In Against Absolute Goodness Richard Kraut aims to show that absolute goodness (or badness) is not reason-giving; it plays no role in justifying or requiring certain attitudes and no role in reasoning about what to do (p. 6). However, the notions of ‘a good R’ (e.g. a good play) and ‘good for S’ do justify certain attitudes and play important roles in practical reasoning. This book is unique in addressing the many ways in which absolute goodness has been put to use by Western moral philosophers from Plato to contemporary philosophers such as David Velleman and Jeff McMahan. The writing style makes the arguments accessible even to an introductory student, and the topics touched on arise in metaethics, ethical theory and applied ethics.

Kraut offers an argument by elimination to show that absolute goodness is not reason-giving. He begins in chapter 9 with the example of pain, which most of us would agree is bad for the one who has it. When practically reasoning about pain (e.g. whether to make it go away etc.) we attend to the fact that it is bad for the one in pain but we do not also attend to whether it is absolutely bad (p. 45). Moreover, if we did take both considerations into account (its being bad for someone and its being bad absolutely bad) this would be ‘double counting – assigning pain more disvalue than it actually has’ (p. 46). Kraut claims that this is not something special about pain, but that nothing is ‘doubly valuable or disvaluable’ (p. 47). He points out that the ‘intensification of reasons is what normally happens when two reason-giving properties are present in the same situation. When no such intensification occurs, we must wonder why. The explanation, I suggested, is that absolute goodness is not a genuine reason’ (p. 79).

Kraut proceeds to consider various contexts in which one might have thought that absolute goodness is reason-giving, such as whether it is a reason to do what is not good for another (chapters 14–15) or a reason for our attitude towards beauty and our choice to bring it about in the world (chapters 16–18). He also considers whether the following are absolutely good: love (chapter 19), equality (chapter 24) and the value of persons (chapters 21, 25, 27, and appendices A and B). In each case, Kraut attempts to persuade the reader via examples that absolute goodness is either too weak a reason to override what is good/bad for others, or is redundant relative to what is good for others and thus ‘in effect it can be either entirely or largely ignored in our practical reasoning’ (p. 80).

If absolute goodness is never itself reason-giving, does it then pass the buck? Kraut affirms that ‘the work of specifying a reason in favor of something is done by other features, not by its goodness and value. . .for a thing to have the property of being good or valuable is simply for there to be some reason or other to value it’ (p. 54). Absolute goodness ‘never adds to the weightiness of more specific reasons’ and so ‘for practical purposes we can ignore it, because we will never go astray if we direct our attention to those more concrete reasons’ (p. 59).

However, when it comes to the notions ‘a good R’ and ‘good for S’, these are not, according to Kraut, buck-passing notions (p. 60). And yet, he also claims that each notion ‘is a supervenient or consequential property, and that when it is instantiated, that is because a further underlying property is also instantiated’ (p. 60). For example, that a play ‘is a good play does not add to the force of the argument that has already been made for seeing the play by citing the specific factors that make it a good play (characterization, plot, and so on)’ (p. 58). Rather the non-evaluative underlying property ‘gives a reason for accepting’ that it is a good play (p. 61). And so it is not clear why Kraut believes that ‘to say it is a good play goes far beyond saying merely that there is some reason to see it’ (p. 57). Instead, it seems, at least in the case of ‘a good R’, that it is as buck-passing as absolute goodness.

In the case of ‘good for S’, Kraut claims that there is no underlying non-evaluative property that can do its justificatory work (p. 57). For example, the mere fact that something is pleasurable for Y does not entail that it is good for S (p. 56). That's true. But later Kraut claims,

Do such biological properties entail that something is good for S? If so, then ‘it is good for S’ does not add to the weightiness of these underlying reasons and so for practical purposes it seems it could be ignored.

Kraut discusses a variety of philosophers, but I’ll focus on just two in this review: Aristotle and Kant. He points out that Aristotle's use of ‘without qualification’ is not meant to indicate absolute goodness (appendix F). That seems right, but Kraut did not discuss Aristotle's notion of ‘the fine’ or ‘noble’ (καλόν), and I wonder what he might say about it. According to Aristotle, ‘the things pertinent to choice [are] three. . .what is fine, what is advantageous, and what is pleasant’ (Nicomachean Ethics, 2002, 1104b30–33). Aristotle is clear that the fine is not reducible to or replaceable by what is advantageous (good for), but rather constrains the pursuit of what might otherwise be advantageous. Perhaps, then, ‘the fine’ is where one could locate absolute goodness in Aristotle: it is neither buck-passing nor replaceable by the notion of ‘good for’, and it is reason-giving for praising, admiring and taking one, rather than another, course of action.

According to Kant, the good will is absolutely good. To convince us otherwise, Kraut offers an example in which ‘one's agency is. . .so ineffectual that one cannot actually produce what is good for others or oneself’ and one is filled with ‘pain and anguish’. Kraut concludes that life ceases to be worth living even though ‘one's rationality remains intact’ (p. 129). But in this thought experiment, does one have a will any longer, if one is wholly ineffectual throughout one's lifetime? It is one thing to be ineffectual in a given circumstance and another for this to occur throughout one's life. It seems one's rationality has been reduced. One is able merely to think about what it would be good to do. But a will, according to Kant, is a ‘capacity to act in accordance with the representation of laws, that is, in accordance with principles’ (Groundwork of the Metaphysics of Morals (1998), p. 23; the italics are mine). It is not mere rationality, but a good will that Kant holds to be absolutely good.

Kraut suggests that taking into consideration what is good for oneself and others is a sufficient condition for having a good will (pp. 123–4), but Kant would disagree. Rather one's will is good when the principles of one's will are the hypothetical and categorical imperatives, where the latter unconditionally commands that we treat practical rationality as an end (Groundwork, p. 429). While this might sometimes involve pursuing what is good for others (Groundwork, p. 430), it needn’t, and it in fact prohibits paternalism. Kraut, unlike Kant, grounds respect for practical rationality in what is good for others (p. 148), and hence, when the two conflict, Kraut endorses forms of paternalism that Kantians reject. I’m uncertain how far Kraut's paternalism extends. Would he endorse paternalism in the case of Dax? Dax had no false beliefs and refused painful medical treatments (lasting several years) that were likely to lead to a long life worth living. His refusal was ignored and he did come to have a long life worth living, though he never wavered at any time during his life in thinking that his choice should have been respected rather than doing what was good for him (‘Case Presentation: Donald (Dax) Cowart Rejects Treatment – And Is Ignored’, 2000).

Kraut points out that philosophers appeal to the absolute goodness of sophisticated cognitive capacities, such as practical rationality, to explain why creatures with this capacity have a greater value than creatures without this capacity (chapter 25). Kraut, on the other hand, attempts to explain this difference in value by appeal to the notion of ‘good for’. According to Kraut, ‘the advantages [the wasp] has are not as good for it as are the advantages of a human being who is living as well as a human being can’ (p. 153, see also p. 164). One way to understand this claim is as a statement about the ‘richness or poverty of those capacities’ (p. 154), which seems to be a statement about which capacity is more valuable (period) rather than a statement about whether and to what degree the capacity is good for the one who has it. Alternatively, the claim can be understood to be on a par with Kraut's claim that ‘human beings can be healthier than dogs’ (p. 158), which I find quite puzzling.

It was only by thinking through the many arguments raised by Kraut that I was able to begin formulating for myself in what ways absolute goodness might or might not be a necessary notion in moral philosophy. Not only does Against Absolute Goodness skilfully question the role of absolute goodness in moral philosophy, while also highlighting the importance of the notion of ‘good for S’, but it also contains many helpful distinctions and interesting side arguments. For example, Kraut distinguishes absolute goodness from objective goodness (p. 22), and what is instrumentally and non-instrumentally good for someone (chapter 7). He also argues that ‘good for S’ cannot be reduced to, and is not dependent on, what is absolutely good (pp. 31, 33). I think readers will find it rewarding to examine both the main and supplementary arguments in this book.