I’ve just finished a draft of a review of Toni Rønnow-Rasmussen’s book, Personal Value. After some soul-searching, I decided I couldn’t give it the thumbs-up that I’d been hoping to when I first cracked the cover. However, there are some good ideas in it, which I hope I’ve emphasized in the review.
In her critique of Moore’s (1903, p. 55) suggestion that one might answer the question “What is good?” with “Books are good,” Judith Jarvis Thomson (1997, p. 276) asks what it could mean to say that books are “just plain” good or bad. Aren’t they rather good to read, or in teaching philosophy, or for weighing down papers? This point is apropos in a review of Toni Rønnow-Rasmussen’s Personal Value in two ways. One might worry, first, about his facile dismissal (p. 7) of the notion that such first-order attributions of goodness (e.g., being good to, good at, good in, and so on) have anything to do with the “genuine evaluative meaning” of the term. This might then lead one to ask whether Rønnow-Rasmussen’s book is good in any way at all. After some struggle, my own answer to this question is a reluctant “yes.”
Rønnow-Rasmussen’s chief aim is to argue for a particular way of carving up the space of values. He wants to demonstrate that it’s worthwhile to countenance personal values, which are good for some agent or other, in addition to impersonal values, which are good, as he says, period. In his effort to show this, he relies primarily on two methods: argument by ostension, and the drawing or undermining of distinctions: between intrinsic and extrinsic values, between instrumental and final values, between the supervenience base for and the constitutive ground of values, between object-given and state-given reasons, between identifying and justifying reasons, and between agent-relative and agent-neutral reasons. While his positive proposal is interesting and somewhat novel, its development is marred by superficial summaries of the existing literature, jarring non sequiturs, sloppy argumentation, superfluous formalism, and jingoistic citation patterns.
Personal Value comprises three parts. Chapter 1 sets the scene by drawing several value distinctions and pointing to plausible examples to support them. Chapters 2 through 8 are the meat of the book, in which Rønnow-Rasmussen articulates, clarifies, and defends a fitting-attitude (aka buck-passing) analysis of personal value, according to which an “object x’s value for a person a (i.e. x’s personal value), consists in the existence of normative reasons for favouring/disfavouring x for a’s sake” (p. 47). In chapter 2, the fitting-attitude analysis of value in general is introduced through its historical progenitors, Franz Brentano and A. C. Ewing. Chapter 3 presents Rønnow-Rasmussen’s take on what has come to be known as the wrong-kind-of-reason objection to the fitting-attitude analysis. In chapter 4, which unfortunately spans only nine pages, Rønnow-Rasmussen introduces the fitting-attitude analysis of personal value. Since this analysis relies on the hitherto unanalyzed notion of favouring an object for someone’s sake, chapter 5 attempts to clarify this notion. Next, chapters 6, 7, and 8 place the current proposal in the context of others’ discussions of personal value (Hurka, Regan, Rosati) and welfare (Darwall, Heathwood). Chapters 9 and 10, which take up the distinction between agent-relative and agent-neutral reasons and the concept of value-bearer pluralism, respectively, make up the third and final section. For all I can tell, neither of these chapters bears more than a glancing relation to the purported topic of the book, so I will not discuss them further.
Before turning to Rønnow-Rasmussen’s definition of personal value, it’s useful to summarize two of the key distinctions from chapter 1: the difference between instrumental and final value, and the difference between intrinsic and extrinsic value. Something has instrumental value when it is good in virtue of what it enables (e.g., a ladder) or prevents (e.g., a prophylactic). Less helpfully, something has final value when it is good non-instrumentally, or, as Rønnow-Rasmussen puts it, “for its own sake.” Examples include health, welfare, and pleasure. Next, something is intrinsically good when all the properties that make it good are internal to it (e.g., a just society), and extrinsically good otherwise (e.g., a rare stamp, Napoleon’s hat, an unspoiled old-growth forest). Rønnow-Rasmussen persuasively argues that these distinctions are orthogonal; in particular, he cites convincing examples of extrinsic final values – things that are good for their own sake but only in virtue of their relational properties. Dialectically, this is a wise move, since personal values as he goes on to analyze them turn out to be necessarily extrinsic, even when they are final.
The fitting-attitude analysis of value grounds the evaluative in the normative: what’s good is whatever there is all-things-considered reason to favour, where favouring is an umbrella notion for all pro-attitudes, including desire, preference, admiration, respect, and love. What’s bad is whatever there is all-things-considered reason to disfavour, where disfavouring includes all con-attitudes, including aversion, hatred, contempt, disgust, and fear. This analysis has much to recommend it. However, it faces objections. One, which I have spelled out elsewhere (Alfano 2009), arises from the fact that it is sometimes fitting to harbor contrary sentiments towards the same object. One might have all-things-considered reason to respect and fear a powerful person; one might have all-things-considered reason to feel love and disgust towards one’s baby when it has diarrhea. Rønnow-Rasmussen’s response is to say that, when such cases arise, the object in question is neither good nor bad – which seems to me to conflate fitting warranted ambivalence with warranted indifference.
According to another, more widely discussed, objection (the “wrong-kind-of-reason” objection), the fitting-attitude analysis misses the mark in both directions: one can have all-things-considered reason to favour something that is valueless or negatively valued, and one can have all-things-considered reason not to favour something that is positively valued. In most such cases, “it is seemingly the pro-attitudes themselves, rather than the objects, that are valuable” (p. 34). What’s valuable in these cases is not the object of favouring, but the favouring itself. Rønnow-Rasmussen presents contrived examples of such cases (e.g., an evil demon threats you with eternal torture unless you admire a saucer of mud, p. 34), but plausible examples are not far to seek. Placebo effects, Jamesian will-to-believe cases, and beginnings of romantic relationships have the required features. Your respect for a doctor might contribute to your recovery from an illness; your pride in your own ability might give you the confidence to hit a home run; your love for another person might induce him to requite the love. In all of these cases, though, it is the pro-attitude itself, rather than the object of the pro-attitude, that provides a reason.
After considering and rejecting the suggestion that the right kind of reason is always object-given (Parfit 2001, p. 21) rather than state- or attitude-given, Rønnow-Rasmussen concludes that the wrong-kind-of-reason problem remains unsolved. He then refuses to revise the fitting-attitude analysis to make favouring for the right reason a primitive, presumably because the notion of a right reason is obscure, but also refuses to draw the further conclusion that the problem is insoluble. This open-minded, corrigible approach extends throughout the book. The author frequently reminds his readers that he is not trying to defend the fitting-attitude analysis (or his extension of it to personal value) against all comers. Instead, his work should be seen as an exploration of some of the positive features of the analysis.
Assuming, then, that the various objections to the fitting-attitude analysis of value can be defused, the proposal that Rønnow-Rasmussen advances for analyzing personal value is quite attractive. Whereas something is good if and only if favouring it is warranted, something is good for some agent a if and only if favouring it for a’s sake is warranted. The chief innovation here is to relativize to the agent not in the subject but in the content of favouring. Rønnow-Rasmussen is not claiming that what’s good for you is what you have all things-considered-reason to favour, but rather that what’s good for you is what anyone has all-things-considered reason to favour for your sake. Thus, your humility may be good for you even though it would be gauche for you to favour it. This novel way of understanding personal value immunizes Rønnow-Rasmussen’s view against criticisms of the notion of good for developed by G. E. Moore (1903), Thomas Hurka (1987), Donald Regan (2004), and Connie Rosati (2008).
It seems promising to understand personal value in terms of warranted for-someone’s-sake favouring, which inevitably leads to the question of what it means to favour something for someone’s sake. The interested reader will at this point be frustrated. Presumably, the most natural way to approach the question would be to articulate a theory of what it means to favour something for something’s sake, and then to narrow the account to an analysis of favouring for someone’s sake. Avocados are good for humans, but not good for rabbits. Honey is good for adults, but not for babies. Sleeping pills are good for insomniacs at night, but not for crane operators on the job. Such an approach would then lead to a discussion of human nature, since the nature of an object seems in large part to determine what’s good for it. Rønnow-Rasmussen, however, doesn’t even consider this approach. Moreover, given the unsystematic way he approaches the question, “the prospects of specifying the characteristic features” of a for-someone’s-sake attitude are, as he himself admits, “bleak” (p. 56).
At the very least, though, the warrant for favouring something for a’s sake must refer to a’s properties asjustifiers or identifiers (p. 56). Reasons make favouring something fitting (or not). Somewhat crudely, we might say that a reason for xing is whatever could be successfully invoked in a satisfactory justification of xing. You might admire someone for her wit. The question, “Why do you admire that author?” could be satisfactorily answered by saying, “She’s a wit.” You might love someone because she’s your mother. The question, “Why do you love her?” could be satisfactorily answered by saying, “She’s my mother.” Given such attitudes, you might protect a book from fire because you admire its author, or care for a stranger for your mother’s sake. In the former case, your admiration will be more or less justified depending on the quality of the author’s writing; the properties of the agent that warrant favouring her are thus justifiers. In the latter case, your love will not depend on what your mother is like but on who she is; the properties of the agent that warrant favouring her are thus identifiers. Rønnow-Rasmussen plausibly claims that for-someone’s-sake attitudes that rely on identifiers are “more interesting” than those that do not because they can be “expected to have a [more] substantial impact on a person’s life” (p. 56).
One last point before wrapping up. Since the distinction between personal and impersonal values cuts across the distinction between instrumental and final values, a final personal value on Rønnow-Rasmussen’s analysis is something that it’s fitting to favour for its own sake for someone’s sake. This idiom is admittedly infelicitous (p. 58), but that does not necessarily make it incoherent. After considering several potential interpretations, Rønnow-Rasmussen urges that the best way to construe the assertion “a favours x for its own sake for b’s sake” is not as the attribution of two attitudes to a (favouring in a final way and favouring for b’s sake) but rather as the attribution of a single complex attitude:
the first ‘sake’ in ‘favouring x for its own sake for b’s sake’ specifies that the value accruing to x need not be conducive to something else that is valuable. […] Similarly, the second ‘sake’ indicates that the favouring attitude is not instrumental. Our attitude is to b, and it is final in the sense that it does not depend on some other favouring. (p. 62)
Scant clarity is achieved in this way, however, since both final and personal value are then negatively defined. Something has final value when it is to be favoured non-instrumentally, and something has personal value when the person for whose sake it is to be favoured is viewed non-instrumentally. Something thus has final personal value when it is warranted to favour both it and the person for whose sake it is favoured non-instrumentally.
One might also complain of the apparent inconsistency between Rønnow-Rasmussen’s claim that the fitting-attitude analysis “neither mentions nor refers to the properties in virtue of which the object is valuable” (p. 25) and his subsequent admissions that “we are to favour [something] on account of its properties” (p. 37) and that we love people “in virtue of their properties” (p. 65). Or the inconsistency between his claim that, according to the fitting-attitude analysis, “it is properties other than the value property that provide reasons to [favour]” (p. 24) and his later contention that something’s being good can sometimes itself provide a reason to favour it (pp. 43-45). To do so, however, would obscure the worthwhile aspects of Rønnow-Rasmussen’s effort.
Alfano, M. (2009). A danger of definition: Polar predicates in moral theory. Journal of Ethics and Social Practice, 3:3, 1-13.
Hurka, T. (1987). ‘Good’ and ‘good for’. Mind, 96, 71-73.
Moore, G. E. (1903). Principia Ethica. Cambridge University Press.
Parfit, D. (2001). Rationality and reasons. In Egonsson, Josefsson, Petterson, & Rønnow-Rasmussen (eds.), Exploring Practical Philosophy: From Action to Values. Aldershot: Ashgate Publishing. Pp. 17-41.
Regan, D. (2004). Why am I my brother’s keeper? In Wallace, Pettit, Scheffler, & Smith (eds.), Reason and Value: Themes from the Moral Philosophy of Joseph Raz. Oxford: Clarendon. Pp. 202-230.
Rosati, C. (2008). Objectivism and relational good. Social Philosophy and Politics, 25:1, 314-349.
Thomson, J. J. (1997). The right and the good. The Journal of Philosophy, 94:6, 273-298.