1. Introduction
Kant asserts that property right is merely secured in the civil condition (MM §8, 6: 255; §9, 257; §10, 259; §14, 263; §16, 267).Footnote 1 Consequently, there can be no conclusive juridical possession prior to the establishment of a civil condition. However, Kant also introduces the notion of provisional possession in the state of nature, describing it as possession ‘in anticipation of and preparation for’ rightful possession (MM §9, 6: 257). The key to understanding this concept lies in clarifying what Kant means by ‘preparation’ and ‘anticipation’.
Previous commentators have observed that the formation of intelligible possession is historical, maintaining that it originates from empirical possession based on priority in time. They further suggest that this historical process must be rationalized to attain normativity. However, they overlook a fundamental point: achieving this rationalization requires an abstraction from all empirical conditions, and this abstraction is essential to preserving the historical dimension of possession. This abstraction is made possible through a specific conceptual tool that serves as a bridge – the Pure Concept of Understanding of Possession (der reine Verstandesbegriff eines Besitzes).
This paper offers a new interpretation of provisional possession by drawing on the long-overlooked §7 of the Doctrine of Right. I argue that provisional possession should be understood in light of the pure concept of understanding of possession. This concept functions as a mediating type between empirical and intelligible possession, enabling the application of a certain concept of reason to empirical objects.
In section 2, I address previous research regarding provisional possession. In section 3, I provide an overview of the structure of Kant’s theory of property, emphasizing that one of Kant’s main philosophical concerns is to dispel the misunderstanding that property is grounded in empirical conditions. In section 4, I reconstruct the problem, arguing that the key to understanding provisional possession lies in explaining the transition from empirical to intelligible possession. In section 5, I elaborate on this approach, demonstrating that provisional possession comprises two components: an empirical title grounded in original possession in common and the pure concept of understanding of possession. In section 6, I explore the possibility of interpreting the pure concept of understanding of possession as a type. Finally, section 7 will provide conclusions and reflections.
2. Provisionality as rationalizationFootnote 2
Rafeeq Hasan (Reference Hasan2018), James Messina (Reference Messina2019: 456; Reference Messina2021: 77), and Helga Varden (Reference Varden, Gomes and Stephenson2024: 417–8) provide similar summaries of the scholarly debate on provisionality in Kant’s theory of property. Hasan’s summary serves as a representative example. According to him, there are two main approaches to understanding provisionality. One group of scholars (e.g., Sharon Byrd and Joachim Hruschka Reference Byrd and Hruschka2010; Paul Guyer Reference Guyer and Timmons2002) advocates for a weak notion of provisionality, arguing that ‘agents in the state of nature can make rightful claims’ (Hasan Reference Hasan2018: 852). On this view, the weaker the provisionality, the closer it comes to conclusiveness, implying that property rights almost exist in the state of nature, with the state functioning merely as a tool to defend these rights. This perspective is reminiscent of Locke’s theory of property. In contrast, another group (e.g., Katrin Flikschuh Reference Flikschuh2000; Alan Brudner Reference Brudner2011; Wolfgang Kersting Reference Kersting1983) supports a strong notion of provisionality, maintaining that the juridical state does not merely defend property rights but is integral to their determination – a view that aligns more closely with Hobbes. Hasan rightly suggests that a reconciliatory stance is plausible, one that understands provisionality as neither too strong nor too weak. He argues that, on the one hand, provisional possession must have juridical significance, and some aspects of natural conditions must be respected by conclusive possession. On the other hand, provisional possession must inevitably transition to juridical possession.
Hasan is correct in recognizing that, in the state of nature, provisional possession entails a normative element – though not rightful possession – which must be respected. This recognition urges us to focus on how rightful possession is formed, namely, by tracing the history of rightful possession in the state of nature. By this, we mean: that ‘justice allows persons to hold what they hold depends (in part) upon how they came to hold it’ (Messina Reference Messina2021: 73).
There are different accounts of the history of rightful possession. For instance, Stone and Hasan outline a two-stage process, arguing that provisionality lies in a second step – ‘acquiring’ rather than ‘having’ (established by the postulate) – which signifies the impossibility of pure private right in the state of nature (Stone and Hasan Reference Stone and Hasan2022: 94ff). In contrast, Messina maintains that provisionality is already present in a first step, namely, following from the postulate of practical reason (Messina Reference Messina2021: 72ff). I will evaluate these perspectives in detail later. For now, we can identify at least two points of consensus: first, the historical process must begin with a special kind of unilateral empirical possession – that is, possession based on priority in time and authorized by the postulate (Messina Reference Messina2021: 74; Stone and Hasan Reference Stone and Hasan2022: 68); second, respecting this history requires accounting for how normative factors develop from the initial act of possession. In other words, a process of rationalization is necessary (Messina Reference Messina2021: 76).
However, some commentators either overlook or at least fail to explicitly acknowledge that provisionality must be understood from two points of view. That is, we must not only outline how provisional possession lacks the full status of intelligible possession but also demonstrate how it contains normative elements in contrast to mere empirical possession. According to this latter suggestion, the development of possession should be understood as moving from unilateral empirical possession to the provisional possession and finally to the intelligible possession.
Moreover, to my knowledge, all commentators have overlooked a crucial aspect of the problem: the process of rationalization must be realized through abstraction rather than incorporation. As Kant explicitly states in the Doctrine of Right, this should act as a guiding principle:
For (as was established in the Critique of Pure Reason) in an a priori theoretical principle an a priori intuition would have to be put under the given concept; and so something would have to be added to the concept of possession of the object. However, in this practical principle one proceeds conversely, and all conditions of intuition which establish empirical possession must be removed (disregarded), in order to extend the concept of possession beyond empirical possession and to be able to say: any external object of my faculty of choice can be counted among the rightfully mine if (and only insofar as) I have control of it without being in possession of it. (MM §6, 6: 251–2)
As this passage illustrates, Kant imposes seemingly paradoxical requirements on rationalization: on the one hand, a special kind of empirical possession must be considered, while on the other hand, empirical conditions must be radically removed to extend possession beyond mere empirical possession. I argue that these paradoxical requirements can be perfectly reconciled by introducing a special conceptual tool as a bridge – the Pure Concept of Understanding of Possession.
Before digging into this argument, however, I will first attempt to outline Kant’s overall account of property in my own terms.
3. Setup: two features of property
3.1 Intelligible versus empirical possession
We must first examine some fundamental concepts in Kant’s theory of property, as these will provide the necessary foundation for understanding the nature of normative property in his framework. The beginning lies in the critical distinction between empirical and intelligible possession:
The expression “an object is external to me” can, however, mean either that it is an object merely distinct from me (the subject) or else that it is also to be found in another location (positus) in space or in time. Only if it is taken in the first sense can possession be thought of as rational possession; if taken in the second sense it would have to be called empirical possession. – Intelligible possession (if this is possible) is possession of an object without holding it (detentio). (MM §1, 6: 245–6)
Regarding empirical possession,Footnote 3 Kant uses the expressions ‘constraint’ (MM §4, 6: 248), ‘holding’ (§2, 6: 246), and ‘continuous occupation’ (§7, 6: 254). This type of possession is based on spatial coexistence or temporal simultaneity. For instance, one may physically hold an apple in hand, or there may be simultaneity between one person’s faculty of choice and another’s performance of an action through that faculty (§4, 6: 247–8)Footnote 4 . In summary, empirical possession depends on empirical conditions, namely, those of space and time (§4, 6: 247–8; §7, 253–4) and constitutes a form of ‘holding’.
However, this concept is insufficient for defining property. For example, if you are not physically holding your bicycle, it would still be unlawful for someone to ride it without your permission. Thus, defining property requires a ‘merely rightful connection of the subject’s will with that object, independent from any relation to it in space and time’ (MM §7, 6: 253–4). This constitutes a negative definition, emphasizing that the normativity of property arises from its independence from empirical conditions. Kant refers to this as intelligible possession. Unlike empirical possession, intelligible possession has no corresponding sensible intuition (MM §6, 6: 252), but it can be easily understood through the above example.
One point worth noting, relevant to our argument later. Kant distinguishes innate from acquired right:
Of rights, as (moral) capacities for putting others under obligation, i.e., as a legal ground (titulum) for doing so. The highest division of these is into innate and acquired right. The former is that which belongs to everyone by nature, independently of any juridical act; the latter is that for which such an act is required.
The innate mine and thine can also be called the inner mine and thine (meum vel tuum internum); for the external must always be acquired. (MM, 6: 237)
The concept of innate right refers to an individual’s right by nature to freedom, equality, self-mastery (MM, 6: 237–8). This right pertains to what is considered internal mine and thine. In contrast, acquired rights require a juridical act – such as a private claim, contract, or law (MM §10, 6: 260) – to determine what constitutes a case of external mine and thine. Property rights, therefore, fall under the category of acquired rights. For example, a cup, as an external object, is not by nature my external possession. The juridical possession of this cup must be established through complex juridical acts (initially through my private claim, but followed – as I will be arguing – by a promise made collectively).
Kant further maintains that a priori claims regarding empirical possession are analytic: when someone physically holds an object, any seizure of that object by another person actually violates his right to control his own body, which follows analytically from his innate right to freedom. By contrast, a priori claims regarding intelligible possession are synthetic: even if someone does not physically touch another person’s body, they can still violate his acquired right to an external object as an external mine (MM §4, 6: 247–8; §6, 249–50)Footnote 5 . Thus, intelligible possession, as an acquired right, extends beyond the innate right to one’s person (MM §6, 6:250).
3.2 Relationship among persons versus subjects and external things
Kant’s definition of intelligible possession draws our attention to will:Footnote 6
Now, the leaving out or disregarding (abstracting from) these sensible conditions of possession as a relation of the person to objects that have no obligation, is nothing other than the relation of a person to persons, to bind them all, with regard to the use of the things, by the will of the first person, insofar as his will conforms with the axiom of outer freedom with the postulate of the capacity [to use external objects of the faculty of choice], and with the lawgiving of the will of all thought as a united a priori. (MM §17, 6: 268)
Strictly speaking, the notion of a right to a thing, as so phrased, can be somewhat misleading. It appears to suggest that external objects are themselves under obligation to their possessor, which is absurd since external things, lacking will, cannot bear obligations. Instead, what Kant describes is not a right to a thing but rather a right against all other persons with respect to that thing (MM, 6: 241; §11, 260–1). This obligation is nothing more than the universal promise made by others not to use one’s property without permission. This formulation constitutes a positive definition, which emphasizes omnilaterality in imposing obligations as the source of normativity.
This definition raises a fundamental question: Why does property involve the will of all others? Kant addresses this through three key concepts: the postulate of practical reason, the concept of permissive law, and the idea of original possession in common. These three concepts collectively characterize a shared foundational condition (Messina Reference Messina2021: 75; Byrd and Hruschka Reference Byrd and Hruschka2006: 272ff). The postulate goes as follows:
Therefore it is an a priori presupposition of practical reason to regard and treat any object of my faculty of choice as an objectively possible mine or thine. (MM §2, 6: 246)
The key word here is ‘possible’.Footnote 7 The postulate merely renders external things attainable (Stone and Hasan Reference Stone and Hasan2022: 73) rather than fully determining any private act of possession. Kant explicitly states that this postulate does not presuppose any individual’s faculty of choice but merely enables one to think of something as an object of one’s faculty of choice (MM §2, 6: 246). In other words, the postulate establishes the possibility of possessing external things, but at this stage, no act of private possession is yet determined. For Kant, private property is inherently exclusive,Footnote 8 meaning that no one else may use another’s property without the owner’s permission. However, since all agents are initially authorized to use external things, it follows that when an individual seeks to incorporate an external object into their possession, they must necessarily engage with the wills of all others.Footnote 9
Messina (Reference Messina2021) proposes a stronger reading of the postulate, arguing that it can generate a kind of general will and thereby rationalize a unilateral act of possession. His primary textual evidence comes from the following passage:
In this way, for example, taking possession of a separate piece of land is an act of the private faculty of choice, but without acting only on one’s authority. The possessor bases his act on the innate common possession of the surface of the earth and on the general will corresponding a priori to it, and permitting private possession on it (since otherwise, unoccupied things would in themselves and according to a law be made things that belong to no one); and originally acquires, through first possession, a definite piece of land, resisting with right (iure) anyone else who would prevent him from making private use of it. Yet since he is in a state of nature, he cannot do so through the law (de iure) because there does not exist any public law in this state. (MM §6, 6: 250)
Messina interprets this passage as follows: ‘Kant claims that taking possession is sanctioned (1) by an innate possession in common of the surface of the earth, and (2) a general will to possess. Each of these points, properly understood, follows directly from his argument for the postulate’ (Reference Messina2021: 73–4). However, Messina’s claim that (2) follows from the postulate is not supported by this passage. The ‘private possession’ mentioned here is not possession merely in accordance with the postulate, but rather conclusive rightful possession. This explains why Kant later states that such possession cannot be achieved in the state of nature. At least, ‘the general will’ that follows from the postulate according to Messina’s reading is still not the general will in the civil condition to determine and secure rightful possession.
3.3 Conclusion: possession is conclusive within the civil condition
Kant concludes that if property is essentially the relationship between one person and all others, this could not ultimately be secure until in the civil condition:
Now, a unilateral will cannot serve as a coercive law for everyone with regard to possession that is external and hence contingent, since that would infringe upon freedom in accordance with universal laws. Therefore it is only a will putting everyone else under obligation, hence only a collectively general (common) and powerful will, that can provide everyone that assurance. (MM §9, 6: 256, also see: §8, 255; §9, 256–7; §10, 259; §14, 263; §16, 267)
The concept of security (Sicherheit) in Kant’s property theory is not employed in the Hobbesian sense of protection from sudden death. Instead, it refers more generally to the security of juridical relationships among persons. According to Kant, the civil condition provides two different factors in securing these relationships.
The first factor is legitimacy. Based on the innate right to freedom (MM, 6: 237–8), Kant asserts that ‘if someone enacts something regarding another, it is always possible for him to do the other wrong, but he can never do wrong in what he decides upon with regard to himself’ (MM §46, 6: 313). Consequently, it is illegitimate – or an unreasonable demand (Anmaßung, MM §8, 6: 255) – to impose obligations on others solely through one’s private will. Instead, only the general will, which is both united (thus omnilateral) and self-legislating (MM §47, 6: 316), possesses the legitimacy to impose obligations on all individuals. Thus, in determining property rights, once an individual has made a private claim by declaring through their private will that something is their property, the corresponding obligation arising from this claim must be legitimately imposed on all through the general will.
The second key factor is force, which ensures the security of all rights. Force is necessary to enforce the obligations imposed by the general will, ensuring that they are fulfilled and that property rights are effectively secured.Footnote 10
It is only within the civil condition that we can expect a general and simultaneously powerful will capable of determining and securing rights. This implies that property rights can be conclusive only in the civil condition (Guyer Reference Guyer2024: 41ff).Footnote 11
This understanding allows us to see why Stone and Hasan’s (Reference Stone and Hasan2022) project ultimately fails. They correctly observe that ‘when Kant speaks of “provisional right,” it is always with reference to this metacondition (omnilaterality)’ (Stone and Hasan Reference Stone and Hasan2022: 70). They are also right in positioning provisional possession as occupying an intermediate stage between private right lacking the general will and private right secured by the general will (p. 94). However, their further claim that provisional possession ‘marks the impossibility of pure private right and thus opens property to requirements other than its own – requirements concerning ownership in relation to the whole system of rights’ (pp. 94–5) is merely rhetorical. Merely emphasizing the impossibility of omnilateral possession in the state of nature does not constitute provisionality; it merely results in tautology. Moreover, the impossibility of pure private right does not necessitate a distinct concept to express it – it is already self-evident in Kant’s framework. Thus, the normativity of provisional possession cannot be defined solely in terms of impossibility; it requires further elaboration to spell out its role within Kant’s system.
4. Reconstructing the question: taking empirical possession into account
The distinction between property right and empirical possession forms the central theme of the first six sections of the Doctrine of Right. This point becomes even clearer when we analyse the formation of intelligible possession.
4.1 Intelligible possession starts from empirical possession
As discussed in section 2, many scholars agree that intelligible possession has an empirical beginning. In this section, I aim to identify textual evidence to support this claim.
Kant outlines three moments of original acquisition: apprehension, declaration, and appropriation:
The elements (attendenda) of original acquisition are therefore: 1) The apprehension of an object that belongs to no one, since otherwise it would conflict with another’s freedom in accordance with universal laws. This apprehension is taking possession of the object of the faculty of choice in space and time, so that the possession in which I put myself is possessio phaenomenon. 2) The declaration (declaratio) of my possession of this object and of my act of the faculty of choice to exclude everyone else from it. 3) The appropriation (appropriatio), as the act of an externally and generally lawgiving will (in idea) through which everyone is bound to agree with my faculty of choice. (MM §10, 6: 258–9)
Although there may be controversy regarding whether the final stage of this process is rightful possession, all commentators agree that the first stage, apprehension, clearly pertains to empirical possession. This is because apprehension involves possession in the realm of phenomena, which is necessarily grounded in the conditions of space and time (Stone and Hasan Reference Stone and Hasan2022: 69; Messina Reference Messina2021: 76; Ripstein Reference Ripstein2009: 154). Another crucial point is also clear: as Kant states in the following passage, ‘the conclusion “the external object is mine,” is correctly drawn from sensible to intelligible possession’ (MM §10, 6: 259). This statement explicitly indicates that intelligible possession begins with empirical possession. The relationship becomes even clearer in light of the following quotation:
In summary, the way to have something external as one’s own in the state of nature is physical possession which has in its favor the rightful presumption that it will be made into rightful possession… (MM §9, 6: 257)
In the natural condition, physical possession carries a rightful presumption that it can be transformed into intelligible possession. This explains why Kant asserts that although possession based solely on private will is not yet rightful, all acquisition is ultimately ‘derived from’ private will and empirical possession, ‘since original acquisition can proceed only from a unilateral will’ (MM §10, 6: 259).
Kant’s insistence that intelligible possession originates from physical possession is straightforward: the original acquisition of property cannot emerge from nowhere. In the state of nature, the private claim that something is one’s property means that one must first take empirical control of it. This act of control serves as the initial step in the process of transforming empirical possession into juridically recognized possession. And of course, the external object possessed is a sensible object (MM §17, 6: 268).
However, this raises an important question: Are all forms of empirical possession admissible as the initial step of juridical possession? If not, what kind of empirical possession holds the rightful presumption necessary to be transformed into rightful possession? Kant provides further clarification, identifying a specific form of empirical possession that qualifies: possession grounded in priority in time.
Taking possession (apprehensio), as beginning to hold a corporeal thing in space (possessionis physicae), conforms with the law of everyone’s outer freedom (hence a priori) under no other condition than that of priority in time, that is, only as the first taking possession (prior apprehensio), which is an act of the faculty of choice. (MM §14, 6: 263. Also see: §6, 250–1; §9, 257; §10, 259)
But Kant also asserts that possession in the state of nature requires the sanction of the general will (MM §44, 6: 312) to become juridical possession. And there is also a further point to consider. We have argued that intelligible possession begins with empirical possession in fact. Why do we emphasize ‘in fact’ rather than ‘in right’? As discussed in section 3.1, empirical possession cannot extend beyond one’s innate right to their body, while intelligible possession pertains to something external, ‘external his’, that goes beyond innate rights. How, then, can intelligible possession in fact originate from empirical possession, while intelligible possession in right cannot simply be regarded as an extension of the latter?
4.2 Task: getting from empirical possession to intelligible possession
Before addressing these issues, we should first outline in detail the process from empirical possession to intelligible possession. As Kant asserts, possession in the state of nature, which has not yet received the sanction of the general will, is what he refers to as provisional possession (MM §44, 6: 312). We have already demonstrated, through textual evidence, that intelligible possession originates from empirical possession. Therefore, the key to understanding provisional possession lies in carefully explicating the process through which empirical possession transitions into intelligible possession.
Kant outlines this process in the Doctrine of Right as follows:
The empirical title of acquisition was physical taking-possession (apprehensio physica), based on the original community of land. Since there is only possession in appearance to put under possession in accordance with rational concepts of right, a title of intellectual taking-possession (setting aside all empirical conditions of space and time) must correspond to this empirical title of acquisition …The rational title of acquisition, however, can lie only in the idea of a will of all united a priori… (MM §15, 6: 264)
The term Title (Titel) in German is the translation of the Latin word titulus. At MM, 6: 260, Kant equates the German term Rechtsgrund (legal ground) with titulus. Here Kant outlines three kinds of titles: those based on the act of a unilateral (facto), bilateral (pacto), or omnilateral (lege) faculty of choice. Corresponding to these titles, private rights are divided into three categories: a right to a thing, a personal right, and a personal right in the manner of a right to a thing. Additionally, at MM, 6: 301, Kant uses the term title as the legal ground for imputation. Accordingly, title holds at least two meanings within Kant’s legal philosophy. First, it serves as a ground to determine a kind of right – constituting the right itself. Second, it acts as the legal ground that a judge may use to render a verdict. In the context of the quoted passage, Kant refers to the first meaning. Here, the two titles not only establish that the right in question is a right to a thing but also identify the possessor and the external object as being so related prior to the establishment of intelligible possessionFootnote 12 .
If we interpret title in this way, we can respond to Stone and Hasan’s claim that there is no specific type of property right in the state of nature and that provisional property cannot be regarded as a morally valid and special type of right (Reference Stone and Hasan2022: 55). The concept of provisional possession is not merely a conceptual device for elucidating conclusive possession (ibid.). Rather, the concept of title suggests that there must be actual legal grounds in the state of nature, which are relatively independent of conclusive possession but serve as a necessary reference point for it.
Kant unfolds the concept of rightful possession into two components: the empirical title and the intellectual (rational) title. These correspond to the stages of generating intelligible possession:
-
A. Original community
-
B. Empirical title
-
C. Intellectual (rational) title
-
D. Intelligible possession
The step (from C to D) is clarified in Kant’s setup of property rights. The rational title rests solely on the idea of the general will, and once this title is secured by the juridical state, it becomes conclusive intelligible possession. The remaining transitional points require further clarification:
From A to B: Empirical title is derived from original community
From B to C: Rational title corresponds in some way to empirical title
Thus, given that provisional possession serves as a preparation for intelligible possession, the essential task in understanding provisional possession is to account for the generative process from empirical to intelligible possession. As argued in section 3.1, empirical possession is conceptually distinct from intelligible possession. If two conceptually separate entities are to interact, a bridging concept is necessary. This explains why provisional possession is positioned between these two. However, as an intermediate concept, provisional possession must maintain relative independence from both empirical and intelligible possession. It cannot be reduced to either of the other two.
5. Rationalization
5.1 Step A: from original community to empirical title
In the Doctrine of Right, Kant attempts several times to explain provisionality through original community and the postulate of practical reason:
This prerogative [of right from empirical possession] arises, instead, because anyone has the capacity, by the postulate of practical reason, to have an external object of his faculty of choice as his own, and so any holding of an external object is a condition whose conformity with right is based on that postulate by a previous act of will; and so long as this condition does not conflict with another’s earlier possession of the same object he is provisionally justified… (MM §9, 6: 257, also see: §6, 251; §14, 263; §17, 268)
It is reasonable to interpret this paragraph as a partial account of provisionality. In this context, reference to provisionality means that if no one claims earlier possession than you, you can provisionally possess an object. Since according to the postulate and the principle of original community, external things are originally ‘possibly yours and mine’ and held in common by all. If external things have not yet been privately possessed by anyone, they can legitimately be possessed by anyone. As Kant puts it: Beati possidente (‘Blessed are the possessors’; MM §6, 6: 251). Effectively, Kant explains the first aspect of provisionality – the transition from original community to empirical title – through the postulate (MM §14, 6: 263).
This framework helps clarify why Kant insists that empirical possession based on priority in time carries a juridical presumption. Since all individuals originally possess all external things in common, it would be juridically wrongful to prevent the first person from claiming external things as her property. The postulate of pure practical reason does not prohibit the use of external things that have not yet been privately owned.Footnote 13 While the criterion for judging priority is based on empirical conditions, the legitimacy of such possession is not derived from these conditions. Instead, it is conferred by the postulate of pure practical reason.Footnote 14
Thus, I argue that the postulate itself is insufficient to fully account for provisionality. To support this claim, it is worth examining Messina’s position in detail. He asserts that, in accordance with the postulate, a form of omnilaterality is already established, as ‘I am ready for the civil condition so far as I recognize that others’ similar claims are also grounded in the postulate (which is omnilaterally willed)’ (Reference Messina2021: 75). However, the alleged omnilaterality extends beyond what the postulate actually implies. In effect, it is not the postulate itself but the omnilateral willing of the postulate that establishes normativity. Thus to secure this omnilaterality, there must be a foundation other than the postulate. Furthermore, Messina acknowledges that ‘all of this is true despite the fact that rights unilaterally or multilaterally acquired are provisional until a civil condition is realized’ (p. 77), which means he concedes that the omnilaterality in question is not that of intelligible possession. This distinction raises two further questions: first, it remains unclear whether the former omnilaterality is genuinely normative. As Kant states, the mere omnilaterality of recognition in the state of nature may still be illegal (MM §42, 6: 307–8). In other words, simply recognizing mutual claims does not constitute normativity. Second, even if some normativity is established, it remains disconnected from intelligible possession. Intelligible possession, as Kant defines it, entails the legitimacy of imposing obligations on all others (MM §8, 6:2 56). However, the omnilaterality Messina describes is merely a recognition of mutual claims, which does not establish the authority to impose obligations. These two forms of omnilaterality should not be conflated.
Both Messina’s and Stone and Hasan’s arguments move too quickly in attempting to establish normative omnilaterality in the state of nature. If true omnilaterality can only be fully realized in the civil condition, then while merely stating its impossibility in the state of nature is tautological, attempting to identify an alternative form of omnilaterality in the state of nature introduces additional theoretical difficulties – particularly in explaining its relation to intelligible possession. However, before addressing omnilaterality, there is still important work to be done to clarify the criteria of possession in the state of nature. One crucial issue is the continued reliance on time as an empirical criterion. Messina claims, ‘with these qualifications in mind, we have it that, in advance of a rightful condition, persons acquire objects by being the first to take them into their possession and remaining ready for a rightful condition’ (Reference Messina2021: 76). But relying on an empirical criterion will not work. For instance, what happens when an individual who was the first to claim the object in question no longer physically holds it? A new rational title must be established to continue the possession – one that moves beyond mere empirical possession. As the quotation in section 4.2 demonstrates, empirical title must further be rationalized in order to be transformed into a rational title.
5.2 Step B: rational title corresponds to empirical title
The question can also be phrased as follows: How can a concept of right (intelligible possession), as a concept of reason, be applied to empirical possession? Kant addresses this issue in §7 of the Doctrine of Right, titled ‘Application of the Principle of the Possibility of the External Mine and Thine to Objects of Experience’. This application is achieved through the pure concept of understanding of possession (hereinafter referred to as PUP), which functions as the necessary bridge between empirical and intelligible and possession.
The central text reads as follows:
The concept of merely rightful possession is not an empirical concept (dependent upon conditions of space and time) and yet it has practical reality, that is, it must be applicable to objects of experience, cognition of which is dependent upon those conditions. – The way to proceed with the concept of a right with respect to such objects as the possible external mine and thine is the following: the concept of a right, which lies in reason alone, cannot be applied directly to objects of experience and to the concept of empirical possession, but must first be applied to the pure concept of understanding of possession in general. Thus instead of holding (detentio), which is an empirical representation of possession, one must think of the concept of having, which abstracts (abstrahieren) from all spatial and temporal conditions, and of the object as only being in my control (in potestate mea positum esse); so then the expression external does not mean existing in a place other than where I am, or that my decision and acceptance are occurring at a different time from the making of the offer, rather it means only an object distinct from me. Now, practical reason, by its law of right, demands that I apply the mine and thine to objects not in accordance with sensible conditions but apart from (abgesehen von) them, since it concerns a determination of the faculty of choice in accordance with laws of freedom, and it also demands that I think of possession of them in this way, since only a concept of the understanding can be subsumed under concepts of right. (MM §7, 6:252–253, my emphasis)
Byrd and Hruschka mention the pure concept of understanding of possession, but they incorrectly attribute this concept to what they call physical possession (Reference Byrd and Hruschka2010: 107–8, 110–1). According to them, ‘Possession can also be (2) having something under my control, such as by locking the doors and windows to my house when I leave it. Kant calls this possession “possession as a pure concept of the understanding”’ (p. 107). This misattribution likely stems from their misunderstanding of the term control (Gewalt). Indeed, Kant does use control in some contexts to describe empirical relations. For instance, he states:
I cannot call mine a wife, a child, a servant, or, in general, another person because I am now in charge of them as members of my household or have them within its restraining walls and in my control and possession, but only if, although they have withdrawn from such constraint and I do not (empirically) possess them, I can still say that I possess them merely by my will, as long as they exist somewhere or at some time, hence merely rightfully. (MM §4, 6: 248)
This example highlights Kant’s distinction between empirical and intelligible possession. In the case of personal right in the manner of a right to a thing, control based on empirical conditions – such as spatial proximity – is distinct from possession by one’s will. When the empirical conditions cease, the empirical possession also ends.
However, Kant also uses control to describe intelligible relations:
Any external object of my faculty of choice can be counted among the rightfully mine if (and only insofar as) I have control of it without being in possession of it. (MM §6, 6: 252, also see: §10, 258; §15, 264; §17, 268)
Here, control clearly characterizes an intelligible connection.
To conclude, the term control is neutral, referring broadly to a certain sort of connection between an agent and an object. The attribution of this connection lies in whether this connection is grounded in empirical conditions or supersensible relations. Thus, in the preceding quotation, the abstraction of control from spatial and temporal conditions cannot be attributed to physical possession.
In the preceding quotation, Kant’s argument could be unfolded into three parts:
A. The concept of merely rightful possession could not directly be applied to empirical objects. For such a concept of reason, there is no corresponding sensible intuition that could be added to (MM §6, 6: 252).
B. Through PUP, possession abstracts from the conditions of time and space. Here, having is understood not as physical possession but as control in abstraction from empirical conditions. External no longer refers to spatial or temporal distinctions but simply to an object’s distinctness from the subject.
C. PUP enables the subsumption of empirical possession under the concept of intelligible possession. This serves as the essential bridge, allowing Kant to reconcile empirical acts of possession with the concept of intelligible possession.
Regarding A, ‘there is no corresponding sensible intuition’ has been explained in section 3.1. About empirical objects, in the quotation, Kant clearly states that the concept of right should be applied to ‘objects of experience and to the concept of empirical possession’ – if there is no relation between empirical possession and intelligible possession, why does Kant take it up to deal with this tricky task of application? Here we encounter another supplementary proof for our argument in section 4.
C is the aim Kant wants to achieve.
B is the essential step that PUP facilitates as a condition for C and requires further remarks.
5.2.1 Mere having
PUP relates to a condition of mere having, which represents a foundational connection apart from empirical possession. As discussed in Section 3.1, holding – possession based on empirical conditions – relies on physical acts, such as grasping an apple. However, beyond these empirical conditions lies a more abstract relationship: control, a connection generated by the agent’s intentional relationship with the object. While this connection may originate from empirical possession, its significance lies in its abstraction from empirical factors. However, since it lacks the omnilateral obligations imposed by the general will, which are necessary for establishing conclusive juridical relations, such mere having does not yet constitute a juridical relationship. Thus, mere having occupies a conceptual space between empirical possession (rooted in physical conditions) and intelligible possession (grounded in omnilateral obligations). It remains distinct from both.
5.2.2 External in the juridical sense
In the juridical sphere, externality is conceptually distinct from empirical externality. According to Kant’s Critique of Pure Reason, we represent objects as external to us through the form of our outer sense, namely space (A22/B37). However, this spatial representation is unsuitable for juridical legislation. While theoretical cognition requires adding sensible intuition to concepts, juridical application removes all sensible intuitions to extend the concept of possession beyond empirical constraints (MM §6, 6: 251–2).
Externality in the juridical sense is necessary because property rights, as part of acquired rights, require that objects be represented as external his rather than internal his. This distinction ensures that property falls under the sphere of acquired right, separate from innate right. Through PUP, externality signifies something distinct from the agent, abstracted from spatial or temporal conditions. This abstraction enables a proper juridical framework for conceptualizing the relationship between agents and objects, aligning it with the requirements of property rights in the juridical sphere.
5.2.3 Abstraction
To address the puzzle posed in section 4.1 – how intelligible possession in fact can originate from empirical possession while in right it cannot simply be regarded as an extension of the latter – we must clarify Kant’s concept of abstraction. Abstraction, as Kant conceives it, does not imply that A contains B in a covert way and that B is merely extracted from A. Such a process would fall under the definition of analyticity (A6/B10). Instead, abstraction involves removing all sensible conditions associated with a concept. As noted in section 3.1, if empirical possession is defined exclusively by its dependence on sensible conditions, then abstraction leaves no residual connotations of empirical possession after these conditions are removed. I propose that abstraction should be understood differently: it allows us to regard the same phenomenon from a new perspective. Through abstraction, we move beyond the empirical conditions of possession and focus on a more fundamental and authentic characterization – mere having or control. In this sense, intelligible possession in fact does originate from empirical possession. However, through the process of abstraction, we move closer to a juridical conception of possession, wherein empirical possession is reinterpreted through the lens of practical reason. This juridical reinterpretation bridges the gap between the sensible and supersensible realms, enabling us to approach intelligible possession without reducing it to an extension of empirical possession.
5.2.4 How does PUP help to generate intelligible possession?Footnote 15
As argued in section 3, there are two definitions of rightful possession: a negative definition (ND), which is based on independence from empirical conditions, and a positive definition (PD), which is based on omnilaterality. By distinguishing between the PUP and rightful possession in the quotation, Kant appears to indicate a gap between ND and PD. This suggests that there is a form of possession (subsumed under PUP) that satisfies ND but does not yet satisfy PD. This gap, in turn, creates the conceptual space for provisional possession.
On the one hand, possession in accordance with PUP satisfies ND, meaning that it is an act of possession independent of empirical conditions as mere having. On the other hand, possession in accordance with PUP remains unilateral in the state of nature, as it lacks universal legislation, and therefore does not yet satisfy PD. However, in terms of normativity, possession in accordance with PUP must be closer to rightful possession than mere empirical possession, since it already disregards empirical conditions.
In this sense, PUP contributes to the establishment of normativity by overcoming the empirical conditions that function as obstacles. It also introduces a new conceptual framework for thinking about the relationship between the agent and their possession. This shift in perspective lays a crucial foundation for universal juridical legislation.
To conclude, I argue that PUP is a normative framework for rationalizing and characterizing what is empirically established through possession based on priority in time. It serves as a precursor to intelligible possession in the state of nature.
To support this claim, let us consider three hypothetical cases in the state of nature.Footnote 16
Case 1: A person sits on a gold nugget without claiming it is hers.
Case 2: The same person sits on the same gold nugget, claims it as hers and, first and foremost, she was the first to claim the nugget. Assume her claim is legitimate, without detracting from others’ freedom under universal law.
Case 3: After case 2, the same person sits next to the gold nugget.
The question has now become: When conflicts arise, how should a judge rule when we enter the civil condition?
Note that case 1 is not a property case, since no act of possession occurs. In case 2, we encounter a kind of normative possession in the state of nature. However, what matters is not the empirical condition (priority in time) but a relationship of possession conceived in terms of PUP. This difference is more evident when we turn to case 3. Case 3 illustrates the distinct role of PUP in the state of nature: in the sense of empirical possession, the person no longer possesses the gold nugget, since she is not physically holding it. However, insofar as the case is subsumed under PUP, she still possesses the nugget. PUP thus defines possible relationships of possession in the state of nature, even when the possessor is no longer in physical contact with the object. This ensures the continuity of prior possession’s normativity, as we finally make the transition to a civil condition.
In a court setting, the only requirement for the claimant would then be to present proof – for example, a sign on the nugget – demonstrating that they had previously possessed it before any other disputants. This proof would then determine the legitimacy of their claim.
5.2.5 Preparation and anticipation
Through the pure concept of understanding of possession (PUP), at least three goals are achieved:
-
1. The conception of the relationship between subjects and objects based on empirical conditions is excluded.
-
2. External things remain distinct from me, irrespective of empirical conditions.
-
3. In some sense, I still possess external things.
These three points correspond to key aspects of intelligible possession: (1) is the precursor of intelligibility, (2) is the precursor of externality, and (3) addresses the juridical issues related to possession. Consequently, through PUP, possession achieves the necessary abstraction and preparation for intelligible possession.
Why is this abstraction necessary? As we have established, juridical possession originates from empirical possession in fact. However, due to the fundamental mismatch between the supersensible nature of intelligible possession and the sensible nature of empirical possession, empirical acts cannot directly be subsumed under the concept of reason in question. As we have argued in section 2, following Kant’s guiding principle, in order to establish the objective reality of the concept of right, we must remove all empirical intuitions from this concept (MM §6, 6: 252). PUP takes this task on. On the one hand, it starts with empirical possession, allowing it to be connected with and yet purified of empirical conditions. On the other hand, as relevantly considered after abstraction, external things remain distinct from me and under my control – not in terms of spatial and temporal conditions, but in a way that makes them ready to be juridically and practically considered.
At this stage, however, I still cannot yet have any external things as juridically external mine, as such determination requires the sanction of the general will within the civil condition. Nevertheless, this abstraction sets the stage for universal juridical legislation, which Kant refers to as anticipation.
Thus, while PUP performs negative and preparatory tasks, its abstraction is indispensable for enabling empirical acts to be subsumed under the concept of right. This explains why, immediately after discussing abstraction, Kant proceeds to provide a precise definition of intelligible possession in the subsequent paragraphs (MM §7, 6: 253–4; §17, 6:268).
6. Implication: PUP as a type?
To further support my reading, we can identify concepts in Kant’s philosophy that perform a function similar to that of PUP. For instance, in the Critique of Practical Reason (CPrR), Kant asserts that to apply a concept of reason to which no corresponding intuition can be added, a type is required.Footnote 17 This section will tentatively explore whether PUP could be considered a kind of type in this sense.
Kant explains the role of a type in the CPrR as follows:
But no intuition can be put under the law of freedom (as that of a causality not sensibly conditioned) – and hence under the concept of the unconditioned good as well – and hence no schema on behalf of its application in concreto. Thus, the moral law has no cognitive faculty other than the understanding (not the imagination) by means of which it can be applied to objects of nature, and what the understanding can put under an idea of reason is not a schema of sensibility but a law, such a law, however, as can be presented in concreto in objects of the senses and hence a law of nature, though only as to its form; this law is what the understanding can put under an idea of reason on behalf of judgment, and we can, accordingly, call it the type of the moral law. (CPrR, 5: 69)
Based on this quotation, I argue that PUP can be regarded as a kind of type.
The type in Kant’s philosophy addresses a fundamental difficulty: mediating between the supersensible and sensible realms (CPrR, 5: 68). It facilitates applying a concept of reason to empirical objects. In the Critique of Practical Reason, the type enables the application of the moral law, allowing empirical acts to be subsumed under it for moral appraisal (5: 69). Unlike the Transcendental Deduction in the Critique of Pure Reason, the challenge of applying these concepts of reason arises from the absence of a corresponding sensible intuition (5: 68). This lack of intuition precludes the possibility of resolving the problem through a schema.
This case closely parallels the situation with PUP. First, PUP serves as a framework for mediating between the supersensible (intelligible possession) and the sensible realm (empirical possession) (MM §7, 6: 253). Kant employs PUP to apply the concept of intelligible possession, allowing empirical possession to be subsumed under the concept of right (ibid.). However, in applying the concept of right, there remains no relevantly corresponding sensible intuition (§6, 6: 252). What is even more striking is that Kant states that the type is a product of the understanding rather than the imagination (CPrR, 5: 69), thereby distinguishing it from the schema (A140/B179). Similarly, PUP is also a product of the understanding (MM §7, 6: 253).
Moreover, it is legitimate to introduce an additional type in juridical theory. In the Critique of Practical Reason, Kant identifies at least three types: the law of nature (CPrR, 5: 69), the nature of the sensible world (5: 70), and happiness (ibid.). Adding PUP as a type in the Doctrine of Right is not at all unusual, let alone questionable. More importantly, these types share similar features: the nature of the sensible world serves as a type due to its form of universal lawfulness (5: 70), and PUP, abstracted from empirical conditions, aligns with universal legislation (MM §7, 6: 253).
I do not hereby aim to resolve broader questions, however, such as the precise nature of the type in question or its relationship to the schema, but merely to explore the possibility of regarding PUP as a type in the relevant sense, as these issues would require substantial further inquiry and likely one or two dedicated papers. However, if this initial attempt proves viable, then PUP provisionally secures a legitimate foothold within Kant’s philosophical system.
7. Conclusion and reflection
Provisional possession can be divided into two distinct parts. The first part is grounded in the postulate of practical reason, which confers an empirical title on empirical possession based on priority in time. The second part is based on the pure concept of understanding of possession (PUP). Through abstraction from empirical conditions, the empirical title transitions to the rational title.
The main advantage of my reading is that it is clearly articulated in two respects. Provisional possession, understood as involving subsumption under PUP, possesses normativity in contrast to empirical possession, yet it does not amount to rightful possession. Moreover, my interpretation is compatible with most existing arguments that assert that the empirical history of just property should be respected. Rather than contradicting these views, my approach supplements them by introducing an additional conceptual tool, namely, PUP, which helps bridge the gap between empirical possession and rightful possession within Kant’s framework.
The weakest part of my argument lies in the absence of direct textual evidence that explicitly identifies possession in accordance with PUP as ‘provisional possession’. This attribution relies on interpreting Kant’s definitions of provisional possession, particularly with respect to his notions of ‘preparation’ and ‘anticipation’. However, even if one rejects the claim that PUP is part of provisional possession, my argument still highlights the significance of PUP as a vital concept in Kant’s theory of property, deserving greater scholarly attention.
If PUP is understood as the preparation for and anticipation of intelligible possession, two important implications emerge. First, in the absence of the general will, there can be no conclusive right in the state of nature – only preparation for such a right. However, possession in terms of PUP must be respected by the ensuing civil condition. Second, provisional possession may exist even within the civil condition. For example, if A contracts to sell a house to B, but the transfer has not yet been registered with the Housing Management Office, the contract may still serve as a title or legal ground for a judge’s decision. Through PUP, B exercises control over the house. In this sense, provisional possession is not itself a right but rather the legal foundation for a right – one that the representative of the general will must recognize and refer to.
Acknowledgements
I am grateful to the audiences at the Kant Online Reading Group, the St Andrews Kant Colloquium, the 14th Bonn International Kant Congress, and the Youth Salon at Chongqing University for their insightful questions and suggestions on an early version of this paper. I would like to thank Jens Timmermann, Kate Moran, and Martin Brecher for allowing me to cite their translation draft. I am especially thankful to Jens Timmermann, who encouraged me to write this paper and provided valuable feedback during its revision, as well as to Frederick Rauscher, Weiyang Kang, and Jiaxian Liu for their constructive suggestions. This research was supported by the CSC-St Andrews PhD scholarship. I would also like to thank the anonymous reviewers and the editors of this journal for their thoughtful comments and guidance.
Open access
