Nandi Theunissen: “Must We Be Just Plain Good? On Regress Arguments for the Value of Humanity”. Précis by Richard Kraut

Welcome to what we expect will be a very interesting and productive discussion of Nandi Theunissen‘s “Must We Be Just Plain Good? On Regress Arguments for the Value of Humanity.” The paper is published in the most recent edition of Ethics and is available through open access here. Richard Kraut kindly agreed to contribute a critical précis and it appears immediately below. Please join in the discussion!

Précis by Richard Kraut:

Nandi Theunissen’s goal is to reject a regress argument often used by Kantian philosophers for the conclusion that for anything to have relational value, there must exist beings who have non-relational value – who are “ends in themselves” – and that human persons are among those who are such ends.  Her method is to survey a variety of ways in which the regress argument might be formulated, and to pose problems for each of them. In the end, she arrives at a positive conclusion: we can make good sense of the idea that we are ends in ourselves by recognizing that we have relational value. We matter not because we are “just plain good” (as her title puts it) but because we are good for someone.  Good for whom?  Quite often, for other human beings; but above all, she holds, one’s status as an end rests on the fact that one is good for oneself: one is “at the center” of one’s own life, which has the potential to be a good life for oneself.

Her discussion begins with a reminder of Christine Korsgaard’s formulation of the regress argument, which seemed to rest on the metaphysical premise that value must have a first cause; but which she later reformulated to read that we cannot be agents unless we confer value on ourselves. This is a transcendental move, not a regress argument, and so, in search of the latter, Theunissen turns next to regress arguments formulated by David Velleman and Joseph Raz. Here the basic idea is that being good for (being advantageous or beneficial) is not necessarily of normative significance. Washing plates is good for having clean ones, and clean plates serve some further purpose, and so on – but there would be no value in seeking these goods unless there is a different kind of value: not good for, but good in itself. Persons (valuers) might have instrumental value, but what puts an end to the regress is that, because they are valuers, they also have non-relational value.

This claim must be distinguished from the thesis that for anything to be instrumentally beneficial, there must be at least one good that is good for someone by being a component of that individual’s well-being and in that sense non-instrumentally beneficial. That thesis was affirmed by Plato and Aristotle; but what is at issue for Theunissen is whether a regress argument can show that not only are there relational values (the components of well-being and the means of achieving it), but a property human beings have in virtue of which their well-being takes on normative significance.

In Section 3, she schematizes the regress argument into five premises and a conclusion. The main idea is that there are chains of value-dependence that must come to an end with something whose “value does not depend on being good for other things.” Valuers meet this description and therefore “they can be the final node in a chain of dependence.” (I was surprised by “can be,” which seems weaker than what Kantians typically have in mind. Don’t they generally hold that it is only rational willing that supplies the condition for anything having value?)

Theunissen surveys, in Section 4, several ways of casting doubt on the regress argument formulated in Section III. They prove unsuccessful,  but there is one that brings her close to her final analysis of where the argument goes astray. In  Section 4.5, she suggests that chains of value-dependence might have a circular structure. As she puts the idea, “relational goods are valuable because they can be good for valuers, and valuers are valuable because they are good for other valuers.”  The “structure of value” (as she calls it) allows X to have value by being valuable-to-Y and Y to have value by being valuable-to-X.

In Section 5, she notes that this circular structure is at odds with what she calls the “metaphysical principle” that the world must contain something positively and non-relationally valuable (something that is good period), if anything is to have relational value. Against this, she adapts an idea of Earl Connee, who notes that instrumental means might all borrow their value from their making the world less bad; nothing need be good, for that to happen. Theunissen transforms this idea into a challenge to the Kantian regress argument, by pointing out that human beings might be valuable to each other simply by diminishing each other’s suffering. In a world where we all are good for each other in this way, value would be entirely relational. People need not have positive non-relational value for there to be duties to aid others.

In  Section 6, she moves to her favored response to the Kantian regress argument for the non-relational value of persons. She agrees that there is something to be said for the circular structure proposed in Section 5: The thought might be expressed by saying: “I am because we are,” “a person is a person through other people.” But Theunissen proposes that we should be not only for others – we should also be for ourselves. Simply by being a benefactor of oneself one can bring the chain of justification and the threatened regress to an end. As she puts it, “we should relate to people always with a view to their being a center of a life to which they bear a special relation.” Moral philosophy, she insists, is not confined to the study of duties owed to others; Kant himself insists upon duties to oneself. She adds: “in order to benefit another we must truly have something to give … Who is the good person?  She is the person who benefits … Who does she benefit? Others, of course, but also, and importantly, herself.  She has something to offer another because she is a master of one – she has knowledge of how to be well in her own life.” Contrary to the Kantian idea endorsed by Velleman and Raz, the chain of dependence of relational goods can come to an end with a reflexive relation. There need be no non-relative value.

These are some of the main currents of thought in this remarkably rich and provocative exploration of the structure of value. I look forward to their further elaboration in her forthcoming book, The Value of Humanity. In what remains, I raise a few questions for discussion.

First, what is wrong with a regress? Simply from the premiss that A has value only if B does, and B has value only if C does, and so on without end, it does not follow that nothing has value. Theunissen notes in Section 4.4: “It is something of a reflex to find infinite regresses intolerable, though what is amiss is not always fully made out.” She replies: “it seems implausible for the existence of value to depend on there being infinitely many things.” True – but after all, there are infinitely many things (e.g. numbers), and so we could accept this as a condition on the existence of value.  Surely the important point is that if the value of everything we did were merely instrumental, there would no reason to do anything. If there is any value at all in the world, some of it must be non-instrumental.

Second, Theunissen seems to set aside an issue that some philosophers would put at the center of her topic. In Section 5, she notes that for the Kantian, the value of alleviating pain is dependent on the value of the individual who is experiencing the pain. By contrast, for a follower of Bentham or Moore, pain is always a bad thing, to whomever it occurs, and so there is always a reason, from an impartial perspective, for anyone to wish that it not occur. Theunissen remarks: “To my mind each side should permit the other their basic foundational assumptions, and the positions evaluated on their own terms.” But can she remain neutral in this way? She seems to help herself to the assumption that the well-being of human beings (or persons or valuers) – particularly one’s own well-being – has normative significance even if it might be the case that the well-being of animals has less or none. But that seem to leave unanswered a legitimate question: what gives your own well-being this special status (or what gives human beings a higher status than that of other creatures)?

A third question concerns a hypothetical individual who is near death and suffering a great deal. He is, in his present condition, good for no one – not for others, and not for himself. Having sunk into passivity, the only question that remains is how much he will suffer. Surely he ought to be comforted and his pain diminished, even though he is good for no one.

Fourth, I take Theunissen to be defending the idea that well-being ought to occupy a central role in moral philosophy and in ordinary practical thought – and doing so without entering into the debate about what exactly well-being consists in. She holds that what is good for oneself should matter to oneself. But might there be important goals (for example, personal perfection, manifested in artistic or intellectual accomplishment) that are not in one’s interest (or anyone else’s) but should be pursued nonetheless? That question, I suggest, cannot be satisfactorily answered in the absence of a theory about what well-being is. We cannot be sure how much it matters without knowing what it includes.