Analytic Formulations. Philosophical analysis, es-
pecially in its British twentieth-century forms, has been
applied in various ways to problems of ethical theory.
G. E. Moore's Principia Ethica (1903), with its
common-sense analysis, reached a position in many
respects analogous to the phenomenological one. The
autonomy of the moral is central to his account. The
basic predicate of morals is “good” in the sense of
intrinsic good. This names a simple quality which
cannot be identified with any descriptive predicate,
whether psychological, such as pleasure or what one
desires; or metaphysical, such as what God wills; or
historical, such as what evolution unfolds. To identify
good with any of these “natural” qualities or predicates
is to commit the naturalistic fallacy. Moore's chief
demonstration of its fallaciousness is the so-called
open-question argument—that if you identify good
with such a descriptive content it is always possible
meaningfully to ask of this content whether it itself
is good. Thus to ask if what God wills is good, or if
pleasure is good, is not to ask a meaningless question
or to affirm that pleasure is pleasure. Though Moore
regards this as establishing the simple nonnatural char-
acter of good, and a domain of values intuitively
grasped as having a worth independent of whatever
the actual state of existence may be, it is more plausible
to see his argument as establishing the permanent
possibility of critical evaluation for any proposed con-
tent.
With respect to right and obligation, Moore's answer
is utilitarian in form. To judge an act as right is to
say that it will cause the world to be better than it
would be on any possible alternative act.
Contrasting relations of the right and the good were
proposed by Moore's contemporaries at Oxford, who
also employed the method of the conceptual analysis
of ordinary moral beliefs or convictions. H. A. Prichard
reversed Moore's relation. The particular judgments of
obligation are the primary material; we know directly
in the particular case what our obligations are, and
we generalize them in rough rules. The notion of good
is derivative: a good man is a man who does what is
right, and the good consists in those goals that a good
man pursues. On the other hand, Sir David Ross took
the strikingly different path of analyzing the right and
the good as coordinate independent ideas. Our duty
in a particular case is hard to work out, but the prima
facie rules which tell us that lying and stealing and
so on are wrong are themselves intuitively evident.
That something is good, or even yields the greatest
good, does not mean that it is our duty to do it; we
may be bound by a stringent duty such as a promise
to a man on his death bed to carry out his wishes, but
acting in accordance with his wishes may not yield
the greatest good we could disinterestedly conceive.
Ross's common-sense analysis reflects quite accurately
the conflicts between duty and interest in ordinary life.
It simply acknowledges the tension of the right and
the good, or of justice and utility, or of duty and
interest, by whatever names distinguished, and takes
for granted that the good is what has to be sacrificed
in cases of tension.
While these contrasting patterns each claimed to be
the correct analysis of ordinary moral concepts and
convictions, it is apparent that they also establish defi-
nite priorities in policy and conduct. To define the right
in terms of the good involves a readiness to evaluate
moral rules critically in terms of the welfare they bring
or frustrate in practice. The separation of the right
and the good or the primacy of the right has the more
conservative potential in giving priority to maintaining
the stability of the existing moral pattern.
Analytic formulations moved in two somewhat
different directions in the mid-twentieth century. One
was toward more formal logical analysis, the other to
more informal contextual linguistic analysis.
The formal analysis was prompted by the rapid
development of logical techniques as well as the pres-
tige which logical positivism attached to formal con-
struction, while disparaging ordinary language as
enshrining the mistakes and myths of the past. The
most prominent work relevant to ethics has stemmed
from the field of deontic logic, in which such concepts
as permissible, imperative, ought, and others of the
same group of right and obligation are analyzed and
systematized in logical fashion. Thus if “permissible”
is taken as a primitive term, “X is obligatory” would
be translated into “It is not permissible not to do X.”
Differences between the operations and trans-
formations permitted in the ordinary propositional
calculus and those in the deontic system are carefully
explored. This is a rapidly growing field of analysis
today. While the right was first dealt with, recent work
has turned also to the good, and axiological systems
have been developed using “better” as a primitive
term. The question of interpretation of such systems,
and of the ways of establishing or verifying statements
in these systems, would raise afresh all the problems
of the right and the good. At present it is the logical
commplexities that stimulate interest.
The positivist analysis of meaning and verification
also had a different impact on ethical theory. Since
the meaning of a term was taken to lie in the mode
of verification implied, and truth was established
either by showing a logical proposition to be analyti-
cally true or an empirical proposition to predict cor-
rectly the course of sense-experience, there seemed no
place left for the ethical propositions of the intuitionist
approach, any more than for propositions of religion
or aesthetics. All these were accordingly denied any
cognitive status; they do not assert anything, but ethical
terms rather serve a noncognitive or practical function,
providing vehicles for expressing or giving vent to
emotions. To say “Stealing is wrong,” argued A. J. Ayer
in his
Language, Truth and Logic (1936), is equivalent
to saying “Stealing!!” in a tone of horror. The differ-
ence between “good” and “right” or between any
ethical terms, lies in the kind and strength of the
emotion conveyed. Ethical statements are therefore
neither true nor false. In a development of the emotive
theory, C. L. Stevenson focused on disagreement in
attitude, as distinguished from disagreement in belief,
as the central moral phenomenon, and analyzed ethical
statements as largely persuasive in effort—practical
attempts to bring about agreement in an emotive way.
To resolve an ethical issue is thus causally to secure
agreement in attitude, not cognitively to establish a
truth.
The distinctive feature of the emotive theory was
not the recognition of the role of emotion in ethics;
this had been a commonplace of the eighteenth-century
theorists who stressed the moral sentiments as against
the Cambridge Platonists who had looked for intellec-
tual ethical axioms. And Westermarck, in his Ethical
Relativity (1932), had recently expounded the view
that ethical beliefs were generalizations of the retribu-
tive feelings, with “wrong” and “bad” resting on the
sterner retaliatory feelings, and “right” and “good” on
the kindlier retributive feelings of gratitude. The dis-
tinctive element was the tie-in with the presumed
correct use of language, and the claim that indicative
forms in “X is good” or “X is wrong” are incorrect
syntactical expressions whose proper form would be
“Would that everybody desired X” as Russell at one
point analyzed “good,” and “Don't do X,” as Carnap
translated moral rules into imperatives. Some, such as
Reichenbach, stressed the more voluntaristic element,
the commitment component in the will-act, in ethical
statements.
The contextual mode of analysis as a systematic
procedure in ethics emerged from such antecedents
under the impact of the revised conception of meaning
that followed upon Ludwig Wittgenstein's later work.
A term was to be understood not by seeking a single
definition expressive of its essence, but by examining
linguistic uses, by seeing carefully how one might come
to learn the use of the term. No one form of unity
was antecedently presupposed; there might be an ulti
mate plurality, or a loose unity in a kind of family
resemblance. In a reverse of the positivist attitude to
ordinary language as a blundering to be superseded
by careful formalization, the new mode of analysis
showed the greatest respect for ordinary language as
a repository of the wisdom and experience of the ages
in communication and interpersonal relations. Accord-
ingly, in ethics, it canvassed the field of the uses of
moral terms and turned up a multitude of differences,
as far apart as expressing feelings and preferences,
expressing decisions, advising, persuading, evaluating
and promulgating. R. M. Hare concentrated on the
commending use of “good,” and the use of “ought”
to indicate the need for a decision. J. L. Austin explored
the performatory uses of language, and in morals the
actual assignment of obligation and responsibilities,
J. O. Urmson the grading uses of “good.” By the time
that G. H. von Wright explored the variety of uses
in his Varieties of Goodness (1963), it was a mark of
lack of philosophical sophistication to ask for “the
meaning of 'good.'”
Contextualism probed even more minutely into con-
text differences. Thus the differentiation in personal
pronoun with “ought” was found to make a difference
in use; for example, “I ought” was sometimes declared
to express a decision, “You ought” to be prescriptive
as addressed to someone in particular, “He ought” to
be evaluative. Thus the kind of term became less im-
portant than the kind of function being performed. But
even evaluation differed (as Toulmin showed in his The
Place of Reason in Ethics, 1950) as one was looking
for the application of a rule in a particular case, ques-
tioning a rule within a moral code, and questioning
a limiting principle in terms of which codes were
themselves adjudged.
The relation of the right and the good underwent
changes in these developments. At first the distinction
was between dentological terms and teleological terms,
and the question of their relation was expressed as
whether deontological statements presupposed teleo-
logical statements—that is, whether ought-assertions
were meaningful only if you assumed certain purposes
in the background. In the language of functions, re-
placing that of statements, the question was the rela-
tion of prescribing to evaluating. With the multi-
plication of contexts and functions it became less a
question of assigning a usage to one or another function
than of exploring the concrete structure of each func-
tion, whatever language it employed. In effect, all the
functions could be seen as contextually differentiated
modes of reflective criticism.
