should act as though it is. The permissive law of practical reason authorizes agents to use coercion to create the condition under which that possibility can be made actual. The permissive law
gives us an authorization that could not be got from mere concepts of right as such, namely to put all others under an obligation, which they would not otherwise have, to refrain from using certain objects of our choice because we have been the first to take them into our possession. (6:247)
The permissive law entitles agents in the state of nature to coercively protect their provisional acquisitions:
[T]he possibility of acquiring something external in … a state of nature … is a principle of private right, in accordance with which each is justified in using that coercion which is necessary if people are to leave the state of nature and enter the civil condition, which can alone make any acquisition conclusive. (6:264; emphasis mine)
Prior to the civil condition … external objects that are mine or yours must therefore be assumed to be possible, and with them a right to constrain everyone with whom we could have any dealings to enter with us into a constitution in which external objects could be mine or yours. (6:256; emphasis mine)
Key to a proper understanding of KPC is Kant's argument that in the state of nature using force to protect my property claims is only rightful if it takes one particular form: coercing others to join a state with me. But before turning to this idea, it is worth noting that in the passages just cited Kant clearly denies strong provisionality. Coercively defending my property in the state of nature is not a wrong to be rendered rightful. Rather, Kant stresses that agents are ‘justified’ in deploying coercion in the state of nature. As long as they are forcing people into the state; they have a ‘right to constrain.’Footnote 32
It is natural to read these passages as providing the following argument in support of weak provisionality: What authorizes agents to coerce others to exit the state of nature can only be a right that they already have, so what renders such coercion legitimate is their pre-existing right to property. Thus, the state merely gives greater shape and specificity to natural right.
5. Anticipatory provisionality
Yet the authorization to coerce people to leave the state of nature and enter the civil condition is unlike any other authorization. As we saw in Section 3, in the civil condition each individual forgoes his right to private judgment and submits to an authoritative determination of right expressed through law. In the case of property, this means that the ultimate decision as to mine and thine rests firmly with the state (6:323). So the authorization to coerce others to enter into the public condition is an authorization to coerce others into a condition in which I no longer have unilateral coercive authority. As Kant writes:
The way to have something external as one's own in a state of nature is physical possession which has in its favor the rightful presumption that it will be made into rightful possession through being united with the will of all in a public lawgiving, and in anticipation (Erwartung) of this holds comparatively as rightful possession. (6:257)
This does not mean that my original act was a wrong that must be righted. But it does indicate that my state of nature right to make property claims has a curious form. When Kant suggests here that provisional rights are only ‘comparatively’ rightful, he indicates that KPC is a form of explanation meant to elucidate the logic of defect: provisional rights are defective instances of rights, without for all that being wrongs.Footnote 33
Understanding KPC as an illustration of defective right accords with Kant's general strategy of philosophical explanation. In the Critique of Pure Reason’s (CPR) “Doctrine of Method” Kant writes:
Philosophy is full of faulty definitions, especially of definitions which, while indeed containing some of the elements required, are yet not complete. If we could make no use of a concept till we had defined it, all philosophy would be in a pitiable plight. But since a good and safe use can still be made of the elements obtained by analysis so far as they go, defective definitions, that is, propositions which are properly not definitions, but are yet true, and are therefore approximations to definitions, can be employed with great advantage … It is desirable to attain an adequate definition, but often very difficult. The jurists are still without a definition of their concept of right.Footnote 34
In other words, Kantian philosophy does not begin with the full definition of any of the concepts it aims to elucidate. Rather, it begins with an incomplete definition that gradually acquires greater determinacy as the inquiry proceeds. Why this must be so is a subject for another paper.Footnote 35 All I wish to signal here is that Kant indicates that the concept of right must be subject to a form of analysis in which a “defective definition” is gradually corrected.Footnote 36 The final sections of this paper attempt to explain why provisional right is defective.
5.1. State of nature property claims: the correction of a defect
In §16 Kant writes that in the state of nature there is a ‘rightful capacity of the will to bind everyone to recognize the act of taking possession and of appropriation as valid, even though it is only unilateral’ (6:267; emphasis in original). But if Recht is defined as reciprocal, rather than unilateral, coercion (6:231), then what could it mean to have a unilateral right to coerce? Moreover, while §16 says that we have such a unilateral right, §9 states that ‘merely unilateral’ possession is unlawful, since lawful possession ‘can be found only in a general will’ (6:257).
Further attention to §16 may help resolve the appearance of contradiction:
Provisional acquisition … needs and gains the favor of a law (lex permissiva) for determining the limits of possible rightful possession. Since this acquisition precedes a rightful condition and, as only leading to it, is not yet conclusive, this favor does not extend beyond the point at which others (participants) consent to its establishment. But if they are opposed to entering it (the civil condition), and as long as their opposition lasts, this favor carries with it all the effects of acquisition in conformity with right, since leaving the state of nature is based upon duty. (6:267)
To make sense of this passage we need to distinguish more clearly than Kant does between two different kinds of property claims in the state of nature. One represents a defective form of a rightful property claim, i.e. a partial actualization of right, the other is a straightforward wrong. In the first case, to claim property in the state of nature just is to be unilaterally authorized to use coercive force because I am aiming at a condition in which I no longer have unilateral authority. As Kant writes, the act of acquiring property must be ‘in conformity with the idea of a civil condition, that is, with a view to it and to its being brought about, but prior to its realization’ (6:264). In the second case, other agents shirk their duty to enter the state. Here, I still have unilateral authorization to defend my rights claim, but that is only because other agents have committed a moral wrong by opting to remain in the state of nature. I can rightfully impose a unilateral obligation in response to a unilateral display of force. As Kant writes: ‘a subject who is ready for [the public condition] resists with right those who are not willing to submit to it and who want to interfere with his present possession; for the will of all others except for himself, which proposes to put him under obligation to give up a certain possession, is merely unilateral’ (6:257; emphasis in original). In defending my property, I am not arrogating a power to myself that I deny to others – this would clearly conflict with the logic of right – but defending myself against a moral wrong.
But why exactly is this a moral wrong? Recall that the postulate licenses the extension of my bodily right to external objects of choice. In so doing, it thereby also expands the class of wrongs. So just as I can protect my body against coercive threats, I can protect my property against the wrong done to me by agents who refuse to form a state.
Based on these passages, as well as the overall movement from §8 to §16, anticipatory provisionality understands KPC as follows: when all agents are acting in such a way as to bring about a public condition, property claims are made legitimate by the law they aim to constitute. When others thwart the public condition, your unilateral property claim is justified because it is a form of defense against wrongful activity.Footnote 37 You have a right to make property claims in the state of nature (pace strong provisionality). But your right to make such claims is not the right to have your property simply confirmed by the state (pace weak provisionality). Rather, it is a right to a public determination of what the norms of individual ownership amount to.
In sum, KPC reveals that the full actualization of right has two components: (1) A moral authorization to coerce that (2) requires systemic interaction between wills. A provisional right to property is a static way of marking this normative progression. KPC shows that there are ways of properly moving from 1 to 2 and ways of arresting that sequence: in the state of nature one can acquire property in a way that aims to establish more complete conditions of right, or one can acquire property in a way that thwarts those conditions.
5.2. ‘Doing wrong in the highest degree’
Nevertheless, a conceptual worry remains. The interpretation advanced above says that by refusing to enter into a state with me, other agents wrong me in the use of my property. Thus, I have a right to coercively defend my property as a form of self-defense. But if I do not fully have a right to property until we enter a state, then how could an agent in the state of nature wrong me by attacking my property?
I do not think that Kant provides a fully satisfactory response to this question. However, there is an underdeveloped idea in DR that may be of help. Kant writes:
Given the intention to be and to remain in this state of externally lawless freedom [i.e. the state of nature], human beings do one another no wrong at all when they feud among themselves, for what holds for one holds also in turn for the other, as if by mutual consent … But in general they do wrong in the highest degree by willing to be and to remain in a condition that is not rightful, that is, in which no one is assured of what is his against violence. (6:307–308; emphasis mine)
It is difficult to see how Kant can make room for a form of wronging that is not a wrong against a specific person.Footnote 38 Wronging is defined as correlative to right (6:231), and right inheres ‘only [in] the external and indeed practical relation of one person to another’ (6:230; emphasis mine). Nevertheless, there is something intuitively plausible about the thought that other agents wrong me indirectly. They do me wrong by perpetuating a condition in which a rightful resolution of conflict is impossible.Footnote 39 In the state of nature each of us is authorized to use violence to secure what is ours, but we are also to be held morally accountable for not seeing that there was another option: instituting a condition under which violence was no longer necessary.
6. Conclusion: property redistribution in the Kantian state
What exactly is the Kantian state's attitude towards private property? Once again, attention to passages from DR is not sufficient to answer this. On one hand, Kant writes that ‘provisional acquisition is true acquisition’ (6:264), suggesting that one has a claim on whatever one has acquired in the state of nature. On the other hand, Kant argues that the state can take away property from the rich, corporations, and the church if these groups come to wield disproportionate social power. In general, the Kantian state has the power to tax “the wealthy to provide the means of sustenance to those who are unable to provide for even their most necessary natural needs” (6:326).Footnote 40 Kant writes that the sovereign is the ‘supreme proprietor’ of the land, since he has the ‘right to assign to each what is his’ (6:324). Moreover, Kant argues that only in a public condition can ‘each can be assigned conclusively what is his’ (6:341). If the form of the state changes, ‘someone who thereby loses his title and precedence cannot say that he was deprived of what was his, since he could call it his only under the condition that this form of state continued’ (6:370).
Weak provisionality tends toward a libertarian view on which the state merely secures whatever property individuals have already amassed.Footnote 41 By contrast, strong provisionality tends toward an egalitarian view on which the state can redistribute property if disparity in holdings leads to asymmetrical forms of sociopolitical power that are inconsistent with equal external freedom.Footnote 42 On these two views, Kant's position is either that the state merely ratifies inequality, or that the very notion of private property is a formal notion up for continual redefinition by courts and legislatures.
With respect to property redistribution, anticipatory provisionality suggests that while the Kantian state can redistribute, it must operate with a prima facie assumption in favor of pre-state holdings: ‘Possession in anticipation of and preparation for the civil condition, is provisionally rightful possession’ (6:257; emphasis in original). It makes clear that UPR is the single normative principle running throughout each stage of Kant's argument, acquiring different determinations. The very same principle that explains: (1) why we have a right to property also explains (2) why that right can only be realized in a state, and also explains why (3) the state can redistribute that property if my ownership threatens to subject others to my will. But if your provisional right to property prior to the state in genuinely a provisional right, then (4) it is a right to enter a civil conditions that is constrained in various ways. For one, there is a prima facie assumption in favor of what people have already claimed, unless the state can provide convincing reasons, grounded in the same principles that licensed acquisition in the first place, that the current distribution of property threatens external freedom. So if (5) an individual feels a government taking to be unlawful, they have a right to adjudication of the dispute in the courts or legislatures, and possibly even to some reasonable degree of compensation.
More concretely, the Kantian approach might explain how the principle that grounds redistribution, also grounds the Fourteenth Amendment's Due Process Clause (no state shall ‘deprive any person of life, liberty, or property, without due process of law’) and the Fifth Amendment's Takings Clause (‘nor shall private property be taken for public use, without just compensation’).Footnote 43 Of course, prior to actual judicial and legislative decisions, there is no way to decide whether a government taking was unlawful, or what reasonable compensation amounts to.
In addition to according with constitutional doctrine, the notion of a prima facie entitlement allows us to preserve two attractive thoughts, both present in DR: (1) The state ought to redistribute if disparity in possession leads to situations of domination, i.e. cases in which one person is subject to the arbitrary power of another (e.g. 6:326).Footnote 44 (2) The state ought to respect the pre-existing property claims of indigenous groups who live in proximity to it, for example by not settling indigenous lands without express permission (6:353).Footnote 45
Let us put 1 and 2 together and generalize somewhat. There is a burden on the Kantian state in favor of claims made by members of disenfranchised groups who had significant pre-state holdings—e.g. indigenous peoples, those living in occupied territories, and racial minorities suffering from a history of land dispossession. Members of other disenfranchised groups—including refugees, guest workers, and recent immigrants—may be entitled to increased protections on the basis of their claims that a public condition has been unable to implement a system of right consistent with the freedom of all.Footnote 46 With respect to both groups, the measures of possible redress are diverse. They include reduced tax burdens and perhaps even reparations. Ultimately, anticipatory provisionality's vision of the Kantian state broadens one's sense of the forms of disempowerment it ought to redress. This is surely as salient an argument as any in favor of the interpretation.
Acknowledgements
Versions of this paper have been presented at: University of Leipzig, Sciences Po Bordeaux, the American Philosophical Association, Northwestern University, the New School for Social Research, Princeton University, and the University of Chicago. I thank audiences there for discussion. I have been fortunate to have had helpful conversations on this material with Lucy Allais, James Conant, Timothy Waligore, JP Messina, Ben Laurence, Anton Ford, Clinton Tolley, and Daniel Smyth. For detailed comments on written drafts I would like to thank first and foremost Frederick Neuhouser, who offered line-by-line comments, as well as Reed Winegar, Aaron Jaffe, Robert Pippin, Helga Varden, Daniel Brudney, Martha Nussbaum, and the anonymous reviewers at the Canadian Journal of Philosophy. Special thanks to Martin Stone for monthly conversations on Kant's Doctrine of Right, from which I have profited enormously. Some of the ideas in this paper should be regarded as the products of our joint labor, though there is no doubt that Stone would disagree in many places with my way of putting things. We explore Kant's notion of provisional right in more detail in a co-authored work in progress. I have had help with research and editing from three undergraduate research assistants: Aaron Cooper-Lob, Samuel Wohlforth, and Humberto Martinez.
