by Bart Schultz
cilable tension between two ultimate and comprehensive principles of rationality.
But this is not to say that any dual-source view must fail to provide any practical
guidance. To take two extreme cases: when I can promote a very great good at
a very small cost to myself, other things being equal my strongest reason overall
is to make the sacrifice; likewise, if I can add to the overall good only a little at very great cost to myself, other things being equal my strongest reason is not to
make the sacrifice. In other words, the strength of the reasons grounded in the
simple thought or the separateness of persons varies according to the good or bad
at stake. The dilemma of practical reason is not quite what Sidgwick took it to be.
It arises most starkly in those cases where I can produce a great increase in overall good at a great cost to myself. Here the simple thought and the separateness of
persons pull hard against one another. The problem here is essentially a Hegelian
(or Freudian) one. The intuitions about rationality and reasonableness we consult
in such cases will have been shaped by an upbringing in a culture itself imbued
with a particular understanding of the relative strengths of the reason to promote
the good and the reason to promote one’s own good.
Arguably, Sidgwick would have appreciated the reference to Hegel and
felt himself equal to dealing with it, providing his own account of moral
maturation. At any rate, on Crisp’s intricately developed line, the question
of “whether morality permits one to pursue one’s own good at the expense
of the overall good” invites the following response:
[T]his gets things the wrong way round. If we are asked what morality consists
in, we can identify it if we wish with the reason to promote the overall good. But
there is no need for any notion of morality, prior to the reason to promote the
good and the competing reason based on the separateness of persons, that will
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
rule on whether one is permitted or required to act on certain reasons on certain
occasions.”
In addressing Shelley Kagan’s penetrating attempt to defeat all such
reasons arising from the personal point of view, Crisp develops some
strikingly Sidgwickian themes:
But what about the practical implications of the dual-source view? Am I seriously
arguing that we should see killing and letting die as on a par, and be prepared
to kill in pursuit of our own good? Am I suggesting that this is how we should
bring up our children? These consequences may be so counterintuitive that the arguments for the dual-source view will have to be rejected.
The dual-source view does not have these practical implications. First, since
we have been brought up to accept common sense morality, and since we live in a
culture based on common sense morality, killing is likely to be far more psycho-
logically and socially costly than letting die. Secondly, these facts militate against educating any individual child to use the dual-source view in practice. Further,
it would probably be a mistake for all of us to begin educating children to be practical dual-source theorists. Human beings are not creatures of pure reason.
We have an evolutionary background and an emotional make-up which cannot
be ignored in moral theory. In particular, we show a particular concern for those
visibly near us, and for what we do to them. It may well be that these concerns, though they might not withstand close intellectual scrutiny, are somehow central
to our becoming and continuing to be rational agents. The risk that this is so
would be sufficient to justify not radically changing the moral education of our
children. What is needed is common sense morality with a far greater emphasis
on the importance of distributive justice and personal generosity.
As Crisp notes, these arguments “parallel those for a ‘split-level’ version
of utilitarianism.” Clearly, they resonate powerfully with Sidgwick’s views
about the potential limitations of even a more highly evolved utilitarian so-
ciety, absent any cosmic ordering, and the nature of moral maturation and
cautious utilitarian reform. Crisp is, in effect, picking up the Sidgwickian
project without the parapsychological and theistic or Buddhist options.
Doubtless there is much to be said for all these interpretations and re-
constructions of Sidgwick’s dualism, which collectively ought to convey
something of the continuing relevance of the issues raised by the Methods.
Beyond a certain point, however, it is just very difficult to say, for exam-
ple, whether Sidgwick shifted his views on the question of “permissive”
reasons, given that he did not use such terminology.
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Consensus versus Chaos
But at least on the question of whether Sidgwick’s axiomatic grounding
of egoism is really as elliptical as Shaver claims, it must be owned that a
great many commentators have followed Sidgwick in moving too quickly to
identify “prudence” with egoism, when giving a summary exposition of the
dualism. Even Crisp, for example, flatly states that Sidgwick “took it as self-
evident that one ought to aim at one’s own good on the whole, accepting
that this good was merely a particular part of the general good.” And
Marcus Singer, in a recent work, simply states that the “fundamental
principles of Philosophical Intuitionism are the intuitively self-evident
axioms of prudence or egoism, justice, and rational benevolence.”
However, in a footnote, Singer does call attention to the interpretation
anticipating Shaver’s advanced by Sidgwick’s student W. R. Sorley, in his
A History of English Philosophy:
It would appear . . . that this dualism was not adequately tested by [Sidgwick] and that it really arises from the ambiguity of the term prudence. Prudence may mean
either “regard for one’s own good on the whole” or (what is not the same thing)
the principle that “hereafter as such is neither less nor more valuable than now.”
Both forms of statement are used by Sidgwick; but only the latter has a claim to
express an absolute ethical principle; and it is not inconsistent with the axiom of
benevolence.
This would certainly suggest some powerful support for treating the move
from axioms to egoism with much greater caution.
Furthermore, Schneewind’s account, which remains the most ex-
tensive, ends up reformulating Sidgwick’s axiom of prudence to read
“Maximizing the agent’s own good is an ultimate right-making charac-
teristic” and his axiom of benevolence to read “Maximizing the universal
good is an ultimate right-making characteristic.” His claim is that “these
formulations seem to express Sidgwick’s understandin
g of the two prin-
ciples involved in the dualism of the practical reason, and they reveal its
structure more plainly than his own statements do.” If the world does
not have the requisite moral order, then it is “logically impossible” for
both of these to be true, for “it cannot be true that it is actually right to do
an act maximizing own-good and not actually right to do it.”
Thus we have found the contradiction, removable by a factual proposition, which
lies at the heart of Sidgwick’s problem. The urgency of the difficulty it creates
can perhaps be brought out by recalling that Sidgwick has tried throughout the
Methods to discover what reason demands of action when applied under the most
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
fundamental conditions of human life. What he finds at the end is that because of
one such basic and undeniable fact about human life, practical reason inevitably
makes contradictory demands on action. If this is not a formal contradiction
within reason itself, its bearing on Sidgwick’s real hope for philosophical ethics
is sufficiently devastating to make it clear why he thinks his endeavour ends in
failure.
And Phillips would also seem to have a point insofar as he suggests
that Sidgwick was also exercised – justifiably or not – about the simple
indeterminacy of permissive reasons on both sides, because this still, to his
mind, amounted to a failure to provide an ultimate unification of practical
reason. Consider Sidgwick’s pointed and quite characteristic criticism of
John Grote’s position:
The non-critical part of Mr. Grote’s book I can scarcely call constructive. It is not even a sketch of a system; it is a collection of sketches. He considers that utilitarians are right in the general assertion (carefully explained to be meaningless) that all
action is aimed at happiness. But he would distinguish the study of the general
effects of Conduct on happiness, from the enquiry into the principles of Duty,
or right distribution of happiness, and from the investigation of the Virtues, or
generous dispositions, which must be left freely to follow their special altruistic
aims, and not made to depend on a utilitarian first principle. What the last two
methods are to be, and how the three enquiries are to be harmonized, Mr. Grote
does not clearly explain. In his desire to comprehend the diversity of human
impulses, he has unfortunately neglected the one impulse (as human as any)
which it is the special function of the philosopher to direct and satisfy: the effort after a complete and reasoned synthesis of practical principles.
Here again one feels the force of the (frustrated) ambition that was be-
hind the Methods, and the refusal to “compress the world into a system.”
What Irwin complained of as an ungrounded demand for clarity at key
points in the Methods was also in large measure a demand for determinate-
ness, for the type of clear guidance that Sidgwick lamented losing along
with his faith.
But in any event, the illuminating point here is that even Schneewind
requires a reconstruction of Sidgwick’s axioms in order to make sense
of the dualism. This, too, supports Shaver’s general account, though
on Schneewind’s reading, Sidgwick’s pursuit of harmonization is clearly
much more of a necessity. Shaver, in fact, admittedly takes his point of de-
parture from Schneewind’s analysis, agreeing with him that it explains – in
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Consensus versus Chaos
a way that Broad’s account does not – much of what Sidgwick says about
how the axioms are merely “consistent with and needed to prove utili-
tarianism, and that the issue between rational egoist and utilitarian turns
on the rationality of taking up the universe’s point of view.” Certainly, as
we have seen, Sidgwick was extremely insistent that the axioms were too
abstract to provide much practical guidance about anything – everything
needed still to be filled in.
It would be rash, then, to deny what Shaver, Schneewind, and appar-
ently Sidgwick himself – in his more considered statements – all expressly
claim in this connection: namely, that the axiomatic basis of egoism is
insufficient in itself to render egoism truly rather than apparently self-
evident, much less of the highest certainty. And yet it is also possible to
think that Shaver has gone too far in discounting the force of the argu-
ment for rational egoism in Sidgwick’s work, and in painting a picture of
what Sidgwick “was really getting at” that comports too easily with the
impartialist attempt to defeat egoism. Schneewind’s work supports just
such a critique.
To be sure, Shaver’s view has great advantages. It makes admirable
sense of Sidgwick’s tendency to describe his own views as “utilitarian,”
without much qualification. And it suggests how, taking moral theory as
a going project, Sidgwick could have continued to develop his account of
the self-evident grounding of utilitarianism, getting beyond the treatment
of it as only “apparently” self-evident, without then running into the
problem of a similar development of the egoistic principle producing a
conflict – an impossible conflict – of genuinely self-evident propositions
of the highest certainty. After all, Sidgwick manifestly aspired to greater
certainty in this department, even if he did not find it. Furthermore, it
helps to explain how so many of those inspired by the Methods and/or
by Sidgwick himself – from Rashdall, Moore, and Russell down to Baier,
Kagan, Singer, and Shaver – could take this as the obvious direction for
the progress that Sidgwick sought but failed to find.
On the other side, however, Sidgwick clearly did lean toward a nonre-
ductionist view of personal identity that undercut a number of potential
challenges to rational egoism, as the following chapter will show. His ar-
ticulation of moral theory was, for better or worse, steeped in the religious
orientation of his youth, which had in effect involved a form of reconcili-
ation. And he did tend, as Crisp suggests, to wield the distinction passage
as an independent argument for at least the personal point of view, and
P: IJD,JRQ ,FhN,GLS
cB.xml
CY/Schultz
February ,
:
Henry Sidgwick: Eye of the Universe
not simply as an objection to utilitarianism. Consider his response, in the
very paper presenting the distinction passage – the relatively late “Some
Fundamental Ethical Controversies” – to Rashdall’s
objection that he fails
to reconcile duty and self-interest because he assigns a “different end to
the individual and to the race.” On Sidgwick’s rendering, Rashdall, try-
ing to avoid the paradox of altruism apparent in Green’s account of the
common good, held that if
“it is pronounced right and reasonable for A to make sacrifices of his own happiness
to the good of B,” as this must be equally right and reasonable for B, C and D, “the
admission that altruism is rational” compels us to conceive “the happiness which
we ought to seek for society,” not as mere happiness but as “moral happiness.” The
ultimate end, for the race as well as for the individual, thus becomes composite:
it consists of a higher good, Virtue, along with a lower good, Happiness, the two
being so related that in case of conflict the higher is always to be preferred to the lower.
Sidgwick grants “to the full” Rashdall’s starting point, the basic charge
that he “assigns a different end to the individual and to the race.” But
he is “unable to see why it constitutes a difficulty, since the individual
is essentially and fundamentally different from the larger whole – the
universe of sentient beings – of which he is conscious of being a part: just
because he is conscious of his relation to similar parts of the same whole,
while the whole itself has no such relation.” Thus,
[W]hile it would be reasonable for the aggregate of sentient beings, if it could act collectively, to aim at its own happiness only as ultimate end – and would be reasonable for an individual to do the same if he were the only sentient being in
the universe – it is yet actually reasonable for an individual to make an ultimate sacrifice of his happiness for the sake of the greater happiness of others, as well
as reasonable for him to take his own happiness as ultimate end; owing . . . to the double view which he necessarily takes of himself as at once an individual
essentially separate from other individuals, and at the same time essentially a part
among similar parts of a larger whole.”
However odd it may seem, Sidgwick does here imply that the dualism
would also be overcome by the destruction of all sentient beings save one,
though this is obviously not the type of harmonization he favors. But at
any rate, Sidgwick’s use of the argument against Rashdall shows that he,
