by Bart Schultz
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
ultimate rational end for himself,’ and ‘ for him all-important’? He does not attempt to define them; and it is largely the use of such undefined phrases which causes
absurdities to be committed in philosophy.
G. E. Moore, Principia Ethica
The logical contradiction involved in Egoism has been powerfully argued by
von Hartmann in his criticism of Nietzsche and Max Stirner. . . . More recently Mr. Moore has incisively expressed the difficulty as follows: ‘What Egoism holds,
therefore, is that each man’s happiness is the sole good – that a number of different things are each of them the only good thing there is – an absolute contradiction!
No more complete and thorough refutation of any theory could be desired. Yet
Professor Sidgwick holds that Egoism is rational,’ a conclusion which he pro-
ceeds to characterize as ‘absurd’ ( Principia Ethica, , p. ). I should agree with him that the position is self-contradictory in a sense in which universalistic Hedonism is not, and that with all his subtlety Sidgwick failed altogether to
escape what was really an inconsistency in thought, even if he escaped an actual
or formal contradiction. But to point out this logical contradiction does not seem
to me quite so easy and final a way of refuting Sidgwick’s position as it does
to Mr. Moore for these reasons: () The Egoist with whom Professor Sidgwick
is arguing would probably not accept Mr. Moore’s (and my own) conception of
an absolute objective good, though I should admit and have contended . . . that if he fully thought out what is implied in his own contention that his conduct
is ‘reasonable’ he would be led to that conception. () Sidgwick only admitted
that the Egoist was reasonable from one point of view – reasonable as far as he
goes, i.e. when he refuses to ask whether his judgements are consistent with what
he cannot help recognizing as the rational judgements of other men, and limits
himself to asking whether he can make his own judgements consistent with them-
selves from his own point of view. No doubt Sidgwick ought to have gone on
to admit that this imperfectly reasonable point of view was not really reasonable
at all, and to some extent he has done this in his last Edition. And () after all,
even if we admit that the Egoist is unreasonable, there remains the question ‘Why
should he care to be reasonable?’ It was largely the difficulty of answering this
question on universalistic Hedonist principles which drove Professor Sidgwick
to admit a ‘dualism of the Practical Reason,’ and I am not sure that the question
has been very satisfactorily answered by Mr. Moore who, though he is no Hedo-
nist, appears to be unwilling to give the good will the highest place in his scale
of goods.
Hastings Rashdall, The Theory of Good and Evil
Although many may hope that Shaver is right in claiming that “neither
Hobbes nor Sidgwick provides good arguments for rational egoism” and
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Consensus versus Chaos
that “Sidgwick suggests good arguments against it,” his suggestion that
most philosophers after Sidgwick have been inclined to reject egoism,
and that this view has been only weakly articulated for much of the last
century, may well seem puzzling. True, Moore, Rashdall, Ross, and
Prichard were quite hostile to any such view, and many of the most promi-
nent movements of the so-called great expansion in substantive moral
theory – Rawlsianism chief among them – have shared various neo-Kantian
assumptions about practical reason. But there is a fairly impressive consen-
sus (including Shaver) that Moore and the rest responded to the dualism
with arguments that were as unsatisfactory as they were curt – thus, for
J. L. Mackie, “egoism can coherently resist any such proof by adher-
ing to the use of such two-place predicates as ‘right,’ ‘ought,’ and ‘good
for’: Objectivity and universalization with respect to these are powerless
against it.” Moore, according to Mackie, was guilty of the worst sort
of effrontery when he chastised Sidgwick for failing to define such ex-
pressions as “the ultimate rational end for himself,” given how Moore
himself had so insisted on the indefinable nature of “good.” And Moore’s
claim that “good” must be a one-place predicate, absolute, was nothing but
sheer assertion, or an unargued assumption that “an undefined one-place-
predicate ‘good’ can be straightforwardly meaningful, but a two-place-
predicate ‘good for’ or ‘all-important for’ cannot.” And as Skorupski has
put it,
What is clear to him [Sidgwick] is that the egoistic principle can be stated in
a rational and universal form. Of course an egoist who thinks his own good is
the only good thing, the only thing that everyone has reason to promote, can be convicted of attaching irrational significance to his good as against that of others.
Such an egoist thinks his own good the only thing that is ‘agent-neutrally’ good,
the one thing that provides everyone with reasons for action. (The term is not
used by Sidgwick; it comes from more recent moral theory.) But egoism need
not appeal to the idea of the agent-neutrally good. The egoist may instead hold simply that his own good is the only good relative to him – this is not a tautological doctrine. And he can put this in universal terms by saying that everyone ought to
pursue what is good relative to them, namely their own good.
It now becomes clear that hedonism is a doctrine about what a person’s good
is. To advance from it to utilitarianism we need at least to add that every person’s
good is agent-neutrally good. The rational egoist can block our considerations at this point, unless we can make it plausible that reasons as such are agent-neutral.
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
Even if one claims, as Skorupski does, that pure practical reason as such
does rest with agent-neutral reasons, the vanquishing or subordination of
agent-relative reasons, including those deployed by rational egoists, is not
generally regarded as simply the correction of an obvious mistake. Indeed,
many defenders of agent-neutral reasons – including both Skorupski and
Thomas Nagel – end up in partial retreat, allowing that agent-relative
reasons cannot be altogether discounted. Crisp goes even further down
the Sidgwickian path, defending a dual-source view of practical reason
that admittedly incorporates “a version of what Henry Sidgwick called
‘the dualism of practical re
ason.’ ” And this allows the rational egoist
considerable room for maneuver, as Samuel Scheffler, long associated with
a similar dualistic account, has also observed. Thus, Hurka, in a recent
defense of perfectionism, framed his argument by explaining that in “the
absence of a compelling argument that goodness must be understood in
the . . . Moorean way, I will assume with Sidgwick that claims about agent-
relative goodness are coherent, and ask how they may affect the recursive
account of self-interest and altruism.”
To be sure, to cast Sidgwick’s dualism in this way is to invite again the
question of why he narrowed the contenders down to utilitarianism and
egoism. That is, Moore’s claim was that the expressions “my own good”
and “good for me” are misguided because such talk can only mean that
“something which will be exclusively mine, as my own pleasure is mine”
is also “good absolutely,” but if “it is good absolutely that I should have it, then everyone else has as much reason for aiming at my having it as I have
myself.” If the Sidgwickian counter to this is simply that there is such a
thing as agent-relative goodness – such that, as Hurka puts it, “the question
is only whether there are different ultimate ends for different people” –
then egoism is only one variety of the challenge to agent-neutral reasons,
Broad’s “self-referential altruism” being one of many other options.
This is not to deny that, on Sidgwick’s rendering, as we have just seen,
egoism houses many different alternatives, from Aristotle to Hobbes to
Goethe. Still, for all its richness in representing “the personal point of
view,” egoism plainly does not encompass all agent-relative reasons, and
when Sidgwick broaches the matter of nonegoistic agent-relative reasons,
he is often less than perspicuous. Interestingly, however, he also on occa-
sion gestures toward further indeterminacies in practical reason, conflicts
between agent-neutral reasons, as when he invokes the Socratic complaint
about the injustice of a universe in which virtue goes unrewarded. And it
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Consensus versus Chaos
may be in part for this reason that some commentators have suggested
that what really bothered Sidgwick was not merely the specific challenge
of egoism, but the general indeterminacy of practical reason.
In this connection, it should be added that the contrast between
Sidgwick and Moore is not quite as simplistic as the just-quoted passages
make it sound. As Hurka spells it out,
[T]he concept of Sidgwick’s that Moore rejected was not well-being but agent-
relative goodness. And his main reason for rejecting it was his belief that goodness
is an unanalysable property. If goodness is this kind of property, it is hard to see
how an object can have it “from one point of view” but not “from another”; surely
it must either have the property or not. Compare squareness. An object cannot be
square from one point of view but not from another; it is either square or not. (The
object can look square from one point of view but not from another, but looking
square is not the same as being square.) So it must be with goodness if that is a
simply property. But Sidgwick held that goodness can be analyzed, in particular
as what a person ought to desire, and it is perfectly possible to say that what each
person ought to desire is different, say, just his own pleasure.
Plausibly, this is the conflict, though as remarked in the last section, it
is hardly obvious why Sidgwick’s “ought” did not share various features
of the Moorean “good” in its idealizing. Still, his greater metaethical
caution made a difference, allowing the cogency of agent-relative reasons
generally.
For all that, it is difficult to deny that Sidgwick was fairly obsessed with
the varieties of egoism, and this concern has struck many as apt, given
the power of egoism as a source for agent-relative reasons, the “personal
point of view.” Thus, it is not surprising that Kurt Baier, for example, in
his recent account of the subject, should suggest that rational egoism is
“the most deeply entrenched normative theory of egoism” and that “the
jury on this case is still in disarray.” Sidgwick, he allows, was engaged with
just this form of the theory, in a weak version admitting that even if it is
always rational to act out of self-interest, acting against one’s interest may
also be rational.
In fact, in his major work, The Rational and the Moral Order, Baier
addresses Sidgwick’s views at length, particularly the dualism of practical
reason:
How serious is Sidgwick’s problem? In my view . . . it is quite serious, especially if one starts, as many do, from Sidgwick’s unfortunate formulation of it. Thus, his
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
bifurcation of practical reason need not give rise to a contradiction, even on those
occasions when, as surely sometimes happens, the two principles do conflict. A
contradiction would arise only if both the reasons they supported were what I
called requiring rather than merely permissive, and even then only if they were
indefeasible. If even only one of them is merely permissive, if it is simply perfectly rational or reasonable, say, to act for one’s own good, but not necessarily always
irrational or unreasonable not to act for one’s own good, then one could act for the
general good and contrary to one’s own without its necessarily being the case that
when the two principles offer reasons for incompatible actions, one both ought
to act for the universal good and not for one’s own, and that one ought to act for one’s own and not for the universal good.
However, as Sidgwick seems to have sensed, there is still a problem even if this
is granted. For if my argument . . . is sound, then Sidgwick’s position would allow that it may always be in accordance with reason to promote the universal good and
always in accordance with reason to promote one’s own good and that, when one
cannot do both, it is in accordance with reason to do either. Nevertheless, Sidgwick also appears to have thought, and it would seem to agree with common sense, that
moral reasons, which he took to be those based on the universal good, defeat,
if not all other kinds, surely at least prudential ones. . . . Sidgwick seems to have grasped this much, even if perhaps only obscurely. For in various places in which
he produces arguments designed to persuade the Egoist to see the rationality of
Universal Hedonism, their thrust is always to show not merely that the Universal
Hedonist is also or equally rational, but that the Egoist ought to give up his position and become a Universal Hedonist. The thrust of his argument “from t
he
Point of View of the Universe,” for example, appears to be that everyone should
look at things from that point of view, and that anyone who does must adopt the
principles of Universal Hedonism as defeating that of Egoistic Hedonism when
the two conflict. Thus Sidgwick seems to have sensed the need for a demonstration
that moral reasons have a greater defeating force than prudential ones, hence his
argument from the point of view of the universe. In any case, whether or not he
sensed it, he is surely wrong in his claim . . . that a completely rational morality requires a demonstration of a “harmony” (i.e., coextensionality) between the
maxims of prudence . . . and of benevolence.
These remarks both situate Sidgwick’s dualism as a live issue and
indicate some different ways of tackling it. What is perhaps especially
instructive is that, like Moore and Rashdall, Skorupski and Shaver, Baier
is basically drawn to Sidgwick’s universalizing challenge to the egoist: why
is your good so special? How, then, does one stop on the slippery slope
before reaching the point of view of the universe? Thus, for Baier, the
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Consensus versus Chaos
better alternative to any attempted proof of the harmony or moral order
of the universe is the demonstration that when morality and prudence
conflict, “the requirements of morality defeat those of prudence.”
But again, a more thoroughly Sidgwickian view is possible. Crisp argues
that “Sidgwick is sometimes described as a utilitarian. But it is more precise
to ascribe to him a version of the dual-source view, held on the basis of the
neutral argument for the existence of options.” Quoting the distinction
passage, he explains that (contra Shaver) it makes “a clear appeal to the
separateness of persons as grounding a counterbalance to the reason to
promote the good.” And for Crisp,
Sidgwick was rightly pessimistic about the reconciliation of Rational Egoism and
Utilitarianism. His version of the dual-source view contains at its heart an irrecon-