When Kant claims that existence of rational beings is of absolute value, that persons are not merely subjective but objective ends (Groundwork, 4: 428) and that humanity possesses dignity (Groundwork, 4: 435), some read him as a value realist asserting that there is something of value independently of whether any being, including itself, does or could recognize the value of that being. This independent value would constrain moral beings’ behaviour in that the categorical imperative demands that actions be such that they would make this value the necessary end not only of actions but of the very moral law itself. To put the claim in its starkest terms: were something other than rational beings to possess this independent absolute value or dignity, the categorical imperative would have to demand that moral beings treat that other thing as an end in itself, because by itself the categorical imperative would not provide this ultimate value but can only conform itself to it. Any absolute value would be indemonstrable since providing reasons for the value would ground that value on something else and thereby violate its absoluteness.
Oliver Sensen presents the most thorough argument against this type of interpretation yet to appear in the literature. He shows why value in Kant must be derived from the categorical imperative. He provides interpretations of Kant’s uses of seemingly value-indicating terms such as ‘end in itself’ and ‘dignity’ that treat them as non-normative descriptive terms. He describes his book as ‘an attempt to see how far one can read Kant without invoking a foundational value, and it seems to me that one can do so all the way down’ (3).
Despite the title, dignity as such is only half of the story of this book. The first half argues directly against value realism before Sensen turns to his account of dignity. Chapter 1 provides an argument against the kind of value realism that takes value to be an intrinsic property of human beings that must be recognized by moral agents. This chapter has a metaethical focus, largely abstracted from the question of the particular nature of humanity that is of value. He offers several reasons. Among them he notes that Kant does not include the value of humanity when listing goods such as happiness and perfection in his delineation of heteronomous principles in philosophy; here Sensen claims that Kant’s list is intended to exhaust all possible conceptions of value, but it seems limited to those conceptions of value independent of the will itself, a possibility Sensen treats in his second chapter. He also argues that there is no place for it in Kant’s epistemology since human beings would have no access to any independent value property in objects given that we can neither sense any value property directly nor intuit it in any non-sensible way, and if we discern it indirectly through our responses we can only understand it in terms of pleasure. He also correctly rejects a claim that an ontological superiority of thing-in-itself over appearance implies any value hierarchy.
If value is not an independent non-natural property in single objects, Sensen argues, it can be one of only three things: a relation among objects (which he rejects since these relations can be understood only in instrumental terms), a fact about what rational beings do value (which he rejects as merely empirical) or ‘a prescription of what one should value’ (28, cf. 32–3). He takes this third option, holding that the faculty of practical reason deeming something necessary gives it value. I think he is correct here, but his choice of words obscures the point. ‘Value’ should not be defined as ‘what one should value’, since even this latter prescriptive formulation retains the term ‘value’, albeit as a verb rather than as a noun. Rather, he means that ‘value’ should be defined only in terms of what reason dictates, so to say that humanity is absolutely valuable is simply to say that practical reason dictates that rational beings always be treated as a limiting condition for maxims.
The second chapter argues that a second possible way to understand value, one that takes value to be ‘internal to one’s own will’ (53), falls short for two main reasons. First, even if these kinds of arguments are successful they do not prove that this value holds prior to and independent of reason’s identification of them, and second that these arguments are in fact not successful in proving their claim that an objective value is internal to the will. Sensen criticizes the interpretations by Korsgaard, Wood and Guyer, who see this inner value in relation to the ability rationally to set ends or freely to choose them, and Dean and Kerstein, who see this inner value as the goodness of a good will. In each case, however, Sensen claims that even if these interpretations were successful, they do not conflict with his own main claim that the value at stake is something that one is rationally committed to and not something prior to and independent of practical reason.
Chapter 3 provides a novel reading of the formula of humanity to show that it supports the claim that respect for others is a command of reason not grounded in anything else. Sensen goes to the heart of the dispute by interpreting Kant’s claim that ‘The ground of this principle is: rational nature exists as an end in itself’ (Groundwork, 4: 428) as not invoking any independent value. Sensen equates end in itself with freedom – a descriptive rather than prescriptive identification – making Kant’s claim ‘rational nature exists as free’, or simply, ‘rational nature is free’, as the ground for the categorical imperative. The formula of humanity then reads roughly ‘treat free beings not as a means but as free’. He finds his main support not in the Groundwork itself but in the course lecture Naturrecht Feyerabend that Kant gave while writing the Groundwork. In Feyerabend Kant declares that rational beings are ends in themselves not merely because they have reason but because they have freedom. Sensen stresses the way that Kant there, and less clearly in the published works, argues that nothing determined only by natural causal laws could be anything but a means, so only a being free from natural causation could be free and thus an end in itself. In the Groundwork humanity is defined as freedom in terms of the capacity to be moral. The formula of humanity then states that your maxim must be one that other free beings could adopt as well and is then more closely linked to the main categorical imperative centring on universality than normally thought. Free beings simply are ends in themselves who must all be capable of agreeing to the maxim and so should be treated as ends in themselves; the non-normative reading of ‘end in itself’ as free is the basis for the ascription of value to persons as ends in themselves by reason.
I am not convinced that such a non-normative reading of ‘end in itself’ is feasible. The passages Sensen invokes can better be read as Kant declaring that rational beings are to be treated as ends in themselves because of the fact that they are free rational agents, implying that humanity as freedom is of value. One may then ask why freedom would ground the value status of an end in itself. Sensen could point out, as some interpreters do in a way Sensen acknowledges is compatible with his approach, that humanity is a free exercise of the rational will, and that since the categorical imperative stems from the rational will, the rational will is commanding itself to create and maintain itself. This deflationary interpretation would allow that ‘end in itself’ is not descriptive but still understand it to be a normative status resulting from reason’s command rather than an independent value. This move would also respond directly to the value realist who asks why the formula of humanity would demand that moral agents treat humanity always as an end if humanity is not independently valuable in itself. Sensen could stress that it is no coincidence that the categorical imperative stemming from reason places the highest value on the exercise of reason through action in conformity with the categorical imperative. It would be quite a coincidence if there were an independent value property for the free exercise of reason in action that happens to be the only possible basis for a law stemming from reason itself. If the value realist is right and there is no further reason for that value to be assigned to humanity (because by hypothesis the value is sui generis), then it would appear that the rational will simply lucked out that it is able to command as a moral imperative the maximization of itself. I think Sensen generally agrees with this reasoning but wishes further to read ‘end in itself’ as a property in the moral object (freedom in rational beings) rather than a moral status imposed by reason (the limitations on treatment) and hence sees end in itself in descriptive rather than prescriptive terms.
Further support for Sensen’s reading of ‘end in itself’ as a non-normative description is found in the second half of the book where Sensen provides a fascinating account of the concept of dignity. He deepens his overall argument by showing that Kant had available and used a conception of dignity that is not an intrinsic value nor even a value assigned by the moral law. Following Cicero in particular and the Stoics in general, Kant uses the term ‘dignity’ to indicate an elevation in status. The dignity of mathematics (Critique of Pure Reason, A464/B492) and the dignity of a teacher (Religion, 6: 162) are examples of this usage. Chapter 4 compares this conception of dignity with the contemporary usage of this term in which human dignity is supposed to be an intrinsic value that grounds respect and human rights. In contrast the traditional understanding of dignity is based on a non-normative elevation of human beings over the remainder of nature in virtue of certain capacities humans possess, particularly reason and freedom (153). Only after this distinction is made is morality brought into the picture, and generally as a claim that one has a duty to actualize one’s dignity to the fullest. The elevation is not a hierarchy of value but only a factual distinction and does not serve as a basis of rights but instead of duties, primarily duties to oneself. Sensen traces this conception from Cicero through Pope Leo the Great to Pico della Mirandola, referencing a large amount of literature in the history of ideas unknown to most Kant scholars. Anyone reading this chapter will greatly benefit from the account it provides in seeing how Kant understood this crucial term.
In the remainder of chapter 4 and all of chapter 5 Sensen then shows how Kant used the term ‘dignity’ in this way in various writings, with a focus on the Groundwork and Doctrine of Virtue. These latter are particularly challenging, given passages such as ‘inner value, that is, dignity’ (Groundwork, 4: 435) and ‘dignity (an absolute inner value)’ (Doctrine of Virtue, 6: 435). Sensen takes passages such as these to be discussing the dignity of morality, not human beings, above all other values that have prices, so the dignity is not itself a moral value but is the elevation of moral value over non-moral value. He catalogues the 111 times Kant uses the term ‘dignity’ and shows that in many of them Kant can be understood only to mean elevation over others in a non-moral sense, as in the dignity of the state authorities (Doctrine of Right, 6: 315) and the dignity of philosophy itself (Critique of Pure Reason, A62/B86). He also notes that Kant does not mention the dignity of humanity in crucial places when attempting to justify the moral law (e.g. Groundwork, III), a strange omission for what some take to be the foundation of the moral law.
Given the sheer number of passages in question, some of his readings are strained. In the core Groundwork passages I do not think that Sensen explains away all traces of value in the term ‘dignity’. But I also do not think that he has to in order to succeed in the broader aim of the book, that is, to show that Kant understands moral value as something stemming from the command of reason through the categorical imperative rather than as an inherent property of beings or wills. In this one can retain the link between value and dignity as elevation that Sensen allows in the first part of the book and can understand humanity as having its value assigned by the rational will in elevating humanity over all other possible ends. Sensen has provided a successful detailed sustained argument against a value realist interpretation of Kant and sets the starting line for the next stage in this debate.
