In
this book, Stephen Darwall endeavors to set out a new theory of welfare. The
main tenets of the book can be sketched as follows.

First,
Darwall presents what he calls a rational care theory of welfare
(hereafter RCTW): "what it is for something to be good for someone just
is for it to be something one should desire for him for his sake, that is,
insofar as one cares for him" (p. 8). This theory has, among other things,
a noteworthy consequence, namely: "the normativity [of welfare] is not (…)
the agent-relative kind of rational preference. It is rather an agent-neutral normativity
grasped from the perspective of someone who cares for the person" (p. 45).
RCTW is presented and upheld (notably against some rivals) in the first two
chapters (pp. 1-49).

However,
the notion of care used in the definition of RCTW is a conceptual primitive,
and cannot be defined in terms of welfare (on pain of circularity). Therefore,
how could we refer to care without presupposing any reference to welfare? Darwall
needs to show how this is possible: this is the task of the third chapter (pp.
50-72). Relying both on philosophers like Hume and Adam Smith and on recent
psychological works on empathy and sympathy, he attempts to explain how care
can be considered as a "natural kind", and thus how we may understand
and identify it (this strategy has unwelcome consequences, see below).

Finally,
Darwall sets out in the last chapter (pp. 73-104) a partial normative theory of
welfare, called the "Aristotelian thesis". According to this thesis,
what is good for human beings is engaging into activities in which "we come
into appreciative rapport with agent-neutral values" (p. 75), that is
activities where one appreciates the worth or merit of things with worth or
merit (this chapter, though very interesting in its own right, is perhaps less innovative
that RCWT, and I leave it aside in this review).

 

The
book is slim, but extremely condensed. It is highly stimulating: Darwall has
some very penetrating things to say about welfare, and his views deserve
without doubt serious consideration. However, the book is also at times
disappointing: some arguments deserve arguably more thorough developments, and Darwall’s
prose is sometimes heavy. Moreover, the book supposes a quite good deal of
knowledge in moral philosophy: for example, I fear that non-specialists will be
at pains to follow Darwall in his critique of Sidgwick and his references to
contemporary theories (notably pp. 25-43), such as the self-interest theory and
various versions of the "present-aim theory" (the deliberative versus
the critical one). Admittedly, these theories are briefly defined, but Darwall’s
discussion is so compact that I don’t see how readers not already familiar with
the problems at stake could gain much from it.

 

RCTW
concerns the metaethics of welfare, not normative ethics. In other words: RCTW
doesn’t answer to "what things are good for humans, or (for example)
breeding animals?" (there’s no need to limit the notion of welfare to
humans, and it is a strenght of Darwall’s approach to have shortcomings in
environmental ethics (p. 21)). What it provides is an analysis of the concept
of welfare (or of "well-being", or of "good for someone"):
it aims at clarifying the meaning of sentences of the form "x is
good for S".

Darwall
puts RCTW in various different ways (pp. 4, 7, 8, 9, 12, 31, 45, 46, 53, 71,
83).  Whatever version we choose, it is clear that the idea underlying his
theory is to stress the connection of the concept of welfare to caring and
concern (p. 31). Integrating the notion of care into our understanding
of welfare is an improvement over the standard view of welfare as pure
interest: a person’s good is not the sum of what she takes an interest in.
Indeed, "we should distinguish between how much a person values or takes
an interest in something (or would rationally do so), on the one hand, and its
benefit to him or contribution to his good, welfare, or interest,
on the other" (pp. 3-4). We all know people may value or desire things
that are not good for them: valuing or pursuing such things leads to a
decrease, not an increase, of their welfare. Moreover, we may have rational
desires which make no contribution to our welfare, for at least two kinds of
reason. First, because their fulfillment is too far to exert some influence
upon us: for example, we may be interested in the survival of the planet or the
happiness of our children long after our death, but such things make no
contribution to our welfare (p. 53). Second, because some rational desires may
harm us: self-sacrifice is an example (if the sum of our rational desires were
the same thing as our welfare, then self-sacrifice would not be sacrifice
at all). By defining the good for S as what is desirable for S from
the perspective of someone (included S) who cares for S, Darwall
hits the mark: his theory explains the gap between what is for a person’s good
and what he prefers or rationally prefers.

Moreover,
RCWT provides an interesting account of the normativity of welfare. If someone
cares for S (or for himself) and desires his good, then his reasons for
promoting S‘s welfare are not relative to S (namely, they do not
depend on S‘s own preferences and values), but they are agent-neutral:
what benefits the cared is not only good for him; it is, or should be, a
good thing absolutely: "a person’s good is intrinsically normative,
not for the agent herself, but for anyone who cares for her, herself
included" (p. 20). For instance, when a father wants his son to listen to
a musical masterpiece, he wants his son to get an appreciative rapport to
aesthetic values which are, in principle, available to anyone, and are not only
valuable for some individual.

 

            But these
successes have a cost. The problem is not that Darwall’s definition is, in a
way, awkward — the expression "insofar as" is strange, because if it
is understood as "to the extent to", RCWT makes no sense (the good
understanding is "provided"). RCWT’s main weakness comes from the
notion of "care". This has not escaped Darwall’s awareness, and he
treats this problem with some length.

"Care"
cannot be defined in terms of "welfare" (that would be circular). But
according to Darwall, "it is enough if care or concern exists as a natural
psychological kind for us to refer to" (p. 12). Darwall takes the model of
"water": we can use a term like "water" without a prior
definition to refer to the natural stuff in the rivers and lakes (and our
definition, or rather understanding, of water, may differ from the
understanding of the chemist who studies water — and it is precisely this last
understanding which fixes the extension of "water", at least according
to the causal theory of reference Darwall seems to follow here). Darwall
believes we can do with "care" the same thing as with
"water", but this strategy is, I fear, hopeless.

To
put things briefly, Darwall faces the following problem. "Caring for
someone" means, at any rate from an "unscientific" point
of view, "having concern for her, desiring her good". If Darwall
wants RCTW to avoid circularity, he must show that there exists a natural
psychological kind, called "care", that has roughly the same
extension as what we call "care", but that can be defined
without having the least recourse to notions like "welfare" or
"good". I don’t think referring to notions like empathy and sympathy
(chapter 3) achieves this aim.

Anyway,
Darwall has a quite restricted choice. The connection between care and welfare
is either necessary or contingent. In the first case, RCWT is circular, unless
we explain "good" in terms of "care", whereas one may think
that "care" should be explained in terms of "good". Truly, Darwall
has a quite smart answer to this objection (pp. 9-11), even though I am unsure
about its success (in any case, this drives us backwards, to the possibility of
a nominal definition of "care" which escapes all reference, implicit
or explicit, to the good and welfare of the person cared for). In the second
case, Darwall must acknowledge that there may be (i. e. it is
conceivable) instances where someone cares for someone else without having any
desire about her good.

 

All
this doesn’t mean that Darwall’s project fails. It seems to me that, as a theory
purporting to define welfare, the analysis upholding RCTW doesn’t
completely succeed. But Darwall has rightly brought to light the conceptual
connections between care and welfare (even if they do not amount to an
explanation of one by the other): the way he approaches the topic of welfare is
often illuminative and his insights here should guide future studies on the
subject.

 

© 2005 Guillaume
Dye

 

 

Guillaume Dye, Ph.D., (University of Paris I Pantheon-Sorbonne), Paris, France.

Categories: Philosophical, Ethics