436
PARTISAN REVIEW
natural quality as its meaning because there was no room for such
mysteries in their philosophy; they held (against Bentham
and
Moore)
that ethical sentences are "descriptively meaningless." They were fol–
lowed by moderates who held that ethical sentences might have des–
criptive meaning but that
this
was not the key to ethical discourse. The
newer group called attention to "emotive meaning"
(pace
I.
A.
Richards), a kind of Christmas cheer generously sprinkled on sentences
like "I approve of this" and "This conduces to the greatest happiness
of the greatest number" in order to produce a shiny ethical judgment.
It
was said that one can support the descriptive component of ethical
sentences by citing empirical evidence ("Look, Ma, I'm approving"),
but that one could hardly prove a cheer or glow. At best one can cause
others to cheer and glow in the same way, but this causal process is not
an argument in the traditional sense. It is not a matter of supplying
premises from which conclusions follow; it is rather a matter of stimu–
lating people to respond agreeably. Ethical persuasion on this view
must always involve a non-rational link which forms neither an induc–
tive nor a deductive argument,
and hence n·ot an argument at all.
The most recent phase of this effort to preserve ethics against
excessive scientism is that of certain younger Oxford moralists who
hold that reasoning does take place in ethics, that it is not just a matter
of stimulating others to adopt your own attitudes and to cheer on your
side, but a real, honest-to-goodness rational argument that just happens
to be neither inductive nor deductive in nature. Once again naturalism
is attacked, because this newer group holds, along with Moore and all
varieties of logical positivists, that ethical sentences cannot be translated
without loss of meaning into descriptive sentences and hence supported
in a typically scientific way. Instead we have non-translational bridges
of inference that conduct us from factual premises to ethical con–
clusions, bridges that are neither deductive nor inductive, neither logical
nor empirical.
What we have here is a position strongly reminiscent of Pascal's
reasoning by heart. Inductive arguments have
their
rules, deductive
arguments have
theirs,
and ethics has a logic logicians never know. Now
if Baumgardt's version of Bentham is correct, Bentham would appear
to hold a view quite similar to the modish one just described. The point
is that the Benthamite inference from "This conduces to the greatest
happiness of the greatest number" to "This is right" cannot be a de–
ductive inference
if
Bentham denies, as Baumgardt says he does, that
"right" is
synonymous
with "conduces to the greatest happiness of the
greatest number." But neither can it be like the inductive inference
