1. Introduction
Kant says that ‘rational nature [including human beings] exists as an end in itself’ (G, 4: 429).Footnote 1 Something that is an end in itself has not merely ‘relative worth’ but ‘inner worth, that is, dignity’ (innern Werth, d. i. Würde) (G, 4: 435). Therefore, in Kant’s view, human beings have dignity. Kant interpreters have different views on what Kant considers the ground of human dignity. Some argue that human dignity has its basis in some feature of human beings. Others contend that humans in themselves lack dignity, but dignity can be attributed to them because morality demands respect for humanity. In this article, I question the assumption implicit in these approaches that any one aspect of Kant’s account of dignity, whether humanity or the moral law, can be taken as the primary ground of human dignity. Against these ‘primacy’ approaches, I outline an alternative that is, at least prima facie, consistent with Kant’s major ethical writings. I argue that, for Kant, human dignity can be seen to be grounded in a reciprocal relationship between the dignity of the moral law and the dignity inherent in the human constitution. Specifically, Kantian human dignity combines the dignity of the moral law, the dignity of human personhood construed as rational inner purposiveness and the dignity attaching to humans taking up the moral law and striving to follow it.
In section 2, I offer a classification of the existing interpretations of Kant’s concept of human dignity and point to textual grounds on which they might be critiqued. In sections 3 and 4, I will then defend an alternative position, which I call the ‘non-disjunctive interpretation’. Kant does not offer a systematic treatment of human dignity. In section 3, I isolate four different strands relating to dignity in Kant’s writings. In section 4, pace existing interpretations of Kantian dignity, I argue that human dignity is grounded in the reciprocal relationship between the dignity of the moral law and the dignity intrinsic to humanity. Finally, in section 5, I indicate briefly that the non-disjunctive view not only resolves a key aporia in the literature, but also opens up the possibility of a new interpretation of Kantian value as multivocal. In addition, I suggest how the non-disjunctive approach might be applied in particular cases.
2. What gives humans dignity: Kant and his interpreters
In his practical philosophy, Kant aims to ‘seek out and establish the highest principle of morality’ (G, 4: 392). This principle is the categorical imperative: act only on the principle that you can prescribe as a universal law (G, 4: 421). Kant describes two different criteria on which human choice (Willkür) is made – the categorical imperative (henceforth CI), and the hypothetical imperative. One acts out of ‘motive’ (Bewegungsgrund) and follows CI if one wills an action to fulfil an ‘objective end’ (Zweck) (G, 4: 427). This end is provided by pure reason, or reason unmixed with anything sensory (ibid.). To act on the basis of CI is an end in itself. In contrast, a hypothetical imperative is an if-then imperative to perform some action if one wants to achieve a particular end (G, 4: 414). If one desires happiness, then one must perform actions that bring about this end (G, 4: 418–19). Such actions are performed out of ‘incentive’ (Triebfeder); and Kant calls them ‘subjective’ or empirical, because desire is non-uniform across individual human beings (G, 4: 427).
Kant introduces the concept of dignity in the context of explicating CI:
Reason therefore relates (beziehen auf) every maxim of the will as universally lawgiving to every other will, and also to every action in relation to itself, and this indeed not for the sake of any other practical motive (Bewegungsgrund) or future advantage, rather out of the idea of dignity of a rational nature which obeys no law but that which it gives itself at the same time. (G, 4: 434)
In this passage, rational nature has dignity because it can legislate the moral law to itself and to all other rational natures. Consequently, dignity relates to morality and rationality, but also to humanity for the following reason. Everything has either a price or dignity (G, 4: 434–5). What has price has an ‘equivalent’, but dignity has no equivalent (ibid.). Human inclinations and needs have market price, but an ‘end in itself (Zweck an sich selbst) does not merely have a relative worth (Werth), i.e., a price, rather an inner worth, i.e., dignity (Würde)’ (ibid.). Further, for Kant, we should treat the humanity in our person, and that of others, as an end in itself and never merely as a means (G, 4: 429). Therefore, on Kant’s view, human persons have dignity because they are ends in themselves possessing an inner worth that transcends market price.
Kant interpreters are divided on what Kant considers to be the foundation of human dignity. I arrange these interpreters into two categories based on what they take to be the primary feature grounding human dignity.Footnote 2 Interpreters in the first category converge on the idea that human dignity can be traced to some feature of human beings (2.1). These commentators can be further sub-divided depending on which feature of human beings gives them dignity – some individual human capacity (2.1.1–2.1.3), or actually acting on the basis of good will (2.1.4). Commentators grounding Kantian human dignity in some human capacity do not converge on the same capacity. Christine Korsgaard and Allen Wood foreground the human capacity to set ends (2.1.1). Others emphasize the human capacity to be moral (Henry Allison, Jens Timmermann, Barbara Herman) (2.1.2). Paul Guyer claims that the human capacity for freedom gives humans dignity (2.1.3). Against the view that some human capacity grounds human dignity, Richard Dean proposes that humans can have dignity only if they realize their capacity for morality, i.e. act morally (2.1.4). Commentators in the second category contend that humanity in itself does not have dignity, and that dignity is attributed to human beings only because the moral law demands it (Oliver Sensen) (2.2).Footnote 3 I now discuss these positions in more detail, and cite textual reasons for why they might be inadequate.
2.1 Humanity as locus of human dignity
2.1.1 Capacity to set ends
Korsgaard argues regressively that the capacity of humans to set an end based on what is valuable to them makes humanity a ‘source of value’ (Reference Korsgaard1996: pp. ix–x), and enables humans to conceive of themselves as having dignity (272–3). Similarly, Wood (2008: 92) argues that our ‘basic act’ as rational (human) beings is the ‘act of setting ends and regarding them as good’, which ‘necessitates our representing ourselves as already ends [i.e., as having dignity]’.
The Korsgaard/Wood line has been widely challenged on textual grounds. First, the mere possession of the capacity to set ends cannot give humans dignity, because humans also set non-moral ends, and dignity for Kant rests solely on morality (Sensen Reference Sensen2011: 62; Timmermann Reference Timmermann2006: 74–6; Allison Reference Allison2011: 231; Dean Reference Dean2006: 28–9). Korsgaard’s regress argument has also been criticized (Timmermann Reference Timmermann2006: 74–5; Allison Reference Allison2011: 225; Kerstein Reference Kerstein, Christoph and Dieter2006: 206–10). Finally, Wood’s claim that we ‘esteem’ the capacity to set ends contradicts Kant’s statement that rationally setting ends leads to hatred rather than esteem of reason, because reason often detracts from the achievement of happiness (Sensen Reference Sensen2011: 76, 77–8).
2.1.2 Capacity for morality
Allison (Reference Allison2011: 210, 218) takes the capacity for morality as the ‘property’ that makes humanity an end in itself. Here the capacity for morality is the ‘capacity to recognize and obey the categorical imperative’ (207 n.), and only means that humans must stand under moral laws – i.e. can recognize and obey CI – not that they actually obey them (218). Timmermann says that rational beings differ from non-rational ones in their ‘capacity to be moral’, and what makes the former ‘wonderful’ is ‘pure practical reason … [or the] capacity to choose moral ends freely’ (Reference Timmermann2006: 85; 2007: 101). Herman (Reference Herman1993: 238) argues that rational nature involves the ‘capacity to be its own original source of reasons’. Since the CI is the only possible ‘self-given’ principle, autonomy must be the ‘capacity to act morally’, and rational nature must have a ‘unique kind of value’, or dignity, to the extent it is ‘capable of morality’ (ibid.).
While Kant does say explicitly that humanity has dignity to the extent it is capable of morality, he also says that morality (Sittlichkeit) itself has dignity – ‘morality (Sittlichkeit) and humanity, to the extent it is capable of [morality], [is] that which (dasjenige) alone has dignity’ (G, 4: 435). Here it appears that for Kant both morality and the human capacity for morality can independently be attributed dignity. This claim is inconsistent with the capacity-for-morality interpretation according to which only the human capacity for morality, and not morality, forms the ground of Kantian dignity. Thus, the capacity-for-morality interpretation seems unable to account for at least one of Kant’s statements regarding dignity.
2.1.3 Capacity for freedom
Guyer (Reference Guyer1998: 34) argues that the ‘ultimate source of dignity’ is the ‘freedom of our will … [that] both allows and requires us to use our reason’. Even the capacity to set ends is a ‘fundamental manifestation of our freedom’ (Guyer Reference Guyer2000: 151). In support of his view, Guyer (Reference Guyer1998: 33) cites Kant’s statement from the Feyerabend lectures that ‘[t]he inner value of the human being rests on his freedom, that he has his own will’, while admitting that his view lacks direct textual support.
It is unclear if Guyer takes freedom as the ‘weak’ or ‘simple’ freedom of choice, or the ‘strong’ freedom required to follow the moral law (Dean Reference Dean2006: 234–5). Guyer (Reference Guyer2000: 156) indicates that Kant sometimes takes only freely chosen acts adhering to the moral law, which require strong freedom, to be valuable (cf. Dean Reference Dean2006: 238ff.). If strong freedom underwrites human dignity for Guyer, then his view seems similar to the Timmermann/Herman interpretation that freely choosing the moral law gives us dignity. But this similarity also makes it potentially vulnerable to the objection directed at this interpretation (section 2.1.2). On the other hand, Guyer (Reference Guyer2006) seems also to suggest that weak freedom has unconditional value. This is because, despite his view that weak freedom is an ‘anthropological and psychological fact’ that cannot ground morality (189), Guyer says that treating humanity as an end in itself requires the ‘preserv[ation]’ of ‘freedom as an ongoing condition’ (197). Consequently, he considers the free choice of suicide to be immoral because it ‘destroys the possibility of any further free acts’ (ibid.). Here it is unclear what sort of freedom is to be preserved. If free choice means any free act, moral or immoral, then the claim that weak freedom makes humanity an end in itself could be attributed to Guyer. But Guyer’s position would then be inconsistent with Kant’s grounding of human dignity in our freely setting moral ends at G, 4: 429–36. Hence, Guyer’s view in general requires further clarification.
2.1.4 Good will
Dean (Reference Dean2006: 40) argues that having a good will makes humanity an end in itself, because humanity and good will are identical for Kant. This view has been criticized on several grounds. First, it presumes without justification that being valuable is identical to being morally good in Kant’s system (Allison Reference Allison2011: 215). Second, it cannot accommodate Kant’s claim that we should respect even the vicious person (Sensen Reference Sensen2011: 92; Allison Reference Allison2011: 214 n.; Kerstein Reference Kerstein, Christoph and Dieter2006: 217). Third, contra Dean, it has been argued that humanity must be something ‘in’ all human beings, and so good will is non-identical with humanity, since not all humans possess good will (Denis Reference Denis2010: 127–8; for Dean’s response to this criticism, see 2013: 193).
2.2 Moral law as source of human dignity
Unlike the interpretations discussed in section 2.1, Sensen denies that any feature of humanity can ground Kantian human dignity, whether it is a human capacity or the good will. Instead, humans have dignity because the moral law demands respect for humanity – ‘Human beings have a dignity because … the moral law, as articulated in the Formula of Humanity, demands respect for human beings’ (Sensen Reference Sensen2011: 199–200). Sensen also says that the ‘absolute worth of human beings is secondary to and depends on a morally good will’ (190). Therefore, on Sensen’s view, the moral law is the primary ground of human dignity.Footnote 4
Sensen’s claim that morality rather than humanity grounds human dignity for Kant requires further justification.
First, Kant says: ‘Now morality is the condition (Bedingung) under which alone a rational being can be an end in itself. … Therefore, morality and humanity, insofar as it is capable of morality, is that which alone has dignity (Also ist Sittlichkeit und die Menschheit, so fern sie derselben fähig ist, dasjenige, was allein Würde hat)’ (G, 4: 435). For Sensen (Reference Sensen2011: 185), this passage ‘link[s] the concepts of “morality”, “dignity” and “worth”’ and shows that ‘[i]t is morality that has an inner worth’. However, this passage can also be read to show that both morality and humanity together make for human dignity, as I now argue.
(a) The ‘and’ in ‘morality and humanity, insofar as it is capable of morality’ could be taken to combine two independent characteristics – morality and humanity. This is because the dasjenige in the statement is ambiguous, and could plausibly refer to the conjunction of morality and humanity.
(b) It is unclear how Kant’s statement that morality might be the condition under which alone rational nature can have dignity should be interpreted. Sensen seems to think that condition here means sufficient condition, and therefore concludes that morality is the sufficient condition for human dignity. But one can resist Sensen’s conclusion if we take condition to mean necessary condition. In this case, morality would indeed be the necessary condition for attributing dignity to any rational being, including human beings. However, one could deny that morality is the sufficient condition for human dignity if we accept the claim in (a) that morality and humanity taken together constitute human dignity.
Hence, from (a) and (b), the passage that Sensen cites to support his view that morality alone grounds human dignity can plausibly be interpreted in a way that counters his own view.
Second, Kant says that persons possess dignity not insofar as they are subordinated to the moral law, but ‘insofar as [they are] lawgiving in relation to the moral law, and [are] subordinated (unterordnen) to the [moral law] only because of this [lawgiving]’ (G, 4: 440). Here dignity is not grounded in our subordination to the moral law, but arises insofar as we are able to give ourselves the law. This problematizes Sensen’s claim that human dignity is based solely on the dignity of the moral law, since it raises the possibility that Kantian dignity could be grounded either in lawgiving alone, or in both the moral law and lawgiving.
In another passage, Kant says: ‘For nothing has worth (Werth) except that which the law determines. The lawgiving itself, however, which determines all value (Werth), must just for this reason (eben darum) have dignity (Würde). … Autonomy is therefore the ground of dignity of human and every other rational nature’ (G, 4: 436). Here it might appear that lawgiving, which determines all value, has dignity, because (‘just for this reason’) the law determines all value. This reading supports Sensen’s claim that human dignity has its basis solely in the moral law. However, contra Sensen, this passage could also be interpreted as saying that while the moral law, which determines all value, might contribute to lawgiving, lawgiving/autonomy, as Kant says explicitly here, is the ground of human dignity.Footnote 5
In sections 2.1–2.2, I described existing interpretations as taking either the moral law or some feature of humanity as the primary ground of Kantian human dignity, and showed that none of these interpretations accommodates all of Kant’s statements on dignity. In what follows, I argue further against this primacy approach. In section 3 I show that, as Sensen argues, Kant attributes dignity to the moral law. However, Kant also ascribes dignity to various features of humanity – moral personhood, giving oneself the moral law and virtue. Subsequently, in section 4, I explore how these various senses of Kantian dignity relate to each other, arguing that the moral law and the human constitution relate to each other reciprocally in grounding human dignity.
3. The multiple strands of Kantian human dignity
I demonstrate that human dignity, in Kant’s view, has multiple aspects. First, the dignity of the moral law (3.1). Second, the dignity of human persons metaphysically not anthropologically construed, and interpreted as the non-substitutable and irreducible rational inner purposiveness that allows access to the moral law (3.2). Third, humanity has dignity to the extent it can give itself the moral law and strive to follow it (3.3).
3.1 Dignity of the moral law
Kant attributes dignity to the moral law. First, the direct evidence for this claim. Kant says that we should ‘contemplate the dignity of the pure law in us (contemplatione)’ (MdS, 6: 397). In addition, at G, 4: 435, the moral law is said to have dignity (‘morality and humanity, insofar as it is capable of [morality] … alone has dignity’). Second, the more indirect support. Kant asserts that two things fill the mind with ‘admiration’ (Bewunderung) and ‘awe’ (Ehrfurcht), i.e. ‘the starry heaven above me, and the moral law in me’ (KpV, 5: 161). While the term ‘dignity’ is absent here, the descriptions ‘awe’ and ‘admiration’ are at the very least suggestive of Kant’s use of that term to express something ‘elevated’ (Sensen Reference Sensen2011: 177). Next, the ‘moral law is holy (inviolable (unverletzlich))’, and while the ‘human being’ (Mensch) is ‘unholy enough, the humanity (Menschheit) in his person must be holy for him’ (KpV, 5: 87). Here, the term ‘holy’ seems interchangeable with ‘dignity’, because Kant says elsewhere that humanity in one’s person has dignity or absolute value (e.g. MdS, 6: 422–3). Hence, for Kant, the moral law has dignity.
3.2 Dignity of human persons
What is the relationship between human personhood and dignity for Kant? In 3.2.1 I show that Kant does not ascribe dignity to the empirically observable human person composed of both sensibility and rationality, but only to the rationality in such a person. On this basis, I argue in 3.2.2 that, metaphysically speaking, human personhood can be interpreted as an irreducible and non-substitutable rational inner purposiveness. This purposiveness enables human persons to access the moral law and constitutes their dignity.
3.2.1 Two senses of person
The ‘formula of humanity’ (FH) states that we should treat the humanity in our own person and that of others as an end and never merely as a means (G, 4: 429). Since FH is a formulation of CI, and CI is a synthetic a priori proposition (G, 4: 454), the notion of person here must be metaphysical, not anthropological. Metaphysics for Kant is a ‘system of a priori cognition merely (bloss) out of concepts’ (MdS, 6: 216). The metaphysics of morals is the ‘critique of pure practical reason’ (G, 4: 391). It involves ‘anthroponomy’ or the unconditional rules given by reason, as opposed to ‘anthropology’ which is concerned with human nature as it is empirically observable (MdS, 6: 405–6).
Kant uses the term ‘person’ anthropologically, but also in a metaphysical sense. Metaphysically speaking, as I now show, Kant views the human person in two different senses: the empirically observable person comprising both sensibility and rationality (henceforth Person1); and the same person regarded only in its rational, and not sensible, aspect (Person2).
The distinction between Person1 and Person2 is present in Kant’s major ethical writings. At KpV, 5: 86–7, he says that the ‘person’ (Person) belonging to the sensible world is subjected to the moral law by his own ‘personality’ (Persönlichkeit). Personality belongs to the intelligible world, and is therefore free from the mechanism of nature (ibid.). In this passage, Person2 subjects Person1 to the moral law. The Person1–Person2 distinction is also implicit in the G, 4: 455, claim that the ‘better person transfers (versetzt) himself into the standpoint of a member of the intelligible world’. Here the ‘better’ person is Person2, and the comparative ‘better’ indicates the existence of the sensible-cum-rational person (Person1).
Kant also makes several references to the Person1–Person2 distinction in MdS. He says that the ‘concept of the human being’ has ‘two different senses’ if viewed from two different standpoints (MdS, 6: 418). A human being as a ‘natural being with reason (homo phenomenon) can be causally determined by his reason to act in the sensible world’, and here the concept of obligation does not arise (ibid.). But the same human being viewed from the perspective of his ‘personality, that is, as a being endowed with inner freedom (homo noumenon)’ can be put under obligation to himself, i.e. to the humanity in his person (ibid.). In addition, Kant says that homo noumenon is ‘personality without physical attributes’, while the human being is the ‘same subject represented as affected by physical attributes’ (6: 239). On the basis of these passages, one can argue that the human person viewed as both sensible/physical and rational – a ‘natural’ being with reason – is Person1, and has no moral obligation. In contrast, if the same human person observes himself through his practical reason, that is, ‘according to the humanity (Menschheit) in his person’ (6: 379–80 n.; also KpV, 5: 87), then such a person, or Person2, can be put under moral obligation.
Similar passages can be found elsewhere in MdS, but also in Religion. The ‘person in general’ (Person überhaupt) observes himself as a sensible being (Person1) as opposed to the rational human being, i.e. a person with ‘duties which his own reason has imposed on him’ (Person2) (MdS, 6: 435). In the same text, the moral person acts freely in consonance with the moral law and is ‘accountable’ (Zurechnung) for its actions, while psychological personality enables the consciousness of one’s identity in different circumstances (6: 223). Since psychological human identity across circumstance must include sensibility and reason (6: 418), this must refer to Person1, while the moral/rational person as such is Person2.Footnote 6 Finally, in Religion, ‘humanity’ is both living (or sensible) and rational (Person1), and ‘personality’ is rational and capable of ‘accountability’ (Person2) (R, 6: 26).
Next, Kant ascribes dignity to Person2, not Person1. He says that, as homo phenomenon, human beings have ‘ordinary value’ like animals (Person1), but construed as homo noumenon, or ‘person … as the subject of morally practical reason’ (Person2), humans are ‘exalted beyond price’, and have ‘dignity (an absolute inner worth)’ (MdS, 6: 434–5; also 439 n.). For instance, in committing suicide, one makes ‘less worthy’ (abwürdigen) the humanity in one’s person, i.e. ‘homo noumenon which was entrusted (anvertrauen) to the human being viewed as homo phenomenon’ (6: 422–3). Similarly, a human being is holy if he observes himself through his practical reason, i.e. ‘according to the humanity (Menschheit) in his person’ (6: 379–80 n.; also KpV, 5: 87). Hence, the human person viewed as Person2 has dignity, but not when viewed as Person1.
In sum: Person1 is the sensible-cum-rational human person possessing only ordinary value. Person2 is the rational-moral aspect of Person1, or the humanity in Person1. Unlike Person1, Person2 can be put under obligation and has dignity.
3.2.2 Person as rational inner purposiveness
In light of the distinction between Person1 and Person2, FH reads: act in a way that you use the humanity in your person (Person2) and that of others as an end and never merely as a means. In what sense is Person2 an end, or, as Kant says more frequently, ‘end in itself’ (Zweck an sich selbst)? I argue that Person2 is the rational inner purposiveness (Zweck) that enables the human person to access the moral law, and distinguishes human persons from all other natural beings. The irreducibility and non-substitutability of this rational inner purposiveness make Person2 a self-existent end in itself and give it dignity.
First, Kant’s claims at G, 4: 396, arguably provide a good characterization of the rational inner purposiveness of Person2. If nature wanted us to be happy, it would have given us a ‘natural instinct’ (Naturinstinkt) for it, because reason cannot fulfil this purpose. If nature has been ‘purposive’ (zweckmässig) in distributing predispositions, and if we have been ‘allocated’ (zutheilen) reason as a practical ability that can ‘influence’ the will, the ‘true [rational] determination’ of the will would involve ‘producing … [a] will that is good in itself’. So the good will is the purpose, and its achievement through the ‘cultivation (Cultur) of reason’ is entirely in tandem with nature even if this means completely forsaking happiness. In acting from good will, reason receives its own kind of ‘satisfaction’ – that of fulfilling a purpose that pure reason has set for itself.
Based on this argument at G, 4: 396, the rational purposiveness (Zweck) able to make the will follow the moral law presupposes the purposiveness of nature. It follows that rational beings have their purposes ‘naturally’ built into themselves. Combining this proposition with the FH claim that humanity in one’s own person (Person2) is the end in itself (Zweck an sich selbst), Person2 can be interpreted as a rational inner purposiveness that is an end in itself, and that has the goal of achieving a good will.Footnote 7 Further, since the moral law is implicit in the notion of the good will, Person2 must be presumed to be able to access the moral law. Finally, one can take this rational inner purposiveness constitutive of Person2 to be aligned with higher purposiveness (höheren Zwecke), since pure reason is at issue in both cases. Higher purposiveness relates to ascertaining good and evil through pure reason, and taking this activity to be prior to the pursuit of happiness (KpV, 5: 61).
Second, Person2 as rational inner purposiveness must be considered a self-existent end, and therefore as having dignity, for two reasons. (i) As mentioned, nature itself is taken to have endowed Person1 with this purposiveness, i.e. endowed Person1 with Person2.Footnote 8 (ii) Person2 is non-substitutable and irreducible. Kant says that objective purposes are ‘things (Dinge) whose existence is an end in itself, and indeed such [an end] in whose stead no other end can be posited for which it could (stehen sollten) serve as means because without this [objective purpose] nothing of absolute worth could be found anywhere’ (G, 4: 428). On this basis, he concludes that persons are ends in themselves (ibid.). So, on Kant’s view, persons as objective purposes cannot be substituted for any other object (‘in whose stead no other end can be posited’). Here ‘person’ must be Person2 because objective purpose in the human case relates to the production of the good will (G, 4: 396). It follows that Person2 as end/purpose cannot be substituted, that is, it cannot serve merely as a means for any end associated with any other entity. The non-substitutability of Person2 also makes it irreducible, which means that it cannot be reduced to any other end in nature. Kant may have had this irreducibility in mind when he says that a human being should preserve his life ‘simply by virtue of his quality as a person (Qualität als Person)’ (MdS, 6: 422), and that a person is an ‘absolute unity’ (6: 278). That is, persons exist qualitatively speaking as absolute or irreducible unities. It is in this sense that Person2 can be considered an end in itself – an absolute unity that is qualitatively irreducible and non-substitutable with regard to any other end external to it. Naturally, at least in the human case, Person1 (as sensible/physical) is required for Person2 to exist, and Person2 expresses itself in making Person1 act in accordance with the good will (see 3.2.1 above). Consequently, a human person is an objective purpose if Person2 unifies the whole person. Such a person has absolute worth or dignity (G, 4: 428).
Hence, from 3.2.1 and 3.2.2, Person2 can be construed as the rational inner purposiveness that is non-substitutable and irreducible and that enables access to the moral law. It unifies the human person for the sake of the fulfilment of the higher purposiveness that nature has set for human beings: to act in accordance with the moral law. This constitutive structure of personhood has dignity.Footnote 9
3.3 Dignity of lawgiving
I have argued that for Kant the moral law has dignity (3.1), and human persons have dignity if unified in accordance with the rational inner purposiveness associated with Person2 (3.2). In addition, Kant attributes dignity to ‘giving oneself the [moral] law’ (selbstgesetzgebend) (henceforth ‘lawgiving’) (3.3.1), and to ‘virtue’ (Tugend) (3.3.2). This is because for Kant, humans can either choose the moral law or satisfaction of inclination. Giving oneself the moral law involves elucidating this law and willing actions in accordance with it. Further, to be moral, human beings must also be virtuous, i.e. strive to overcome hindrances preventing them from acting on the moral law. This is required because humans can esteem the law without following it.
3.3.1 Lawgiving
Human choice, Kant says, is ‘affected but not determined by human [sensory] impulses’, and it can be determined purely by reason (‘can still be determined to actions by the pure will’) (MdS, 6: 213). Yet human beings do not automatically follow CI, because they either do not ‘know’ (wissen) that actions derived from CI are good, or choose principles that are at odds with it (G, 4: 414). So following CI requires both knowing it and choosing to act on its basis. Since Kant takes the good will as already ‘present’ (beywohnet) in the ‘natural and healthy understanding (Verstand)’ (G, 4: 397; cf. 4: 454, KpV, 5: 36, MdS, 6: 438), knowing the moral law requires its ‘elucidation’ (aufklären) (G, 4: 397). Knowing the law, however, is insufficient: one must also give oneself the law. The will is not merely subordinate to the law, but subordinate to the ‘law which it gives itself’, and of which it is itself the ‘author’ (4: 431).
For Kant, lawgiving has dignity in two senses. First, the individual person giving herself the law can be said to have dignity: ‘[T]he lawgiving (Gesetzgebung) itself, which determines all value, must have dignity, i.e., unconditional, incomparable value. … Autonomy is therefore the ground (Grund) of dignity of human and every rational nature’ (G, 4: 435–6). Here, individual persons giving themselves the law (‘the lawgiving itself’) are said to have dignity. Similarly, Kant says that we attribute ‘sublimity’ and ‘dignity’ to the person fulfilling all her duties, not merely because she is subject to the moral law, but to the extent she gives herself the law (4: 440). Second, rational persons can legislate for all persons in the kingdom of ends. As Kant says, rational nature is an end in itself owing to the ‘suitability’ (Schicklichkeit) of its maxims for ‘universal lawgiving’, i.e. it can give laws that are valid for all persons in the kingdom of ends, and this lawgiving gives it ‘dignity’ (Würde (Prärogativ)) (4: 438).
3.3.2 Virtue
Kant says that human beings require virtue because their natural impulses are ‘hindrances’ (Hindernisse) in following the moral law (MdS, 6: 380; similarly, ‘hindrance’ at MdS, 6: 441, G, 4: 397, KpV, 5: 79). A lack of virtue can coexist with the ‘best will’ (besten Willen) (MdS, 6: 408), because humans have a ‘propensity’ to evil – frailty, impurity and depravity (R, 6: 29–30). While this propensity to evil is natural and cannot be entirely extirpated through human strength, human beings are ‘capable of fighting (bekämpfen)’ and ‘defeating’ (besiegen) these hindrances (MdS, 6: 380; ‘overcome’ (überwiegen), R, 6: 37). This requires virtue (Tugend (virtus, fortitudo moralis)), which is the ‘ability and resolve based on reflection (der überlegte Vorsatz) to resist the strong but unjust opponent [through] fortitude (Tapferkeit (fortitudo))’ (MdS, 6: 380.; also bekämpfen at MdS, 6: 405). The reflective resolve required for virtue consists in the ‘moral strength of the will’ to implement the moral law (6: 405; ‘strength of resolution’ at 6: 477, 384). The result is ‘self-constraint’ (Selbstzwang), and virtue (unlike right) ‘rests on free self-constraint alone’ (6: 383; also 394, 396). The moral purpose of all humans is to acquire virtue, which entails ‘progress[ing]’ from one perfection to another’ (6: 446; cf. KpV, 5: 32–3, 84; R, 6: 47–8).
Unlike autonomous lawgiving, human virtue for Kant is not merely an ‘aptitude’ (Fertigkeit) for following the law through the representation of the law (MdS, 6: 407). This is because beings with no desire to violate the moral law can autonomously give themselves the moral law. The virtuous person is not merely autonomous but ‘autocratic’, which involves the ‘not immediately perceived (wahrgenommenes) consciousness of the ability (Vermögen) to become master (Meister) over his inclinations that are intractable (widerspenstig) [with respect] to the law’ (MdS, 6: 383). According to Kant, the virtuous person, like the autonomous lawgiver, can also be attributed dignity. He says that the ‘morally good disposition or virtue’ (Tugend) has dignity, because its lawgiving makes it part of the kingdom of ends (G, 4: 435–6). Further, in MdS, Kant says that ‘if the dignity of virtue is not raised above everything, then the concept of duty itself disappears, and is reduced (zerrinnen) to mere pragmatic rules; since then the nobility of the human being disappears in his own consciousness, and he is “for sale” and can be bought for a price that seductive inclinations offer’ (6: 483, cf. 397).Footnote 10
In sum, from 3.3.1 and 3.3.2, dignity in the Kantian system can be attached to lawgiving, and to virtue as the reflective resolve to overcome our predisposition to evil.Footnote 11
4. The non-disjunctive interpretation
In sections 3.1–3.3, I isolated the various ways in which Kant employs the concept of dignity: the dignity of the moral law (DMoralLaw); the dignity of human persons as rational inner purposiveness (DPerson2); the dignity in giving oneself the moral law (DLawgiving), and in striving to follow the moral law (DVirtue). I now show (4.1) that all these senses of dignity are implicit in Kant’s argument at G, 4: 420–36. Based on this claim, in 4.2 I provide support for the non-disjunctive view that human dignity within Kant’s system combines DMoralLaw, DPerson2, DLawgiving and DVirtue. More specifically, human dignity for Kant is grounded in the reciprocal relationship between the dignity of the moral law and the dignity intrinsic to human beings.
4.1 Human dignity at G, 4: 420–36
Kant lays out multiple formulations of CI at G, 4: 420–36. I show that DMoralLaw, DPerson2, DLawgiving and DVirtue are either implicitly or explicitly present in these formulations.
DMoralLaw seems to be implicit in the first two formulations of CI: (i) The first two formulations presuppose a conception of the moral law without reference to humanity. In the first formulation (FUL) – act so that your action can simultaneously be willed as a universal law (G, 4: 421) – the moral law applies to all rational beings who act ‘according to principles, or a will’ (4: 412; KpV, 5: 32), and not merely to human beings. Therefore, the moral law is characterized in terms of rationality, not humanity. The second formulation is the formula of the law of nature (FLN): ‘Act as if the maxims of your action through your will could be (werden sollte) made into universal laws of nature’ (4: 421). Here the moral law is characterized in a way that goes beyond humanity, because it is formulated in terms of the concept of nature construed as a set of universal laws (ibid.), and humans form only one part of nature. Thus, FUL and FLN are formulated in terms of features that transcend humanity. (ii) The moral law has dignity sans any connection to humanity (see section 3.1). (iii) Therefore, since humanity is not the basis on which the moral law is characterized in (i) and (ii), FUL and FLN can be said to presuppose the dignity of the moral law abstracting from humanity (DMoralLaw).
The dignity of persons as rational inner purposiveness (DPerson2) appears to be implicit in the third formulation, FH – act so that you use the humanity in your person (and that of another) as end (Zweck) and never merely as a means to another end (G, 4: 429) – for the following reasons. (i) The ground of FH is that ‘rational nature exists as an end [or purpose] (Zweck) in itself’ (ibid.). Humans as rational beings ‘represent themselves’ (sich vorstellen) as ends (or purposes) in themselves, and take this as the ‘subjective principle’ in choosing actions (ibid.). Moreover, all that is an end in itself has dignity (4: 434–5; see section 2), although Kant does not mention ‘dignity’ in articulating FH (Sensen Reference Sensen2011: 144). Therefore, the ground of the FH claim that we must treat humanity as an end in itself is that human (rational) nature is an end in itself, i.e. has dignity. (ii) In section 3.2, I argued that human persons are ends in themselves to the extent that they exemplify rational inner purposiveness (Person2), and as such possess dignity (=DPerson2). From (i) and (ii): DPerson2 can be seen as implicit in FH.
The dignity of lawgiving (or DLawgiving) is explicitly mentioned in the fourth formulation. This is the formula of autonomy (FA) which ‘follows from’ FLN and FH (G, 4: 431). FLN forms the objective ground, because it formulates the CI as if it were a natural law. FH is the subjective ground of CI because all rational beings represent themselves as ends in themselves. In FA, the ‘idea of the will of every rational nature [is taken] as a universal lawgiving will’, which enables the ‘harmoniz[ation]’ of the will (subjective) and universal practical reason (objective) (ibid.). The will is subject to the universal moral law, but it must ‘give itself the law’ such that it can ‘regard’ (betrachten) itself as the ‘author’ of this law (ibid.). This is the ‘autonomy of the will’ (4: 433), which Kant explicitly says gives humans dignity (4: 434).
Finally, the dignity of virtue (DVirtue) is arguably presupposed in the formula of the kingdom of ends (FKE) – if all rational persons were to legislate CI autonomously, the kingdom of ends would be achieved (G, 4: 435). The ‘morally good disposition or virtue’ (die sittlich gute Gesinnung oder die Tugend) is supposed to ‘procure’ (verschafft) a ‘share’ (Antheil an) for rational beings in universal lawgiving (4: 435–6). This presupposes that a rational being has maxims that allow for universal lawgiving, but also that ‘the [rational being] subordinates (unterwirft) itself [to these maxims] at the same time’ (ibid.). In this passage, a virtuous rational being ‘procures’ a role in establishing the kingdom of ends by ‘subordinating’ itself to the moral law. Since both procurement and subordination are active verbs, it can be said that virtue is not some sort of default condition, but requires some doing. In the human case, virtue means the reflective resolve to overcome the non-rational impediments to following the moral law (see section 3.3). Consequently, FKE can be said to presuppose DVirtue.
To summarize: at G, 4: 420–36, FUL and FLN presuppose the dignity of the moral law. FH, FA and FKE presuppose the dignity of persons as rational inner purposiveness, the dignity of lawgiving and the dignity of virtue respectively.
4.2 DMoralLaw and DHumanity: A reciprocal relationship
If G, 4: 420–36 incorporates the various notions of dignity (section 4.1), then this passage can help us explore how DMoralLaw, DPerson2, DLawgiving and DVirtue might relate to each other. I propose that these different aspects of dignity can be shown to come together in Kant’s characterization of human dignity. First, I argue that DMoralLaw, DPerson2, DLawgiving and DVirtue are irreducible with regard to each other; and that DPerson2, DLawgiving and DVirtue can be conceptually subsumed under the dignity of humanity (DHumanity). Therefore, DMoralLaw and DHumanity cannot be reduced to each other, which further means that human dignity must somehow unite them (4.2.1). Second, I argue that, in Kant’s system, the relationship between DMoralLaw and DHumanity can be considered reciprocal since DMoralLaw and DHumanity can be seen to presuppose each other (4.2.2).
4.2.1 Human dignity relates DMorallaw and DHumanity
(1) DMoralLaw, DPerson2, DLawgiving and DVirtue can be considered irreducible in relation to each other.
(i) Without the moral law, persons could exist but not Person2. This is because Person2 is higher rational purposiveness internally connected to the moral law. However, it could be argued that the moral law could be present without Person2 in the Kantian system. The moral law in FUL and FLN is characterized in a way that transcends humanity (see section 4.1). But how does the moral law, metaphysically speaking, transcend humanity? Is the moral law embedded in the structure of things, or merely constitutive of the structure of reason? Kant says that we cannot answer this question. The consciousness of the moral law belongs to the ‘intellectual world’ construed as the realm of ‘pure activity … [or] what reaches consciousness immediately and not through the affection of the senses’, and humans have no ‘cognizance’ of this intellectual world (KpV, 5: 46).
However, as I now argue, the moral law can be considered independent of Person2 within the Kantian framework. For Kant, the moral law is embedded in the structure of reason, but it needs to be made explicit or brought to consciousness. Several passages attest to this claim. In Kant’s view, the moral law is present in ‘common human reason’, and enables the distinction between good and evil. Nevertheless, practical philosophy is required for common human reason to gain ‘information (Erkundigung) and distinct instruction regarding the source of its [moral] principle and its correct determination with respect to (Gegenhaltung) maxims based on need and inclination’ (G, 4: 404–5). Similarly, we need to ‘determine (bestimmte) concepts of morality and freedom’ or we would be unable to ascertain what to take as noumenon (KpV, 5: 6). Finally, awareness of the pure will or moral consciousness requires ‘attend[ing] to the necessity with which reason prescribes it to us’ (5: 30). Hence, the moral law could somehow be present indistinctly in Person1, but this person can consciously access the moral law only as Person2 (see 3.2.2). It follows that, conceptually speaking, the moral law is not reducible to Person2, since it can be present independent of Person2.Footnote 12
(ii) While Person2 allows access to the moral law, it is a rational inner purposiveness independent of the moral law. This is because if Person2 as a particular sort of structure were absent, the moral law could be present in some way (4.2.1(1)(i)), but remain beyond the reach of rational (human) persons. So Person2 is not reducible to the moral law.
(iii) The moral law and Person2 could both occur, but lawgiving might still be impossible. One might, for instance, imagine a being with access to the moral law (so exemplifying Person2), but with such strong non-rational inclinations that it could never give itself the moral law.Footnote 13 Obviously, such a being could not be virtuous either, because it need not overcome any hindrances to following the law if it cannot even give itself the law. Thus, lawgiving/virtue cannot be reduced to the moral law or to Person2. Further, virtue cannot be reduced to lawgiving, because one could give oneself the law without possessing the reflective resolve, or virtue, to act in tandem with the moral law.
(iv) Kant attributes dignity to the moral law, rational personhood and lawgiving/virtue (section 3). From (i)–(iii), DMoralLaw, DPerson2, DLawgiving and DVirtue cannot be reduced to each other.
(2) DPerson2, DLawgiving and DVirtue can be subordinated to the broader category of dignity of humanity (DHumanity). This is because each of these can be viewed as constituting different aspects of DHumanity (Person2, lawgiving and virtue) (4.2.1(1)(ii)–(iii)).
(3) Hence, from (1) and (2), DMoralLaw, DPerson2, DLawgiving and DVirtue are irreducible to each other, and DPerson2, DLawgiving and DVirtue can be subsumed under DHumanity. Consequently, DMoralLaw and DHumanity are irreducible vis-à-vis each other, which entails that the concept of human dignity must involve some sort of combination of the two.
4.2.2 Reciprocal relationship between DMorallaw and DHumanity
Since DMoralLaw and DHumanity cannot be reduced to each other, human dignity must involve a relationship between them (4.2.1). I now argue that DMoralLaw presupposes DHumanity, and DHumanity presupposes DMoralLaw. Consequently, within the Kantian system, human dignity can be construed as the reciprocal relationship between DMoralLaw and DHumanity (DPerson2, DLawgiving and DVirtue). Further, since DMoralLaw, DPerson2, DLawgiving and DVirtue are implicit in FUL/FLN, FH, FA and FKE respectively (4.1), the reciprocal relationship between DMoralLaw and DHumanity points to a way of coherently relating the various formulations of CI. This bolsters the non-disjunctive view, because it implies that this account of human dignity can plausibly fit into Kant’s overall account of the CI.
(1) DHumanity (DPerson2, DLawgiving and DVirtue) presupposes DMoralLaw.
(i) Unlike purely rational beings, it does not suffice for human beings merely to give themselves the moral law. Instead, they must strive to follow the moral law in the face of non-rational human inclination. This striving gives humans dignity (DVirtue) – the notion of dignity implicit in FKE.
(ii) Since virtue would be otiose in the absence of lawgiving, DVirtue presupposes autonomous lawgiving which has a dignity of its own (DLawgiving) – FA.
(iii) Lawgiving would not be an issue for human beings if human persons could never be construed as exemplifying Person2, i.e. exemplifying rational inner purposiveness. So DLawgiving and DVirtue must presuppose DPerson2 – FH.
(iv) DPerson2, DLawgiving and DVirtue must presuppose the moral law, because the concepts of Person2, lawgiving and virtue would not arise if the moral law were not implicated in each of them. Therefore, DMoralLaw is a necessary condition for DHumanity (DPerson2, DLawgiving and DVirtue) – FUL and FLN.
(2) DMoralLaw presupposes DHumanity (DPerson2, DLawgiving and DVirtue).
(i) DMoralLaw alone cannot constitute human dignity in the absence of the structural features of humanity that enable Person2, lawgiving and virtue – FUL and FLN.
(ii) Without the rational inner purposiveness associated with Person2, rational beings could not access the moral law. So DMoralLaw requires DPerson2 – FH.
(iii) Even if Person2 existed and could access the moral law, rational human beings would be unable to take up the law if they could not autonomously give themselves the law. Consequently, DMoralLaw and DPerson2 require DLawgiving – FA.
(iv) Even if rational human beings could access the moral law, and take up this law, it would mean little if they lacked the reflective resolve to follow the law in opposition to non-rational impulses. Therefore, DMoralLaw, DPerson2 and DLawgiving require DVirtue – FKE.
(3) As we have just seen, DMoralLaw requires DHumanity and DHumanity presupposes DMoralLaw. But DMoralLaw and DHumanity cannot be reduced to each other, and so human dignity must involve combining them (section 4.2.1). Therefore, human dignity can be interpreted as grounded in the reciprocal relationship between DMoralLaw and DHumanity (non-disjunctive view).
In sum, in sections 4.2.1 and 4.2.2, I have argued that the ground of human dignity within the Kantian system is a reciprocal relationship between the dignity of the moral law and the dignity of humanity (dignity of Person2, dignity of lawgiving and dignity of virtue), where all these various aspects of human dignity are irreducible vis-à-vis each other.
(4) As we have also seen, Kant’s view of human dignity is interwoven with the various formulations of CI (section 4.1). Thus any interpretation of Kantian human dignity must accord with a consistent account of the relationship between these formulations. I now show briefly that if we accept the non-disjunctive interpretation, we can offer a coherent account of how the different formulations of CI are related to each other. This further bolsters the non-disjunctive view.
The concept of dignity is implicit in FUL, FLN, FH, FA and FKE (4.1), and human dignity is the reciprocal relation between DMoralLaw and DHumanity (4.2). On this basis, FKE presupposes FA which presupposes FH that in turn presupposes FUL/FLN (4.2.2(1)); and FUL/FLN presupposes FH which presupposes FA that in turn presupposes FKE (4.2.2(2)). Therefore, FUL/FLN, FH, FA and FKE all presuppose each other reciprocally, which implies a systematic relationship between these formulations.Footnote 14
5. Concluding remarks
I have argued that Kant’s statements regarding dignity are consistent with the claim that the ground of human dignity involves a reciprocal relationship between the dignity of the moral law and the dignity of humanity. Specifically, it includes the dignity of the moral law, the dignity of personhood as rational inner purposiveness, and the dignity of giving oneself the moral law and striving to follow it.Footnote 15 Existing interpretations take some feature of Kant’s account of human dignity as primary as compared to other features, whether it is the capacity to set ends, capacity for morality, capacity for freedom, the good will or the moral law. In contrast, the non-disjunctive interpretation avoids this primacy approach with implications for both understanding and applying Kant’s theory of human dignity.
(i) Interpreters differ on whether merely possessing a rational nature gives humans dignity (Wood Reference Wood1999: 115–18; 2008: 85ff.; Hill 1992: 43–4); or if human dignity requires the full realization of rational nature, or good will (Dean Reference Dean2006: 40–2). From the non-disjunctive perspective, the G, 4: 429, claim that ‘rational nature is an end in itself’ is merely generic, because rational nature has dignity to the extent it is implicated in specific capabilities of humanity – Person2, lawgiving and virtue. At the same time, fully realized rational nature also has dignity since such a nature is Person2 manifesting itself by way of giving oneself the law and virtuously abiding by it.Footnote 16 Consequently, the non-disjunctive interpretation can accommodate Kant’s MdS, 6: 463, claim that even the vicious must be respected. This is because the vicious person has DPerson2 even though neither DLawgiving nor DVirtue, while the moral person possesses DPerson2, DLawgiving and DVirtue.
(ii) The non-disjunctive view opens up the possibility of an alternative interpretation of Kant’s theory of value. According to Reference KantKant, dignity means absolute, or unconditional and incomparable, value (MdS, 6: 435; G, 4: 436). Contemporary discussions on Kantian value revolve around the question of whether human beings confer value on things, or if things have value independent of human beings.Footnote 17 Sensen (Reference Sensen2011: 38; 2015) argues that Kant means ‘absolute inner value’ not as a metaphysical property, but merely as a way of describing what perfectly rational beings would value based on the moral law.Footnote 18 This view presumes that Kantian value is univocal – that there is one sort of value which can be either a metaphysical property or humanly prescribed. In contrast to Sensen, one could argue that, on the non-disjunctive view, DMoralLaw, DPerson2, DLawgiving and DVirtue represent different notions of value, each requiring a different analysis. This implies that value could be both metaphysically real and humanly prescribed. For instance, DMoralLaw could potentially be valuable as something metaphysically real,Footnote 19 as might be DPerson2, while DLawgiving/DVirtue entail value arising from human prescription. Hence, while Sensen conceives of Kantian value univocally, the non-disjunctive interpretation raises the possibility that value for Kant is multivocal.Footnote 20
(iii) In applying Kant’s concept of dignity to particular cases, the non-disjunctive interpretation can form a reasonable alternative to existing interpretations of Kantian dignity. Consider Kerstein’s position that applying the standard interpretation of Kantian dignity, i.e. that rational persons have unconditional and incomparable value in all circumstances, leads to counterintuitive results. To illustrate his view, he discusses (2014: 226) the case of a journalist who discovers financial irregularities in a large company, and who must now decide whether or not to shoot and kill a security officer of the company who is poised to kill him. Kerstein says that, according to the standard interpretation, killing the security officer Y would violate Y’s dignity.
The non-disjunctive interpretation, on the other hand, can coherently avoid this counterintuitive result. First of all, Y lacks DLawgiving and DVirtue, because his vicious action threatens the journalist’s life and therefore the latter’s dignity as exemplifying Person2. The choice, then, is between a vicious and valueless action and the dignity of a person. So the journalist would be within his rights to kill Y to preserve his dignity as a person (Person2), if killing Y is the only option to save his life. Further, in this situation, Y can be killed without violating Y’s dignity as a person (DPerson2). Kant’s account of punishment provides a clue to understanding how this might be possible. How a criminal is put to death determines if her dignity has been violated. Having the criminal torn by dogs or quartered violates the dignity of the criminal’s person (DPerson2) (MdS, 6: 463, see also 331–2), while giving him, say, a quick and painless death does not. Therefore, if the journalist’s anticipatory killing is analogous to the death penalty, then the non-disjunctive view, at least at first sight, can provide a viable framework for applying Kant’s concept of dignity.Footnote 21
