Grice, P. (2001) - The Conception of Value.

THE CONCEPTION OF VALUE
The Conception of Value
PAUL GRICE
With an Introduction by
JUDITH BAKER
CLARENDON PRESS · OXFORD

1991
Oxford University Press, Wa/ton Street, Oxford ox2 6oP
Oxford New York Toronto
Delhi Bombay Calcutta Madras Karachi
Petaling ]aya Singapore Hong Kong Tokyo
Nairobi Dar es Salaam Cape Town
Melbourne Auckland
and associated companies in
Berlin I badan
Oxford is a trade mark of Oxford University Press
Published in the United States

by Oxford University Press, New York
© Kathleen Grice 1991
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without
the prior permission of Oxford University Press
This book is sold subject to the condition that it shall not, by way
of trade or otherwise, be lent, re-sold, hired out or otherwise circulated
without the publisher's prior consent in any form of binding or cover
other than that in which it is published and without a similar condition
including this condition being imposed on the subsequent purchaser
British Library Cataloguing in Publication Data

data available
Library of Congress Cataloging in Publication Data

Grice, H. P. (H. Paul)
The conception of value/Paul Grice: with an introduction by
Judith Baker.
Includes bibliographical references and index.
1. Values. I. Title.
B1641.G481 1991 121'.8-dc20 90-26891
ISBN 0-19-824495-9
Typeset by Cambrian Typesetters

Frimley, Camberley, Surrey
Printed in Great Britain by
Biddies Ltd.
Guildford & King's Lynn
Foreword
Paul Grice wrote and delivered his Carus lectures on the

conception of value in 1 9 8 3 . He returned to the topic of
those lectures, and in particular to the construction of
value, in the course of responding to essays written in his
honour. Shortly before he died, he asked me to publish the
lectures, but with the argument for value sketched in his
' Reply' replacing the one presented in the third and final
lecture. He regarded his later attempt as a clarification,
better representing his thought. Unfortunately no straight
forward substitution is possible. It seemed better to allow
the reader direct access to the work of Paul Grice, and so
his Carus lectures are published below in their entirety,
followed by the relevant section of the 'Reply'. Since he
and I had discussed the problems over the preceding
decade, my Introduction describes one way to understand
the most difficult of Grice's new ideas about value, and the
relationship between the third lecture and the ' Reply'.
The lectures themselves were edited by comparing the
manuscript (to be deposited in the Paul Grice Archives,
University of California at Berkeley) and a rough type
script version he had made in 1 9 8 3 . Aside from minor
typographical emendations, the two previously published
essays-the Reply and an address to the American
Philosophical Association-are unaltered.
In the footnotes, editorial notes and insertions are
indicated by enclosure in square brackets.
J. B.
Con ten ts
Introduction by ]udith Baker 1
The Carus Lectures on the Conception of Value
1 . Value and Objectivity 23

2 . Relative and Absolute Value 47
3 . Metaphysics and Value 69
Reply to Richards, Final Section:
Metaphysics, Philosophical Psychology, and Value 93
Method in Philosophical Psychology (From the

Banal to the Bizarre) 121
Index 1 63
Introduction
Judith Baker
H. P. GR I C E ' S CO N S TR U CT I O N O F V A L U E
The three works o f Paul Grice collected in this volume

were selected in order to present his metaphysical defence
of value. The Cams lectures and the selection from the
'Reply to Richards' take value as their topic. The APA
address, 'Method in Philosophical Psychology (from the
Banal to the Bizarre) ', an essay in philosophical psychology,
is included for its exposition of Humean Projection, a
metaphysical 'routine' Grice made the basis for the
construction of value. This introduction will attempt to
clarify some of the basic elements of Grice's constructivist
theory and the relationship between the third Cams lecture
and the 'Reply'.
In 1 98 3 Paul Grice gave the Cams lectures, the Concep
tion of Value. The lectures, particularly the third and last,
on the metaphysical construction of value, were difficult to
understand. He returned to the topic in his reply to essays
written in his honour. Although the later account was
meant to clarify and improve the earlier argument, the
lectures remain difficult. The one on the construction of
value is an essay in metaphysics and not on or within ethics
or theory of value. It is indeed not clear how those theories
get attached to the work. One difficulty is that both the
Cams lecture and the 'Reply' call on a number of
philosophical themes which interested Grice for a long
time, but which were not among the more widely
publicized aspects of his work. What is referred to in the
2 Introduction
Cams lectures as Humean Projection explained the growth
of conceptualization and representation in progressively
more complex creatures. It was part of an enterprise in
creature-construction, first introduced ill the Urbana
lectures of 1970-1 on 'Language and Reality' (un
published) . It was fully developed and played a large part
in the APA address. Discussion of this address is limited
here to the topic of Humean Projection, but it is an
enormously rich essay. It attempted both to understand
psychological concepts as theoretical concepts explaining
behaviour and to account for the privileged position of
some of those concepts. The third Cams lecture asserts,
and its argument relies on, an interdependence of the
notions of finality and value. Like the discussion of
freedom, the discussion of finality occupied Grice for
many years, but it was perhaps not fully accomplished.
The importance of the philosophical attempt invites us,
however, to work to understand the lectures.
For they are a modern effort to provide a metaphysical
foundation for value. Value j udgements are viewed as
objective, value is part of the world we live in, none the less
value is constructed by us. Grice's story puts together a
number of seemingly divergent elements and attempts to
make them work together. We inherit, or seem to inherit,
the Aristotelian world in which objects and creatures are
characterized, perceived, understood in terms of finality
features, namely, what they are supposed to do. We are
thereby enabled to evaluate by reference to, or relative to,
function and finality. That is not surprising. However,
Grice maintains that the legitimacy of such evaluations
rests ultimately on an argument for absolute value. Grice
thinks that vitalistic phenomena can be explained by
mechanistic, cybernetic notions, but he insists that
explanations in terms of finality and attributions of
relative value are none the less both legitimate and
irreducible. This claim depends on the thesis that the
reducibility and substitutability of finality features may be
Introduction 3
blocked in all areas if there is one area in which attribution

of finality has no substitute. The non-substitutable attribu
tion of finality accompanies absolute value and so must
rest on Grice's argument for absolute value. Value is itself
overtly constructed, but not from such naturalistic items as
desires or desires to desire. Our attitudes, which provide
the basis for value, are irreducible modes of "thinking
valuable" or "feeling-valuable". (But you must hear the
hyphens between the terms, which indicate that we do not
have at the start an object or the representation of an
object. )
This introduction will take as its focus what Grice calls
construction routines. They are essential to the argument
for the construction of objective value. The attempt to
understand them raises very general questions about
Grice's conception of transcendental arguments and objec
tivity. In what follows, the metaphysical operations are
interpreted in the light of the prototypical Humean
Projection, not as invented procedures but as natural
dispositions to take the world to be a certain way. So
interpreted, they can be seen to carry certain of the features
of the objectivity sought: consensus via commonality of
conception, and also non-arbitrariness, for the conceptual
ization does not depend upon individual will or whim-.
Routines answer a criticism of the alleged objectivity of
constructed value : they remove a threatened gap between
the justified acceptance of a construction thesis and
legitimate belief in its truth .
It may help to understand the role of the routines in
Grice's theory of value to have an outline, a sketch of the
formal structure of the third Carns lecture. I include a list
of the main elements, although one cannot expect to find
the steps in the argument immediately understandable
from such a display. It is offered as a reference, to which
one can return. I have made use of material from the
'Reply' where it helps to clarify the argument. In particular,
value of a kind is introduced in the ' Reply' as an important
4 Introduction
step in the defence of absolute value. Although the steps
are not easy to understand, the project can be simply
stated: Grice's goal is to show that some value judgements
are objective, and to do this by displaying their construc
tion. Two routines are involved in the construction : one,
Humean Projection, accounts for the projection of attitudes
onto the world, the other, provocatively called Meta
physical Transubstantiation, for a new kind of creature or
substance. They work together for the legitimacy of
projection, as constructing value or placing value in the
world depends upon the status of the projector.
According to Grice, a value judgement--one that is

objective-has the following construction history :
1 . Such a judgement starts off life as a human attitude, a
mode of "thinking-valuable".
2. Hume attributed to the mind a tendency 'to spread
itself upon objects', to project into the world items which
properly or primitively considered are only features of our
states of mind. 1 Grice delighted in the description, but he
wanted to reverse Hume's implication. What Hume
diagnosed as a source of illusion and confusion becomes,
under specified conditions, a legitimate construction pro
cedure. Grice's Humean Projection occurs by stages :
intermediate stages reflect a growing logical complexity,
the new concept is embedded within a more general
psychological attitude, then within another embeddable
1' 'Tis a common observation, that the mind has a great propensity to
spread itself on external objects, and to conjoin with them any internal
impressions, which they occasion, and which always make their appearance
at the same time that these objects discover themselves to the senses'
(Treatise, 1. iii. xliv).
Later, writing of reason and taste, Hume said that 'The one discovers
objects as they really stand in nature, without addition or diminution; the
other has a productive faculty, and gilding or staining all natural objects
with the colours, borrowed from internal sentiment, raises in a manner a
new creation' (Enquiry, Appendix 1). This text has also captured the
attention of Simon Blackburn, Spreading the Word (Oxford: Oxford
University Press, 1984), 145.
Introduction 5
operator. At the final stage, what characterized thinking
characterizes the object of thinking. "Thinking-valuable"
becomes judging something to be of value. The legitimacy
of this move depends on two conditions : that there is a
rational demand for the new concept of value (indeed, that
there is a rational demand for each new stage) and that at
the final stage there are 'truth conditions, or something
analogous' for the judgements in which value occurs.
3. The rational demand for absolute value is not fully
argued for. This legitimating condition is variously de
scribed as a theoretical motivation or purpose in the Cams
lectures, and a rational demand in the 'Reply'. In the latter
work, Grice refers to what is most likely a transcendental
or, his preferred term, metaphysical argument as the
supporting consideration.
4. Grice provides not truth conditions but something
analogous for statements of value. He answers to the
demand for absolute value by telling a 'trouble-free' or
good story of how we might construct it. The legitimacy of
substituting a convincing story for truth conditions is
derived from a principle Grice dubs the metaphysical
principle of supply and demand; he meant this title,
entertaining as it is, with high seriousness. It is a principle,
he said, 'which allows credence to a set of procedures the
adoption and use of which will most satisfactorily meet a
rational need'.
5 . The story begins :
(a) by characterizing living creatures, including
humans, essentially, in terms of finality features. What
the creatures are is understood in terms of what they do;
certain of their states and behaviour are such that the
creature aspires to them, sets out to do them (and, in the
case of humans, consciously so) .
(b) Finality or function provides us with a standard
for evaluating both behaviour and creatures. A theory of
conduct, or of behaviour, can be generated from the
6 Introduction
starting-point of function ; and a creature that does what
it is supposed to do is good qua that kind of creature.
Grice claims, explicitly in the 'Reply', implicity in the
third Cams lecture, that relative value judgements
depend upon finality; we can legitimately evaluate
creatures, given finality features and the derivable
theory of behaviour. 2
(c) A human being has as an accidental property a
complex attribute called rationality, defined as (at least
partly) the capacity and concern that its attitudes be
well-grounded, or validated. A thoroughgoing concern
to have reasons brings with it a concern to find some
attitudes, decisions, beliefs, absolutely validated (or of
absolute value) .
(d) This accidental property becomes the essential
property of, and thereby provides for, a new kind of
creature, persons. Human beings redistribute their
properties, 'reconstitute' themselves as persons, as
essentially rational, via a routine called Metaphysical
Transubstantiation.
(e) We can then characterize a person who is good at
her function or metier of finding reasons, as good qua
person.
(f) It is characteristic of humans and persons to
2 Relative value judgements depend upon finality, but finality must be a

genuine, 'non-Pickwickian' concept. Grice devotes half of the third Carus
lecture to the question of the substitutability of finality by cybernetic or
mechanistic notions. I cannot discuss this large question in this paper, but
the form of his argument is the following: vitalistic phenomena form a
domain in which both explanations in terms of finality and cybernetic
notions are acceptable. If, however, there is an area in which cybernetic or
mechanistic explanations cannot do the work of finality, we may regard
finality as non-substitutable in all areas. He proceeds to argue that the
characterization of person in terms of finality features provides for such an
area. And that this is because persons, as essentially rational beings, are free.
I conjecture that "free " does the following work: the behaviour of persons
who seek ultimate justifications goes beyond what is biologically useful; the
principles or procedures for answering their demand to find ultimate
reasons must be constructed by them; a rewrite in mechanistic terms thus
seems to be precluded.
Introduction 7
'project' their 'thinking-valuable' ; they j udge and speak
of the objects of their thinking as valuable. But the move
is legitimated in the case of a person good at her metier;
her value is transmitted to what she "thinks-valuable " .
(g) The transmitted value may b e viewed a s absolute.
If the value of persons were relative, the value transmitted
would be relative. But persons as good qua members of
a kind may be viewed as candidates for primary
examples of what is of absolute value. Grice argues for
granting them that status on the basis of the meta
physical principle of supply and demand. The proj ection
routine or procedure is a way by which we might extend
the range of value judgements if we could start with
persons being of absolute value. We could then get the
absolute value we as rational beings legitimately
demand.
The operation flauntingly named Metaphysical Tran

substantiation is a procedure for the redistribution, but not
the invention, of properties. What was accidental becomes
essential. Rationality, which in Grice's account attaches
only accidentally to humans, becomes via this construction
routine an essential property of persons.3 Although there is
no new stuff in the world, the world is meant to be richer
as a result of this metaphysical routine. If the title has had
its full provocative effect, the reader needs some distance
from what is supposed to happen to humans in their
.l 'They [humans] do not alter the totality of attributes which each of
them as humans possess, but they redistribute them; properties which they
possess essentially as humans become properties which as substances of a
new psychological type called persons they possess accidentally; and the
property or properties called rationality, which attached only accidentally to
humans, attaches essentially to persons. While each human is standardly
coincident with a particular person (and is indeed, perhaps, identical with
that person over a time), logic is insufficient to guarantee that there will not
come a time when that human and that person are no longer identical, when
one of them, perhaps, but not the other, has ceased to exist.' ( 'Reply to
Richards', Philosophical Grounds of Rationality: Intentions, Categories,
Ends, ed. Richard Grandy and Richard Warner (Oxford: Oxford University
Press, 1 98 6 ) , 1 02 ; see p. 1 14 below.)
8 Introduction
passage to personhood. Mo�e modest e� ample s must
illustrate how such a metaphysical construction works and
with wha t justice we may think of this procedure, in
general, as a natural disposit ion.
Here are some mundane examples of something like
metaphysical transubstantiation, three candidates for new
entities that come into being in th at way. First, sheepdogs.
Suppose that there are dogs which have the herding
characteristic, that a dog which attached itself to a
shepherd exhibited this feature, and that the shepherd
came to exploit and develop what he found in this animal.
The dog has as an accidental property a capacity and
desire to herd. To conceive of this property as essential
would be to think of an animal whose j ob or function it is
to herd sheep, and this of course is what we mean by a
sheepdog. But so far, although we have introduced an
expression whose use is evident we could not, says Grice,
claim that a new kind of entity had come into being. We
would have only 'a new way of speaking' .
What more is needed, to move from new words to new
entities ? It is plausible to suppose that there is a tendency
to reconceive things in terms of their usefulness to us, to
shift the emphasis of the properties which characterize
them. The new way of speaking arises from or is rather
part of it. If there is such a natural disposition, or
construction routine, at work, it creates a new kind of
entity according to Grice only if there is legitimate and
legitimating motivation ; thus a theoretical purpose, which
would be served if we could show that we can extend to
the new entity needed explanations. The following motiva
tions seem relevant to establish such a purpose : first, we
would like to be able to extend to the dog those forms of
explanation which involve attitudes toward work pre
viously restricted to humans. We need to explain the
sheepdog's behaviour using such notions as recognition of
a task, eagerness to get to it, perception that help is needed.
We may explain a particular old sheepdog's longevity as
Introduction 9
we do a person's, in terms of its desire to get its work done.

The new kind of animal is one that essentially co-operates
with a human in a job, that can indeed be said to have a
job.
Second, we will want to justify treating our eo-worker
animals in a way different from the way in which we treat
wild animals. That is, we may want to extend not only
prinCiples of explanation, but also practical principles.
It is not in itself the use we make of an animal with
herding instincts which brings into being a new kind of
entity, nor even the training which devel ops the animal's
natural capacities. Although our use and training may
bring into the world a particular example of the kind, the
construction of the kind is distinct; it is possible only given
the shift in concept coupled with a legitimate demand for
an entity of the given type. There is an additional
observation to be made here. The shift in conception in the
case of sheepdogs may be what a purposive human
requires in order to engage in the activities involving the
animals. What can then be viewed as a necessary pre
supposition for engaging in those activities backs (or can
be used to back) the claim that a new kind of entity has
been constructed.
Reference to two additional examples will allow us to
raise questions about the two kinds of motivation for the
introduction of new entities. Pets and tools exemplify the
redistribution of properties typical of Metaphysical Tran
substantiation. However, the legitimating motivation
seems purely practical. We would like to extend practical
principles, possibly, in the case of pets, via an extension of
the notion of legitimate expectations. A pet's expectations
could then be understood to derive from the kinds of
dependency we previously held ourselves responsible for
only with respect to humans.
The first tools may have been coincident with found
natural objects which exhibited features useful to humans.
The practical need to conserve the most apt of these, to
10 Introduction
keep other humans from casually disposing of them, and
to enjoy the advantages of trading for them and improving
them, provides practical motivation for the new kind of
entity, for the redistribution of properties, such that a tool
is essentially what is able to do certain work for us. Tools
are essentially valuable to us, and so whatever practical
principles apply to guarding, conserving, and caring for
what is valuable get extended to them.
It might be argued that there is no mystery (and hence
nothing metaphysically exciting) in our changing behaviour
and practices, in extending practical principles, in these
two examples. For all that is at issue is the way we treat the
object, not the object's behaviour. This is why purely
theoretical motivation (as in the sheepdog example) seems
to provide stronger support for a new entity. It is not clear
how much weight one should give this criticism: first, it is
only our conception and our subsequent behaviour which
is at issue in the case of those social constructions like
money which we accept as part of the world. Second,
where the constructed entities have their place in the world
of practice or action, the appropriate justification may well
exemplify what Grice thought to be one pattern of
transcendental arguments : some area of local rationality,
or as we might paraphrase, some forms of behaviour and
human activity would be unintelligible and unavailable to
us without the disputed entity. Grice himself shifts, as
noted, in calling first for theoretical motivation, and then
for a rational demand, in order to legitimize construction
of a new entity.
How are we to understand the application of this
routine in the case of persons ? Although Grice does not
raise the question, if his argument is to go through, it must
be possible to see at work here a natural disposition to
conceive of the world (or of ourselves) in certain ways. His
claim is that there is a rational demand for absolute value
which can be satisfied by finding a being whose essence it
is to establish and to apply forms of absolute value, and
Introduction 11
that such a being can be obtained by Metaphysical
Transubstantiation. It is accompanied by a suggestion-for
an-argument to show that there is a rational demand for
absolute value. Once the rational need or demand for such
an entity has been established, Grice can invoke the
'Metaphysical Principle of Supply and Demand'. Certainly
Grice meant to provide justification: for the construction
routine of Metaphysical Transubstantiation and so for the
entity constructed. The suggested argument for absolute
value seems to be his answer to why humans would engage
in this routine.
Although complicated by the fact that two constructions
are involved-absolute value and persons-the general
pattern is clear. We need absolute value, and persons can
provide it. Although it is not likely that anyone is actually
motivated to engage in the construction routine by such a
thought, this does not show Grice's account to be faulty.
His theory, which describes both the justifying principles
and the forms of construction, may be seen as displaying
our ability to construct without supposing the routines to
be actual, historical, processes . It is an ability we must
have to satisfy our rational demands. But the philosophical
task of constructing value cannot be viewed as one of
simply showing how it is possible for value to be
constructed. For that there is absolute value is contested,
and the philosopher needs to do more to establish that a
new entity has been constructed.
It may be that the construction routines, understood as
natural ways to conceive or reconceive the world, can do
much to account for the objectivity of our construction.
The rationale which Grice offers as justification for the
routine of Metaphysical Transubstantiation, that it will get
us absolute value, will need to be supplemented by an
arguably natural tendency to redistribute the property of
rationality. Plausible motivation must be found which can
arguably be attributed to most or many, if not all, humans.
For a human performs Metaphysical Transubstantiation
12 Introduction
only on herself. And if Grice's metaphysical essay is to
provide a foundation for ethics, we clearly want it to be the
case that most, if not all, humans are persons and so of
absolute value.
The difficulty does not lie in finding some natural
disposition on our part to re-evaluate our own charac
teristics. No doubt individuals may fairly commonly shift
in their sense of what is central to who they are ; but one
needs to find rationality the focus of such a change over a
common enough ground. Two examples seem likely
candidates for Grice's redistribution of properties . One's
integrity might be at issue in a given moral dilemma, and
provoke a reconception-cum-commitment on the part of
the agent to be a certain sort of person, namely, one for
whom ultimately justifiable principles for action and
thought are central. A given commitment to theoretical
endeavour might also be interpreted in this way. Certainly
moral behaviour and theoretical practice would provide
admirable examples of a common human activity in which
a demand for a new entity could arise. But what is needed
is a way to see how it is that the reconception prompted by
issues of integrity makes those activities intelligible and
possible.
A third candidate may be more helpful. It is important
to remember that Grice would not have regarded what is
called 'rational choice theory' as an adequate study of
rationality. He understood rationality to include a concern
for justified or validated attitudes and decisions, a concern
which questions the desirability or propriety of ends as
well as the means taken to achieve them. Humans possess
rationality as an accidental property. But Grice does not
say what difference it would make for this property of
rationality to be essential, apart from the attribution of
value which is made possible by the change. A speculation :
although humans look for justification, are motivated to
find reasons for their attitudes, even ultimate ones, we
cannot expect them to be fully thoroughgoing in their
Introduction 13
search. Should the principles they propose for themselves
or the search for them seriously conflict with self-interest,
they will not withstand the challenge. Only an individual
with an essential or central concern to have justified
attitudes can reasonably be expected to withstand the
conflicts posed by self-interest.
Here is a suggestion for the kind of redistribution of
one's properties we are looking for. When we engage in a
co-operative venture with other humans which is based on
agreement, and where compliance contains elements of
trust, our activity is intelligible only if we conceive of
ourselves and the others as essentially rational. Although
in the act of identifying myself with this kind of group I
only construct one given individual instance, myself, I do
this via construction of the type. An additional benefit of
the example of trust is that, even if we suppose ( for reasons
not discussed) that a human must perform the routine of
Metaphysical Transubstantiation on herself, the construc
tion envisaged in trust requires conceiving of other humans
as persons and so treating them as persons.
The underlying ideas here are familiar: insofar as I think
of the general truth 'we are all alike, we all stand to gain by
doing this together', I need impute nothing more than self
interest to myself and others. I can think our capacity to
engage in, and act on the basis of, reason-giving to be
essentially prompted by and limited by our self-interest.
But if I believe myself and others to be the kind of creature
concerned with standards of acceptability which are not
inherently limited by the criterion of self-interest and
preservation, if I think we are concerned to find acceptable
rules of conduct which are maximally general and which
can be justified to other rational beings, then I think of us
as that new kind of species, persons.
I need not, of course, use the term "persons " : insofar as
my expectations that others will keep their part of the
agreement extends beyond what could be expected on the
basis of their calculation of self-interest, I exercise trust.
14 Introduction
That others are essentially rational, concerned to justify
their actions, would make that trust intelligible, as well as
justifiable. There is good reason to suppose not only that
we often trust others in this way, but that there is such a
natural tendency or disposition on the part of human
beings. For such a disposition is biologically useful; to get
on with the business of life in a community some modicum
of trust is essential. This is a good example for the thesis
that the construction routines are best viewed not as
invented procedures but as natural tendencies to take the
world to be a certain way.
We may now turn to the issue of Humean Projection.
Hume attributed to the mind a tendency 'to spread itself
upon objects', to project into the world items which
properly or primitively considered are only features of our
states of mind. What served as a diagnosis of one source of
illusion becomes, under specified conditions, a legitimizing
procedure in Grice. It is important to remember that many
instances of projecting our wishes and attitudes onto the
world do not carry reason or objectivity with them. We
find things and people bleak or hateful, awesome, ace,
charming, and likeable; we use the vocabulary of our
passions to describe the world, but this does not by itself
alter what there is.
Grice thought of Hume's description of the mind's
disposition to project as prototypical of the procedure he
envisaged. He described two important applications of
Humean Projection. Their differences arise mainly from
the purposes for which Grice deployed them. In the APA
address he wished to legitimize the attribution to creatures
of concepts which were unavailable to their predecessors.
It is part of the story of the development of more
sophisticated psychological creatures. It offers a partial
explanation of what it is to be more developed, by
describing greater internalization or representational
capacity in a creature. Internalized expressions included
connectives, quantifiers, temporal modifiers, mood-
Introduction 15
indicators, modal operators, names of psychological states
like 'judge' and 'will'. In the Carus lectures and the 'Reply'
the procedure was used to deliver legitimated concepts.
At stage 1, some initial concept, like that expressed by
the word 'or' or 'not' or 'value', is seen as an extrinsic
modifier of a psychological attitude (APA) , or as an
intuitive but unclarified element of our vocabulary
('Reply') . At stage 2 we reach a specific mental state like
"or-thinking" (disjoining) , "not-thinking" (rejecting,
denying) , and "value-thinking" (valuing, approving) .
These states must be manifested in behavioural responses
appropriately connected to the initial concept: for example,
" or-judging A, B" must generate behaviour connected to
that of " judging A or judging B" .
At stage 3, reference to these specific states is replaced
by a more general psychological verb like 'think' or
'judge', coupled with an operator which appears within,
and is restricted to, the scope of the general verb, as in
'thinking p or q' or 'thinking it valuable to learn Greek'.
Only at stage 4 is the restriction removed, so that the
operator can appear in subordinate clauses. Grice warns
that stage 3 will be apparent and not really distinct from
stage 2, that we will have only a notational variant of an
expression, unless certain conditions are met. But these
conditions differ in his two essays.
In the APA address, what is required to reach a real third
stage, for more than notational change, is that an
embedded operator must occur within the scope of
another embeddable operator. Thus 'x judges (if A or B,
C) ' . ' Or-thinking A or B' is accorded the status of judging
that A or B only when it can be embedded within another
operator, here the conditional judgement. But the move
from one .stage to another must itself, for Grice, be
motivated by a theoretical purpose. Grice speculates: there
is the demand for a relatively high grade of inference, in
which the passage from premisses to conclusion is en
compassed, so to speak, within a single thought.
16 Introduction
The specific benefit of the procedure described in the
APA address is that at stage 3, although we cannot yet
claim full objectivity for value (what is thought to be of
value lacks syntactico-semantical freedom from being
'thought-that' ), judgements of value are not translatable
back into preferences or imperatives or any other approving/
volitional attitudes from which they are being constructed.
This is arguably due to the more complex linguistic and
mental structure in which we find them. It has often been
noted that meta-ethical substitutes of favourings and
imperatives do not allow for the kind of argumentation
which our value judgements permit. Non-translatability
does not guarantee that there is no counterpart move in the
language of approval for every move in the language of
value. It does however put the onus on the anti-objectivist
to make out such a claim.
In the 'Reply' Grice requires that one also reach stage 4.
Seemingly, it is because 'the permissible complexities [can]
be made intelligible' (p. 9 8 ) only via the provision of
something like truth conditions for statements within
which the operators appear. Legitimate motivation is
demanded as well as truth conditions or their analogues ; a
point or purpose must be served by invoking a new stage.
In the Carns lectures and the 'Reply', truth conditions or
some analogous valuation is called for to reach stage 4.
The likely candidate for accreditation is the philosophical
story Grice himself tells. This would permit us to attribute
to persons, at least, value as members of a kind, where the
statement that persons are of value will be detachable from
its psychological attitude.
Of course one will think that many other things are of
value. To get to the objectivity of those judgements, Grice
invokes the following principle: Humean Projection, if
performed by a good person in conditions of freedom,
allows for the transmission of (absolute, non-relative)
value from the person to what seems to him valuable.
Grice thought his principle an analogue of the Aristotelian
Introduction 17
'what seems good to the good man is good. ' The principle
may be viewed as itself an application of the metaphysical
principle of supply and demand; it is a procedure for
securing rational demands that is seemingly 'trouble-free'.
As rational beings we demand justification for our projects
and attitudes ; the standards for acceptance do not,
however, come ready-made, but must themselves be
constructed. Persons, who work at their job, who are
concerned to find principles acceptable to rational beings,
get the benefits of their job: what seems good to them is.
Grice was amply aware of the problems left unsolved by
his account. This introduction will address only two
objections it might be natural to raise to the solutions he
offered. The first arises from the procedure of Humean
Projection, the second concerns both construction routines.
One may worry that the application of Humean Projection
involves a dangerous circle: value is a suspect term, hence
has to be put through the mangle to be legitimated; but it
can only be legitimated if we suppose the creature at the
mangle to be of absolute value. The 'Reply' offers the
means to respond: Grice refers to the principle he once
invented, though he did not establish its validity, a
principle he labelled Bootstrap. 'The principle laid down
that when one is introducing the primitive concepts of a
theory formulated in an object language, one has freedom
to use any battery of concepts expressible in the meta
language, subject to the condition that counterparts of
such concepts are subsequently definable or otherwise
derivable in the object-language' (p. 93) . He could appeal
to Bootstrap, to argue that when value as member of a
kind is attributed to persons, the term is taken from the
theoretician'<; language, but is viewed as primitive, and
unanalysed in the language of humans whose concepts are
the object of study. This is legitimate, because it is
redeemed once value has been put through the mangle;
truth conditions or something analogous having been
specified for statements of value.
18 Introduction
The second objection contests the legitimating role
played by a 'theoretical purpose' and maintains that
playing a part in a theory and being objective are quite
separate, and that the fact that, for example, persons might
be needed for a certain theory does not go very far towards
establishing their objectivity. The criticism gets its force
from a position taken in the instrumentalist-realist dispute
within the philosophy of science. Now there are certain
disanalogies in the work and aims of Grice and those
disputants which need to be noted at the start, but none
the less there is a real issue here, namely, what the
difference is between scientific and metaphysical construc
tion, and so what differences with respect to the objectivity
of the constructed entities. First, Grice might have been
happy if theories of value were 'anti-realist' but had as
much objectivity as parts of natural science and mathe
matics.
Second, the kind of entities which are the focus of the
dispute in philosophy of science-unobservable or, better,
unobserved entities, are not really the kind of entities
which concerned Grice. He once noted in a seminar that
some theoretical entities in science, that is, entities inferred
but not observed or thought to be observable, have
become at a given time observable or recognized to be so.
But note that there are two entirely different ways in which
this has been held to come about. There is the naive view,
held by some philosophers and the vast majority of
microscopists, that successive improvements in instru
ments have made whole domains of objects observable,
starting with sperm and cells in the seventeenth century
and, for example, the structure of man-made, integrated
circuits on silicon chips in our own day. Then there is the
idea, entertained by some philosophers, that if certain
theories are true, then objects that we have always been
able to observe are identical with what we formerly
thought of as unobservable. For example, certain visible
crystals are now held to be single molecules, and certain
Introduction 19
flashes of light are held b y NASA to b e high-energy
electrons. If the scientific theories are correct, and a
Kripke-like theory about identity is correct, then we were
always able to see some molecules and some electrons.
Evidently what counts as observable is not plain sailing.
In the first type of case, in which we extend the limits of
observability by making better instruments, it is contingent
that we should have become able to observe the sperm or
make and observe chips. In the second type of case, on one
theory of identity, the flashes are not only identical but
also necessarily identical with electrons, but this is known
only a posteriori. Grice suggested in contrast a quite
different kind of theoretical item, such that it was both
necessary that things of that kind should be observable,
but also that it could be known a priori, in the course of
theory construction, that they should be observable
otherwise they would not fulfil the demands of the theory.
Grice did not propose this curious kind of necessary a
priori observability as the distinguishing feature of meta
physically constructed items. It is however tempting and
an obvious line to pursue if we confine ourselves to the
construction of entities within practical theories. It may be
very important for action that persons are directly known
and are the kinds of creature that we can know a priori to
be directly knowable.
Grice was aware of a need to distinguish scientific from
metaphysical construction-what he referred to in the
'Reply' as, respectively, hypothetization and hypostatiza
tion. In the 'Reply' he emphasized the importance of
construction routines and claimed that we have meta
physical routines of construction, but that there are no
such procedures for science, or that when there are, there is
a move to metaphysical construction. Grice did not
expand on this remark, but it is plausible that the presence
of routines is important because there is a difference in who
does the constructing; not a small scientific elite, or group
of experts, but people in general.
20 Introduction
One might, however, still object: although routines may
distinguish between the two types of construction, the
most we get for our metaphysical entities is justified
acceptance of them, not true belief. The difference between
the two is integral to the dispute. Grice was himself
concerned that transcendental or metaphysical argument
might establish only the first, and weaker, consequence.
Now Grice asserts in the ' Reply' that in some conditions
acceptance may confer truth, namely, within the area of
metaphysical argument. On the supposition of an intoler
able breakdown of rationality if we do not accept a certain
thesis, then the demand for acceptance would be sufficient
to confer truth on what is to be accepted. 'Proof of the
pudding comes from the need to eat it, not vice versa'
(p. 96) .
Why should we agree to this ? We need to see how the
argument might go in specific cases to see just what factors
are at work. Two features of the earlier example of
sheepdogs raise an important question : the new entity,
sheepdog, would serve what Grice called a theoretical
purpose, permitting us, for example, to explain its
behaviour in terms of awareness of its tasks. But at least
some of the characteristics which serve as explanation may
also be viewed as those presuppositions necessary for any
rational agent who works with and trains the animal. They
make the agent's action intelligible. Moreover, the rational
action in question is a common human activity. Both of
these factors help to make it plausible that a new entity has
been constructed. The issue they raise here is whether they
aren't also underpinnings of the metaphysical argument
which threatens a breakdown of rationality. Grice thinks
metaphysical arguments are practical, that they articulate
what are rational demands. The interpretation here is in
line with and, I believe, supports this view: the meta
physical argument for the construction of value is indeed
practical. For the reconceptions of construction routines
arise from our interests and the demands are made from
Introduction 21
within a practice, from the need to make our common
behaviour intelligible. On this interpretation a theoretical
purpose may legitimate construction, but only when the
construction routine itself is part of a common human
activity.
Independently of the appeal to metaphysical argument,
Grice has available a second response to the challenge that
only necessary acceptance, not truth, is required for his
constructions. The routines themselves may be called on to
fill the alleged gap between the two. In the example of
values, it might be argued that we have both a rational
justification for value and the natural tendency to project
our attitudes on the world; here there is no gap between
justified acceptance of a thesis that there are things of
value and belief in its truth. The scientist who accepts a
certain constructed entity may be committed to a research
programme, to thinking in terms of such an entity. But
ontological reservations can have some expression when
she is not on her job, when it is not up to her to act as a
scientist. Whereas if indeed we value something, our
attitude is naturally projected onto the world; it does not
result from a commitment nor is it confined to those areas
in which action is up to us. Whenever we think of what we
value, we have the attitudes and emotions which go with
it. What difference could belief in truth bring?
In the closing paragraph of the 'Reply', Grice acknow
ledged that his defence of value 'bristled' with problems
unsolved or incompletely solved. He was avowedly happy
rather than dismayed to leave his philosophical friends
with work to do.
The Carus Lectures on the
Conception of Value
1. V A L U E A N D O BJECT I V I T Y
The title which I have chosen for these lectures embodies,

as I am sure you will have noticed, an ambiguity of a
familiar type, an act-object ambiguity. The title-phrase
[The Conception of Value] might refer to the item, whatever
it may be, which one conceives, or conceives of, when one
entertains the notion of value ; again, it might refer to the
act, operation, or undertaking in which the entertainment
of that notion consists, and of which the conception (or
concept) of value, in the first sense, is the distinctive object.
My introduction of this ambiguity was not accidental : for
the precise nature of the connection between, on the one
hand, the kind of thinking or mental state which is found, at
least in primary instances, when we make attributions of
value, and, on the other, the kind of item (if any) which
serves as the characteristic object of such thinking is a
matter which I regard as quite central to a proper study of
the notion of value ; my concern with it, moreover, is not an
idiosyncracy, but has been shared by very many of the
philosophers who, throughout the ages, have devoted
themselves to this topic. Indeed a full understanding of the
relationship between this or that fundamental form of
thinking and the item, or class of items, which is, or at least
might claim to be, a counterpart in extra-mental reality of
that form of thinking seems to me a characteristic end of
metaphysical enquiry. So it will not, perhaps, surprise you
when I suggest, first that one should be ready to pay
24 The Carus Lectures
attention not merely to the special [peculiar] character of
the central questions about value but also to their general
character, that is, to their place on the map of philosophical
studies and their connection with other questions which are
also represented in that map ; and second that we should be
ready, or even eager, if we can, to provide any answers
which may initially find favour in our eyes with a suitable
metaphysical backing. To do this might be a way, and might
even be the only way, to remove the bafflement of certain
people (of whom I know several) who are extremely able
and highly sophisticated philosophers, particularly in the
region of metaphysics, but who say, nevertheless, that they
'really just don't understand ethics'. I suspect that what they
are lacking is not (of course) any competence in practical
decision-making, but rather a dear picture (if one can be
found) of the nature of ethical theorizing and of its proper
place in the taxonomy of the enquiries which make up
philosophy.
To decide whether and to what extent the kind of global
approach which I have in mind would be appropriate in a
treatment of fundamental problems about the nature of
value, it is obviously desirable to have a reasonably well
defined identification of those problems. To judge from
the philosophical literature, prominent among such issues
are questions about the objectivity of value (or of values)
and questions about the possibility of defending or
rebutting scepticism about value (or values); and no
sooner has so much been said than it becomes evident that
methodological uncertainties arise at the very outset of our
investigations. For it is far from dear whether the two sets
of questions to which I have just alluded are identical with
one another or distinct; are questions about objectivity the
same as, or different from, questions about the possible
range of scepticism ? And if the questions are the same,
which way do the identities run ? Is the case for scepticism
to be equated with the case for objectivity, or with the case
against objectivity ?
Value and Objectivity 25
I myself, in these lectures, plan to pursue my investigation
of the conception of value by addressing myself, in the first
instance, to questions about objectivity in this region and
to the relation of such questions to questions about
scepticism. And since my own pre-reflective leanings are in
the direction of some form or other of objectivism, I shall,
with at least a faint hope of determining whether these
leanings are defensible and (indeed) whether they are
coherently expressible, begin (hut I hope not end) by
considering the ideas of two recent anti-objectivists. Today
it is the turn of the late J. L. Mackie; 1 tomorrow I shall
turn to Philippa Foot. 2
'There are no objective values' says Mackie (p. 1 5 ). Let us
try to outline the steps which he takes in order to elucidate
and defend this 'bald statement' (as he calls it) of his
central thesis concerning the status of Ethics. First of all, he
makes it clear that in denying objectivity to values he is not
just talking about moral goodness, or moral value (in the
strictest sense of that phrase), but it referring to a
considerable range of items which could be called "values " ;
to items which could b e 'more loosely called moral values
or disvalues', like 'rightness and wrongness, duty, obliga
tion, an action being rotten and contemptible, and so on' ;
also to an unspecified range of non-moral values, 'notably
aesthetic ones, beauty and various kinds of artistic merit.'
He suggests that, so far as objectivity is concerned, 'much
the same considerations apply to aesthetic and to moral
values, and there would be at least some initial implaus
ibility in a view which gave the one a different status from
the other'. I find myself in some uncertainty at this point
about the extent of the range of values with the status of
1 [J. L. Mackie, Ethics: Inventing Right and Wrong (Harmondsworth,
Middlesex, and New York: Penguin Books, 1977), esp. eh. 1. All the
quotations from this book were taken without change, with one exception:
when quoting from Mackie's p. 17 (p. 31 below), Grice underlined 'not'.]
2 [Especially 'Morality as a System of Hypothetical Imperatives' in
Philippa Foot, Virtues and Vices and Other Essays in Moral Philosophy
(Berkeley and Los Angeles: University of California Press, 1978).]
which Mackie is concerned, and perhaps partly in conse
quence of this uncertainty I am not sure whether his
suggestion is that, so far as relates to objectivity, it is
implausible not to assign the same status to moral and to
aesthetic values, or whether it is the seemingly much
stronger suggestion that, so far as relates to objectivity,
plausibility calls for the assignment of the same status to
all values. I shall return to this question.
Mackie envisages three very different reactions to his
initial 'bald statement' : that of those who see it as false,
pernicious, and a threat to morality; that of those who see
it as a trivial truth hardly worth mentioning or arguing for;
and finally that of those who regard it as 'meaningless or
empty', as raising no real issue. Before going further into
his elaboration and defence of his anti-objectivist thesis, I
shall find it convenient to touch briefly on his treatment of
the last of these reactions. Mackie (pp. 2 1 -2) associates
this reaction with R. M. Hare, who claimed not to
understand what is meant by "the objectivity of values "
and not to have met anyone who does . Hare's position is
(or was) that there is a perfectly familiar activity or state
called "thinking that some act is wrong" to which
subjectivists and objectivists are both alluding, though the
subjectivist calls this state " an attitude of disapproval "
while the objectivist calls it " a moral intuition " ; these are
just different names for the same kind of thing and neither
can be shown to be preferable to the other. As I understand
Mackie's understanding of Hare, this stand-off is ensured
by the fact that the subjectivist has at his disposal a
counterpart move within his own theory for every move
which the objectivist may try to make in order to provide a
distinguishing, and justifying, mark for his view of values
as objective. The objectivist, for example, may urge that if
one person declares eating meat to be wrong and another
declares it to be not wrong, they are, both in reality and on
his theory, contradicting each other: to which the subject
ivist may retort that though on some subjectivist accounts
they cannot, perhaps, be said to be contradicting each
other, they can be said to be negating (or disagreeing with)
one another : if, for example, one (A ) is expressing or
reporting the presence of disapproval of meat-eating in
himself (A ), and the other (B) its absence in himself (B),
this would be a case of disagreement or negation ; and who
is to say that contradiction rather than "negation " is what
the facts demand ? Again, suppose the objectivist claims,
with respect to the persons A and B, one of whom thinks
meat-eating wrong and the other of whom thinks it not
wrong, that he alone (not the subjectivist) is in a position
to assert (as we should wish to be able to assert) that one of
them has to be wrong; Hare's subjectivist, it seems, replies
as follows:
1. Someone (x) thinks that A judges wrongly that meat
eating is wrong
x disapproves A's judgement that meat-eating is
wrong
x disapproves A 's disapproval of meat-eating
x non-disapproves meat-eating
(-�l
2. Someone x thinks that B judges wrongly that meat
eating is not wrong
x disapproves B's judgement that meat-eating is
not wrong
x disapproves B's non-disapproval of meat-eating
x disapproves of meat-eating
(� )
3. Any person x must either disapprove or non
disapprove of meat-eating [disapproval might be either
present or absent in him] .4 So,
3 [The arrow appears to be Grice's shorthand way of saying that Hare's
subjectivist could hold all the above assertions to have the same force, or
that some are successively weaker than their predecessors. No matter what
the subjectivist holds on this point, the move from ( 1), (2), (3), to (4) is
invalid.]
4 [Grice took full advantage of the convention of parentheses and
apparently used square brackets for his more important parenthetical
remarks.]
4. Any person x must judge that either A or B judges
wrongly.
Hare adds the following further consideration (quoted
by Mackie) :
Think of one world into whose fabric values are obj ectively
built, and think of another in which those values have been
annihilated. And remember that in both worlds the people in
them go on being concerned about the same things-there is no
difference in the 'subjective' concern which people h ave for
things, only in their 'objective' value. Now I ask 'what is the
difference between the states of affairs in these two worlds ? '
Can a n y answer b e given except 'None whatever' ?
Mackie seems to m e not to handle very well this attempt

at the dissolution of debates about objectivity. He concen
trates on the final invocation of the indistinguishability of
the two worlds, the one with and the one without objective
values ; and he makes three points against Hare. His first
comment is that Hare's appeal to the two allegedly
indistinguishable worlds does not prove what Hare wants
it to prove; all that it does is to underline the point (made
by Mackie himself) that it is necessary to distinguish
between first-order and second-order ethics, and that the
j udgements or other deliverances which fall within first
order ethics may be maintained quite independently of any
judgement for or against the objectivity of values, which
will fall within second-order ethics ; it does not show, as
Hare would like it to, the emptiness or undecidability of
such questions about objectivity. That such questions are
not empty is, according to Mackie, indicated by his two
further comments ; first, that were beliefs in the objectivity
of values admissible, they would provide us with a
justificatory backing for our valuations, which we shall
otherwise be without; and second, that were the world
stocked with objective values, we would have available to
us a seemingly simple way of acquiring or changing our
directions of concern ; one could simply let the realities of
the realm of values influence one's attitudes, by 'letting
one's thinking be controlled by how things were'. Hare's
failure to allow for such considerations as these is laid by
Mackie at the door of Hare's "positivism " , which is
comparable with that of a Berkeleian who insists that
appearances might be just as they are whether or not a
material world lies behind them (or under them) .
I am unimpressed. Mackie's first point relies crucially on
a deployment of a distinction between first- and second
order ethics which is a central part of this theoretical
armament, but whose nature and range of legitimate
employment I find exceedingly obscure. I shall postpone
further comment until I return to this element in Mackie's
apparatus. As for Mackie's other points, "positivism" is, I
agree, a bad word, and accusatory applications of it are
good for an unreflective giggle. But I suspect that many
would regard an unverifiable backing for the propriety of
our concerns as being little better than no backing at all.
And while it might be held that objective values, should
they exist, might exercise an influence on our subjective
states, it is by no means clear to me that this is an idea
which an objectivist would, or even should, regard with
favour. Mackie seems to me, moreover, to have missed the
real weakness in Hare's argument ( at least, as presented by
Mackie) . The execution of the second stage of Hare's
'duplication procedure' relies essentially, but not quite
explicitly, on the idea that with regard to any particular
" content" cj>, anyone must either disapprove cj> or not
disapprove cj>. This is indeed, as Hare says, a tautology,
but unfortunately it does not entail the premiss which he
needs so that his argument will go through; that premiss is
that for any cj>, anyone either has an attitude of dis
approval with respect to cj> or an attitude of non
disapproval with respect to cj>. This is not a tautology,
since absence of disapproval only amounts to an attitude
of non-disapproval if some further condition is also
fulfilled, e.g. that the person concerned has considered the
matter.
The upshot of this discussion is that I am prepared to
concede that Mackie is right, though not for the right
reasons, when he claims that Hare's attempt to establish
that there is no real issue between objectivists and their
opponents fails. To make this concession, however, is to
condemn only Hare's attempt to show that there is no real
issue ; I remain perfectly free, should further argument
point that way, to revive a " dissolutionist" position in a
new or modified form. I turn now to the task of trying to
identify more precisely the thesis about which objectivists
and anti-objectivists are to be supposed to disagree; and I
shall start by trying to get clear about what Mackie regards
as the thesis which, as an anti-objectivist, he is concerned
to maintain. First of all, it is an important part of Mackie's
position to uphold the existence of a distinction between
first-order and second-order topics (questions, ethical
j udgements) and to claim that, though both first-order and
second-order questions may fall within the province of
ethics, his anti-objectivist thesis, like all questions about
the status of ethics, is of a second-order rather than a first
order kind. First-order ethical judgements are said to
include both such items as evaluative comments about
particular actions, and also broad general principles, like
the principle that everyone should strive for the general
happiness or that everyone should look after himself. By
contrast, 'a second-order statement would say what is
going on when someone makes a first-order statement, in
particular whether such a statement expresses a discovery
or a decision, or it may make some point about how we
think and reason about moral matters, or put forward a
view about the meanings of various ethical terms' (p. 9).
Mackie holds there to be a considerable measure of
independence between the two realms (first-order and
second-order) ; in particular, " moral scepticism" may
belong to either of the two realms and 'one could be a
second-order moral sceptic without being a first-order one,
or again the other way round. A man could hold strong
moral views, and indeed ones whose content was

thoroughly conventional, while believing that they were
simply attitudes and policies with regard to conduct that
he and other people held. Conversely, a man could reject
all established morality while believing it to be an objective
truth that it was evil and corrupt' (p. 16).
A second salient feature of Mackie's version of anti
objectivism (or moral scepticism) is that it is a negative
thesis. 'It says that there do not exist entities or relations of
a certain kind, objective values or requirements, which
many people have believed to exist' (p. 1 7) . On some views
which have been called objectivist, an objectivist position,
despite its positive guise, would turn out to be intelligible
only as the denial of some position which would bear the
label of " subjectivist" , e.g. as the denial of the contention
that value statements are reducible to, or really amount to,
the expression of certain attitudes like approval or
disapproval. On such an interpretation, of the pair of
terms, " objectivism" and "subjectivism" (or " non
objectivism " , if you like) , it would be the latter term which
would be, perhaps despite a negative garb, what used to be
called in Oxford (with typical artless sexism) the "trouser
word " . But, for Mackie, " objectivist" is not a crypto
negative term. A third salient feature is closely related to
the foregoing; the assertion or denial of objectivism is not,
like some second-order ethical theses, a semantic thesis
(about the meaning of value terms or the character of value
concepts), nor is it a logical thesis (e.g. about the structure
of certain types of argument) , but it is an ontological
thesis ; it asserts (or denies) the existence of certain items in
the world of reality. Fourth and last, since Mackie's moral
scepticism is proclaimed by him not to be a thesis about
the meaning of what moral j udgements or value statements
assert, but rather about the non-presence of certain items
in the real world, it seems to be open to him to hold that
the real existence of values is implied by, or claimed in,
what ordinary people think and say, but is nevertheless not
in fact a feature of the world, with the result that the
valuations spoken or thought by ordinary people are
systematically and comprehensively false. This is in fact
Mackie's position ; his view is what he calls an " error
view " : 'I conclude, then, that ordinary moral judgements
include a claim to objectivity, an assumption that there are
objective values in just the sense in which I am concerned
to deny this' (p. 3 5 ). He compares his position with regard
to values with that adopted by Boyle and Locke with
regard to colours. The suggestion is (I take it) that Boyle
and Locke regarded it as a false, vulgar belief that things in
the real world possess such qualities as colour ; real things
do indeed possess ·c ertain dispositions to give us sensations
of colour, and also possess certain primary qualities (of
shape, size, etc.) which are the foundations of these
dispositions . But neither of these types of item, which
provide explanations for our sensations of colour, is to be
identified with particular colours, or colour; indeed,
nothing is to be identified with a particular colour. And the
situation with values is analogous.
This leaves us with two questions calling for answers:
( 1 ) Why does Mackie hold that claims to objectivity are
incorporated in ordinary value judgements ? (2) Why does
he hold that these claims are false ? With regard to the first
question, one should perhaps first produce a bit of
preliminary nit-pic.k ing. Mackie himself wants to hold that
a claim to objectivity is incorporated in the ordinary value
j udgement; such a claim is therefore presumably part of
the meaning of such value j udgements (or the sentences in
which they are expressed) ; and it does not seem to be, or to
be regarded by Mackie as being, a platitude that such a
claim is included. Mackie cannot therefore consistently
assert that his anti-objectivism is not a thesis about the
meaning of value averrals; the most he can claim is that
though it contains a thesis about meaning, it is not
restricted to a thesis about meaning. More importantly, his
view that a claim to objectivity is incorporated in an
ordinary value judgement seems t o rest, perhaps somewhat
insecurely, on his suggestion (pp. 32-4) that there are two
leading alternatives to the supposition that it is the
function of ordinary value judgements to introduce object
ive values into discourse about conduct and action : non
cognitivism, which (broadly speaking) characterizes value
averrals not as statements but rather as expressions of
feelings, wishes, decisions, or attitudes ; and naturalism,
which treats them as making statements about features
which are objects of actual or possible desires. Both
analyses leave out, and are thought by the ordinary user of
moral language to leave out, in one way or another 'the
apparent authority of ethics'. The ordinary man's dis
comfort is relieved only if he is allowed to raise such
questions as 'whether this course of action would be
wrong in itself. Something like this is the everyday
objectivist concept of which talk about non-natural
qualities is a philosopher's reconstruction' (p. 3 4).
Mackie has two arguments, or bundles of argument, on
which he relies to support his thesis that the objectivist
elements, which according to him are embedded in
ordinary value judgements, and in consequence the value
judgements which embed them, are false. He calls these
arguments the argument from relativity and the argument
from queerness, and considers the second more important
than the first. The premiss of the argument from relativity
is the familiar range of differences between moral codes
from one society to another, from one period to another,
and from one group or class to another within a complex
community. That there exist these divergences is, according
to Mackie, just a fact of anthropology which does not
directly support any ethical conclusion, either first-order
or second-order. But it may provide indirect support for
such conclusions; Mackie suggests that it is more plausible
to suppose that moral beliefs reflect ways of life than the
other way around: people (in general) approve of
monogamy because they live monogamously, rather than
live monogamously because they approve of monogamy.
This makes it easier to explain the divergences actually
found as being the product of different ways of life than as
being in one way or another distorted perceptions of
objective values. The counter-suggestion that it is open to
the objectivist to regard the divergent beliefs as derivative,
as the outcome of the operation of a single set of agreed
upon, very general principles on diverse circumstantial
assumptions, is dismissed on the grounds that often the
divergent beliefs do not seem to be arrived at by derivation
from general principles, but seem rather to arise from
'moral sense' or 'intuition'.
The second argument, the ' argument from queerness'
consists in an elaboration, along not wholly unfamiliar
lines, of the contention that the objectivist, in order to
sustain his position, is committed to 'postulating value
entities and value-features of quite a different order from
anything with which we are acquainted' and also to
attributing to ourselves, in order to render these entities
and features accessible to knowledge, a special faculty of
moral intuition, a faculty 'utterly different from our
ordinary ways of knowing anything else'. In this connec
tion he focuses particularly on the so-called relation of
supervenience, which has to be invoked in order to
account for the connection of non-natural features with
natural features, and the dependence of non-natural
features upon natural features. The presence of super
venience in particular cases involves the application of a
special sort of " because " ; 'but just what in the world is
signified by this " because " ? '
Before I try to estimate the merits and demerits of
Mackie's position and of the arguments by which he seeks
to support it, there seem to me to be two directions of
enquiry which are important in themselves, and which
could be conveniently attended to at this point, particularly
as consideration of them might help to give shape to an
evaluation of Mackie. First of all, there are (as Mackie
observes) several different possible interpretations of the
notion of objectivity, most of them mentioned by him at
least in passing, but not all of them ideas which he is
concerned to develop or apply. I think it might be useful to
enquire what kind or degree of unity, if any, exists between
these different readings of the notion of objectivity.
Second, I find myself in considerable uncertainty about the
connection or lack of connection between attributions (or
denials) of objectivity and the adoption (or rejection) of
scepticism in one or other of its forms. Does scepticism
reside in the camp of the non-obj ectivist (e.g. Mackie) or in
that of the objectivist, or (perhaps) sometimes in one and
sometimes in the other ?
As regards the notion of objectivity, we have first the
interpretation which seems to be the one singled out by
Mackie, according to which to ascribe objectivity to a class
or category of items is to assert their membership in the
company of things which make up reality, their presence in
the furniture of the world. We might call this sort of
objectivity, metaphysical objectivity, and it is the kind of
objectivity most commonly supposed to be claimed by
realists for whatever it may be that they are realists about.
A main trouble with this kind of objectivity is the difficulty
in seeing what it is that the objectivist could be claiming;
whether, for example, in attributing objectivity to numbers
or to material things he is doing anything more than
shouting and banging the table as he says 'numbers exist'
or 'material things are real . ' If the proposition that
numbers exist is a consequence of the proposition that
there is a number between three and five, what is the
objectivist asserting that anyone would care to deny ? That
numbers (or values) do not just exist, they really exist?
And what does that mean ? To escape this quandary, it is
not uncommon to take the course which Mackie rejects,
namely, to understand 'values (or numbers) are objective'
as really negative in character, as a denial of the suggestion
that values (or numbers) are reducible, by means of one or
another of the possible varieties of reduction, to members
of some class of items which are not values (or numbers) ,
t o (for example) natural features which find favour, o r to
classes. Or, maybe, not any and every form of reducibility
would be incompatible with objectivity, but only the kind
of reducibility whose direction is to states of mind,
attitudes, or appearances, to subjective items like approvals
or seeming valuable. An objectivist would now be a
resister, an " anti-dissolutionist", one who seeks to block
certain moves to reach a theoretical simplification or
economy with regard to the constituents of the world. The
objectivist's prime opponent may however be a dissolu
tionist not in this commodious sense, but in a different and
perhaps even more commodious sense. This opponent may
be one who seeks not to dissolve the target notion (value,
number, material thing, or whatever) into some one or
more different and favoured items or categories of item,
but rather, in one or other of a multitude of diverse ways,
to dissolve the target notion altogether, to dissolve it into
nothing; he may be a 'nihilistic dissolutionist' . He may
suggest that belief in the application of the target notion is
a mistake, one which characteristically or inevitably grips
the unschooled mind; or that such beliefs can claim only
some relativized version of truth (like truth relative to a set
of assumptions, or to a set of standards) , not absolute
truth. Mackie himself allows to some value j udgements
'truth relative to standards', even though by implication he
seems to deny to them " absolute " truth [whatever the
ordinary man may think] . Again, the anti-objectivists may
wish to suggest not that attributions of the target notion
are mistakes but rather that they are inventions, or perhaps
myths (that is to say, inventions which are backed by
practical motivation, perhaps derived from the utility of
such inventions towards the organization of some body
of material ; in the case of values (perhaps) the body of
material might be rules or principles of conduct) . As myths
(or as the stuff of which myths are made) they might have
fictive reality, o r b e " as if" real, without possessing reality
proper. Or again, the target notion might be held by the
anti-objectivist to be a construct (or a construction) :
though possessing (or belonging to) reality, values might
be held to lack (or fail to inhabit) primary or original
reality ; they would belong to an extension of reality
provided by us. By contrast, an obj ectivist about values
would attribute to them primary or original reality. [I
should say at this point that in my view such ideas as are
now being raised, that is, distinctions between " as if" or
fictive entities, real but constructed entities, and primary
or original reality, are among the most important and also
the most difficult problems of metaphysics. The obscurity
in this area is evidenced by the fact that constructed (non
original) reality might be conceived by some as possessing
objectivity and by others as failing to possess objectivity ;
for some, deficiency in objectivity precludes truth (at least
unqualified truth) ; for others, value claims might be true
(in some cases) even though values (as constructed items)
lack objectivity.]
It might seem that the wheel, in turning, has now
reached the point from which its turning began ; for the
notion of primitive (unconstructed) reality might be
regarded as the same notion as the hazy notion of " out
thereness " or of " being really real " which typified the
metaphysical objectivist. It might also seem that the new
'interpretation' of objectivity is scarcely if at all less hazy
than the earlier one. In an attempt to dispel the mists a
little, one might offer the notion of causal efficacy as an
index of metaphysical objectivity. Items might be accorded
the ribbon of metaphysical objectivity just in case they
were capable of acting upon other items, and attributes or
features might be regarded as objective just in so far as
they were attributes or features in virtue of the possession
of which one item would causally influence another, in so
far as they helped to explain or account for the operation
of such causal influences. A special case of the fulfilment of
this condition for objectivity would, in my view, be the
capacity, possessed by some objects and some of their
attributes, for being perceived, or exercising causal influ
ence on a percipient qua percipient. Now the idea of
connecting objectivity with causal efficacy seems to me one
which has considerable intuitive appeal, indeed much the
same kind of appeal as that which may have sustained Dr
Johnson in his violent and protracted, though vicarious,
assault on Bishop Berkeley. The adequacy, however, of
this criterion of objectivity would be seriously, if not
fatally, impaired should it turn out that the distinction
between what is primitive and what is constructed applies
within the scope of causal efficacy-if, that is to say, causal
efficacy itself were to be sometimes primitive and some
times constructed. It is my suspicion that this would indeed
turn out to be the case. There would then, perhaps, be no
quick recognition-test for objectivity ; there would be no
substitute for getting down to work and building the
theory or system within which the target notion would
have to be represented, and seeing whether it, or its
representation, does or does not occupy in that theory an
appropriate position which will qualify it as objective.
On the approach j ust considered, then, decisions about
the objectivity of a given notion would involve the
examination and, if necessary, a partial construction of a
theory or system in which that notion (or a counterpart
thereof) appears, to see whether within such a system the
notion in question (or its counterpart) satisfies a certain
condition. The operation of such a decision-procedure
would be torpedoed if the requisite theory or system could
not be constructed, if the target concept were not theory
amenable. The merits of an allegation that a given notion
was not theory-amenable might depend a good deal on
what kind of a theory or system was deemed to be
appropriate ; it would be improper (taking heed of Aristotle)
to expect a moralist to furnish a system which allowed for
the kind of demonstration appropriate to mathematics.
But one kind o f anti-objectivist (who might also be a
sceptic) might claim that for some notions no kind of
systematization was available; in this sense, perhaps,
values might not be objective. It may be (as I think my
colleague Hans Sluga has argued) that Wittgenstein was
both sceptical and anti-objectivist with regard to sensa
tions. In this sense of objectivist, an objectivist would only
have to believe in theory-amenability; he would not have
to believe in the satisfaction, by his target notion, of any
further condition within the appropriate systematization.
One further interpretation of obj ectivity noted by
Mackie is one which I shall not pursue today. It connects
objectivity with (so-called) categorical imperatives as
distinct from hypothetical imperatives, and with the
(alleged) automatic reason-giving force of some valuations.
Since this idea is closely related to Miss Foot's theories, I
shall defer consideration of it.
I have listed a number of different versions of the idea of
obj ectivity, and have tried to do so in a way which exhibits
connections between them, so that the different versions
look somewhat tidier than a mere heap . But many of the
connections seem to me fairly loose ['such-and-such a
notion might be taken as an interpretation of so-and-so'],
and I see little reason to suppose many tight, logical
connections between one and another version of objec
tivity.
So much for the panoply of possible interpretations of
the notion of objectivity. I turn now to the second of the
general directions of enquiry with regard to which I
expressed a desire for enlightenment. How is obj ectivity
related to scepticism ? Speaking generally, I would incline
towards the idea that scepticism consists in doubting or
denying something which either is a received opinion, or
else, at least on the face of it, to some degree deserves to be
a received opinion. In the present context we are of course
concerned only with philosophical scepticism ; and, without
any claim to originality, I would suggest that philosophical
sceptics characteristically call in question some highly
general class of entity, attribute, or kind of proposition ;
what they question are categories, or what, if we took
ordinary language as our guide, would be categories. To
adduce more seeming platitudes, the objectivist is,
compared with the anti-objectivist, a metaphysical infla
tionist; there are more things in his heaven and earth than
an anti-objectivist Horatio would allow himself to dream
of. And so, it is standardly thought, it is Horatio who is the
sceptic and the objectivist who is the target of scepticism;
and (often) Horatio remedies his own initial scepticism by
'reducing' the suspect items to their appearances or
semblances : he takes the phenomenalist cure. But here the
issue becomes more complex than is ordinarily supposed:
for there are to my mind not less than two forms of
scepticism, which I will call "Whether ? " scepticism and
"Why ? " scepticism. It may be true that the run-of-the-mill
objectivist, on account of his inflationary tendencies,
provokes "Whether ? " scepticism, and that the sceptic who
seeks to remedy his own initial scepticism by taking a dose
of phenomenalism is not himself open to "Whether ? "
scepticism. But it may also be true that the phenomenalist
is a proper target for "Why ? " scepticism; for he, has left
himself with no way of explaining the phenomena into
which he has dissolved the entities or attributes dear to the
objectivist. And it may be that the objectivist, if only his
favoured entities or attributes were admissible and access
ible to knowledge, would be in a position to explain the
phenomena; and, further, that this capability would be
unaffected by the question whether the phenomena are
related to possible states of the world (like sensible
appearances) or to possible action (like approvals). If only
he could be allowed to start, the objectivist could (under
one or another interpretation) 'explain' in the one area
why it seems that so arid so is the case, and in the other
why do so and so (e.g. why pay debts ) .
The foregoing message, that both the true-blue, con-
servative, and inflationary objectivist and the red, radical,
and deflationary phenomenalist or subjectivist run into a
pack of sceptical trouble, of one kind or another, and that
more delicate and refined footwork is needed seems to me
to be the front-page news in the work of Kant. It also
seems to me that Mackie, by being wedded to if not rooted
in the apparatus of empiricism, has cut himself off from
this lesson. Which is a pity.
However, I must move to somewhat less impressionistic
comments on Mackie's position. These comments will fall
under three heads:
1. The alleged commitment of 'vulgar valuers', in their
valuations, to claims to objectivity.
2. The separation of value judgements into orders, with
the assignment to the second order of questions or
claims about the status of ethics ; and the remedi
ability of an apparent incoherence in Mackie.
3. The alleged falsity of claims of objectivity.
I should say at once that though I think that the
considerations which I am about to mention show that
something has gone wrong (perhaps that more than one
thing has gone wrong) in Mackie's account, the issues
raised are so intricate, and so much bound up with (so far
as I know) unsolved problems in metaphysics and semantics,
that I simply do not know what prospects there might be
for refurbishing Mackie's position.
1. It seems to me to be by no means as easy as Mackie
seems to think to establish that the 'vulgar valuer', in his
valuations, is committed to the objectivity of value(s) . It is
not even clear to me what kind of fact would be needed to
establish such a commitment. Perhaps if the vulgar valuer,
when making a valuation, (say) that stealing is wrong,
were to say to himself ' and by "wrong" I mean objectively
wrong', that would be sufficient (at least if he added a
specification of the meaning of " objective " ) . But nobody,
not even Mackie would suppose the vulgar valuer to do
that. Mackie relies, in fact, on the alleged repugnance to
the valuer of the two main rivals to an objectivist thesis
about value. But even if this were sufficient to show that
the vulgar valuer believes in an objectivist thesis about
value, it would not be sufficient to show that an objectivist
interpretation is built into what he means when he judges
that stealing is wrong. There are other ways of arguing
that a speaker is committed to an interpretation, for
example, that he has it subconsciously (or unconsciously)
in mind, or that what he says is only defensible on that
interpretation. But the first direction seems not to be
plausible in the present context, and Mackie is debarred
from the second by the fact that he holds that what the
vulgar valuer says or thinks is not defensible anyway.
To illustrate the fiendish difficulties which may arise in
this region, I shall give, in relation to the valuation that
stealing is wrong, four different interpretative supposi
tions--each of which would, I think, have some degree of
philosophical appeal-and I shall add in each case an
estimate of the impact of the supposition on the assign
ment of truth value to the valuation.
(a) There is a feature W which is objective but

provably vacuous of application ; a vulgar valuer, when
he uses "wrong " , is ascribing W. Conclusion: vulgar
valuation 'stealing is wrong' invariably false.
( b) A vulgar valuer thinks (wrongly) that there is a
particular feature W which is objective, and when he
uses "wrong" he intends to ascribe this feature, even
though in fact there is no such feature. Conclusion:
obscure, with choice lying between false, neither true
nor false but a miscue, and meaningless (non-significant) .
(c) A vulgar valuer is uncommitted about what
feature "wrong" signifies ; he is ascribing whatever
feature it should in the end turn out to be that " wrong"
signifies. Conclusion: assignment of truth value must
await the researches of the semantic analyst. ·
(d) A vulgar valuer is uncommitted about what
feature "wrong" signifies ; truth value is assigned in
advance of analysis by vulgar methods, and such
assignment limits the freedom of the semantic analyst.
Conclusion: truth value assigned (as stated) by vulgar
methods.
2. The idea, to which Mackie subscribes, of separating

valuations into orders as a step towards the elucidation of
an intuitive distinction between " substantive " and " formal "
questions and theses in ethics plainly has considerable
appeal ; it seems by no means unpromising to regard
" substantive " theses about values as being first-order
valuations (statements) , and to regard " formal " theses in
ethics, like theses about the logic of value, or the meaning
of value terms, as being a sub-class of second-order theses,
and to regard theses about the status of ethics as also
falling within this subclass-to treat them, that is to say, as
theses about first-order valuations. [Such second-order
theses, of course, though necessarily about valuations, may
or again may not themselves be valuations.] But Mackie's
deployment of this idea plainly runs into trouble. For
according to Mackie, vulgar valuations incorporate or
entail claims to objectivity ; claims to objectivity, according
to him, since they fall within, or imply theses belonging to,
the class of claims about the status of ethics, are second
order claims ; and so, since (presumably) what incorporates
or entails a second-order thesis is itself a thesis of not lower
than second-order, vulgar valuations are of at least second
order. But vulgar valuations, as paradigmatic examples of
substantive value theses, cannot but belong to the first
order, which is absurd. Now I can suggest an explanation
for the appearance on the scene of this incoherence. As I
mentioned earlier, among the possible versions of the
notion of objectivity are what I called a positive version
and a negative version. The positive version, that to
attribute objectivity to some item is to proclaim that item
to 'belong to the furniture of the world', is firmly declared
by Mackie to be his version ; and it is, as I have remarked,
obscure enough for it to be possible (who knows ? ) for
attributions of objectivity to belong to the first order. The
negative version, that to attribute objectivity to something
is to deny that statements about that thing are in this or
that way eliminable or " reducible " , is plainly of second (or
higher) order; and despite his forthright assurances,
Mackie may have wobbled between these two versions.
But to explain is neither to justify nor to remedy: and I
have the uneasy feeling that Mackie's troubles have a
deeper source than unclarities about application of the
notion of order. His "error-view" about value has an
Epimenidean ring; it looks a bit as if he may be supposing
vulgar valuations to say of themselves that the value which
they attribute to some item or items is objective ; and I feel
that it may be that such self-reference, though less
dramatic, is no less vitiating than would be saying of
themselves that they are false. It is true that Mackie
regards vulgar valuations as being, in fact, comprehensively
false ; but it is evident that he expects and wants that falsity
to spring from the general inapplicability of the attribute
being ascribed by such valuations to themselves, not from
a special illegitimacy attending a valuation's ascription of
the attribute to itself.
3 . I find myself quite unconvinced (indeed unmoved) by
the arguments which Mackie offers to support his claim
that values are not objective or ( should one rather say ? )
that there are n o objective values. The first argument from
relativity he regards as of lesser importance than, and
indeed as ultimately having to appeal to, the second
argument, the argument from queerness. This argument
(so it seems to me) seeks to make mileage out of two bits of
queerness : first, the queerness of the supposition that there
are certain "non-natural " value-properties which are in
some mysterious way " supervenient upon " more familiar
natural features ; and second, the queerness of the supposi-
tion that the recognition of the presence of these non
natural properties motivates us, or can motivate us,
without assistance from any desire or interest which we
happen to have. What strikes me as queer is that the
queernesses referred to by Mackie are not darkly concealed
skeletons in objectivist closets which are cunningly dragged
to light by him ; they are, rather, conditions proclaimed by
objectivists as ones which must be accommodated if we are
to have a satisfactory theoretical account of conduct, or of
other items qua things to which value may be properly
attributed. So while these queernesses can be used to
specify tasks which an objectivist could be called upon,
and very likely would call on himself, to perform, and
while it is not in advance certain that these tasks can be
successfully performed, they cannot be used as bricks to
bombard an objectivist with even before he has started
to try to fulfil those tasks. It is perhaps as if someone were to
say, 'I seriously doubt whether arithmetic is possible ; for if
it were possible it would have to be about numbers, and
numbers would be very queer things indeed, quite in
accessible to any observation ' ; or even as if someone were
to say, 'I don't see how there can be such a thing as
matrimony; if there were, people would have to be bound
to one another in marriage, and everything we see in real
life and on the cinema-screen goes to suggest that the only
way that people can be bound to one another is with
ropes.'
Conception of Value
2. RE L A T I V E A N D ABS O L U T E V A L U E
My enterprise in this lecture will, initially at least, be to

take up and pursue a version of the notion of objectivity
which is mentioned by Mackie, but which I have so far
deliberately kept out of the limelight. At the conclusion of
a short discussion of categorical and hypothetical imper
atives, we find (p. 29) the following statement:
So far as ethics is concerned, my thesis that there are no obj ective
values is specifically the denial that any such categorical element
(in moral j udgments) is objectively valid. The objective values
which I am denying would be action-directing absolutely, not
contingently (in the way indicated) upon the agent's desires and
inclinations.
The language is not wholly clear; but what is seemingly

being asserted is that Mackie's denial of objective values is
tantamount to a denial that there are any absolutely
action-directing values, despite what may be claimed in
ordinary moral j udgements. This thesis seems to be a close
relative of a well-known position advanced by Philippa
Foot, who has discussed it at some length, and to whom I
shall turn my attention in a moment. First, however, let
me present the question at issue in a slightly more
comprehensive way. There seem to be a number of fairly
well-publicized dichotomies, to which the objectivity or
non-objectivity of values may be closely related. These
include the dichotomy of categorical and hypothetical
imperative, the dichotomy of moral value and non-moral
value, the dichotomy of absolute value and relative value,
and the dichotomy of unconditional and conditional value.
The questions at issue seem to me to concern the relation
of each of these dichotomies to others in the list. I hope to
return to this array of questions after a hopefully succinct
presentation of what I take to be Philippa Foot's views.
The following I think would be a fair summary (in my
language) :
1 . Hypothetical imperatives are distinguished by the
existence of an associated "let out" or "extrication "
condition. This will consist in the existence of an associated
end, a lack of desire for which will remove from the
potential agent all reason to carry out the injunction
contained in the imperative. The imperative 'If you want a
good dinner, you should eat at the White House' leaves me
cold if I have no interest in food.
2. The widespread belief that moral imperatives are
categorical, in that they have a reason-giving force that is
independent of any desire on the part of the potential
agent, is mistaken. There are no such automatic reason
giving forces, and so no categorical imperatives .
3 . Though there are no categorical imperatives, there
are some "non-hypothetical uses " of " ought" , where a
disclaimer of interest would have no extricating effect.
These occur in " oughts " of etiquette, conduct in games,
and possibly (colloquially) in moral statements. But the
reason why disclaimers of interest have no effect here is
that non-hypothetical uses of ought are (atypically) not
reason-giving at all, and so there is here nothing to be
extricated from.
4. So if moral " oughts " are to be reason-giving at all,
they must be interpreted (or re-interpreted) as expressing
hypothetical imperatives, depending on some end (like
human happiness) which decent people can be counted on
to be concerned about.
5 . To regard moral precepts as categorical imperatives
must be to base morality on reason; anti-moral behaviour
Relative and Absolute Values 49
would have to be represented as counter-rational. But
there is nothing irrational in immorality ; no contradiction
or self-defeating behaviour is (characteristically) present.
6. We do not want moral " oughts" to be ipso facto
motivating or compelling, regardless of interest or inclina
tion or desire. We want volunteers rather than conscripts
in moral service.
7. In an earlier version, morality had to be partially
justified by reference to the happiness of the agent. In a
later version, concern for the welfare of others, as part of
one's own happiness, demands a consequential concern for
morality, with a view to the welfare of others.
Now before I get too heavily involved in substantive
issues, it might be a good idea for me to pay a little heed to
the structural aspects of the region under debate : let us
have a look at the girders before covering them with
cement. There seem to me to be not less than six
dichotomies which are under review, though not every
philosopher would regard all of them as well founded.
Some philosophers would regard some of them as not
distinct from one another, and (I hope) all philosophers
would regard some or even all of them as obscure, perhaps
even intolerably obscure. These dichotomies are (or
include) :
1 . objective-non-objective (or perhaps, subjective),
(entity, value), a dichotomy or cluster of dichotomies
on which I have already spent some time.
2. categorical-hypothetical ( imperative)
3 . absolute-relative (value)
4. moral-non-moral (value, imperative, etc.)
5 . unconditional-conditional (value, etc.)
6 . underived-derived (value)
Today's Special (2. Cats and Hypes), like other members
of the bunch, calls (even clamours) for interpretation.
(a) A blind logical nose might lead us (or be led) to the
assumption of a link between hypothetical imperatives and
hypothetical statements (propositions) . Such a link no
doubt exists, but the. most obvious version of it is plainly
inadequate. At least one other philosopher besides myself
has noticed that 'If he molests the children, you should
have him arrested' is unlikely to express a hypothetical
imperative; and that even if one restricts oneself to cases in
which the antecedent clause specifies a want, we find pairs
of examples like:
If you want to go to Chicago, you should travel by AA
via Cleveland.
If you want to go to Philadelphia, you should see a
psychiatrist.
where it is plain that one is, and the other is not, the
expression of a hypothetical imperative (I won't tell you
which ) .
( b ) A less easily eliminable suggestion, yet one which
would still interpret the notion of a "hypothetical imper
ative" in terms of that particular logical form to which the
names "hypothetical" and " conditional " attach, would be
the following. Let us assume that it is established, or
conceded, as legitimate to formulate conditionals in which
not only the consequents (apodoses) are couched in some
mood (mode) other than the indicative, as in conditional
commands ('If you see the whites of their eyes, shoot
(fire) ' ) but also the antecedents (protases), or some part
(clause) of them; in which case all of the following might
be admissible conditionals :
If let the cat b e taken to the vet, then let it be put in a
cage.
If let the cat be taken to the vet and there is no cage
available, then let Martha put it on her lap.
If the cat is sick, let it be taken to the vet.
If this suggestion seems rebarbative, think of these quaint
conditionals (when they are quaint) as conditionalized
versions of arguments, such as
Let the cat go to the vet, so let it be put in a cage.
Let the cat go to the vet; there isn't a cage, so let
Martha put the cat on her lap.
and then maybe the discomfort will be reduced.
(c) Among conditionals with an imperatival or "voli
tival " consequent, some will have " mixed" antecedents
(partly indicative, partly imperatival) and some will have
purely indicative antecedents (like the last of my three
examples) . I might now give a provisional definition of the
terms categorical and hypothetical imperative. A hypo
thetical imperative is either a conditional the consequent of
which is imperatival and the antecedent of which is
imperatival or mixed (partly indicative, partly imperatival),
or it is an elliptical version of such an imperative. A
categorical imperative is an imperative which is either not
conditional in form, or else, if it is conditional, has a purely
indicative antecedent.
Quick comments :
(i) The structures which I am offering as a way of
interpreting hypothetical and categorical imperatives do
not, as they stand, offer any room for the appearance of
practical modalities like ought and should, which are so
prominently visible in the standard examples of those
kinds of imperatives : the imperatives suggested by me are
really imperatives : they conclude 'do such and such', not
'you/one ought to do such and such'. But maybe my
suggestion could be modified to meet the demand for the
appearance or occurrence of ought (etc.), if such occur
rence is needed.
(ii) It would remain to be decided how close the
preferred reading of my 'deviant' conditional imperatives
would be to the accepted interpretation of standard
hypothetical imperatives. But even if there were some
divergence, that might be acceptable if the 'new' imper
atives turned out to embody a more precise notion than the
standard conception.
(iii) There are, I think, serious doubts of the admissibility

of conditionals with non-indicative antecedents which will
be to my mind connected with the very difficult question
whether the indicative and the other moods are co
ordinate, or whether the indicative mood is in some crucial
sense prior to the other moods . I do not know the answer
to this question.
A third interpretation of the distinction between cate
gorical and hypothetical imperatives would, like the first
two, be formal in character, and would link the categorical
hypothetical distinction, in relation to imperatives, with
one of the other listed dichotomies, that between absolute
and relative value. Hypothetical imperatives would be end
relative value attributions, and might be analogous to
evidence-relative probabilities; categorical imperatives
would not be end-relative. In my eyes this suggestion has
the great merit that the idea of the relativization of value
might (would) apply to other kinds of relativization than
relativization to ends ; a notable member of the wider
group of relativizations would be relativization to subjects
of ends, persons who have or who might have ends
( "valuable to me " ) . It is my suspicion that the range of
different kinds of relativization is going to prove enorm
ously important in the clarification of the idea of value ; it
might, for example, turn out that non-relative (absolute)
value has some special connection with some favoured
relativization of the notion of value (e.g. to people) . This
third suggestion might help us to see hypothetical and
categorical imperatives as important in this context. We
might indeed, not inappropriately, use a further member of
the original list of dichotomies, the unconditional
conditional value dichotomy, as a means for expressing the
distinction between value relative to an end, and value not
relative to an end. Then the distinction between absolute
and relative value would include, as a special case, but
would not be restricted to, the distinction between
unconditional and conditional value.
(d) The last interpretation which I shall mention seems
not to be, as its predecessors were, formal in character. It is
close to part of what Kant says on this topic, and it also
either is or is close to the interpretation employed by Foot.
It is a distinction between an imperative being escapable
(hypothetical), through the absence of a particular desire
or concern, and its not being thus escapable (categorical).
If we understand the idea of escapability sufficiently
widely, the following imperatives are all escapable, even
though their logical form is not in every case the same :
You should give up popcorn.
To get slim, you should give up popcorn.
If you want to get slim, you should give up popcorn.
Now suppose that I have no concern to get slim. One
might say that the first imperative is "escaped " , provided
giving up popcorn has nothing else to recommend it, by
being falsified. The second and third would not, perhaps,
be falsified, but they would, in the circumstances, be
inapplicable (to me)-and inapplicability, too, counts as
escape. Categorical imperatives, however, are in no way
escapable.
We should, I suggest, consider not merely imperatives of
each sort, together with the range of possible characteriza
tions of the sorts, but also the possible forms of (practical)
argument into which such imperatives (particular hypo
thetical imperatives) might, on this or that interpretation,
enter, and even forms of (practical) argument which
involve not hypothetical imperatives themselves, but close
relatives of them. To indicate the importance, for a proper
understanding of this thorny area, of a consideration of the
forms of argument into which they may enter and not
merely of the imperatives themselves, I shall give three
such patterns of argument, at least superficially different
from one another, and (so it seems to me) of varying
degrees of breadth of application.
1. (using dichotomy of original-derived value)
To defend the Philosophy Department would be a
good thing. (It is not specified whether the value is
original or derived. )
If to defend the Philosophy Department would be
a good thing, then to learn to use bows and arrows
would be a good thing (as conducive).
So : To learn to use bows and arrows would be a good
thing. (This would be derived value, provided the
second premiss is true.)
2 . It is noble to fight for your country (unconditional
value) .
It is valuable, in the matter of fighting for one's
country, to j oin one of the services (ascription of
conditional value) .
So: Join up ! (We cannot conclude either to uncondi
tional value of joining up (false) nor to conditional
value of j oining up, with respect to fighting for
one's country, since this is one of the premisses.)
3. It is good for me to increase my holdings in oil
shares.
If I visit my father he will give me some oil shares.
So: It is good for me to visit my father. (This argument
purportedly transmits relative value, that is,
subject-relative value.)
Now where does Foot stand in respect of claims about
value ?
It seems to me that the issues on which battle has been
joined within this topic are always (nearly always) related
to different views about the potency of reason (as reflected
in the lesser scope or the larger scope allowed for the
appearance on the scene of reasons) . It is by no means clear
to me where, precisely, Foot stands in this spectrum (if it
matters) , but wherever it is, it is somewhere in the middle.
The stages which I have in mind are presented in order of
decreasing scepticism, or increasing trust in the power of
reason, or, as I would like to be able to put the matter,
increasing trust in the legitimate and efficacious operation
of the concept of value in the conduct of argument.
1. Thoroughgoing scepticism. The notion of value has
no genuine legitimate application in argument; it is never
strictly speaking the case that one should draw such and
such a conclusion from a set of premisses, that it would be
good or valid to draw such and such a conclusion, or bad
not to. We do, of course, as victims of bad habits,
commonly talk that way, and we do, inveterately, throw
around the word " reasons " , but this is only a way of
talking and is not to be taken seriously: it may be
hallowed, but it is not at all holy. The things we say are
either not to be regarded as true, or if true are true only in
some Pickwickian sense of the words employed. There are
strictly speaking no arguments at all, as ( allegedly) it is not
too difficult to 'demonstrate'.
2. Stingy cognitive rationalism. The terms " value" and
" reasons " properly apply, in a non-Pickwickian sense,
only within the confines of the area of factual belief, the
" alethic" area, and even there only subject to strong
safeguards. The only way in which one can find a
conclusion validated or called for by reason (subject to
reasons) is by finding a case in which to deny rather than
accept that conclusion would involve one in contradiction.
3 . Open-handed cognitive rationalism. The crucial
terms ( "value " and " reason " ) have a more general licence
(inductive reasons are, for example, OK) ; but strictly
legitimate application is still confined to the alethic area.
4. Limited cognitive-cum-practical rationalism . The
crucial terms have a liberal authentic application in the
cognitive (alethic) zone, and also a limited authentic
application in the practical zone, where they are limited
(otherwise than merely as a way of talking) to the area of
the relation of means to ends, the area of Aristotelian
bELVO'tll � (whatever that area may be) .
5 . Unlimited cognitive-cum-practical rationalism. No

types of application are subject to sceptical smear.
Now, as I say, I am not really very sure where Foot
stands in this array of stances; I suspect in category 4. I am
also fairly sure that wherever Foot may stand, quite a large
number of philosophers have occupied, or have thought
that they occupied, one of the intermediate positions
bearing numbers between 2 and 4 (inclusive) . The further
suspicion which I would like at this point to voice is that
the adoption of one of these "part-way" positions is
incoherent, that you either have to be a whole-hog sceptic
or else not a sceptic at all : half-hogging is no good. I shall
not attempt to prove this point now; in Lecture 3 I do try
to prove a closely related thesis (that if you get as far as
stage 4 you have to (in some sense of "have to " ) go on to
stage 5 . But there is a stronger and a weaker interpretation
of " have to " ; the stronger interpretation would allege
some form of contradiction in accepting 4 but refusing 5 ,
and I rather doubt if that can be shown. I have hopes,
however, of being able to reach a weaker conclusion, that
to accept 4 and to reject 5 (to hold, for example, that
hypothetical imperatives are all right, but that categorical
imperatives are not) would be wantonly to refuse to satisfy
a legitimate rational demand. But for that you must wait
patiently for a day.
To initiate a substantive discussion of Foot's position, I
ask what there is in it to appeal to us, and again what there
is in it to make us hesitate or recoil ; and in asking these
questions I note that reactions, whether favourable or
unfavourable, seem likely to be strong. It seems to me that
in these discussions a key role is played by the idea of
reason, or of reasons ; it will be some set of considerations
about reasons which will turn some people on, at least to
begin with, and it will be another set of considerations (or
possibly even the same set of considerations) about reasons
which will, at least to begin with, turn other people off. Let
Relative and A bsolute Values 57
u s turn first to the considerations which might engender a
favourable response.
A central view of Foot's (which might indeed have an
extension beyond the realm of the concept of ought, so as
to apply to a larger range of valuations) is that the primary
function (though not its invariable function) of the use of
an " ought" statement is to produce, or to state the
existence of, a reason for a potential agent to perform
some specified action or to occupy some specified position
or situation. She would go on to say, I think, that it has
been, at least since Hume, a commonplace of philosophy
that the existence for someone of a reason to perform an
action or to occupy a position or situation depends on his
having some desire, interest, or disposition of will pointing
in that direction; it is objectionable to suppose that there
are any features the mere recognition of which is sufficient
to provide one with a reason for doing something. The
objectionableness of such a supposition may be of either of
two kinds; the supposition may be disbelievable, or
repugnant to the intellect or judgement; or it may be
distasteful, or repugnant tO the will or to inclination. A
subsidiary argument of Foot's is, I think, one which would
represent the idea that morality consists in a system of
categorical imperatives as distasteful, indeed morally
distasteful ; or at least as less tasteful than the more
Humean alternative. We would rather (she suggests) be
able to think of people as volunteers in moral service, than
be forced to think of them as conscripts, as the more
1
Kantian position would entail.
1 [The following remarks were included as an aside in the manuscript o f
the lecture: ] The kind of moral (or more o r less moral) distaste t o which
Foot briefly alludes is one which I feel that, as someone brought up in the
enlightened 'pinko' (at least on the surface) atmosphere of Oxford, as it used
to be, I understand very well. We are in reaction against our Victorian
forebears; we are independent and we are tolerant of the independence of
others, unless they go too far. We don't like discipline, rules (except for rules
of games and rules designed to secure peace and quiet in Colleges), self
conscious authority, and lectures or reproaches about conduct (which are
usually ineffective anyway, since those whom they are supposed to influence
Now Foot (an old friend of mine) told me on one of the
more recent occasions when we discussed these questions
that she had not intended to attach very much weight to
her mot about 'volunteers and conscripts'. If this is so,
then I think that in one pretty important respect she was
doing herself an injustice. For whether or not it in fact
succeeds on this occasion, it is very much the right kind of
consideration to bring to bear. In the case of some sorts of
valuation, the apparatus for determining whether some
particular target should be accorded favourable or un
favourable valuation cannot sensibly be turned upon itself;
we cannot sensibly ask whether the apparatus for deter
mining pictorial valuation, or our use of this apparatus, is
pronounced by the apparatus itself to be worthy of
favourable pictorial valuation, since neither the apparatus
nor our use of it is a picture. We can ask whether the
standards (so far as we can identify them) applied in
determining whether something is funny, or our applica
tions of these standards, are themselves licensed as being
funny by those very standards : but I very much doubt
whether an affirmative answer would be regarded as a
significant endorsement of those standards . In other
cases-perhaps, for example, with regard to standards of
utility-it may well be that a certificate of conformity to
these standards, received by the standards themselves or by
our use of them, would be properly regarded as an
endorsement of the standards. But if, as is the case with
moral standards, the standards (in advance of any precise
determination of their value) are thought to be paramount
are usually either too sensitive or not sensitive e nough). A bove all we dislike
punishment, which only too often j ust plays into the hands of those who are
arrogant or vindictive . We don 't much care to talk about " values "
(pompous) or "duties " (stuffy, unless one means the duties of servants or the
military, or money extorted by the customs people). Our watchwords (if we
could be moved to utter them) would be 'Live and let live, though not
necessarily with me around ' or ' If you don't like how I carry on, you don't
have to spend time with me. ' With these underlying attitudes, it is not
surprising that we don't find Kant congenial, and that we do very much l ike
Strawson's Freedom and Resentment.
(not susceptible to being overridden) and i t i s also the case
that the standards endorse themselves or our use of them,
then it might, I think, be plausibly suggested that such an
endorsement is specially powerful, to the extent that its
availability might be taken as a relevant interpretation of
the notion of objectivity. And an ethical system which
failed this test would not have much to hope for beyond a
decent burial.
What seems to me wrong with Foot's procedure at this
point is not that it is an unsuitable procedure for
producing rabbits from hats (it is indeed quite suitable) ,
but that on this occasion it does not produce any rabbits.
As one of my colleagues at Seattle (David Keyt) remarked,
once you are in one of the services it does not matter
whether you are one of the volunteers or one of the
conscripts : both are treated alike, and indeed, virtually no
one knows which you are. The fact that a consideration is
motivating independently of any desire one may have does
not imply as a matter either of physical or logical necessity
that one in fact acts in line with it; as Kant and others have
observed, it is only too obvious that all too often one does
not act in line with it. One is not compelled or constrained
unless by " constrained" /" compelled" is meant " rationally
constrained"/" compelled " -and, as Kant suggested, maybe
that kind of constraint/compulsion is just what the doctor
ordered for the free man.
I turn now to the specification of an attempt to represent
the position of Foot's opponent, a champion of 'the
received view' which allows a viable distinction between
categorical and hypothetical imperatives and seeks to
associate moral valuation with categorical imperatives, not
as distasteful but as disbelievable. What I have to say has
an obvious relation to questions which students in ethics
are ready to pose within their first week of classes about
whether and how it is possible to justify ends. I am not sure
that the considerations with which I shall be concerned are
actually voiced by Foot; but that she would assent to them
seems to me to be strongly indicated by her view that
reasons have to be constituted as such by their connection
with desire or interest, and by her refusal (explicitly
avowed to me) to countenance such questions as whether,
independently of any connection with actual desires, a
person should have recognized as a reason something
which he did not so recognize. I shall take a little trouble to
exhibit clearly the structure of the present issue.
It might be suggested that an adherent of the received
view is likely to have a certain picture of practical
reasoning which is, perhaps, redolent of Aristotle. We
decide on the performance of a particular action by
ascribing to it a certain value, which is inherited from some
state of affairs to which the action would be conducive ; the
inherited value will be recognized to have descended
through a sequence of inheritances, starting from some
item whose value is not inherited but original. This picture
raises at once hoary problems about how the original value
comes to be there and how it comes to be detected. One
who adheres to Foot's scheme, however, can lay claim to a
capacity to solve or to bypass these difficulties. It is a
mistake, he can say, to think of practical reasoning as
recognizing the transmission of an original non-relativized
value down a chain of inheritors : what we start with is a
relativized value (relativized to some person or potential
agent) , and it is this value which is (sometimes) transmitted.
So the question of j ustifying ends, otherwise than by
showing them to be actually desired, does not arise.
The legitimacy of a conception of absolute value, of a
kind being denied by the suppositions adherent to Foot's
view, is a main topic of my third lecture, and so what I say
here should be regarded as having a fairly limited aim. It is
designed only to show, or suggest, that should it turn out
to be theoretically desirable to be able to regard absolute
(non-relative) value as attaching to some ends, we should
not be at a loss when it comes to saying how such absolute
value is to be detected, or how rational decision about ends
is possible. While I would not claim to be in a position to
give a tidy, comprehensive theoretical account of the
matter, it seems to me pretty clear that ordinary agents are
thoroughly practised at end-selection. (At this point I draw
heavily upon a paper on Happiness which I am prone to
deliver, wholly or in part, at every possible opportunity . )
B y way of preliminary, three general points seem
appropriate.
1. Ends go around in packs or systems ; so in deter
mining the suitable ends two linked considerations come
into play: the suitability of the end considered as an
individual, and also its suitability when it is considered as a
member of an actual or potential system of ends (whether
in this or that way it does or would fit in).
2. Alterations in and institutions of systems of ends
normally occur as the outcome of revision ; system S' is
substituted for system S which is previously ensconced,
and what S' and S have in common is much more extensive
than the respects in which they differ. As with clothes,
changes are mostly matters of patching; and where a new
suit is acquired, it is usually ready-made by professional
purveyors, like churches or political parties, or private
persons like spouses.
3 . It is (fortunately for us) possible to make revisions in
our system without having to articulate (which we almost
certainly would be unable to do) the contents of the
system. We can consider a possible change and see what
comes to mind, one way or the other, about such a change.
Systems in situ seem (not very surprisingly) to be very
much like the human beings in whom they are situated.
Both change, but in the normal course of events not
usually very rapidly: and when changes occur they tend to
occur according to natural laws or trends : systems and
people grow and develop and sometimes even decay. So
when we look for the properties which commend systems,
we find them to be not unlike those aspects of stability
which commend systems of beliefs ; which according to
Idealists (of the Oxford kind, Bradley, etc.) are such things
as coherence, consistency, and comprehensiveness; and it
is systems deficient in such respects as these which get
modified. Systems which are harmonious, in that the
realization of or pursuit of some elements enhances the
prospects for other elements, are favoured. So are systems
which are (so to speak) teleologically suitable, which bring
into play more fully rather than less fully the capacities and
attributes which are central to one's constitution as a
human being. So, again, are systems which are flexible,
which allow for easy and untraumatic revision where
revision is required.
When we turn to a consideration of individual ends, we
find a variety of procedures which we use to assess the
suitability or unsuitability of suggested or possible ends,
some of which can also be applied to the assessment of
systems of ends. Sometimes we ask whether the adoption
of such and such an end would put us at the mercy of
circumstances beyond our control ; to what extent we
should need what Aristotle called "ektos choregia" like
government grants. Sometimes we enquire about the likely
durability of an actual or suggested interest: 'Would we get
tired of it?', 'How long would we be capable of sustaining
it?', etc. Sometimes we see whether we or someone else can
present us with a favourable (or unfavourable) 'picture' of
life with such and such as one of our ends. Sometimes we
raise second-order questions about the desirability, of one
sort or another, of our having some specified item as an
end ( 'Could I be talked into it?', 'Would it be a useful
interest to have ?', 'Would I look ridiculous if I went in for
that?', etc.) .
The purpose o f this lightning tour o f methods o f end
assessment has not been to present a systematic account of
them, though that would fill a need; it has been intended
merely to indicate that so far from being at a loss when it
comes to the assessment of ends, we seem to have a wealth
of resources at our disposal; so the suggestion that Foot's
position has the advantage of enabling us to dispense with
such assessment would be to try to pull us out of a hole
which we are not in. But there is a further question,
namely, whether the methods which we do use for such
assessment are more in tune with Foot's position or with a
Kantian position. Here I find the outcome not at all clear.
It is not at all clear to me how the criteria which we seem
to apply in the assessment of ends, and of attachments to
them, are to be justified, or even whether they are to be
justified; and our employment of some of them seems
somewhat fluctuating (for example, durability of an
interest as something solid (and so good), or as stolid and
so not good) . It might turn out that though we evaluate
ends, we do not evaluate the criteria by which we evaluate
ends ; and that might favour Foot. But who knows ?
I turn now to a brief delineation of two aspects of Foot's
position which seem to have some tendency to make things
difficult for her. The first was vividly presented in a talk
given by Judy Baker, the relevant passage from which I
shall summarize. If I say to you that the door is closed,
standardly my purpose in saying this to you is to get you to
believe that the door is closed ; there are variant descriptions
which apply to some cases, like reminding you that the
door is closed, letting you know that I am aware that
the door is closed, and so on. But it is natural to think of the
arousal of a belief as the central case. When I utter a
(grammatical) imperative, there is more than one thing I
may be doing; if, as a friend watching you shiver, I say
'Close the door' (in a gentle tone of voice, perhaps) , I could
be advising you to shut the door (since you are cold) : if, as
a parent to a child, I say 'Shut the door', I might be telling
you to shut the door. There are further distinctions which
might be made ; for example, we can distinguish between
an officer saying to a private soldier, 'Fetch the provisions',
when he would be ordering the soldier to fetch the
provisions, and one private soldier saying to another,
'Fetch the provisions', when he might be relaying an order
to fetch the provisions. There will be at least two (maybe
more) main families of operations, telling and advising,
which will each be further differentiated. All of this is
evident to common sense. Now a modalized imperative,
like 'You ought to visit your aunt' or 'You mustn't touch
the flowers' is perhaps not strictly a recipient of the
classifications applicable to unmodalized imperatives, but
it is plausible to suppose that assimilation of the modalized
imperatives to membership of one or another of the
families of imperatives is possible. Now Foot's position
seems to demand that moral judgements (valuation,
exhortation) should be assimilated to the advising family.
But this is repugnant to common sense; common sense
would support an assimilation to the telling family,
particularly perhaps to the sub-family of relaying orders
(in the moral case, from an unspecified and perhaps even
unspecifiable source) .
Difficulty for Foot may also arise ( a s Judy Baker also
suggested) from her treatment of " non-hypothetical uses" ,
which seem to b e represented a s " decayed/degenerate
cases " of utterances of " ought"-statements (and such
like), which in their primary and non-degenerate employ
ment are dependent on the possession of a certain desire or
interest on the part of the addressee or potential agent, but
which are here used even though the speaker may not
attribute to his addressee such an interest. Examples are
etiquette and club rules. The steward may say to a visitor,
'You may not bring ladies into the smoking-room', even
though it is obvious that the visitor does not give a fig for
the club or its rules. The steward may even say, 'I know
you don't care about our rules, but you may not bring
ladies into the smoking-room.' The difficulty for Foot is
alleged to consist in the fact that we are told that though
these are cases in which the original or normal dependence
of such utterances on a potential agent's concern or
interest has been lost, such injunctions are nevertheless still
voiced, perhaps in one or another version of a social
routine ; we have perhaps got used to saying such things. It
is suggested that Foot has not succeeded in making such
utterances understandable, from the point of view of the
utterer. In particular, while the establishment of social
routines or practices is not mysterious, the execution of
them has to be thoroughgoing. There would be some
inconsistency of behaviour in including in a routine
occurrence of a statement that ladies may not be brought
into the smoking-room, a remark to the effect that the
addressee does not have the normally requisite interest,
which would be an open admission that what is taking
place is only the operation of a routine or pretence.
Admittedly these objections only show that Foot's
position is counter-intuitive, is against what people
ordinarily suppose to be the case. It might be none the
worse for that. But most of the time Foot seems to want to
present herself as coming to the aid of the vulgar valuer,
the maker of ordinary moral judgements, in order to
protect him against the attempts of the philosophers to
read into vulgar valuations material which is not there. It
would be uncomfortable to her to have to take the position
of condemning (philosophically) what according to her is
there.
A further difficulty for Foot may arise from this fact that
she seems to me to be liable to a charge of having failed to
distinguish two different interpretations of phrases of the
form "has a reason to " and " has no reason to (for) " , and
of putting forward a thesis about moral judgements, that
they are (or should be) hypothetical imperatives, the
attraction of which depends on a failure to make this
distinction. Suppose that an old lady is struggling up the
stairs with a mass of parcels, that I see her, and that I am
young, able-bodied, and in no particular hurry. I could go
and help her to cope, but I do not, because, as I would say,
'What is there in it for me ? ' I don't care about the minor
distresses of old people, and I don't see any likelihood that
I would be rewarded for helping her or penalized for not
doing so. On one reading of " have a reason to " it may be
that these facts are sufficient to ensure that I have no
reason to help her. But helping such people in such
circumstances is in fact a matter of ordinary decency and
so something we should do. On another reading this may
be sufficient to ensure that I have a reason to help her;
there is a reason for me to help her (its being a matter of
ordinary decency), whether or not I recognize the fact; so I
do have a reason to help her. The charge against Foot
would be that the attractiveness of her case for supposing
moral imperatives to be hypothetical depends first on
equating, perhaps correctly, the application of a moral
imperative to a person with his having a reason (perhaps a
particular kind of reason) for acting in the prescribed way,
and second on attaching to the phrase "having a reason"
the first of the two interpretations j ust distinguished (that
in which the man who exhibits indifference towards a
given line of action would thereby be shown to lack a
reason for such action), thus failing to notice or ignoring
the second interpretation, which is the only reading which
intuition would allow as adequate for the explication of
morality.
Foot might say that she is not confusing the two
readings but denying that there is more than the first
reading. But she would have to argue for this contention,
and it would (I think) be difficult to argue for it in a non
circular way.
I shall conclude this lecture with a brief interim
statement, quite undocumented, about where we are and
what I might expect to find myself trying to do about it. It
seems to me that a whole lot of the trouble that has arisen
for Mackie and for Foot has come out of the reluctance of
each of them, in this or that degree, to allow full weight to
the idea of value as making a bridge between the world of
fact and the world of action. It is my suspicion (at the
present point no more than a suspicion) that to get our
heads clear and keep them clear we shall need to do (at
least) five things :
1 . To pay unrelenting attention to the intimate connec
tion between reason (the faculty) and reasons, and
the intimate association of both with argument and
value.
2. To allow for parity, at least in a rational being, of
cognition and practical faculties; each is equally
guided by reason (rational will) , and each alike
guides reason (rational will) .
3 . To take really seriously a distinction between
rational and pre-rational states and capacities, with
unremitting attention to the various relations
between the two domains.
4. To recognize value as embedded, in some way yet to
be precisely determined, in the concept of a Rational
Being. Value does not somehow or another get in, it is
there from the start.
5. To realize, as one of the fundamental and urgent
tasks of philosophy, the need to reach an under
standing of the way in which the world ( 'pheno
menal') viewed in terms of cause and effect, and the
world ('noumenal') viewed in terms of reasons, fit
together (a classical version of the Problem of
Freedom) .
Perhaps we might, i n tomorrow's lecture, move a little way
in one or two of these directions.
Conception of Value
3. M E T A PHY S I C S A N D V A L U E 1
The main object of this lecture will be to explore the

possibility of providing some kind of metaphysical account
of, and positive backing for, the notion of value; and to
begin it, I think it would be appropriate for me to voice
one or two methodological reflections. But George Myro
has told you a story about me, which has indeed the ring of
truth; so I shall pause to tell you another story which I
hope will ring equally true. When I was still quite a little
nipper, my mother gave me a china beer-mug, which I
liked enormously; in fact I used it not only to drink beer
from, but for a whole lot of other purposes, like carrying
gasoline (or as we used to call it, 'petrol') to a neighbour's
house when I wanted to set the place on fire. One day I
dropped it, and there it lay in fragments. Just at that point
my mother appeared on the scene, and reproached me for
destroying the mug which she had given me and which I so
dearly loved. I said to her, 'Mother [in my family we used
to address one another in somewhat formal style] , mother,
I haven't destroyed anything; I have only rearranged it a
little.' She, being too wise, and insufficiently nimble, to
take me on in dialectical jousting, looked at me sadly and
said, 'You will have to learn as you grow older.' This
1 In this lecture I am especially indebted to Richard Warner, without
whose timely, skilful, and patient piloting the lecture, and so the series
containing it, would not have been written.
encounter, it now seems to me, shaped my later philo
sophical life; it seems that I did, indeed, learn in the end.
For first, the metaphysical programme which I shall be
seeking to follow will be a constructivist programme and
not a reductionist programme. The procedure which I
envisage, if carried out in full, would involve beginning
with certain elements which would have a claim to be
thought of as metaphysically primary, and then to build up
from these starting-points, stage by stage, a systematic
metaphysical theory or concatenation of theories. It would
be no part of my plan to contend that what we end up with
'is really only such and suches', or that talking about what
my enterprises produce is really only a compressed way of
talking about the primary materials, or indeed to make any
other claim of that relatively familiar kind. I suspect that
many of those who have thought of themselves, like
Carnap, as engaged in a programme of construction are
really reductionists in spirit and at heart; they would like
to be able to show that the multifarious world to which we
belong reduces in the end to a host of complexes of simple
ingredients. That is not my programme at all; I do not
want to make the elaborate furniture of the world dissolve
into a number of simple pieces of kitchenware; I hope to
preserve it in all its richness. I would seek, rather, to
understand the metaphysical processes by which one
arrives at such richness from relatively simple points of
departure. I would not seek to exhibit anything as 'boiling
down' to any complex of fundamental atoms.
In order to pursue a constructivist programme of the
kind which I have in mind, I will need three things : first, a
set of metaphysical starting-points, things which are
metaphysically primary ; second, a set of recognized
construction routines or procedures, by means of which
non-primary items are built up on the basis of more
primary items ;. and third, a theoretical motivation for
proceeding from any given stage to a further stage, so that
the mere possibility of applying the routines would not be
Metaphysics and Value 71
itself enough to give one a new metaphysical layer; one
would have to have a justification for making that move; it
would have to serve some purpose.
The way I imagine myself carrying out this programme
in detail (on some day quite a bit longer than today) is
roughly as follows. First, I would start by trying to reach a
full-dress characterization of what a theory is ( an exercise
in what I think of as theory-theory); I would be hoping
that the specification of what theorizing is would lead in a
non-arbitrary way to the identification of some particular
kind of theorizing (or theory) as being, relative to all other
kinds of theorizing, primary and so deserving of the title of
First Theory (or First Philosophy) . I would expect this
primary theorizing to be recognizable as metaphysical
theorizing, with the result that a specification of the
character and content of metaphysics would be reached in a
more systematic way than by just considering whether
some suggested account of metaphysics succeeds in fitting
our intuitive conception of that discipline. The implement
ation of this kind of metaphysical programme would, I
hope, lead one successively through a series of entities
(entity-types), such as a series containing at one stage
particulars, followed by continuants, followed by a
specially privileged kind of continuants, namely substances,
and so on. Each newly introduced entity-type would carry
with it a segment of theory which would supplement the
body of theory already arrived at, and which would serve
to exhibit the central character of the type or types of
entity associated with it.
The application of these programmatic ideas to the
determination of the conception of value would be
achieved in the following way. The notion of value, or of
some specially important or fundamental kind of value,
like absolute value, would be shown as occupying some
indispensable position in the specification of some stage in
this process of metaphysical evolution. Such a meta
physical justification of the notion of value might perhaps
be comparable to the result of appending, in a suitably
integrated way, the Nicomachean Ethics as a concluding
stage to the De Anima. One would first set up a
specification of a series of increasingly complex creatures;
and one would then exhibit the notion of value as entering
essentially into the theory attending the last and most
complex type of creature appearing in that series. The un
Carnapian character of my constructivism would perhaps
be evidenced by my idea that to insist with respect to each
stage in metaphysical development upon the need for
theoretical justification might carry with it the thought
that to omit such a stage would be to fail to do justice to
some legitimate metaphysical demand.
I propose to start today's extract from this metaphysical
story at a point at which, to a previously generated stock
of particulars which would include things which some,
though not all, would be prepared to count as individual
substances, there is added a sequence of increasingly
complex items, which (as living things) would be thought
of by some (perhaps by Aristotle) as the earliest items on
the ascending metaphysical ladder to merit the title of
substances proper. My first metaphysical objective, at this
stage of my unfolding story, would be to suggest a
consideration of the not unappealing idea that the notion
of living things presupposes, and cannot be understood
without an understanding of, the notion(s) of purpose,
finality, and final cause. In my view this idea is not only
. appealing but correct; I recognize, however, that there are
persons with regrettable deflationary tendencies who
would insist that while reference to the notions of finality
or final cause may provide in many contexts a useful and
illuminating manner of talking about living things, these
notions are not to be taken seriously in the metaphysics of
biology and are not required in a theoretical account of the
nature of life and of living things. A finalistic (or
" vitalistic") account of the nature of living things might
take as a ruling idea (perhaps open to non-vitalists as well)
that life consists (very roughly) in the possession of a no
doubt interwoven set of capacities the fulfilment, in some
degree, of each of which is required for the set of
capacities, as a whole, to be retained; a sufficiently serious
failure in respect of any one capacity will result in the
currently irreversible loss of all the rest; and that would be,
as one might say, death. The notion of finality might be
thought to be unavoidably embedded in the notion of life
for more than one reason. One reason would be that if it is
in one way or another of the essence of living creatures
that one has, instead of an indefinitely extended individual
thing, an indefinitely long sequence of living things, each
individual being produced by, and out of, predecessors in
the sequence, then to avoid having individuals which are of
outrageous bulk because they contain within themselves
the actual bodies of all their descendants, it will be
necessary to introduce the institutions of growth and
maturity; and with these institutions will come finality,
since the states reached in the course of growth and
maturity will have to be states to which the creatures
aspire and strive, though not necessarily in any conscious
way. Another way in which it might be suggested that the
notion of finality has to enter will be that at least in a
creature of any degree of complexity, the discharge of its
vital functions will have to be effected by the operations of
various organs or parts, or combinations of such; and each
of these organs or parts will have, so to speak, its job to do,
and indeed its status as a part (a working functional part,
that is to say, and not merely a spatial piece) is determined
by its being something which has such-and-such a job or
function (eyes are things to see with, feet to walk on, and
so forth); and these jobs or functions have to be distin
guished by their relation to some feature of the organism
as a whole, most obviously to such things as its continued
existence. The organism's continuance, though not ordin
arily, perhaps, called a function of the organism, will
nevertheless be required as something which the organism
strives for, in order that we should be able to account for
the nature of the parts as parts ; it will be that thing, or one
of the things, to which in their characteristic ways the parts
are supposed to contribute. It is worth noting, with regard
to the first of these reasons for the appearance of finality
on the scene, that for the idea of actual containment of a
creature within its forebears there is substituted the idea of
potential containment, an idea which can be extended
backwards (so to speak) without any attendant inflation of
the bodies of ancestral creatures.
Perhaps I might at this point make two marginal
comments on the scheme which I am proposing. First, if I
allow myself in discussing the notion of life to ascribe
purpose or finality to creatures, parts of creatures, or
operations of creatures, such purposes or finalities are to
be thought of as detached from any purposers, from any
creature or being, mundane or celestial, which consciously
or unconsciously harbours that purpose or finality. If the
walrus or the walrus's moustache has a purpose, that
purpose, though it would be the purpose of the walrus, or
of its moustache, would not be the walrus's purpose, its
moustache's purpose, or even God's purpose. A failure to
appreciate this point has been, I think, responsible for
some of the disrepute into which serious application,
within the philosophy of biology, of the concept of finality
has fallen. Second, if one relies on finality as a source of
explanation (if one does not, what is the point of appealing
to it? ) , if indeed one wishes to use as modes of explanation
all, or even more than one, of Aristotle's " Four Causes " ,
one should not b e taken to b e supposing that there i s a
single form of request for explanation, a proper response
to which will, from occasion to occasion, be now of this
type and now of that type, and maybe sometimes of more
than one type. It is not that there would be just one kind of
"Why ? " question with alternative and possibly at times
even rival kinds of answer; there would be several kinds of
"Why ? " question, different kinds of question being perhaps
linked to categorially different kinds of candidates for
explanation, and each kind calling for its own kind of
'cause', its own type of explanation. And it might even be
that, so far from being rivals, different types of 'cause'
worked together, or one through another; it might be, for
example, that the operation of final causes demanded and
was only made possible by the operation, say, of efficient
or material causes with respect to a suitably linked
explicandum; the success of a " finality " explanation to a
question asking why a certain organ is present in a certain
kind of organism might depend upon the availability of a
suitable "efficient cause" explanation of how that organ
came to be present.
Nevertheless, these assuaging and anodyne representa
tions are, I fear, unlikely to appease a dyed-in-the-wool
mechanist. Such a one would be able to elicit from any
moderately sensible vitalist an admission that the mere fact
that the presence of a certain sort of feature or capacity
would be advantageous to a certain type of creature offers
no guarantee that the operation of efficient causes would
in fact provide for the presence of that feature or capacity
in that type of creature. So a vitalist would have to admit
that a " finality" explanation would be non-predictive in
character; and at this point the mechanist can be more or
less counted on to respond that a non-predictive explana
tion is not really an explanation at all. While a mechanist
might be prepared to allow that survival is a consequence
(in part) of those features which, as one might say, render a
type of creature fit to survive, and perhaps also allow that
this consequence is a beneficial consequence, or pay-off,
arising from the presence of those features, he would not
be ready to concede that the creature comes to have those
features in order that it should survive. To refine the terms
of this discussion a little: one might try to distinguish
between the question why a certain feature is present, and
the question why (or how) that feature comes to be
present; the mechanistically minded philosopher might be
represented as doubting whether the first question is
intelligible if it is supposed to be distinct from the second,
and as being ready to maintain that if, after all, the two
questions are distinct and are both legitimate, the provision
of an answer to the first cannot be held to require the
existence of an answer to the second, and indeed is
possible at all only given independent information that
that feature has come to be present.
Furthermore, in respect of other areas where a vitalist
(or finalist) is liable to invoke finality as a tool for
explanation, the mechanist will say that one can quite
adequately explain the phenomena which lead the vitalist
to appeal to finality, without having recourse to such an
appeal; one will use a particular kind of explanation by
efficient causes, one which deploys such cybernetic notions
as negative feedback and homoeostasis; as one mechanistic
philosopher, David Armstrong, has suggested, people will
be like guided missiles.
At this the vitalist might argue that to demand the
presence of biological explanation as embodying finalistic
apparatus casts no disrespect on the capacity of "prior"
sciences (such as physics and chemistry) in a certain sense
to explain everything. In a certain sense I can explain why
there are seventeen people in the market-place, if in respect
of each person who is in the market-place I can explain
why he is there, and I can also explain why anyone else
who might have been in the market-place is in fact
somewhere else. But there is an understandable sense in
which to do all that does not explain, or at least does not
explain directly, why there are seventeen people in the
market-place. To fill the " explanation-gap " , to explain not
just indirectly but directly the presence of seventeen people
in the market-place, to explain it qua being the presence of
seventeen people in the market-place, I might have to
introduce a new theory, perhaps some highly dubitable
branch of social psychology; and I suspect that some
scientific theories, like perhaps catastrophe-theory, have
arisen in much this kind of way on the backs (so to speak)
of perfectly adequate, though not omnipotent, pre-existing
theories, in order to explain, in a stronger sense of
"explain " , things which are already explained in some
sense by existing theory.
To this, the mechanist might reply that there is (was)
indeed good reason for us to strengthen our explanatory
potentialities by adding to the science(s) of physics and
chemistry the explanatory apparatus of biology, thus
giving ourselves the power to explain directly rather than
merely indirectly such phenomena as those of animal
behaviour; but it does not follow from that that the
apparatus brought to bear by biology should include any
finalistic concepts or explanations; explanations given in
biological terms are perfectly capable of being understood
in cybernetic terms, without any appeal to concepts whose
respectability is suspect or in question.
In response to this latest tiresome intervention by the
mechanist, I now introduce a reference to something which
is, I think, a central feature of the procedures leading to the
development of a cumulative succession of theories or
theory-stages, each of which is to contain its predecessor.
This is the appearance of what I will call overlaps. In
setting out some theory or theory-stage B, which is to
succeed and include theory or theory-stage A, it may be
that one introduces some theoretical apparatus which
provides one with a redescription of a certain part of
theory A. I may, for instance, in developing arithmetic,
introduce the concept of positive and negative integers;
and if I were to restrict myself to that part of the domain of
this 'new' concept which involves solely positive integers, I
would have a class of formula: each of which provides a
redescription of what is said by a formula relating to
natural numbers; and since the older way of talking (or
writing) would be that much more economical, if it came
to a fight the natural numbers would win; the innovation
would be otiose, since theorems relating solely to positive
integers would precisely mirror already available theorems
about natural numbers. But of course, to attend in a
myopic or blinkered way only to positive integers would
be to ignore the whole point of the introduction of the
'new' class of positive and negative integers, which as a
whole provides for a larger domain than the domain of
natural numbers for a single battery of arithmetical
operations, and so for a larger range of specific arithmetical
laws. That is the ratio essendi of the newly introduced
integers, to extend the range of these arithmetical opera
tions. Somewhat similarly, if one paid attention only to
examples of universal statements which related to finite
classes of objects, one might reasonably suppose the
expressions of universal statements to be equivalent to a
conjunction of, or to be a 'compendium' of, a set of
singular statements ; and indeed, at times philosophers
have been led by this idea, notably in the case of Mill. But
of course, at least part of the point of introducing universal
statements is to get beyond the stage at which one is
restricted to finite classes, and therefore beyond a stage at
which one can regard universal statements as being simply
compendia of singular statements.
In the present connection, one might suggest that
discourse about detached finality might be regarded as
belonging to a system which is an extension or enlarge
ment of a system which confines itself to a discussion of
causal connections (efficient or material), including the
special kind of causal connections with wh ich cybernetics
is concerned ; and that this being so, not only is it not
surprising but it is positively required, for the extension to
be successfully instituted, that finality should enjoy an
overlap with some special form of causal connection, like
that which is the stock-in-trade of cybernetics. To test the
relation between these two conceptual forms ( finality and
" cybernetical " causal connection) , it will be necessary to
see whether there is an area beyond the overlap, in which
talk about finality is no longer mappable onto talk about
caual connection; and the place where it seems to me that
it would be natural to look for this divorce to occur, if it
occurs at all, would be in the area to which attributions of
absolute value belong, and in which talk about finality
goes along with attributions of absolute value. If one then
denies, as Foot and others have done, that there is any
metaphysical region in which absolute value is a lawful
resident, one seems to be cutting oneself off, maybe not
without good reason, from j ust that testing-ground which
is needed to determine the conceptual relationship between
finality and causal connection. To this point I shall return.
As I am talking, in this lecture, about finality, I am
always referring to what I am calling detached finality,
that is, to purposes which are detached from any purposer,
purposes which can exist without there being any conscious
being who has, as his purpose, whatever the content of
those purposes may be. Now there are some distinctions in
relation to finality which I want to make. First, it may be
either an essential property of something or an accidental
property of it that it has a certain finality. My conception
of essential property is of properties which are defining
properties of a certain sort or kind and also, at one and the
same time, intimately bound up with the identity conditions
for entities which belong to that kind. [It might indeed be
taken as a criterion of a kind's being a substantial kind that
its defining properties should enter into the identity
conditions for its members.] The essential properties of a
thing are properties which that thing cannot lose without
ceasing to exist (if you like, ceasing to be identical with
itself).
It is clear that the essential properties of a sort are not to
be identified with the necessary properties of that sort, the
properties which things of that sort must have; for those
properties whose presence is guaranteed by logical or
metaphysical necessity, given the presence of the essential
properties, would not thereby be constituted as essential
properties of the sort, that is, as properties which are
constitutive of the sort. Moreover, attention to the words
of Terry Irwin and Alan Code leads me to suspect that in
the case of sorts of living things, properties which are
essential to a sort might not be invariably present in
instances of the sort, and so might fail to be necessary
properties of that sort. This possibility, if it is realized,
would arise from the fact that membership in a vital sort
may be conferred not (or not merely) by character but by
ancestry ; so a freakish or degenerate instance of such a sort
might even lack some essential feature of the sort, provided
that its parents or suitably proximate ancestors exhibited
that feature.
The link between the two strands in the idea of essential
property, that of being definitive of a kind and that of
constituting an identity condition for members of that
kind, becomes eminently intelligible if one takes Aristotle's
view that to exist and to be a member of a certain kind are
one and the same thing (with appropriate consequences
about the multiplicity contained within the notion of
being) . Now my idea, and I think also the idea of Aristotle,
is that the range of essential properties of this or that kind
of thing would sometimes, or perhaps even always, include
what I might call finality properties, that is, properties
which consist in the possession of a certain detached
finality.
A second distinction which I want to make is between
active and passive finality, a distinction between what it is
that, as it were, certain things are supposed to do, and (on
the other hand) what it is that certain things are supposed
to suffer, have done to them, have done with them, and
such-like, including particularly uses to which they are
supposed to be put. One might, as a bit of j argon, label the
active kind of finality property as metiers or roles. It would
of course be possible, within the area of active finality, to
allow for different versions of activity : the activity of a
tiger, for example, while legitimately so-called, might be,
as one would be inclined to say, activity in a different
'sense o f the word' from the activity of a scholar. My idea

would be that every sort of creature, and every individual
belonging to any such sort, must, in virtue of the fact that
it is a sort of living creatures, or a sort of living creature,
possess as an essential property an active finality. To be a
tiger or to be a human being is to possess as an essential
property the capacity to tigerize or the capacity to
humanize, in whatever those things may be thought to
consist.
I will now try to connect up the material which I have
been very sketchily presenting, with the range or corpus of
metaphysical routines or operations which it will be
proper for a metaphysician to deploy in the course of what
I would call the metaphysical evolution of entities or types
of entity. One of these manoeuvres will be, I think, a
manoeuvre which I call Metaphysical Transubstantiation.
There might well be specifically variant forms which this
operation might take, but the central idea would be that
you might have two entity-types (secondary substances)
such that it is perfectly possible for a particular instance of
one type to have exactly the same properties as a particular
instance of another, for them to be in fact indiscernible;
the selection of properties, however, from that total set
which would be essential to an individual entity qua
member of one substantial type would not be the same
selection as the selection which would be essential to it qua
member of the second type. Therefore it would be a
possibility that something which at one time exhibited the
essential properties of both types should at another, later
time exhibit the essential properties of only one of these
types. There will be an S 1 which existed both at time t 1 and
at time t2 , and an S2 which existed at time t 1 when it was
identical with the aforementioned S 1 ; but at time t2 the S 2
no longer exists and so of course is not then identical with
the S1 • It is this kind of mind-twister which lies at the heart
of Hobbes's problem about the ship and the timber, and
also lies at the centre of what I might call the Grice-Myro
theory of identity, which would allow a thing x and a thing
y to be identical at a certain time but to be not identical at
a different time, when indeed one of the things may have
ceased to exist. Execution of the manoeuvre of Meta
physical Transubstantiation would consist in taking a
certain sort of substance S1 , to which a certain property or
set of properties P would be essential, and then introducing
a type of substance S2 , an instance of which may indeed
possess property or properties P, but if it does, does not
possess them essentially. What will be essential to S2 will
be some other set of properties P ' , properties which even
might attach, though not essentially, to some, or even to
all, instances of S 1 • I am not sure what examples of this
manoeuvre are to be found, but the one which I have it in
mind to make use of is one in which the manoeuvre is
employed to erect, on the basis of the substance-type
human being, a further substance-type person. It is
imperative at this point to remember that the general
principles of entity-construction, which I adumbrated at
the start of this lecture, will dictate that if we are to
suppose the manoeuvre to be executed in order to
generate, in this way, the substance-type person, we must
be able to specify an adequate theoretical motivation for
the enterprise in which the manoeuvre is employed.
The time has now come to consider the introduction,
into a sequence of substantial types being designed by a
metaphysician, of the attribute of rationality, which I shall
take, when I come to it, as consisting, in the first instance,
of a concern on the part of the creature which has it that its
acceptances and perhaps (more generally) its attitudes
which belong to some specifiable particular class should be
well grounded, based on reasons, or (getting closer to the
notion of value) validated; a concern, that is, on the part of
the reason-seeker that the attitudes, positions, and accept
ances which he (voluntarily) takes up should have attached
to them certificates of value of some appropriate kind. The
creature's rationality will consist in the having of this
concern together with a capacity, t o this or that degree, to
give effect to that concern. It will be convenient to consider
the introduction of this rationality into a metaphysical
scheme in terms of the idea of "construction by a genitor" ,
2
which I used in an earlier paper. I will ask first why the
genitor should be drawn at all towards the idea of adding
rationality to any of his sequence of constructed creatures.
The answer to that question might be that what he is
engaged in constructing are the essential and non-essential
features of a sequence of biological types which have to
cope with the world and maintain themselves in being, and
that a certain range of the exercises of rationality would
have biological utility, would improve the chances of a
creature which possessed it. One would have to be careful
at this point not to go too far, and suppose, for example,
that any kind of creature in any kind of circumstances
would be biologically improved by the admixture of a dose
of rationality: gnats and mosquitoes, for example, might
be hindered rather than helped by being rationalized. But
rationality might be a biological boon to creatures whose
biological needs are complex and whose environment is
subject to considerable variation, either because the world
is unstable, or because the world though stable combines a
high degree of complexity with a reluctance to make easy
provisions for its denizens. If a creature's survival depends
on the ability to produce differing responses to a vast and
varied range of stimuli, then it will become more and more
difficult and 'expensive' to equip the creature with a
suitably enormous battery of instincts, and the substitution
of a measure of rationality will be called for.
Two questions now confront us with regard to the
genitor's introduction of rationality. First, is he to be
thought of as introducing a relatively unlimited, un
restricted capacity, a capacity perhaps for being concerned
about and for handling a general range of "Why ? "
2 [APA Address, pp. 12 1-6 1 below; see pp. 140-5 .]

questions, or, indeed, simply of questions ; where the
capacity itself is unlimited, though the genitor's interest in
it is restricted to certain applications of it? Or is he rather
to be thought of as introducing a limited capacity, a
capacity (perhaps) for being concerned about and handling
j ust a small, potentially useful range of questions (just
those which are biologically relevant) ?
The second question is, is the genitor to be thought of as
introducing whatever kind or degree of rationality he does
introduce as an essential characteristic of the substantial
type(s) which he is designing and endowing with it, which
we will think of as Homo sapiens, or as an accidental (non
essential) feature of that type ?
To answer these questions in the reverse order, I
conj ecture, though I cannot establish, that the right
procedure is to think of him as introducing rationality as a
non-essential feature of Homo sapiens, though perhaps as
a feature which, despite its being non-essential, one can be
reasonably assured that instances of Homo sapiens will
possess. My characteristic candour forces me to admit that
I should like to be able to find that the genitor would
install rationality as an accidental feature, because then
there would be scope for a profitable deployment of my
new toy, Metaphysical Transubstantiation, the application
of which by the genitor would deliver as a constructed
substantial type the type person, to which rationality could
be supposed to belong as an essential property; a result
which to my mind would be intuitively congenial. But
wishful thinking is not an argument. Perhaps an argument
could be found for attributing to the genitor the seemingly
circuitous manoeuvre of first instituting a new biological
type (Homo sapiens) to which rationality attaches non
essentially though predictably, and then converting this
new biological type into a further non-biological type
(person) by a subsequent metaphysical operation which
installs rationality as, in this case, an essential feature, by
reflecting that the programme which in the first instance is
engaging the genitor's attention is that of constructing a
sequence, or kingdom, of biological substantial types, and
that rationality is not a feature of the right kind to be a
differentiating essential feature in any type falling under
that programme. If a substantial type is needed to which
rationality attaches essentially, that type must be generated
by a further step.
With regard to the first question, whether the rationality
being introduced by the genitor is to be thought of as an
unlimited or as a limited capacity, while I think it would be
an acceptable general principle that when the genitor
wants to achieve a certain objective, of two capacities
which would achieve that objective he would (or should)
install the weaker capacity on grounds of economy, or so
that nothing which he does should be lacking in motiva
tion or deficient with respect to Sufficient Reason, this
general principle should not apply to cases in which the
weaker capacity could only be generated by initially
building a stronger capacity and then subsequently fitting
in curbs to restrict the initially installed stronger capacity.
Then the suggestion would be that the installation of a
limited rationality would be a case of the second kind,
achievable only by building in additional special restric
tions, and so would be exempted from the scope of the
proposed general principle. I think that reaching a decision
on this issue would be both rewarding and arduous; and I
must regretfully allow the matter to wait.
The way has now been more or less cleared for me to
begin to unveil the main idea lying behind this prolonged
build-up. This idea is that, given the foregoing assump
tions, when the genitor installs rationality into his new
substantial type Homo sapiens in order to further a
biological end, he gets more than he bargained for, in that
the newly installed rationality is capable of raising more
questions than that limited range of questions in the
answering of which the biological utility of that rationality
at least initially consists. The creature which the genitor
creates will not merely be capable of raising and answering
a range of questions about how certain ends are to be
achieved, of exhibiting, that is, what Aristotle called
l>ELVOtY) �, but will also have both the ability and the
requisite concern to raise questions about the desirability
or propriety of the ends or results which his rationality
enables him to realize. That is to say, the genitor has
designed a creature which is capable of asking questions
about the value of ends, and so of enquiring about the
possible availability of categorical imperatives over and
above the hypothetical imperatives which the creature was
initially scheduled to deliver. Of course, to say that the
creature has the capacity and the concern needed to raise,
and desire answers to, certain questions is not to say that
the creature is in a position to answer those questions ;
indeed, we can be sure that initially he will not be in a
position to answer those questions, since the procedures
for getting answers to them have not been designed and
installed in advance, and so will have to be evolved or
constructed, presumably by Homo sapiens himself. But
given that such a creature is equipped to formulate a
legitimate demand for solutions to these questions, then
any set of procedures which it could devise which are
such that there is no objection to them, and which if they
were accepted as proper procedures could then be used
to deliver answers to these questions, will be pro
cedures which it will be reasonable for the creature
to accept as proper, provided that there are no
other equally unobj ectionable candidates with equally
good prospects of delivering answers to the same ques
tions.
Should there indeed be two or more sets of procedures,
where each set could be put to work, with full success, to
provide an answer to a rational demand, and should it be
required, for some reason or other, that one and only one
such set should be left in the field, and should there be no
non-arbitrary way of deciding between the surviving
candidates, then I would allow i t a s rational to decide by
the toss of a coin. But I hope to avoid being faced with
having to accept so hilarious a mode of decision. I might
dub the principle which allows credence to a set of
procedures the adoption and use of which will most
satisfactorily meet a rational need, "The Metaphysical
Principle of Supply and Demand " .
The second part o f my leading idea would b e that, for
reasons which have not yet been specified, the newly
"generated" creature Homo sapiens finds it metaphysically
suitable/fitting to perform the operation of Metaphysical
Transubstantiation on, so to speak, himself, and to set it
up so that the attribution of rationality, which originally
(thanks to the genitor) attaches non-essentially to one
substantial type, namely, Homo sapiens, now attaches
essentially to a different but standardly coincident sub
stantial type, to himself as a person ; and that when we
come to consider the application of such notions as value
and finality to persons, where the notions of value in
question are not questions of relative value but rather of
absolute value, then we find that a 'paraphrase' or
'translation' of whatever it is' we have to say concerning
these notions into some rigmarole or other couched in
terms of causal connections of a cybernetic kind is no
longer available; and this would be where the end of the
overlap is located; when it comes to questions of absolute
value and about persons, the overlap has now ceased. The
reasons why the overlap should end at this point might, I
suspect, turn on the propriety of supposing persons
(essentially rational beings) to be necessarily, and perhaps
for that reason, free.
Before I fill out the attempted justification for the
application of the notion of absolute value on which I have
started, I would like to do two things. I would like first to
introduce a piece of abbreviatory j argon ; and second, I
would like to introduce, or reintroduce, a metaphysical
construction routine which I referred to in a previous
paper,1 which I called " Humean Projection " . The bit of
abbreviatory j argon is the phrase " Mechanistically Substi
tutable " ; I shall call an idea or concept " Mechanistically
Substitutable " when it is one which initially appears not to
be amenable to interpretation in terms agreeable to a
mechanist, but which is found to be, after all, so amenable.
So for an idea or concept to be not mechanistically
substitutable would be for it to resist reinterpretation by a
mechanist.
As regards " Humean Projection '' , its title is perhaps
somewhat misleading, since though some such operation
does seem to be described by Hume, he seemingly regards
it as a way of accounting for certain mistakes which we
make, of a deep-seated variety, rather than as a way of
validating some of the things which we should like to be
able to say. In this respect I think we find Mackie going
along with Hume. As I see it, this operation consists in
taking something which starts life, so to speak, as a specific
mode of thinking, and then transforming it into an
attribute which is ascribed not to thinking but to the thing
thought about and indeed is, in a given case, attributed
either correctly or incorrectly. To take an example with
which I am presently concerned, we might start with a
notion of valuing, or of (hyphenatedly, so to speak)
thinking-of-as-valuable some item x ; and, subject to the
presence of certain qualifying conditions, we should end
up with the simple thought, or belief, that the item x is
valuable; and in thinking of it as valuable, we should now
be thinking, correctly or incorrectly, that the item x has the
attribute of being valuable.
I shall now proceed to the amplification of the idea,
which I introduced a few minutes ago, of seeking to
legitimize, and to secure truth or falsity for, attributions of
absolute value by representing such attribution as needed
in order to fulfil a rational demand.
3 [APA Address pp. 12 1-6 below.]

I shall present this amplified version i n two parts. The
first part is designed to exhibit the structure of the
suggestion which is being made and will refer to, without
specifying, a certain system of hypotheses, or story, to be
called "Story S " , which is to play a central role in the
satisfaction of the aforementioned rational demand. The
second part will specify, in outline, Story S.
1 . The genitor, in legitimately constructing Homo
sapiens (an operation to which, as genitor, he is properly
motivated by a concern to optimize the spread of
biological efficiency and so of biological value ) , will have
constructed a creature which will fulfil five conditions :
(a) it will legitimately demand j ustification, ultimately
in terms of absolute and not relative value, for
whatever attitudes, purposes, or acceptances
(whether alethic or practical) it freely adopts or
maintains.
(b) it will regard any system of hypotheses, or story,
which fulfils the condition of satisfying the afore
mentioned demand ( " Demand D ") without being
itself open to objection, as being a story which is
worthy of acceptance unless this condition is met by
more than one story. (That is, it will regard such a
story as a story which is an admissible candidate for
acceptance. )
(c) i t will hold that if there i s a plurality of admissible
candidates all of which involve a certain supposi
tion, then that supposition is worthy of acceptance.
(d) it will allow that Story S is an admissible candidate
which meets Demand D and that the supposition of
the applicability to the world of the notion of
absolute value which is embodied in Story S would
also be embodied in any other story which would
satisfy Demand D, with the consequence that this
supposition is worthy of acceptance.
(e) that the notions of absolute value and of finality,
which appear within Story S, are not mechanistically
substitutable, and so are authentic and non
Pickwickian concepts.
2. I shall present Story S in the manner of one of those
cosy and comforting question-and-answer sequences
which telephone companies, public utilities, or insurance
companies are liable to give us when they are hoping to
unload on us some new gimmick.
Q 1 • How can we satisfy Demand D , and get non-relative
justification for our purposes, attitudes, and so forth ?
A 1 • By setting up conditions for the successful application
to those purposes, attitudes (etc.) of concept(s) of
absolute value.
Q 2 • How do we do that ?
A 2 • By setting it up so that certain sorts of attitudes (etc.)
have absolute value inasmuch as they are (or would be)
valued by (seem valuable to) a certified value-fixer,
whose valuations are (therefore) eligible for the benefits
provided by Humean Projection.
Q 3 • How do we find one of them ?
A 3 • We find a being whose essence (indeed whose metier)
it is to establish and to apply forms of absolute value.
Q4• How do we get such a being ?
A4• By Metaphysical Transubstantiation from the bio
logical substantial type Homo sapiens, which produces
for us the non-biological substantial type person. We are
now seen to be (qua persons) accredited value-fixers, at
least in relation to ourselves, and (qua creatures which
are both persons and specimens of Homo sapiens)
subjects whose voluntary operations are supposed to
conform to the values we fix for ourselves.
Q 5 • Can we be secured against the risk that the notions of
absolute value and of finality, which appear in Story S,
might turn out to be mechanistically substitutable (in
which case the vitalistic-mechanistic overlap would not
after all have been terminated, and the notions in
question would after all b e 'Pickwickian', not authentic) ?
A 5 • Yes, we can attain this security, provided we can show
that being an accredited value-fixer (in relation to
oneself) requires being free, in something like Kant's
sense of positive freedom; and that if the dovetailed
concepts of absolute value and finality were mechan
istically substitutable, this fact would import into the
genesis of our self-originated attitudes (etc.) a 'foreign
cause' (that is, a cause external to the legislator
cum-agent) which would constitute a barrier to the
presence of freedom, and which would for that reason
fatally undermine the prospects for success for Story S.
I am only too well aware that today's lecture, particularly
in its terminal convulsions, contains much that is obscure,
fragmentary, and ill defended (where indeed it is defended
at all ) . I am also conscious of the likelihood that, were I
(Heaven forbid) after a breather to return to the fray, I
should find myself inclined to produce quite different,
though maybe equally problematic, reflections. But I h ave
some hope that today's offering might provide an adequate
starting-point for one of those interminable sequences of
revisions of which serious theoretical thought seems so
largely to consist.
Rep ly to R ichards� 1
Final Section
M E T A P HY S I C S , P HI L O S O P H I C A L P S Y C H O L O G Y ,
A N D VA L U E
In this final section of my 'Reply to Richards', I shall take

up, or take off from, a few of the things which they have to
say about a number of fascinating and extremely import
ant issues belonging to the chain of disciplines listed in the
title of this section. I shall be concerned less with the
details of their account of the various positions which they
see me as maintaining than with the kind of structure and
order which , albeit in an as yet confused and incomplete
way, I think I can discern in the disciplines themselves and,
especially, in the connection between them. Any proper
discussion of the details of the issues in question would
demand far more time than I have at my disposal ; in any
case it is my intention, if I am spared, to discuss a number
of them, at length, in future writings. I shall be concerned
rather to provide some sort of picture of the nature of
metaphysics, as I see it, and of the way or ways in which it
seems to me to underlie the other mentioned disciplines. I
might add that something has already been said in the
1 [The 'Reply to Richards' is Grice's own contribution to the collection of
essays in his honour, Philosophical Grounds of Rationality: Intentions,
Categories, Ends, ed. Richard Grandy and Richard Warner ( Oxford :
Oxford University Press, 1 9 8 6 ) . The editors provided a synoptic view of
Grice' s work, enabling Grice to comment on his own work and interpreta
tions of it in responding to their essay. 'So I make my contribution a
response to theirs . . . For convenience I fuse the editors into a multiple
personal ity called " Richards " , whose multiplicity is marked by the use of
plural pronouns and verb-forms' (p. 45 ) . ]
94 Reply to Richards, Final Section
previous section of this 'Reply' about philosophical
psychology and rationality, and that general questions
about value, including metaphysical questions, were the
topic of my 1 9 8 3 Carus lectures. So perhaps I shall be
pardoned if here I concentrate primarily on metaphysics. I
fear that, even when I am allowed the advantage of
operating within these limitations, what I have to say will
be programmatic and speculative rather than well ordered
and well argued.
At the outset of their comments on my views concerning
metaphysics, Richards say, 'Grice's ontological views are
at least liberal' ; and they document this assertion by a
quotation from my Method in Philosophical Psychology,
in which I admit to a 'taste for keeping open house for all
sorts and conditions of entities, j ust so long as when they
come in they help with the housework'. I have no wish to
challenge their representation of my expressed position,
particularly as the cited passage includes a reference to the
possibility that certain sorts of entities might, because of
backing from some transcendental argument, qualify as
entia realissima. The question I would like to raise is,
rather, what grounds are there for accepting the current
conception of the relationship between metaphysics and
ontology ? Why should it be assumed that metaphysics
consists in, or even includes in its domain, the programme
of arriving at an acceptable ontology ? Is the answer merely
that that enterprise is the one, or a part of the one, to
which the term 'metaphysics' is conventionally applied,
and so, that a j ustification of this application cannot be a
philosophical issue ?
If this demand for a justified characterization of
metaphysics is to be met, I can think of only one likely
strategy for meeting it. That will be to show that success
within a certain sort of philosophical undertaking, which I
will with striking originality call First Philosophy, is
needed if any form of philosophy, or perhaps indeed any
form of rational enquiry, is to be regarded as feasible or
Reply to Richards, Final Section 95
legitimate; and that the contents of First Philosophy are
identical with, or at least include, what are standardly
regarded as the contents of metaphysics. 2 I can think of
two routes by which this result might be achieved, which
might well turn out not to be distinct from one another.
One route would perhaps involve taking seriously the idea
that if any region of enquiry is to be successful as a rational
enterprise, its deliverance must be expressible in the shape
of one or another of the possibly different types of theory;
that characterizations of the nature and range of possible
kinds of theory will be needed ; and that such a body of
characterization must itself be the outcome of rational
enquiry, and so must itself exemplify whatever require
ments it lays down for theories in general ; it must itself be
expressible as a theory, to be called (if you like) Theory
theory. The specification and j ustification of the ideas and
material presupposed by any theory, whether such account
falls within or outside the bounds of Theory-theory, would
be properly called First Philosophy, and might turn out to
relate to what is generally accepted as belonging to the
subject-matter of metaphysics. It might, for example, turn
out to be establishable that every theory has to relate to a
certain range of subject items, has to attribute to them
certain predicates or attributes, which in turn have to fall
within one or another of the range of types or categories.
In this way, the enquiry might lead to recognized meta
physical topics, such as the nature of being, its range of
application, the nature of predication, and a systematic
account of categories.
A second approach would focus not on the idea of the
expressibility of the outcomes of rational enquiry in
theories but rather on the question of what it is, in such
enquiries, that we are looking for, why they are of concern
to us. We start (so Aristotle has told us) as laymen with
the awareness of a body of facts; what as theorists we
2 In these reflections I have derived much benefit from discussions with
Alan Code.
strive for is not (primarily) further facts, but rational
knowledge, or understanding, of the facts we have,
together with whatever further facts our investigations
may provide for us. Metaphysics will have as its concern
the nature and realizability of those items which are
involved in any successful pursuit of understanding; its
range will include the nature and varieties of explanation
(as offered in some modification of the Doctrine of Four
Causes), the acceptability of principles of logic, the proper
standards of proof, and so on.
I have at this point three comments to make. First,
should it be the case that ( 1 ) the foregoing approach to the
conception of metaphysics is found acceptable, (2) the
nature of explanation and (understood broadly) of causes
is a metaphysical topic, and ( 3 ) that Aristotle is right (as I
suspect he is) that the unity of the notion of cause is
analogical in character, then the general idea of cause will
rest on its standard particularizations, and the particular
ideas cannot be reached as specifications of an antecedent
genus, for there is no such genus. In that case, final causes
will be (so to speak) foundation members of the cause
family, and it will be dubious whether their title as causes
can be disputed.
Second, it seems very likely that the two approaches are
in fact not distinct; for it seems plausible to suppose that
explanations, if fully rational, must be systematic and so
must be expressible in theories. Conversely, it seems
plausible to suppose that the function of theories is to
explain, and so that whatever is susceptible to theoretical
treatment is thereby explained.
Third, the most conspicuous difficulty about the
approach which I have been tentatively espousing seems to
me to be that we may be in danger of being given more
than we want to receive ; we are not, for example, ready to
regard methods of proof or the acceptability of logical
principles as metaphysical matters, and it is not clear how
such things are to be excluded. But perhaps we are in
danger of falling victims to a confusion. Morality, as such,
belongs to the province of ethics and does not belong to
the province of metaphysics. But, as Kant saw (and I agree
with him) , that does not preclude there being metaphysical
questions which arise about morality. In general, there
may be a metaphysics of X without it being the case that X
is a concept or item which belongs to metaphysics.
Equally, there may be metaphysical questions relating to
proof or logical principles without it being the case that, as
such, proof or logical principles belong to metaphysics. It
will be fair to add, however, that no distinction has yet
been provided, within the class of items about which there
are metaphysical questions, between those which do and
those which do not belong to metaphysics.
The next element in my attitude towards metaphysics to
which I would like to draw attention is my strong
sympathy for a constructivist approach. The appeal of
such an approach seems to me to lie essentially in the idea
that if we operate with the aim of expanding some set of
starting-points, by means of regulated and fairly well
defined procedures, into a constructed edifice of consider
able complexity, we have better prospects of obtaining the
explanatory richness which we need than if, for example,
we endeavour to represent the seeming wealth of the world
of being as reducible to some favoured range of elements.
That is, of course, a rhetorical plea, but perhaps such pleas
have their place.
But a constructivist methodology, if its title is taken
seriously, plainly has its own difficulties. Construction, as
normally understood, requires one or more constructors ;
so far as a metaphysical construction is concerned, who
does the constructing ? 'We' ? But who are we, and do we
operate separately or conjointly, or in some other way ?
And when and where are the acts of construction
performed, and how often ? These troublesome queries are
reminiscent of differences which arose, I believe, among
Kantian commentators, about whether Kant's threefold
synthesis (perhaps a close relative of construction) is (or
was) a datable operation or not. I am not aware that they
arrived at a satisfactory solution. The problem becomes
even more acute when we remember that some of the best
candidates for the title of constructed entities, for example
numbers, are supposed to be eternal, or at least timeless.
How could such entities have construction dates ?
Some relief may perhaps be provided if we turn our eyes
towards the authors of fiction. My next novel will have as
its hero one Caspar Winebibber, a notorious English
highwayman born (or so I shall say) in 1 764 and hanged in
1 79 8 , thereby ceasing to exist long before sometime next
year, when I create (or construct) him. This mind-boggling
situation will be dissolved if we distinguish between two
different occurrences ; first, Caspar's birth (or death),
which is dated to 1 764 (or 1 79 8), and second, my creation
of Caspar, that is to say, my making it in 1 9 8 5 fictionally
true that Caspar was born in 1 764 and died in 1 79 8 .
Applying this strategem to metaphysics, we may perhaps
find it tolerable to suppose that a particular great
mathematician should in 1 9 6 8 make it true that (let us say)
ultralunary numbers should exist timelessly or from and to
eternity. We might even, should we so wish, introduce a
'depersonalized' (and 'detemporalized') notion of con
struction ; in which case we can say that in 1 96 8 the great
mathematician, by authenticated construction, not only
constructed the timeless existence of ultralunary numbers
but also the thereby depersonalized and detemporalized
construction of the timeless existence of ultralunary
numbers, and also the depersonalized construction of the
depersonalized construction of . . . ultralunary numbers. In
this way, we might be able, in one fell swoop, to safeguard
the copyrights both of the mathematician and eternity .
Another extremely important aspect of my conception
of metaphysical construction (creative metaphysical think
ing) is that it is of its nature revisionary or gradualist in
character. It is not just that, since metaphysics is a very
R eply to Richards, Final Section 99
difficult subject, the best way to proceed is to observe the
success and failures of others and to try to build further
advance upon their achievements. It is rather that there is
no other way of proceeding but the way of gradualism. A
particular bit of metaphysical construction is possible only
on the basis of some prior material; which must itself
either be the outcome of prior constructions, or perhaps be
something original and unconstructed. As I see it, gradual
ism enters in in more than one place. One point of entry
relates to the degree of expertise of the theorist or
investigator. In my view, it is incumbent upon those whom
Aristotle would have called 'the wise' in metaphysics, as
often elsewhere, to treat with respect and build upon the
opinions and the practices of 'the many' ; and any
intellectualist indignation at the idea of professionals being
hamstrung by amateurs will perhaps be seen as in
appropriate when it is reflected that the amateurs are really
(since personal identities may be regarded as irrelevant)
only ourselves (the professionals) at an earlier stage; there
are not two parties, like Whigs and Tories, or nobles and
the common people, but rather one family of speakers
pursuing the life of reason at different stages of develop
ment; and the later stages of development depend upon the
earlier ones.
Gradualism also comes into play with respect to theory
development. A characteristic aspect of what I think of as a
constructivist approach towards theory development in
volves the appearance of what I call 'overlaps'. It may be
that a theory or theory-stage B , which is to be an extension
of theory or theory-stage A, includes as part of itself
linguistic or conceptual apparatus which provides us with
a restatement of all or part of theory A, as one segment of
the arithmetic of positive and negative integers provides us
with a restatement of the arithmetic of natural numbers.
But while such an overlap may be needed to secure
intelligibility for theory B, theory B would be pointless
unless its expressive power transcended that of theory A ,
1 00 Reply to Richards, Final Section
unless (that is to say) a further segment of theory B lay
beyond the overlap. Gradualism sometimes appears on the
scene in relation to stages exhibited by some feature
attaching to the theory as a whole, but more often perhaps
in relation to stages exemplified in some department of, or
some category within, the theory. We can think of
metaphysics as involving a developing sequence of meta
physical schemes ; we can also locate developmental
features within and between particular metaphysical cat
egories. Again, I regard such developmental features not as
accidental but as essential to the prosecution of meta
physics. One can only reach a proper understanding of
metaphysical concepts like law or cause if one sees, for
example, the functional analogy, and so the developmental
connection, between natural laws and non-natural laws
(like those of legality or morality ) . 'How is such and such a
range of uses of the word (the concept) x to be rationally
generated ? ' is to my mind a type of question which we
should continually be asking.
I may now revert to a question which appeared briefly
on the scene a page or so ago. Are we, if we lend a
sympathetic ear to constructivism, to think of the meta
physical world as divided into a constructed section and a
primitive, original, unconstructed section ? I will confess at
once that I do not know the answer to this question. The
forthright contention that if there is a realm of constructs
there has to be also a realm of non-constructs to provide
the material upon which the earliest ventures in construc
tion are to operate has its appeal, and I have little doubt
that I have been influenced by it. But I am by no means
sure that it is correct. I am led to this uncertainty initially
by the fact that when I ask myself what classes of entities I
would be happy to regard as original and unconstructed, I
do not very readily come up with an answer. Certainly not
common objects like tables and chairs ; but would I feel
better about stuffs like rock or hydrogen, or bits thereof? I
do not know, but I am not moved towards any emphatic
Reply to R ichards, Final Section 101
'yes ! ' Part o f m y trouble i s that there does not seem to me
to be any good logical reason calling for a class of ultimate
non-constructs. It seems to me quite on the cards that
metaphysical theory, at least when it is formally set out,
might consist in a package of what I will call ontological
schemes in which categories of entities are constructively
ordered, that all or most of the same categories may
appear within two different schemes with different ordering,
what is primitive in one scheme being non-primitive in the
other, and that this might occur whether the ordering
relations employed in the construction of the two schemes
were the same or different. We would then have no role for
a notion of absolute primitiveness. All we would use
would be the relative notion of primitiveness-with-respect
to-a-scheme. There might indeed be room for a concept of
authentic or maximal reality; but the application of this
concept would be divorced from any concept of primitive
ness, relative or absolute, and would be governed by the
availability of an argument, no doubt transcendental in
character, showing that a given category is mandatory,
that a place must be found for it in any admissible
ontological scheme. I know of no grounds for rejecting
ideas along these lines.
The complexities introduced by the possibility that there
is no original, unconstructed, area of reality, together with
a memory of the delicacy of treatment called for by the last
of the objections to my view on the philosophy of
language, suggest that debates about the foundations
of metaphysics are likely to be peppered with allegations of
circularity; and I suspect that this would be the view of any
thoughtful student of metaphysics who gave serious
attention to the methodology of his discipline. Where are
the first principles of First Philosophy to come from, if not
from the operation, practised by the emblematic pelican,
of lacerating its own breast. In the light of these considera
tions it seems to me to be of the utmost importance to get
clear about the nature and forms of real or apparent
circularity, and to distinguish those forms, if any, which
are innocuous from those which are deadly. To this end I
would look for a list, which might not be all that different
from the list provided by Aristotle, of different kinds, or
interpretations, of the idea of priority, with a view to
deciding when the supposition that A is prior to B allows
or disallows the possibility that B may also be prior to A,
either in the same, or in some other, dimension of priority.
Relevant kinds of priority would perhaps include logical
priority, definitional or conceptual priority, epistemic
priority, and priority in respect of value. I will select two
examples, both possibly of philosophical interest, where
for differing m and n, it might be legitimate to suppose that
the priority m of A to B would not be a barrier to the
priorityn of B to A . It seems to me not implausible to hold
that in respect of one or another version of conceptual
priority, the legal concept of right is prior to the moral
concept of right: the moral concept is only understandable
by reference to, and perhaps is even explicitly definable in
terms of, the legal concept. But if that is so, we are perhaps
not debarred from regarding the moral concept as valua
tionally prior to the legal concept; the range of application
of the legal concept ought to be always determined by
criteria which are couched in terms of the moral concept.
Again, it might be important to distinguish two kinds of
conceptual priority, which might both apply to one and
the same pair of items, though in different directions. It
might be, perhaps, that the properties of sense-data, like
colours (and so sense-data themselves), are posterior in
one sense to corresponding properties of material things
(and so to material things themselves) ; properties of
material things, perhaps, render the properties of sense
data intelligible by providing a paradigm for them. But
when it comes to the provision of a suitably motivated
theory of material things and their properties, the idea of
making these definitionally explicable in terms of sense
data and their properties may not be ruled out by the
Reply to Richards, Final Section 1 03
holding of the aforementioned conceptual priority in the
reverse direction. It is perhaps reasonable to regard such
fine distinction as indispensable if we are to succeed in the
business of pulling ourselves up by our own bootstraps. In
this connection it will be relevant for me to reveal that I
once invented (though I did not establish its validity) a
principle which I labelled as Bootstrap. The principle laid
down that when one is introducing the primitive concepts
of a theory formulated in an object language, one has
freedom to use any battery of concepts expressible in the
meta-language, subject to the condition that counterparts
of such concepts are subsequently definable or otherwise
derivable in the object-language. So the more economically
one introduces the primitive object-language concepts, the
less of a task one leaves oneself for the morrow.
I must now turn to a more direct consideration of the
question of how metaphysical principles are ultimately to
be established. A prime candidate is forthcoming, namely,
a special metaphysical type of argument, one that has been
called by Kant and by various other philosophers since
Kant a transcendental argument. Unfortunately it is by no
means clear to me precisely what Kant, and still less what
some other philosophers, regard as the essential character
of such an argument. Some, I suspect, have thought of a
transcendental argument in favour of some thesis or
category of items as being one which claims that if we
reject the thesis or category in question, we shall have to
give up something which we very much want to keep ; and
the practice of some philosophers, including Kant, of
hooking transcendental argument to the possibility of
some very central notion, such as experience or knowledge,
or (the existence of) language, perhaps lends some colour
to this approach. My view (and my view of Kant) takes a
different tack. One thing which seems to be left out in the
treatments of transcendental argument just mentioned is
the idea that transcendental argument involves the sugges
tion that something is being undermined by one who is
1 04 Reply to R ichards, Final Section
sceptical about the conclusion which such an argument
aims at establishing. Another thing which is left out is any
investigation of the notion of rationality, or the notion of a
rational being. Precisely what remedy I should propose for
these omissions is far from clear to me ; I have to confess
that my ideas in this region of the subject are still in a very
rudimentary state. But I will do the best I can.
I suspect that there is no single characterization of
transcendental arguments which will accommodate all of
the traditionally recognized specimens of the kind ; indeed,
there seem to me to be at least three sorts of argument
pattern with good claims to be dignified with the title of
'transcendental' .
1 . One pattern fits Descartes's cogito argument, which
Kant himself seems to have regarded as paradigmatic. This
argument may be represented as pointing to a thesis,
namely his own existence, to which a real or pretended
sceptic is thought of as expressing enmity, in the form of
doubt; and it seeks to show that the sceptic's procedure is
self-destructive in that there is an irresoluble conflict
between, on the one hand, what the sceptic is suggesting
(that he does not exist), and on the other hand the
possession, by his act of suggesting, of the illocutionary
character (being the expression of a doubt) which it not
only has but must, on the account, be supposed by the
sceptic to have. It might, in this case, be legitimate to go on
to say that the expression of doubt cannot be denied
application, since without the capacity for the expression
of doubt the exercise of rationality will be impossible; but
while this addition might link this pattern with the two
following patterns, it does not seem to add anything to the
cogency of the argument.
2. Another pattern of argument would be designed for
use against applications of what I might call 'epistemo
logical nominalism' ; that is, against someone who proposes
to admit ys but not xs on the grounds that epistemic
justification is available for ys but not for anything like xs,
Reply to R ichards, Final Section 1 05
which supposedly go beyond ys ; we can, for example,
allow sense-data but not material objects, if they are
thought of as 'over and above' sense-data; we can allow
particular events but not, except on some minimal
interpretation, causal connections between events. The
pattern of argument under consideration would attempt to
show that the sceptic's at first sight attractive caution is a
false economy ; that the rejection of the 'over-and-above'
entities is epistemically destructive of the entities with
which the sceptic deems himself secure ; if material objects
or causes go, sense-data and datable events go too. In some
cases it might be possible to claim, on the basis of the lines
of the third pattern of argument, that not j ust the minimal
categories, but, in general, the possibility of the exercise of
rationality will have to go .
3 . A third pattern of argument might contend from the
outset that if such and such a target of the sceptic were
allowed to fall, then something else would have to fall
which is a pre-condition of the exercise of rationality; it
might be argued, for example, that some sceptical thesis
would undermine freedom, which in turn is a pre
condition of any exercise of rationality whatsoever.
It is plain that arguments of this third type might differ
from one another in respect of the particular pre-condition
of rationality which they brandished in the face of a
possible sceptic. But it is possible that they might differ in a
more subtle respect. Some less ambitious arguments might
threaten a local breakdown of rationality, a breakdown in
some particular area. An argument might hold, for
instance, that certain sceptical positions would preclude
the possibility of the exercise of rationality in the practical
domain. While such arguments may be expected to carry
weight with some philosophers, a really doughty sceptic is
liable to accept the threatened curtailment of rationality;
he may, as Hume and those who follow him have done,
accept the virtual exclusion of reason from the area of
action. The threat, however, may be of a total breakdown
of the possibility of the exercise of rationality ; and here
even the doughty sceptic might quail, on pain of losing his
audience if he refµses to quail.
A very important feature of these varieties of transcend
ental argument (though I would prefer to abandon the
term 'transcendental' and just call them metaphysical
arguments) may be their connection with practical argu
ment. In a broadened sense of 'practical', which would
relate not just to action but also to the adoption of any
attitude or stance which is within our rational control, we
might think of all argument, even alethic argument, as
practical, perhaps with the practical tailpiece omitted;
alethic or evidential argument may be thought of as
directing us to accept or believe some proposition on the
grounds that it is certain or likely to be true. But sometimes
we are led to rational acceptance of a proposition (though
perhaps not to belief in it) by considerations other than the
likelihood of its truth. Things that are matters of faith of
one sort or another, like the fidelity of one's wife or the
justice of one's country's cause, are typically not acc�pted
on evidential grounds but as demands imposed by loyalty or
patriotism ; and the arguments produced by those who
wish us to have such faith may well not be silent about this
fact. Metaphysical argument and acceptance may exhibit a
partial analogy with these examples of the acceptance of
something as a matter of faith. In the metaphysical region,
too, the practical aspect may come first: we must accept
such and such a thesis or else face an intolerable
breakdown of rationality. But in the case of metaphysical
argument, the threatened calamity is such that the accept
ance of the thesis which avoids it is invested with the
alethic trappings of truth and evidential respectability.
Proof of the pudding comes from the need to eat it, not
vice versa. These thoughts will perhaps allay a discomfort
which some people, including myself, have felt with
respect to transcendental arguments. It has seemed to me,
in at least some cases, that the most that such arguments
could hope to show is that rationality demands the
acceptance, not the truth, of this or that thesis. This feature
would not be a defect if one can go on to say that this kind
of demand for acceptance is sufficient to confer truth on
what is to be accepted.
It is now time for me to turn to a consideration of the
ways in which metaphysical construction is effected, and I
shall attempt to sketch three of these. But before I do so, I
should like to make one or two general remarks about
such construction routines. It is pretty obvious that
metaphysical construction needs to be disciplined, but this
is not because without discipline it will be badly done, but
because without discipline it will not be done at all. The
list of available routines determines what metaphysical
construction is ; so it is no accident that it employs these
routines. This reflection may help us to solve what has
appeared to me, and to others, as a difficult problem in the
methodology of metaphysics, namely, how are we to
distinguish metaphysical construction from scientific con
struction of such entities as electrons or quarks ? What is
the difference between hypostasis and hypothesis ? The
answer may lie in the idea that in metaphysical construc
tion, including hypostasis, we reach new entities (or in
some cases, perhaps, suppose them to be reachable) by
application of the routines which are essential to meta
physical construction ; when we are scientists and hypo
thesize, we do not rely on these routines, at least in the first
instance; if at a later stage we shift our ground, that is a
major theoretical change.
I shall first introduce two of these construction routines;
before I introduce the third I shall need to bring in some
further material, which will also be relevant to my task in
other ways. The first routine is one which I have discussed
elsewhere, and which I call Humean Projection. Something
very like it is indeed described by Hume, when he talks
about 'the mind's propensity to spread itself on objects ' ;
but he seems t o regard i t a s a source, o r a product, o f
confusion and illusion which, perhaps, our nature renders
unavoidable, rather than as an achievement of reason. In
my version of the routine, one can distinguish four real or
apparent stages, the first of which, perhaps, is not always
present. At this first stage we have some initial concept,
like that expressed by the word 'or' or 'not', or (to take a
concept relevant to my present undertakings) the concept
of value. We can think of these initial items as, at this
stage, intuitive and unclarified elements in our conceptual
vocabulary. At the second stage we reach a specific mental
state, in the specification of which it is possible, though
maybe not necessary, to use the name of the initial concept
as an adverbial modifier: we come to 'or-thinking' (or
disjoining) , 'not-thinking' (or rejecting, or denying) , and
'value-thinking' (or valuing, or approving) . These specific
states may be thought of as bound up with, and indeed as
generating, some set of responses to the appearance on the
scene of instantiation of the initial concepts. At the third
stage, reference to these specific states is replaced by a
general (or more general) psychological verb together with
an operator corresponding to the particular specific stage
which appears within the scope of the general verb, but is
still only allowed maximal scope within the complement of
the verb and cannot appear in subclauses. So we find
reference to 'thinking p or q' or 'thinking it valuable to
learn Greek'. At the fourth and last stage, the restriction
imposed by the demand that the operators at the third
stage should be scope-dominant within the complement of
the accompanying verb is removed; there is no limitation
on the appearance of the operation in subordinate clauses.
With regard to this routine I would make five observa
tions :
1 . The employment of this routine may be expected to
deliver for us, as its end-result, concepts (in something like
a Fregean sense of the word) rather than objects. To
generate objects we must look to other routines.
2. The provision, at the fourth stage, of full syntactico-
semantical freedom for the operators which correspond to
the initial concepts is possible only via the provision of
truth conditions, or of some different but analogous
valuations, for statements within which the operators
appear. Only thus can the permissible complexities be
made intelligible.
3. Because of observation 2, the difference between the
second and third stages is apparent rather than real. The
third stage provides only a notational variant of the second
stage, at least unless the fourth stage is also reached.
4 . It is important to recognize that the development, in
a given case, of the routine must not be merely formal or
arbitrary. The invocation of a subsequent stage must be
exhibited as having some point or purpose, as ( for
example) enabling us to account for something which
needs to be accounted for.
5. Subject to these provisos, application of this routine
to our initial concept ('putting it through the mangle') does
furnish one with a metaphysical reconstruction of that
concept; or, if the first stage is missing, we are given a
metaphysical construction of a new concept.
The second construction routine harks back to Aristotle's
treatment of predication and categories, and I will present
my version of it as briefly as I can. Perhaps its most proper
title would be Category Shift; but since I think of it as
primarily useful for introducing new objects, new subjects
of discourse, by a procedure reminiscent of the linguists'
operation of nominalization, I might also refer to it as
subjectification (or, for that matter, objectification) . Given
a class of primary subjects of discourse, namely substances,
there are a number of 'slots' (categories) into which
predicates of these primary subjects may fit; one is
substance itself (secondary substance), in which case the
predication is intra-categorial and essential; and there are
others into which the predicates assigned in non-essential
or accidental predication may fall: the list of these would
resemble Aristotle's list of quality, quantity, and so forth.
It might be, however, that the members of my list, perhaps
unlike that of Aristotle, would not be fully co-ordinate; the
development of the list might require not one blow but a
succession of blows ; we might, for example, have to
develop first the category of attribute, and then the
subordinate categories of quantitative attribute (quantity)
and non-quantitative attribute ( quality), or again the cat
egory of event before the subordinate category of action.
Now though substances are to be the primary subjects of
predication, they will not be the only subjects. Derivatives
of, or conversions of, items which start life (so to speak) as
predicable, in one non-substantial slot or another, of
substances may themselves come to occupy the first slot;
they will be qualities of, or quantities of, a particular type
or token of substantials : not being qualities or quantities
of substances, they will not be qualities or quantities
simpliciter. (It is my suspicion that only for substances, as
subjects, are all the slots filled by predicable items.) Some
of these substantials which are not substances may derive
from a plurality of items from different original categories ;
events, for example, might be complex substantials deriving
from a substance, an attribute, and a time.
My position with regard to the second routine runs
parallel to my position with regard to the first, in that here
too I hold strongly to the opinion that the introduction of a
new category of entities must not be arbitrary. It has to be
properly motivated; if it is not, perhaps it fails to be a case
of entity-construction altogether, and becomes merely a
way of speaking. What sort of motivation is called for is
not immediately clear. One strong candidate would be the
possibility of opening up new applications for existing
modes of explanation; it may be, for example, that the
substantial introduction of abstract entities, like properties,
makes possible the application to what Kneale called
'secondary induction', 3 the principles at work in primary
3 William Kneale , Probability and Induction ( Oxford: Clarendon Press,
1949), 104, for example.
Reply to R ichards, Final Section 111
induction. But it is not only the sort but the degree of
motivation which is in question. When I discussed meta
physical argument, it seemed that to achieve reality the
acceptance of a category of entities had to be mandatory ;
whereas the recent discussion has suggested that apart
from conformity to construction routines, all that is
required is that the acceptance be well motivated. Which
view would be correct ? Or is it that we can tolerate a
division of constructed reality into two segments, with
admission requirements of differing degrees of stringency ?
Or is there just one sort of admission requirement, which
in some cases is over-fulfilled ?
Before characterizing my third construction routine I
must say a brief word about essential properties and about
finality, two Aristotelian ideas which at least until recently
have been pretty unpopular, but for which I want to find
metaphysical room. In their logical dress, essential
properties would appear either as properties which are
constitutive or definitive of a given, usually substantial,
kind; or as individuating properties of individual members
of a kind, properties such that if an individual were to lose
them, it would lose its identity, its existence, and indeed
itself. It is clear that if a property is one of the properties
which define a kind, it is also an individuating property of
individual members of a kind, properties such that if an
individual were to lose them, it would cease to belong to the
kind and so cease to exist. (A more cautious formulation
would be required if, as the third construction routine might
require, we subscribed to the Grice-Myra view of identity.)
Whether the converse holds seems to depend on whether we
regard spatio-temporal continuity as a definitive property
for substantial kinds, indeed for all substantial kinds.
But there is another, more metaphysical dress which
essential properties may wear. They may appear as
Keynesian generator-properties, 'core' properties of a
substantive kind which co-operate to explain the pheno
menal and dispositional features of members of that kind.
On the face of it this is a quite different approach ; but on
reflection I find myself wondering whether the difference is
as large as it might at first appear. Perhaps at least at the
level of a type of theorizing which is not too sophisticated
and mathematicized, as maybe these days the physical
sciences are, the logically essential properties and the
fundamentally explanatory properties of a substantial kind
come together; substances are essentially (in the 'logical'
sense) things such that in circumstances C they manifest
feature F, where the gap-signs are replaced in such a way
as to display the most basic laws of the theory. So perhaps,
at this level of theory, substances require theories to give
expression to their nature, and theories require substances
to govern them.
Finality, particularly detached finality ( functions or
purposes which do not require sanction from purposers or
users), is an even more despised notion than that of an
essential property, especially if it is supposed to be
explanatory, to provide us with final causes. I am
somewhat puzzled by this contempt for detached finality,
as if it were an unwanted residue of an officially obsolete
complex of superstitions and priestcraft. That, in my view,
it is certainly not: the concepts and vocabulary of finality,
operating as if they were detached, are part and parcel of
our standard procedures for recognizing and describing
what goes on around us. This point is forcibly illustrated
by William Golding in The Inheritors.4 There he describes,
as seen through the eyes of a stone-age couple who do not
understand at all what they are seeing, a scene in which (I
am told) their child is cooked and eaten by iron-age
people. In the description functional terms are eschewed,
with the result that the incomprehension of the stone-age
couple is vividly shared by the reader. Now finality is
sometimes active rather than passive : the finality of a thing
then consists in what it is supposed to do rather than in
4 [London: Faber & Faber, 1 95 5 . ]

what it is supposed to suffer, have done to it, or have done
with it. Sometimes the finality of a thing is not dependent
on some ulterior end which the thing is envisaged as
realizing. Sometimes the finality of a thing is not imposed
or dictated by a will or interest extraneous to the thing.
And sometimes the finality of a thing is not subordinate to
the finality of some whole of which the thing is a
component, as the finality of an eye or a foot may be
subordinate to the finality of the organism to which it
belongs. When the finality of a thing satisfies all of these
overlapping conditions and exclusions, I shall call it a case
of autonomous finality ; and I shall also on occasion call it
a metier. I will here remark that we should be careful to
distinguish this kind of autonomous finality, which may
attach to substances, from another kind of finality which
seemingly will not be autonomous, and which will attach
to the conception of kinds of substance or of other
constructed entities. The latter sort of finality will represent
the point or purpose, from the point of view of the
metaphysical theorist, of bringing into play, in a particular
case, a certain sort of metaphysical manreuvre. It is this
latter kind of finality which I have been supposing to be a
requirement for the legitimate deployment of construction
routines.
Now it is my position that what I might call finality
features, at least if they consist in the possession of
autonomous finality, may find a place within the essential
properties of at least some kinds of substances (for
example, persons) . Some substances may be essentially 'for
doing such and such ' . Indeed, I suspect we might go
further than this, and suppose that autonomous finality
not merely can fall within a substance's essential nature,
but indeed, if it attaches to a substance at all, must belong
to its essential nature. If a substance has a certain metier, it
does not have to seek the fulfilment of that metier, but it
does have to be equipped with the motivation to fulfil the
metier should it choose to follow that motivation. And
since autonomous finality is independent of any ulterior
end, that motivation must consist in respect for the idea
that to fulfil the metier would be in line with its own
essential nature. But however that may be, once we have
finality features enrolled among the essential pr.o perties of
a kind of substance, we have a starting-point for the
generation of a theory or system of conduct for that kind
of substance, which would be analogous to the descriptive
theory which can be developed on the basis of a substance's
essential descriptive properties.
I can now give a brief characterization of my third
construction routine, which is called Metaphysical Tran
substantiation. Let us suppose that the genitor has
sanctioned the appearance of a biological type called
humans, into which, considerate as always, he has built an
attribute, or complex of attributes, called rationality,
perhaps on the grounds that this would greatly assist its
possessors in coping speedily and resourcefully with
survival problems posed by a wide range of environments,
which they would thus be in a position to enter and to
maintain themselves in. But, perhaps unwittingly, he will
thereby have created a breed of potential metaphysicians;
and what they do is (so to speak) to reconstitute
themselves. They do not alter the totality of attributes
which each of them as a human possesses, but they
redistribute them; properties which they possess essentially
as humans become properties which as substances of a
new psychological type called persons they possess
accidentally; and the property or properties called
rationality, which attaches only accidentally to humans,
attaches essentially to persons. While each human is
standardly coincident with a particular person (and is
indeed, perhaps, identical with that person over a time),
logic is insufficient to guarantee that there will not come a
time when that human and that person are no longer
identical, when one of them, perhaps, but not the other,
has ceased to exist. But though logic is insufficient, it may
be that other theories will remedy the deficiency. Why,
otherwise than from a taste for mischief, the humans (or
persons) should have wanted to bring off this feat of
transubstantiation will have to be left open until my final
section, which I have now reached.
My final undertaking will be an attempt to sketch a way
of providing metaphysical backing, drawn from the
material which I have been presenting, for a reasonably
unimpoverished theory of value; I shall endeavour to
produce an account which is fairly well ordered, even
though it may at the same time be one which bristles with
unsolved problems and unformulated supporting argu
ments. What I have to offer will be close to and I hope
compatible with, though certainly not precisely the same
as, the content of my third Carus lecture. 5 Though it lends
an ear to several other voices from our philosophical
heritage, it may be thought of as being, in the main, a
representation of the position of that unjustly neglected
philosopher Kantotle. It involves six stages.
1. The details of the logic of value concepts and of their
possible relativizations are unfortunately visible only
through thick intellectual smog; so I shall have to help
myself to what, at the moment at least, I regard as two
distinct dichotomies. First, there is a dichotomy between
value concepts which are relativized to some focus of
relativization and those which are not so relativized, which
are absolute. If we address ourselves to the concept being
of value there are perhaps two possible primary foci of
relativization ; that of end or potential end, that for which
something may be of value, as bicarbonate of soda may be
of value for health (or my taking it of value for my health ) ,
o r dumb-bells may be of value (useful) for bulging the
biceps; and that of beneficiary or potential beneficiary, the
person (or other sort of item) to whom (or to which)
5 [Original Version, 1 9 8 3 ; see pp. 69-9 1 above.]

something may be of value, as the possession of a
typewriter is of value to some philosophers but not to me, ·
since I do not type. With regard to this dichotomy I am
inclined to accept the following principles. First, the
presence in me of a concern for the focus of relativization
is what is needed to give the value concept a 'bite' on me,
that is to say, to ensure that the application of the value
concept to me does, or should, carry weight for me; only if
I care for my aunt can I be expected to care about what is
of value to her, such as her house and garden. Second, the
fact that a relativized value-concept, through a de facto or
de jure concern on my part for the focus of relativization,
engages me does not imply that the original relativization
has been cancelled, or rendered absolute. If my concern for
your health stimulates in me a vivid awareness of the value
to you of your medication, or the incumbency upon you to
take your daily doses, that value and that incumbency
are still relativized to your health ; without a concern
on your part for your health, such claims will leave you
cold.
The second dichotomy, which should be carefully
distinguished from the first, lies between those cases in
which a value concept, which may be either relativized or
absolute, attaches originally, or directly, to a given bearer,
and those in which the attachment is indirect and is the
outcome of the presence of a transmitting relation which
links the current bearer with an original bearer, with or
without the aid of an intervening sequence of 'descendants'.
In the case of the transmission of relativized value
concepts, the transmitting relation may be the same as, or
may be different from, the relation which is embodied in
the relativization. The foregoing characterization would
allow absolute value to attach originally or directly to
promise-keeping or to my keeping a promise, and to attach
indirectly or by transmission to my digging your garden
for you, should that be something which I have promised
to do; it would also allow the relativized value-concept of
value for health to attach directly to medical care and
indirectly or by transmission to the payment of doctor's
bills, an example in which the transmitting relation and the
relativizing relation are one and the same.
2. The second stage of this metaphysical defence of the
authenticity of the conception of value will involve a
concession and a contention. It will be conceded that if the
only conception of value available to us were that of
relativized value then the notion of finality would be in a
certain sense dispensable ; and further, that if the notion of
finality is denied authenticity, so must the notion of value
be denied authenticity. A certain region of ostensible
finality, which is sufficient to provide for the admissibility
of attributions of relativized value, is " mechanistically
substitutable " ; that is to say, by means of reliance on the
resources of cybernetics and on the fact that the non
pursuit of certain goals such as survival and reproduction
is apt to bring to an end the supply of potential pursuers,
some ostensibly final explanations are replaceable by, or
reinterpretable as, explanations of a sort congenial to
mechanists. But if the concept of value is to be authentic
and not merely 'Pickwickian' in character, then it is
required that it be supported by a kind of finality which
extends beyond the 'overlap ' with mechanistically sub
stitutable finality ; autonomous finality will be demanded,
and a mechanist cannot accommodate and must deny this
kind of finality ; and so, as will shortly be indicated, he is
committed to a denial of absolute value.
3. That metaphysical house-room be found for the
notion of absolute value is a rational demand. To say this
is not directly to offer reason to believe in the acceptability
of the notion, though it makes a move in that direction. It
is rather to say that there is good reason for wanting it to
be true that the notion is acceptable. There might be more
than one kind of rational ground for this desire. It might be
that we feel a need to appeal to absolute value in order to
justify some of our beliefs and attributes with regard to
relativized value, to maintain (for example) that it is of
absolute value that everyone should pursue, within certain
limits, what he regards as being of value to himself. Or
again, it might be that, by Leibnizian standards for
evaluating possible worlds, a world which contains
absolute value, on the assumption that its regulation
requires relatively simple principles, is richer and so better
than one which does not.
But granted that there is a rational demand for absolute
value, one can then perhaps argue that within whatever
limits are imposed by metaphysical constructions already
made, we are free to rig our metaphysics in such a way as
to legitimize the conception of absolute value; what it is
proper to believe to be true may depend in part on what
one would like to be true. Perhaps part of the Kantian
notion of positive freedom, a dignity which as rational
beings we enjoy, is the freedom not merely to play the
metaphysical game but, within the limits of rationality, to
fix its rules as well. In any case, a trouble-free metaphysical
story which will safeguard the credentials of absolute value
is to be accepted should it be possible to devise one. I have
some hopes that the methodology at work here might link
up with my earlier ideas about the quasi-practical character
of metaphysical argument.
4. On the assumption that the operation of Meta
physical Transubstantiation has been appropriately carried
through, a class of biological creatures has been 'invented'
into a class of psychological substances, namely persons,
who possess as part of their essential nature a certain
metier or autonomous finality consisting in the exercise, or
a certain sort of exercise, of rationality, and who have only
to recognize and respect a certain law of their nature, in
order to display in favourable circumstances the capacity
to realize their metier. The degree to which they fulfil that
metier will constitute them good persons ('good qua'
persons) ; and while the reference to the substantial kind
persons undoubtedly introduces a restriction or qualifica-
tion, it is not clear (if it matters) that this restriction is a
mode of relativization.
5. Once the concept of value-qua-member-of-a-kind
has been set up for a dass of substances, the way is opened
for the appearance of transmitting relationships which will
extend the application of value-in-a-kind to suitably
qualified non-substantial aspects of members of a kind,
such as actions and characteristics. While it cannot be
assumed that persons will be the only original instances of
value-in-a-kind, it seems plausible to suggest that whatever
other original instances there may be will be far less
fruitful sources of such extension, particularly if a prime
mode of extension will be by the operation of Humean
Projection. It seems plausible to suppose that a specially
fruitful way of extending the range of absolute value might
be an application or adaptation of the routine of Humean
Projection, whereby such value is accorded, in Aristotelian
style, to whatever would seem to possess such value in the
eyes of a duly accredited j udge ; and a duly accredited j udge
might be identifiable as a good person operating in
conditions of freedom. Cats, adorable as they may be, will
be less productive sources of such extension than persons.
6 . In the light of these reflections, and on the assump
tion that to reach the goal of securing the admissibility of
the concept of absolute value we need a class of primary
examples of an unqualified version of that concept, it
would appear to be a rational procedure to allot to persons
as a substantial type not j ust absolute value qua members
of their kind, but absolute value tout court, that is to say,
unqualified absolute value. Such value could be attributed
to the kind, in virtue of its potentialities, and to selected
individual members of the kind, in virtue of their achieve
ments.
Such a defence of absolute value is, of course, bristling

with unsolved or incompletely solved problems. I do not
find this thought daunting. If philosophy generated no new
120 Reply to R ichards, Final Section
problems it would be dead, because it would be finished;
and if it recurrently regenerated the same old problems it
would not be alive because it could never begin. So those
who still look to philosophy for their bread-and-butter
should pray that the supply of new problems never dries
up.
Method In Philosophical
Psychology
(From the Banal to the
Bizarre) 1
Preamble
In what follows, I shall be presenting some of my ideas

about how I want to approach philosophical psychology.
My hope is, in effect, to sketch a whole system; this is quite
an undertaking, and I hope that you will bear with me if in
discharging it I occupy a little more of your time than is
becoming in holders of my august office. While I am sure
that you will be able to detect some affinities between my
ideas and ideas to be found in recent philosophy, I propose
to leave such comparisons to you. Though at certain points
I have had my eye on recent discussions, the main
influences on this part of my work have lain in the past,
particularly in Aristotle, Hume, and Kant. I have the
feeling that between them these philosophers have written
a great deal of the story, though perhaps not always in the
most legible of hands.
1 Presidential Address delivered at the 49th Annual Meeting of the

American Philosophical Association in San Diego, 28 March 1975. Earlier
versions of this address were given in lectures at Princeton University in
March 1972, and as a John Dewey lecture at the University of Michigan
in Ann Arbor in April 1974.
1 22 American Philosophical Association
I. Some Focal Problems
I shall begin by formulating, in outline, a sequence of four

particular problems which I think an adequate philo
sophical psychology must be able to lay to rest.
The first problem (Problem A) concerns the real or
apparent circularity with which one is liable to be faced if
one attempts to provide an analysis of central psychological
concepts by means of explicit definitions. Suppose that,
like some philosophers of the not so distant past, we are
attracted by the idea of giving dispositional behaviouristic
analyses of such concepts, and that we make a start on the
concept of belief. As a first shot, we try the following, B 1 : x
believes that p just in case x is disposed to act as if p were
true. In response to obvious queries about the meaning of
the phrase, 'act as if p were true', we substitute, for B � > B 2 :
x believes that p j ust in case x is disposed, whenever x
wants (desires) some end E, to act in ways which will
realize E given that p is true rather than in ways which will
realize E given that p is false. The precise form which such
a definition as B 2 might take is immaterial to my present
purpose, provided that it has two features observable in
B 2 ; first, that a further psychological concept, that of
wanting, has been introduced in the definiens; and second,
that to meet another obvious response, the concept of
belief itself has to be reintroduced in the definiens. For the
disposition associated with a belief that p surely should be
specified, not as a disposition to act in ways which will in
fact realize E given that p is true, but rather as a
disposition to act in ways which x believes will, given that
p is true, realize E. One who believes that p may often fail
to act in ways that would realize E given that p were true,
because he is quite unaware of the fact that such actions
would realize E if p were true; and he may quite often act
in ways which would realize E only if p were false, because
he mistakenly believes that such ways would realize E if p
were true.
Method in Philosophical Psychology 1 23
If we turn to the concept of wanting, which B 2
introduced into the definiens for belief, we seem to
encounter a parallel situation. Suppose we start with W 1 : x
wants E just in case x is disposed to act in ways which will
realize E rather than in ways which will realize the
negation of E. Precisely the same kind of objection as that
just raised in the case of belief seems to compel the
introduction of the concept of belief into the definiens of
W 1 giving us W2 : x wants E j ust in case x is disposed to act
in ways which x believes will realize E rather than in ways
which x believes will realize the negation of E. We now
meet the further objection that one who wants E can be
counted on to act in such ways as these only provided that
there is no E' which he wants more than E; if there is such
an E ' , then in any situation in which E' conflicts with E he
may be expected to act in ways he thinks will realize the
negation of E. But the incorporation of any version of this
proviso into the definiens for wanting will reintroduce the
concept of wanting itself.
The situation, then, seems to be that if, along the
envisaged behaviouristic lines, we attempt to provide
explicit definitions for such a pair of concepts as those of
belief and wanting, whichever member of the pair we start
with we are driven into the very small circle of introducing
into the definiens the very concept which is being defined,
and also into the slightly larger circle of introducing into
the definiens the other member of the pair. The idea
suggested to me by this difficulty is that we might be well
advised, as a first move, to abandon the idea of looking for
explicit definitions of central psychological concepts, and
look instead for implicit definitions, to be provided by
some form of axiomatic treatment; leaving open the
possibility that, as a second move, this kind of treatment
might be made the foundation for a different sort of
explicit definition. Such a procedure might well preserve
an attractive feature of behaviouristic analyses, that of
attempting to explain psychological concepts by relating
124 American Philosophical Association
them to appropriate forms of behaviour, while at the same
time freeing us from the logical embarrassments into
which such analyses seem to lead us.
We are now, however, faced with a further question : if
we are to think of certain psychological concepts as being
implicitly defined by some set of laws (or quasi-laws) in
which they figure, are we to attribute to such laws a
contingent or a non-contingent status ? A look at some
strong candidates for the position of being laws of the kind
which we are seeking seems merely to reinforce the
question. Consider the principle 'He who wills the end
wills the means,' some form of which seems to me to be the
idea behind both of the behaviouristic definitions j ust
discussed; and interpret it as saying that anyone who wills
some end, and also believes that some action on his part is
an indispensable means to that end, wills the action in
question. One might be inclined to say that if anyone
believed (really believed) that a certain line of action was
indispensable to the attainment of a certain end, and yet
refused to adopt that line of action, then that would count
decisively against the conceptual legitimacy of saying that
it was really his will to attain that end. To proceed in this
way at least appears to involve treating the principle as a
necessary truth . On the other hand, one might also be
inclined to regard the principle as offering a general
account of a certain aspect of our psychological processes,
as specifying a condition under which, when our will is
fixed on a certain object, it will be the case that it is also
fixed on a certain other object; to take this view of the
principle seems like regarding it as a psychological law and
so as contingent. My second problem (Problem B) is, then,
this : how (without a blanket rejection of the analytic/
synthetic distinction) are we to account for and, if possible,
resolve the ambivalence concerning their status with which
we seem to look upon certain principles involving psycho
logical concepts ?
To set this problem aside for a moment, the approach
which I am interested in exploring is that of thinking of
certain central psychological concepts as theoretical
concepts ; they are psychological concepts just because they
are the primitive concepts which belong to a certain kind
of psychological theory without also belonging to any
presupposed theory (such as physiological theory) ; and a
psychological theory is a theory whose function is to
provide, in a systematic way, explanations of behaviour
which differ from any explanations of behaviour which
may be provided by (or may some day be provided by) any
presupposed theory (such as physiological theory ) . To
explicate such psychological concepts is to characterize
their role in the theory to which they primarily belong, to
specify (with this or that degree of detail) the laws or
quasi-laws in which they figure, and the manner in which
such laws are linked to behaviour. Now to say this much is
not to say anything very new; I am sure you have heard
this sort of thing before. It is also not to say very much ; all
that has so far been given is a sketch of a programme for
the philosophical treatment of psychological concepts, not
even the beginnings of a philosophical treatment itself.
What is needed is rather more attention to detail than is
usually offered by advocates of this approach ; we need to
pursue such questions as what the special features of a
psychological theory are, why such a theory should be
needed, what sort of laws or generalities it should contain,
and precisely how such laws may be used to explicate
familiar psychological terms. I shall be addressing myself
to some of these questions in the remainder of this address
to you .
But before any more is said, a further problem looms. I
can almost hear a murmur to the effect that such an
approach to philosophical psychology is doomed from the
start. Do we not have privileged access to our own beliefs
and desires ? And, worse still, may it not be true that at
least some of our avowals of our beliefs and desires are
incorrigible ? How, then, are such considerations as these
to be rendered consistent with an approach which seems to
imply that the justifiability of attributing beliefs and
desires to people (and maybe animals) , including ourselves,
rests on the utility of the theory to which the concepts of
belief and desire belong, in providing desiderated explana
tions of behaviour ? This is my third problem (Problem C ) .
M y fourth problem (D : the Selection Problem) is
connected with a different and less radical objection to the
approach which I have just begun to sketch. Surely, it may
be said, it cannot be right to suppose that a conjunction of
all the laws of a psychological theory in which a
psychological concept figures is what should be used to
explicate that concept. Even if we are in a position to use
some of these laws (which is not certain) , we are not, and
indeed never shall be, in a position to use all of them.
Moreover, some particular laws are not going to be
suitable. For all we know, some modification of one or
other of the following laws 2 might be a psychological law,
possibly an underived law:
1. Optimism Law: the more one wants <P the more
likely one is to believe <P .
2. Optimism/Pessimism Law: given condition C i , the
more one wants cp, the more likely one is to believe
cp; given condition C 2 , the more one wants cp, the
less likely one is to believe cp.
These do not seem the right kind of laws to be used to
explicate wanting and believing. We need some selective
principle: what is it ?
II. Some General Aspects of the Relation between

Psychological Theory and Psychological Concepts
First, a preliminary observation : if we are seeking to

explicate psychological concepts by relating them to a
2 Examples by courtesy of Judy Baker.
psychological theory, the theory which we invoke should
be one which can be regarded as underlying our ordinary
speech and thought about psychological matters, and as
such will have to be a part of folk-science. There is no need
to suppose that its structure would be of a sort which is
appropriate for a professional theory, nor even to suppose
that it would be a correct or acceptable theory ; though I
would hope that there would be a way of showing that at
least some central parts of it would be inter.pretable within
an acceptable professional theory in such a way as to come
out true in that theory, and that indeed this demand miAAt
constitute a constraint on the construction of a professional
theory. This is perhaps a latter-day version of a defence of
common sense in this area. 3
Let us unrealistically suppose, for the purposes of
schematic discussion, that we have a psychological theory
which, when set out in formal dress, contains a body of
underived laws, the formulation of which incorporates
two primitive predicate-constants, ] and V; these we want
to correspond, respectively, to two psychological terms
"judging" and "willing", which in their turn will serve as a
regimented base for the explication of such familiar
notions as believing and wanting. Let us think of J and V
as correlated with, or ranging over, " instantiables " ,
leaving open the question whether these instantiables are
sets or properties, or both, or neither. Let us abbreviate the
conjunction of these laws by " L " , and let us ignore for the
moment the obvious fact that any adequate theory along
these lines would need to distinguish specific sub
instantiables under ] and V, corresponding to the diversity
of specific contents which j udging and willing may take as
modifications (as " intentional objects " ) .
How are we to envisage the terms "judge" and "will " as
being introduced ? Two closely related alternative ways
suggest themselves. First, what I shall call the way of
3 This topic will be touched upon somewhat more fully in Section VIII.
Ramsified naming: 'There is j ust one J and j ust one V such
that L, and let J be called " judging" and V be called
"willing " . ' On this alternative, the uniqueness claim is
essential, since "judging" and "willing" are being assigned
as names for particular instantiables. The second altern
ative, which I may call the way of Ramsified definition,
can dispense with the uniqueness claim. It will run : ' (a) x
judges just in case there is a J and there is a V such that L,
and x instantiates ]; ( b) x wills j ust in case there is a J and
there is a V such that L, and x instantiates V. '
We now have available a possible explanation of the
ambivalence we may feel with regard to the status of
certain psychological principles, which was the subject of
Problem B . The difference between the two alternative
procedures is relatively subtle, and if we do not distinguish
them we may find ourselves under the pull of both. But the
first alternative will render certain psychological principles
contingent, while the second will render the same principles
non-contingent. Let me illustrate by using an even simpler
and even less realistic example. Suppose that a psycho
logical law tells us that anyone in state P hollers, and that
we seek to use this law to introduce the term "pain " . On
the first alternative, we have : there is just one P, such that
anyone in state P hollers, and let us call P "pain " . Since on
this alternative we are naming the state P "pain " , if the
utilized law is contingent, so will be the principle 'anyone
who is in pain hollers. ' If we use the second alternative, we
shall introduce "pain " as follows : x is in pain j ust in case
there is a P such that anyone in state P hollers and x is in
state P. On this alternative it will be non-contingently true
that anyone who is in pain hollers. (For to be in pain is to
be in some state which involves hollering.) While this
suggestion may account for the ambivalence noted, it does
not of course resolve it; to resolve it, one would have to
find a reason for preferring one of the alternatives to the
other.
In a somewhat devious pursuit of such a reason, let us
enquire further about the character of the postulated
psychological instantiables-are they, or could they be,
identifiable with physical (physiological) instantiables ?
One possible position of a sort which has been found
attractive by some slightly impetuous physicalists would
be to adopt the first alternative (that of Ramsified naming)
and to combine it with the thesis that the J-instantiable is
to be identified with one physiological property and the V
instantiable with another such property. This position at
least appears to have the feature, to which some might
strongly object, of propounding a seemingly philosophical
thesis which is at the mercy of possible developments in
physiology; and, one might add, the prospects that such
developments would be favourable to the thesis do not
seem to be all that bright. The adaptiveness of organisms
may well be such as to make it very much in the cards that
different creatures of the same species may, under different
environmental pressures, develop different sub-systems
(even different sub-systems at different times) as the
physiological underlay of the same set of psychological
instantiables, and that a given physiological property may
be the correlate in one sub-system of a p articular psycho
logical instantiable, while in another it is correlated with a
different psychological instantiable, or with none at all.
Such a possibility would mean that a physiological
property which was, for a particular creature at a
particular time, correlated with a particular psychological
instantiable would be neither a necessary nor a sufficient
condition for that instantiable; it would at best be
sufficient (and perhaps also necessary) for the instantiable
within the particular kind of sub-system prevailing in the
creature at the time. The unqualified identification of the
property and the instantiable would now be excluded.
Faced with such a difficulty, a physicalist might resort to
various manoeuvres. He might seek to identify a particular
psychological instantiable with a disjunctive property,
each disjoined constituent of which is a property consisting
in having a certain physiological property in a certain kind
of sub-system ; or he might relativize the notion of identity,
for the purpose of psycho-physical identification, to types
of sub-systems ; or he might abandon the pursuit of
identification at the level of instantiables, and restrict
himself to claiming identities between individual psycho
logical events or states of affairs (e.g. ]ones's believing at t
that p) and physiological events or states of affairs (e.g.
]ones's brain being in such and such a state at t) . But none
of these shifts has the intuitive appeal of their simplistic
forebear. Moreover, each will be open to variants of a
familiar kind of objection. For example, ]ones's j udging at
noon that they were out to get him might well be a case of
j udging something to be true on insufficient evidence ; but
(to use Berkeley's phrase) it 'sounds harsh' to say that
]ones's brain's being in such and such a state at noon is a
case of j udging something to be true on insufficient
evidence. For my part, I would hope that a much-needed
general theory of categories would protect me against any
thesis which would require me either to license such
locutions as the last or to resort to a never-ending stream
of cries of 'Opaque context !' in order to block them. If the
prime purpose of the notion of identity is, as I believe it to
be, to license predicate-transfers, one begins to look a little
silly if one first champions a particular kind of identifica
tion, and then constantly jibs at the predicate-transfers
which it seems to allow.
Such considerations as these can, I think, be deployed
against the way of Ramsified naming even when it is
unaccompanied by a thesis about psycho-physical identities.
However unlikely it may be that the future course of
physiological research will favour any simple correlations
between particular psychological instantiables and par
ticular physiological properties, I do not see that I have any
firm guarantee that it will not, that it will never be
established that some particular kind of brain state is
associated with, say, judging that snow is white. If so, then
Method in Philosophical Psychology 131
i f I adopt the first alternative with its uniqueness claim, I

have no firm guarantee against having to identify my
brain's being in some particular state with my j udging that
snow is white, and so being landed with the embarrass
ments I have just commented upon. Since I am inclined to
think that I do in fact have such a guarantee (although at
the moment I cannot lay my hands on it) , I am inclined to
prefer the second alternative to the first.
As a pendant to the discussion j ust concluded, let me
express two prejudices. First, any psycho-physical identi
fications which are accepted will have to be accepted on
the basis of some known or assumed psycho-physical
correlations; and it seems to me that in this area all that
philosophical psychology really requires is the supposition
that there are such correlations. Whether they do or do not
provide a legitimate foundation for identifications seems
to me to be more a question in the theory of identity than
in the philosophy of mind. Second, I am not greatly
enamoured of some of the motivations which prompt the
advocacy of psycho-physical identifications ; I have in
mind a concern to exclude such ' queer' or 'mysterious'
entities as souls, purely mental events, purely mental
properties, and so forth. My taste is for keeping open
house for all sorts and conditions of entities, j ust so long as
when they come in they help with the housework.
Provided that I can see them at work, and provided that
they are not detected in illicit logical behaviour (within
which I do not include a certain degree of indeterminacy,
not even of numerical indeterminacy) , I do not find them
queer or mysterious at all. To £angle a new ontological
Marxism, they work therefore they exist, even though only
some, perhaps those who come on the recommendation of
some form of transcendental argument, may qualify for
the specially favoured status of entia realissima. To
exclude honest working entities seems to me like meta
physical snobbery, a reluctance to be seen in the company
of any but the best objects.
III. The Form of Psychological "Laws "
In this section I shall briefly discuss two formal features

which I think it might be desirable to attribute to some at
least of the laws or quasi-laws of the theory to be used to
explicate psychological concepts.
1 . Aristotle distinguished between things which are so
of necessity and things which are so for the most part, and
located in the second category things which are done.
More or less conformably with this position, I shall suggest
that some or all of the laws which determine psychological
concepts are ceteris paribus laws, which resemble prob
ability generalizations in that they are defeasible, but differ
from them in that ceteris paribus laws do not assign
weights. I envisage a system which without inconsistency
may contain a sequence consisting of a head-law of the
form A 's are Z, and modifying laws of the form A 's which
are B are Z' (where Z' is incompatible with Z), and of the
form A 's which are B and C are Z, and so on. Analogous
sequences can be constructed for functional laws; if A, B ,
and Z are determinables we may have, in the same
system:
1
(a) things which are A to degree a are Z to degree f ( a )
(b) things which are A to degree a and B to degree P
are Z to a degree which is the value of (2 (p, ( 1 (a) )
If we like, we can prefix " ceteris paribus " to each such
generality, to ensure that "A's are Z" is not taken as
synonymous with "All A's are Z. " For any such system a
restriction on Modus Ponens is required, since we must
not be allowed to infer from A,s are Z and x is both A and
B to x is Z if A 's which are B are Z' (where Z' is
incompatible with Z) is a law. This restriction will be
analogous to the Principle of Total Evidence required for
probabilistic systems ; a first approximation might run as
follows : in applying a law of the system to an individual
case for the purpose of detachment, one must select the
law with the most specific antecedent which is satisfied by
the individual case. Two results of treating psychological
laws as ceteris paribus laws which will be attractive to me
are first that it can no longer be claimed that we do not
know any psychological laws because we do not know all
the restrictive conditions, and second, that a psychological
theory may be included in a larger theory which modifies
it; modification does not require emendation.
If a modifying sequence terminates, then its constituent
laws can be converted into universal laws; otherwise not.
If we hold a strong version of Determinism ( roughly, that
there is an acceptable theory in which every phenomenon
is explained) , we might expect the Last Trump to herald a
Presidential Address in which the Final Theory is presented,
entirely free from ceteris paribus laws ; but if we accept
only a weaker version, with reversed quantifiers ( for every
phenomenon there is an acceptable theory in which it is
explained), the programme for Judgement D ay will lose
one leading attraction : other items on the agenda, however,
may prove sufficiently exciting to compensate for this loss.
2 . I suspect that even the latitude given to our prospect
ive psychological folk-theory by allowing it to contain or
even consist of ceteris paribus laws will not make it folksy
enough. There is (or was) in Empirical Psychology a
generality called " The Yerkes-Dodson Law " ; based on
experiments designed to test degrees of learning compet
ence in rats set to run mazes under water, after varying
periods of initial constraint, it states (in effect) that, with
other factors constant, degrees of learning competence are
correlated with degrees of emotional stress by a function
whose values form a bell-curve. I have some reluctance
towards calling this statement the expression of a law, on
the grounds that, since it does not (and could not, given
that quantitative expression of degrees of competence and
of stress is not available) specify the function or functions
in question, it states that there is a law of a certain sort
rather than actually state a law. But whether or not the
Yerkes-Dodson Law is properly so-called, it is certainly
'law-allusive' ; and the feature of being law-allusive is one
which I would expect to find in the psychological 'laws' to
be used to explicate psychological concepts.
IV. Semi-realistic Procedure for Introducing some

Psychological Concepts
I turn now to the task of outlining, in a constructive way,

the kind of method by which, in accordance with my
programme, particular psychological concepts, and lin
guistic expressions for them, might be introduced; I shall
take into account the need to clarify the routes by which
psychological sub-instantiables come to be expressed by
the combination of a general psychological verb and a
complement which specifies content. My account will be
only semi-realistic, since I shall consider the psychological
explanation of only a very rudimentary sample of
behaviour, and even with respect to that sample a proper
explanation would involve apparatus which I omit-for
example, apparatus to deal with conceptual representation
and with the fact that motivations towards a particular
sort of behaviour may be frustrated. But since I am
attempting only to illustrate a general method, not to make
substantial proposals, these over-simplifications should
not matter. I construct my account as if I espoused the first
alternative discussed in Section II (the way of Ramsified
naming) rather than the second, since that considerably
simplifies exposition : a transition to the second alternative
can, of course, be quite easily effected.
Let us suppose that a squarrel (a creature something like
a squirrel) has some nuts in front of it, and proceeds to
gobble .them ; and that we are interested in the further
explanation of this occurrence. Let us call the squarrel
"Toby" . Our ethological observations of Toby, and of
other squarrels, tell u s that Toby and other squarrels often
gobble nuts in front of them, and also other things besides
nuts, and indeed that they are particularly liable to do
some gobbling after a relatively prolonged period in which
they have done no gobbling at all. On the basis of these
observations, and perhaps a range of other observations as
well, we decide that certain behaviour of squarrels,
including Toby's gobbling nuts in front of him, are suitable
subjects for psychological explanation; so we undertake to
introduce some theoretical apparatus which can be used to
erect an explanatory bridge between Toby's having nuts in
front of him, and his gobbling them.
We make the following postulations (if we have not

already made them long ago) :
1 . That the appropriate explanatory laws will refer to
three instantiables P, ], and V which (we stipulate) are to
be labelled respectively "prehend" , "join " , and "will" .
2. That for any type o f creature T there i s a class N of
kinds of thing, which is a class of what (we stipulate) shall
be called "necessities for T" , and that N (in N) is a
necessity for T j ust in case N is vital for T, that is, j ust in
case any member of T which suffers a sufficiently
prolonged non-intake of N will lose the capacity for all of
those operations (including the intake of N), the capacity
for which is constitutive of membership of T [in a fuller
account this condition would require further explication] .
3 . That if N is a necessity for T, then ceteris paribus a
moderately prolonged non-intake of N on the p art of a
member (x) of T will cause x to instantiate a particular sub
instantiable Vi of "will" , x's instantiation of which will
(ceteris paribus) be quenched by a moderately prolonged
intake of N; and (we stipulate) Vi shall be called " will for
N" (or, if you like, "willing N").
4. That for any member x (e.g. Toby) of any creature
type T (e.g. squarrels) there is a class ft of object-types
(which, we stipulate, is a class of what shall be called
"object-types familiar to x" ) ; a class R of relations (a class
of what shall be called "relations familiar to x " ) , and a
class A of action-types (a class of what shall be called
" action-types in the repertoire of x " ) , which satisfy the
following conditions :
(a) that there are two ways (if you like, functions) w1
and w2 such that ceteris paribus if an instance of an
object-type F (familiar to x) is related to x by a
relation R (familiar to x), then this causes x to
instantiate a sub-instantiable of prehending which
corresponds in ways w1 and w2 to F and R
respectively; which sub-instantiable (we stipulate)
shall be called "prehending an F as R " [e.g.
"prehending nuts as in front" ] .
(b) that ceteris paribus if, for some N [e.g. 'squarrel
food'] which is a necessity for x's type, for some A
which is in x's repertoire, and for some F and R
familiar to x, x has sufficiently frequently, when
instantiating will for N, performed A upon instances
of F which have been related by R to x, and as a
result has ceased to instantiate will for N: then x
instantiates a corresponding [corresponding to N,
A, F, R] sub-instantiable of j o in ing; which sub
instantiable shall be called "joining N with A and F
and R " .
We can now formulate a n " overall" law (a " PJV Law " ) ,
i n which we n o longer need to restrict N, A, F, and R to
members of classes connected in certain ways with a
particular creature or type of creature, as follows :
5 . Ceteris paribus, a creature x which wills N, prehends
an F as R, and j oins N and A and F and R, performs A .
Finally, we can introduce " judging" by derivation from
" joining" ; we stipulate that if x j oins N with A and F and R ,
then w e shall speak of x a s judging A , upon-F-in-R, for N
[e.g. judging gobbling, upon nuts (in) in front, for squarrel
food] .
We now have the desired explanatory bridge.
(i) Toby has nuts in front of him ;
(ii) Toby is short on squarrel-food (observed or
assumed) ; so
(iii) Toby wills squarrel-food [by postulate 3, connect
ing will with intake of N] ;
(iv) Toby prehends nuts as in front [from (i) by
postulate 4 (a ) , if it is assumed that nuts and in front
are familiar to Toby] ;
(v) Toby j oins squarrel-food with gobbling and nuts
and in front (Toby judges gobbling, on nuts in
front, for squarrel-food) [by postulate 4 ( b ) , with
the aid of prior observation] ; so, by the PJV Law,
from (iii) , (iv) , and (v),
(vi) Toby gobbles ; and, since nuts are in front of him,
gobbles the nuts in front of him.
I shall end this section by some general remarks about the
procedure j ust sketched.
1. The strategy is relatively simple; we invoke certain
ceteris paribus laws in order to introduce particular
psychological sub-instantiables and their specification by
reference to content: thus " willing N" is introduced as the
specific form of willing which is dependent on the intake
and deprivation of a necessity N, "prehending F as R " as
the variety of prehension which normally results from an F
being R to one, and similarly for " joining " . Then we use a
further " overall" ceteris paribus law to eliminate reference
to the psychological instantiables and to reach the
behaviour which is to be explained. A more developed
account would (no doubt) bring in intermediate laws,
relating simply to psychological instantiables and not also
to features of the common world.
2. The generalities used have at various points the 'law
allusiveness' mentioned in Section III. Every reference to,
for example, 'a sufficiently prolonged deprivation', 'a
sufficiently frequent performance', or to 'correspondence
in certain ways' in effect implies a demand on some more
developed theory actually to produce laws which are here
only asserted to be producible.
3. The psychological concepts introduced have been
defined only for a very narrow range of complements.
Thus, j udging has so far been defined only for complements
which correspond in a certain way with those involved in
the expression of hypothetical judgements (judging is, one
might say, so far only a species of "if-judging" ) ; and
"will " has been defined, so far, only for complements
specifying necessities. Attention would have to be given to
the provision of procedures for extending the range of
these concepts.
4. If the procedure is, in its general character, correct
( and I have not even attempted to show that it is) , the
possibility of impact on some familiar philosophical issues
is already discernible. To consider only "prehension " ,
which i s intended a s a prototype for perception :
(a) "Prehending F as R " lacks the existential implica
tion that there is an F which is R to the prehender ; this
condition could be added, but it would be added. This
might well please friends of sense-data.
(b) The fact that reference to physical situations has
to be made in the introduction of expressions for
specific prehensions might well be very unwelcome to
certain phenomenalists.
(c) The fact that in the introduction of expressions
for specific prehensions a demand is imposed on a
further theory to define functions mapping such pre
hensions on to physical situations might well prove fatal
to a sceptic about the material world. How can such a
sceptic, who is unsceptical about descriptions of sense
experience, combine the demand implicit in such
descriptions with a refusal to assent to the existence of
the physical situations which, it seems, the further
theory would require in order to be in a position to meet
the demand ?
V. Creature Construction
I have so far been occupying myself with questions about

the structure of the kind of psychological theory which
would fulfil the function of bestowing the breath of life on
psychological concepts. I shall move now to asking by
what methods one might hope to arrive at an acceptable
theory of this sort.
One procedure, which I do not in the least despise,
would be to take as a body of data our linguistic intuitions
concerning what we would or would not be prepared to
say when using psychological terms; then, as a first stage,
to look for principles (perhaps involving artificially con
structed concepts) which would seem to generalize the
features of this or that section of our discourse, making
adj ustments when provisionally accepted principles lead to
counter-intuitive or paradoxical results; and, as a second
stage, to attempt to systematize the principles which have
emerged piecemeal in the first stage, p aying attention to
the adequacy of a theory so constructed to account for the
data conformably with general criteria for the assessment
of theories. This would be to operate in two stages
approximately corresponding to what the Greeks called
the " analytic" (or " dialectical" ) and the " synthetic"
procedures.
I am, for various reasons, not happy to confine myself to
these methods. In applying them, many alternative theor
etical options may be available, which will often be
difficult to hit upon. I would like, if it is possible, to find a
procedure which would tell me sooner and louder if I am
on the right track, a procedure, indeed, which might in the
end tell me that a particular theory is the right one, by
some test beyond that of the ease and elegance with which
it accommodates the data. I suspect that a dividend of this
sort was what Kant expected transcendental arguments to
yield, and I would like a basis for the construction of
some analogue of transcendental arguments. I am also
influenced by a different consideration. I am much im
pressed by the fact that arrays of psychological concepts, of
differing degrees of richness, are applicable to creatures of
differing degrees of complexity, with human beings (so far)
at the peak. So I would like a procedure which would do
j ustice to this kind of continuity, and would not leave me
j ust pursuing a number of separate psychological theories
for different types of creatures.
The method which I should like to apply is to construct
(in imagination, of course) , according to certain principles
of construction, a type of creature, or rather a sequence of
types of creature, to serve as a model (or models) for actual
creatures. My creatures I call pirots (which, Russell and
Carnap have told us, are things which karulize elatically ) .
The general idea i s t o develop sequentially the psycho
logical theory for different brands of pirot, and to compare
what one thus generates with the psychological concepts
we apply to suitably related actual creatures, and when
inadequacies appear, to go back to the drawing-board to
extend or emend the construction (which of course is
unlikely ever to be more than partial) .
The principles of pirot-construction may b e thought of
as embodied in what I shall call a genitorial programme,
the main aspects of which I shall now formulate.
1 . We place ourselves in the positio 'n of a genitor, who
is engaged in designing living things (or rather, as I shall
say, operants ) . An operant may, for present purposes, be
taken to be a thing for which there is a certain set of
operations requiring expenditure of energy stored in the
operant, a sufficient frequency of each operation in the set
being necessary to maintain the operant in a condition to
perform any in the set (i.e. to avoid becoming an ex
operant) . Specific differences within such sets will deter
mine different types of operant. The genitor will at least
have to design them with a view to their survival
(continued operancy) ; for, on certain marginal assump
tions which it will be reasonable for us to make but tedious
for m e to enumerate, if a n operant (x) i s not survival
oriented, there is no basis for supposing it to exist at all;
since (by the assumption) x has to be produced in some
way or other by other operants of the type, x will not exist
unless pro-genitors are around to produce it; and if they
are to have the staying power (and other endowments)
required for x's production, x, being of the same type,
must be given the same attributes. So in providing for the
individual x, some provision for the continuation of the
type is implicit. In order to achieve economy in assump
tions, we shall suppose the genitor to be concerned only to
optimize survival chances.
2. Since the genitor is only a fiction, he is to be
supposed only to design, not to create. In order that his
designs should be useful to philosophical psychology, he
must be supposed to pass his designs on to the engineer
(who, being also a fiction, also does not have the power to
create) : the function of the engineer is to ensure that the
designs of the genitor can be realized in an organism which
behaves according to pre-psychological laws (e.g. those of
physics and physiology ) . Since the genitor does not know
about engineering, and since he does not want to produce
futile designs, he had better keep a close eye on the actual
world in order to stay within the bounds of the possible.
3. The mode of construction is to be thought of as being
relative to some very generally framed " living-condition"
concerning the relation of a pirot to its environment; the
operations the capacity for which determines the type of
the pirot are to be those which, given the posited
condition, constitute the minimum which the pirot would
require in order to optimize the chances of his remaining in
a condition to perform just those operations. Some pirots
(plant-like) will not require psychological apparatus in
order that their operations should be explicably performed;
others will. Within the latter class, an ascending order of
psychological types would, I hope, be generated by
increasing the degree of demandingness in the determining
condition. I cannot specify, at present, the kind of
generality which is to attach to such conditions ; but I have
in mind such a sequence as operants which do not need to
move at all to absorb sources of energy, operants which
only have to make posture changes, operants which,
because the sources are not constantly abundant, have to
locate those sources, and (probably a good deal later in the
sequence) operants who are maximally equipped to cope
with an indefinite variety of physiologically tolerable
environments (i.e., perhaps, rational pirots) . Further types
(or sub-types) might be generated by varying the degree of
effectiveness demanded with respect to a given living
condition.
4. Aristotle regarded types of soul ( as I would suppose,
of living thing) as forming a " developing series" . I
interpret that idea as being the supposition that the
psychological theory for a given type is an extension of,
and includes, the psychological theory of its predecessor
type. The realization of this idea is at least made possible
by the assumption that psychological laws may be of a
ceteris paribus form, and so can be modified without
emendation. If this aspect of the programme can be made
good, we may hope to safeguard the unity of psychological
concepts in their application to animals and to human
beings. Though (as Wittgenstein noted) certain animals
can only expect such items as food, while men can expect a
drought next summer, we can (if we wish) regard the
concept of expectation as being determined not by the laws
relating to it which are found in a single psychological
theory, but by the sequences of sets of laws relating to it
which are found in an ascending succession of psycho
logical theories.
5. Since the genitor is only to install psychological
apparatus in so far as it is required for the generation of
operations which would promote survival in a posited
living-condition, no psychological concept can be instanti
ated by a pirot without the supposition of behaviour which
manifests it. An explanatory concept has no hold if there is
nothing for it to explain. This is why 'inner states must
have outward manifestations' .
6. Finally, w e may observe that w e now have t o hand a
possible solution of the Selection Problem ( D ) . Only those
laws which can be given a genitorial justification (which
fall within 'Pure Psychology') will be counted as helping to
determine a psychological concept. It is just because one is
dubious about providing such a j ustification for the
.
envisaged " Optimism " laws that one is reluctant to regard
either of them as contributing to the definition of believing
and wanting.
Kant thought that, in relation to the philosophy of
nature, ideas of pure reason could legitimately be given
only a regulative employment; somewhat similarly, as far
as philosophical psychology is concerned, I think of the
genitorial programme (at least in the form j ust characterized)
as being primarily a heuristic device. I would, however,
hope to be able to retain a less vivid reformulation of it, in
which references to the genitor's purposes would be
replaced by references to final causes, or (more positivistic
ally) to survival-utility; this reformulation I would hope to
be able to use to cope with such matters as the Selection
problem. But (also much as Kant thought with respect
to ideas of pure reason) when it comes to ethics the
genitorial programme, in its more colourful form, j ust may
have a role which is not purely heuristic. The thought that
if one were genitor, one would install a certain feature
which characteristically leads to certain forms of behaviour
might conceivably be a proper step in the evaluation of
that feature and of the associated behaviour.
Let me be a little more explicit, and a great deal more
speculative, about the possible relation to ethics of my
programme for philosophical psychology. I shall suppose
that the genitorial programme has been realized to the
point at which we have designed a class of pirots which,
nearly following Locke (Essay, I I . 27. 8 ) , I might call 'very
intelligent rational pirots'. These pirots will be capable of
putting themselves in the genitorial position, of asking
how, if they were constructing themselves with a view to
their own survival , they would execute this task; and if we
have done our work aright, their answer will be the same
as ours . In virtue of the rational capacities and dispositions
which we have given them, and which they would give
themselves, each of them will have both the capacity and
the desire to raise the further question 'Why go on
surviving ? ' ; and (I hope) will be able to j ustify his
continued existence by endorsing (in virtue of the afore
mentioned rational capacities and dispositions) a set of
criteria for evaluating and ordering ends, and by applying
these criteria both to ends which he may already have, as
indirect aids to survival, and to ends which are yet to be
selected ; such ends, I may say, will not necessarily be
restricted to concerns for himself. The justification of the
pursuit of some system of ends would, in its turn, provide a
justification for his continued existence.
We might, indeed, envisage the contents of a highly
general practical manual, which these pirots would be in a
position to compile; this manual, though perhaps short,
might serve as a basis for more specialized manuals to be
composed when the pirots have been diversified by the
harsh realities of actual existence. The contents of the initial
manual would have various kinds of generality which are
connected with familiar discussions of universalizability.
The pirots have, so far, been endowed only with the
characteristics which belong to the genitorially j ustified
psychological theory ; so the manual will have to be
formulated in terms of the concepts of that theory,
together with the concepts involved in the very general
description of living-conditions which have been used to
set up that theory ; the manual will therefore have
conceptual generality. There will be no way of singling out
a special subclass of addresses, so the injunctions of the
manual will have to be addressed, indifferently, to any very
intelligent rational pirot, and will thus have generality of
form. And since the manual can be thought of as being
composed by each of the so far indistinguishable pirots, no
pirot would include in the manual injunctions prescribing
a certain line of conduct in circumstances to which he was
not likely to be subject ; nor indeed could he do so, even if
he would. So the circumstances for which conduct is
prescribed could be presumed to be such as to be satisfied,
from time to time, by any addressee ; the manual, then, will
have generality of application.
Such a manual might, perhaps, without ineptitude be
called an IMMANUEL; and the very intelligent rational
pirots, each of whom both composes it and from time to
time heeds it, might indeed be ourselves (in our better
moments, of course) .
VI. Type-Progression in Pirotology: Content

Internalization
My purpose in this section is to give a little thought to the

question 'What are the general principles exemplified, in
creature-construction, in progressing from one type of
pirot to a higher type ? What kinds of steps are being
made ?' The kinds of step with which I shall deal here are
those which culminate in a licence to include, within the
specification of the content of the psychological state of
certain pirots, a range of expressions which would be
inappropriate with respect to lower pirots ; such expressions
include connectives, quantifiers, temporal modifiers,
mood-indicators, modal operators, and (importantly)
names of psychological states like "j udge " and "will " ;
expressions the availability of which leads to the structural
enrichment of specifications of content. In general, these
steps will be ones by which items or ideas which have,
initially, a legitimate place outside the scope of psycho
logical instantiables (or, if you will, the expressions for
which occur legitimately outside the scope of psychological
verbs) come to have a legitimate place within the scope of
such instantiables : steps by which (one might say) such
items or ideas come to be internalized.
I am disposed to regard as prototypical the sort of
natural disposition which Hume attributes to us, and
which is very important to him; namely, the tendency of
the mind 'to spread itself upon objects', to project into the
world items which, properly (or primitively) considered,
are really features of our states of mind. Though there are
other examples in his work, the most famous case in which
he tries to make use of this tendency to deal with a
philosophical issue is his treatment of necessity ; the idea of
necessary connection is traced (roughly) to an internal
impression which attends the passage of the mind from an
idea or impression to an associated idea. I am not wholly
happy about Hume's characterization of this kind of
projection, but what I have to say will bear at least a
generic resemblance to what may be found in Hume.
The following kinds of transition seem to me to be
characteristic of internalization.
1 . References to psychological states ('4J-states) may
occur in a variety of linguistic settings which are also
appropriate to references to states which have no direct
connection with psychology. Like volcanic eruptions,
judgings or willings may be assigned both more precise
and less precise temporal locations (e.g. last Thursday or
next Tuesday, in the future or in the past) ; sometimes
specifications of degree will be appropriate; judging [A] or
willing [A] may be assigned a cause or an effect; and
references to such '4J-states will be open to logical
operations such as negation and disjunction. To suppose
that a creature will, in the future, j udge [A] , or that he
either judges [Al or j udges [B] , is obviously not to attribute
to him a j udging distinct from judging [A] , or from judging
[Al and from judging [B] ; nevertheless, we can allow a
transition from these attributions to the attribution of
judgings which are distinct, viz. of "future-judging [A] "
(expecting [A] ) and of "or-j udging [A, B] " . Since these are
new 'ljJ-states, they will be open to the standard range"of
linguistic settings for references to 'ljJ-states ; a creature
may now (or at some time in the future) future-judge [A]
(or past-judge [A] ) . There will be both a semantic
difference and a semantic connection (not necessarily the
same for every case) between the terminology generated by
this kind of transition and the original from which it is
generated; while " x or-judges [A, B] " will not require the
truth of " x judges [A] or x judges [ B] ", there will be a
connection of meaning between the two forms of expres
sion; if a creature x or-j udges its potential prey as being [in
the tree, in the bush] , we may expect x to wait in between
the tree and the bush, constantly surveying both ; or
j udging [A, B] will, in short, be typically manifested in
behaviour which is appropriate equally to the truth of A
and to the truth of B, and which is preparatory for
behaviour specially appropriate to the truth of one of the
pair, to be evinced when it becomes true either that x
j udges [A] or that x judges [B] .
This kind of transition, in which an "extrinsic" modifier
attachable to 'ljJ-expressions is transformed into (or
replaced by) an " intrinsic" modifier (one which is specific
atory of a special 'ljJ-state), I shall call first-stage internal
ization.
2 . It should be apparent from the foregoing discussion
that transitions from a lower type to a higher type of pirot
will often involve the addition of diversifications of a 'ljJ
state to be found undiversified in the lower type of pirot;
we might proceed, for example, from pirots with a
capacity for simple j udging to pirots with a capacity for
present-judging (the counterpart of the simple j udging of
their predecessors) , and also for future-judging (primitive
expecting) and for past-j udging (primitive remembering) .
It may be possible always to represent developmental steps
in this way; but I am not yet sure of this. It may be that
sometimes we should attribute to the less-developed pirot
two distinct states, say judging and willing; and only to a
more developed pirot do we assign a more generic state,
say accepting, of which, once it is introduced, we may
represent judging and willing as specific forms (judicatively
accepting and volitively accepting) . But by whatever route
we think of ourselves as arriving at it, the representation
within a particular level of theory of two or more 'tjl-states
as diversifications of a single generic 'tjl-state is legitimate,
in my view, only if the theory includes laws relating to the
generic 'tjl-state which, so far from being trivially derivable
from laws relating to the specific 'tjl-states, can be used to
derive in the theory, as special cases, some but not all
generalities which hold for the specific 'tjl-states. We
might, for example, justify the attribution to a pirot of the
generic 'tjl-state of accepting, with judging and willing as
diversifications, by pointing to the presence of such a law
as the following: ceteris paribus, if x or-accepts (in mode m
of acceptance) [A, B ] , and x negatively accepts, in mode m,
[A] , then x positively accepts, in mode m, [B] . From this
law we could derive that, ceteris paribus, (a) if x or-judges
[A, B] and negatively judges [A] , then x positively judges
[B] ; and (b) if x or-wills [A, B] and negatively wills [A] ,
then x positively wills [ B ] . ( (a) and (b) are, of course,
psychological counterparts of mood-variant versions of
modus tollendo ponens. )
3 . A further kind o f transition i s one which I shall label
second-stage internalization. It involves the replacement of
an "intrinsic" modifier, which signifies a differentiating
feature of a specific 'tjl-state, by a corresponding operator
within the content-specifications for a less specific 'tjl
state. If, for example, we have reached by first-stage
internalization the 'tjl-state of future-judging [A] , we may
proceed by second-stage internalization to the 'tjl-state of
judging [in the future, A] : similarly, we may proceed from
judging (judicatively accepting) [A] to accepting [it is the
case that A ] , from willing (volitively accepting) [A] to
accepting [let it be that A ] , and from disjunctively judging
(or-judging) [A, B] to judging [A or B] . So long as the
operator introduced into a content-specification has maxi
mum scope in that specification, the result of second-stage
internalization is simply to produce a notational variant:
" "
x judges [A or B] " is semantically equivalent to x
disjunctively judges [A, B] " , and if matters rested here,
there would be no advance to a new level of theory. To
reach a new level of theory, and so to reach 'ljJ-states
exemplifiable only by a higher type of pirot, provision
must be made for 'ljJ-states with contents in the specifica
tion of which the newly introduced operator is embedded
within the scope of another operator, as in " x judges [if A
or B, C] " , or in " x accepts [if A, let it be that B] " . An
obvious consequence of this requirement will be that
second-stage internalizations cannot be introduced singly;
an embedded operator must occur within the scope of
another embeddable operator. The unembedded occur
rences of a new operator, for which translation back into
the terminology of the previous level of theory is possible,
secure continuity between the new level and the old;
indeed, I suspect that it is a general condition on the
development of one theory from another that there should
be cases of "overlap " to ensure continuity, and cases of
non-overlap to ensure that a new theory has really been
developed.
Two observations remain to be made. First, if we
enquire what theoretical purpose is likely to be served by
endowing a type of pirot with the capacity for 'ljJ-states the
specification of which involves embedded operators, one
important answer is likely to be that explanation of the
ranges of behaviour to be assigned to that type of pirot
requires a capacity for a relatively high grade of inference,
in which the passage from premisses to conclusion is
encompassed, so to speak, within a single thought; the
pirot has to be capable, for example, of thinking [since, A,
B, and C, then DJ . This capacity may well require other
capacities, e.g. for evaluation, for higher-order states, and
so on. There are fascinating and difficult questions about
the extent to which a full internalization of one kind of
idea or item carries with it full internalization of other
kinds of idea or item ; but these questions must be deferred
to another occasion.
Second, I am sympathetic to two methodological prin
ciples relating to internalization. One is that, since
principles of parsimony dictate that we introduce no
explanatory tool more powerful than is required to do the
job, when psychological states which have been generated
by first-stage internalization can be seen to be adequate for
explanatory purposes, one should not attribute to a pirot
the states which would be reached by superimposing
second-stage internalization. The other is that, where
possible, one should avoid the fully internalized appearance
of an idea or item if that appearance has not been
preceded, at a lower level of theory, by a first-stage
internalization of the idea or item in question ; full
internalization will be relatively unproblematic if it has
come about in accordance with a general pattern for
internalization. There are certainly, however, exceptions
to the latter principle; one very important kind of
internalization, the internalization of psychological
instantiables such as j udging and willing (as distinct from
the internalization of differentiating features of such
instantiables) is not easily accommodated within the
characterization of procedures for internalization which I
have been sketching; and while the two-stage pattern
seems appropriate for internalization with respect to
tenses, it does not seem appropriate with respect to very
specific temporal modifiers like "three weeks from next
Tuesday" .
Nevertheless, subject to a list o f exceptions which I am
not in a position to formulate, we might gather up the
following provisional list of principles governing internal
ization:
1 . 'ljJ-states generated by internalization should con
form to the restrictions propounded in Section VII ; that is
to say, they should not be assigned to a pirot without (a)
the assignment to the pirot of behaviour the presence of
which they will be required to explain, and (b) a genitorial
justification for the presence of that behaviour in the pirot.
2. If both generic and specific 'ljJ-states are to be
recognized by a theory, each such state should figure non
trivially in laws of that theory.
3. Second-stage internalization should be invoked only
when first-stage internalization is inadequate for explan
atory purposes.
4. Where possible, full internalization should be reached
by a standard two-stage progression.
VII. Higher- O rder Psychological States
In this section I shall focus on j ust one of the modes of

content-internalization, that in which psychological con
cepts (or other counterparts in psychological theory) are
themselves internalized. I shall give three illustrations of
the philosophical potentialities of this mode of internal
ization, including a suggestion for the solution of my third
initial problem (Problem C), that of the accommodation,
within my approach, of the phenomena of privileged
access and incorrigibility.
1 . As I have already remarked, it may be possible in a
sufficiently enriched theory to define concepts which have
to be treated as primitive in the theory's more impoverished
predecessors. Given a type of pirot whose behaviour is
sufficiently complex as to require a psychological theory in
which psychological concepts, along with such other items
as logical connectives and quantifiers, are internalized, it
will I think be possible to define a variety of judging, which
I shall call "judging* " , in terms of willing. I doubt if one
would wish judging '} to replace the previously distinguished
notion of judging; I suspect one would wish them to
coexist; one would want one's relatively advanced pirots
to be capable not only of the highly rational state of
judging* but also of the kind of j udging exemplified by
lower types, if only in order to attribute to such advanced
pirots implicit or unconscious j udgings. There may well be
more than one option for the definition of judging'� ; since
my purpose is to illustrate a method rather than to make a
substantial proposal, I shall select a relatively simple way,
which may not be the best.
The central ideal is that whereas the proper specification
of a (particularized) disposition, say a man's disposition to
entertain his brother if his brother comes to town, would
be as a disposition to entertain his brother if he believes
that his brother has come to town ( cf. the circularities
noted in Section I), no such emendation is required if we
speak of a man's will to entertain his brother if his brother
comes to town. We may expect the man to be disappointed
if he discovers that his brother has actually come to town
without his having had the chance to entertain him,
whether or not he was at the time aware that his brother
was in town. Of course, to put his will into effect on a
particular occasion, he will have to judge that his brother
is in town, but no reference to such a judgement would be
appropriate in the specification of the content of his will.
Here, then, is the definition which I suggest. x judges *
that p just in case x wills as follows : given any situation in
which (i) x wills some end E, (ii) there are two non-empty
classes (K 1 and K 2 ) of action-types such that the perform
ance by x of an action-type belonging to K 1 will realize E
j ust in case p is true, and the performance by x of a
member of K2 will realize E j ust in case p is false, (iii) there
is no third non-empty class (K3 ) of action-types, such that
the performance by x of an action-type belonging to K3
will realize E whether p is true or p is false: then in such a
situation x is to will that he perform some action-type
belonging to K 1 • Put more informally (and less accurately) ,
x judges * that p j ust i n case x wills that, if h e h a s to choose
between a kind of action which will realize some end of his
just in case p is true and a kind of action which will realize
that end just in case p is false, he should will to adopt some
action of the first kind.
2. We might be able to use the idea of higher-order
states to account for a kind of indeterminacy which is
frequently apparent in our desires. Consider a disgruntled
employee who wants more money; and suppose that we
embrace the idea (which may very well be misguided) of
representing his desire in terms of the notion of wanting
that p (with the aid of quantifiers) . To represent him as
wanting that he should have some increment or other, or
that he should have some increment or other within certain
limits, may not do j ustice to the facts ; it might be to
attribute to him too generic a desire, since he may not be
thinking that any increment, or any increment within
certain limits, would do. If, however, we shift from
thinking in terms of internal quantification to thinking in
terms of external quantification, then we run the risk of
attributing to him too specific a desire ; there may not be
any specific increment which is the one he wants; he may
just not have yet made up his mind. It is as if we should like
to plant our (existential) quantifier in an intermediate
position, in the middle of the word 'want' ; and that, in a
way, is just what we can do. We can suppose him (initially)
to be wanting that there be some increment (some
increment within certain limits) such that he wants that
increment; and then, when under the influence of that
want he has engaged in further reflection, or succeeded in
eliciting an offer from his employer, he later reaches a
position to which external quantification is appropriate;
that is, there now is some particular increment such that he
wants that increment.
3. I shall now try to bring the idea of higher-order states
to bear on Problem C (how to accommodate privileged
access and, maybe, incorrigibility) . I shall set out in stages
a possible solution along these lines, which will also
illustrate the application of aspects of the genitorial
programme.
Stage 0. We start with pirots equipped to satisfy unnested
judgings and willings (i.e. whose contents do not
involve judging or willing) .
Stage 1 . It would be advantageous to pirots if they could
have judgings and willings which relate to the
j udgings or willings of other pirots ; for example, if
pirots are sufficiently developed to be able to will
their own behaviour in advance (form intentions for
future action), it could be advantageous to one pirot
to anticipate the behaviour of another by judging that
the second pirot wills to do A (in the future) . So we
construct a higher type of pirot with this capacity,
without however the capacity for reflexive states.
Stage 2 . It would be advantageous to construct a yet
higher type of pirot, with judgings and willings which
relate to its own judgings or willings. Such pirots
could be equipped to control or regulate their own
judgings and willings ; they will presumably be
already constituted so as to conform to the law that
ceteris paribus if they will that p and j udge that not-p,
then if they can, they make it the case that p. To give
them some control over their judgings and willings,
we need only extend the application of this law to
their judgings and willings ; we equip them so that
ceteris paribus if they will that they do not will that p
and judge that they do will that p, then (if they can)
they make it the case that they do not will that p (and
we somehow ensure that sometimes they can do this) .
[It may b e that the installation o f this kind o f control
would go hand in hand with the installation of the
capacity for evaluation ; but I need not concern myself
with this now.]
Stage 3. We shall not want these pirots to depend, in
reaching their second-order judgements about them
selves, on the observation of manifestational be
haviour; indeed, if self-control which involves
suppressing the willing that p is what the genitor is
aiming at, behaviour which manifests a pirot's
judging that it wills that p may be part of what he
hopes to prevent. So the genitor makes these pirots
subject to the law that ceteris paribus if a pirot judges
(wills) that p, then it judges that it j udges (wills) that
p. To build in this feature is to build in privileged
access to j udgings and willings.
To minimize the waste of effort which would be
involved in trying to suppress a willing which a pirot
mistakenly judges itself to have, the genitor may also
build in conformity to the converse law, that ceteris
paribus if a pirot judges that it judges (wills) that p,
then it judges (wills) that p. Both of these laws
however are only ceteris paribus laws; and there will
be room for counter-examples ; in self-deception, for
example, either law may not hold (we may get a
judgement that one wills that p without the willing
that p, and we may get willing that p without judging
that one wills that p-indeed, with judging that one
does not will that p ) .
Stage 4. Let m e abbreviate " x j udges that x judges that p "
b y " x judges 2 that p " , and " x j udges that x j udges
that x judges that p " by " x judges 3 that p " . Let us
suppose that we make the not implausible assump
tion that there will be no way of finding non-linguistic
manifestational behaviour which distinguishes
j udging3 that p from judging2 that p. There will now
be two options : we may suppose that " judge 3 that p "
i s a n inadmissible locution, which one has n o basis
for applying; or we may suppose that " x j udges 3 that
p" and " x j udges 2 that p" are manifestationally
equivalent, j ust because there can be no distinguish
ing behavioural manifestation.
The second option is preferable, if (a) we want to allow
for the construction of a (possibly later) type, a talking
pirot, which can express that it judges 2 that p ; and (b) to
maintain as a general (though probably derivative) law
that ceteris paribus if x expresses that <I> then x j udges that
<j>. The substitution of "x j udges 2 that p" for " <I>" will
force the admissibility of "x judges 3 that p " . So we shall
have to adopt as a law that x judges 3 that p if£ x judges 2
that p. Exactly parallel reasoning will force the adoption of
the law that x judges4 that p if x judges 2 that p .
If we now define "x believes that p " as "x judges 2 that
p " , we get the result that x believes that p if£ x believes
that x believes that p. We get the result, that is to say, that
beliefs are (in this sense) incorrigible, whereas first-order
judgings are only matters for privileged access.
VIII. Are Psychological Concepts Eliminable ?
I have tried to shed a little light on the question 'How are

psychological concepts related to psychological theory and
to psychological explanation ? ' ; I have not, so far, said
anything about a question which has considerably vexed
some of my friends as well as myself. Someone (let us call
him the eliminator) might say : 'You explicitly subscribe to
the idea of thinking of psychological concepts as explicable
in terms of their roles in psychological laws, that is to say,
in terms of their potentialities for the explanation of
behaviour. You also subscribe to the idea that, where there
is a psychological explanation of a particular behaviour,
there is also a physiological explanation; with respect to
pirots, it is supposedly the business of the engineer to make
psychological states effective by ensuring that this condition
holds. You must, however, admit that the explanations of
behaviour, drawn from pirot-physiology, which are
accessible to the engineer are, from a theoretical point of
view, greatly superior to those accessible to the genitor; the
former are, for example, drawn from a more comprehensive
theory, and from one which yields, or when fully developed
will yield, even in the area of behaviour, more numerous
and more precise predictions. So since attributions of
psychological states owe any claim to truth they may have
to their potentialities for the explanation of behaviour, and
since these potentialities are inferior to the potentialities of
physiological states, Occam's Razor will dictate that
attributions of psychological states, to actual creatures no
less than to pirots, should be rejected en bloc as lacking
truth (even though, from ignorance or to avoid excessive
labour, we might in fact continue to utter them) . We have
indeed come face to face with the last gasp of Primitive
Animism, namely the attempted perpetuation of the myth
that animals are animate.'
The eliminator should receive fuller attention than I can
give him on this occasion, but I will provide an interim
response to him, in the course of which I shall hope to
redeem the promise, made in Section II, to pursue a little
further the idea that the common-sense theory which
underlines our psychological concepts can be shown to be
at least in some respects true.
First, the eliminator's position should, in my view, be
seen as a particular case of canonical scepticism, and
should, therefore, be met in accordance with general
principles for the treatment of sceptical problems. It is
necessary, but it is also insufficient, to produce a cogent
argument for the falsity of his conclusion ; if we confine
ourselves to this kind of argumentation we shall find
ourselves confronted by the post-eliminator, who says to
us 'Certainly the steps in your refutation of the eliminator
are validly made ; but so are the steps in his argument.
A correct view of the matter is that from a set of principles in
which psychological concepts occur crucially, and through
which those concepts are delineated, it follows both that
the eliminator is right and that he is wrong; so much the
worse, then, for the set of principles and for the concepts ;
they should be rejected as incoherent. ' A satisfactory
rebuttal of the eliminator's position must, therefore,
undermine his argument as well as overturn his conclu
sion.
In pursuit of the first of these objectives, let me remark
that, to my mind, to explain a phenomenon (or sequence
of phenomena) P is typically to explain P qua exemplifying
some particular feature or characteristic; and so one fact
or system may explain P qua exemplifying characteristic
C 1 while another fact or system explains P qua exemplify
ing C2 • In application to the present case, it is quite
possible that one system, the physiology of the future,
should explain the things we do considered as sequences of
physical movements, while another system, psychological
theory, explains the things we do considered as instances
of sorts of behaviour. We may, at this point, distinguish
two possible principles concerning the acceptability of
systems of explanation :
1 . If system S 1 is theoretically more adequate, for the
explanation of a class of phenomena P will respect to a
class of characteristics C, than is system S2 for the
explanation of P with respect to C, then (other things
being equal) S1 is to be accepted and S2 is to be rejected.
I have no quarrel with this principle.
2. If system S 1 is theoretically more adequate, for the
explanation of P with respect to Ch than is system S2 for
the explanation of P with respect to C2 [a different class of
characteristics] , then (other things being equal) S 1 is to be
accepted and S2 is to be rejected.
The latter principle seems to me to lack plausibility ; in
general, if we want to be able to explain P qua C1 (e.g. as
behaviour), our interest in such explanation should not be
abandoned merely because the kind of explanation avail
able with respect to some other class of characteristics is
theoretically superior to the kind available with respect to
C2 • The eliminator, however, will argue that, in the present
case, what we have i s really a special case of the
application of Principle 1 ; for since any C2 (in C:2 ) which is
exemplified on a given occasion has to be " realized" on
that occasion in some C 1 (some sequence of physical
movements) , for which there is a physiological explanation,
the physiological explanation of the presence of C 1 on that
occasion is ipso facto an explanation of the prese_nce of C2
(the behavioural feature) on that occasion, and so should
be regarded as displacing any psychological explanation of
the presence of C2 •
To maintain this position, the eliminator must be
prepared to hold that if a behavioural feature is, on an
occasion, " realized" in a given type of physical movement
sequence, then that type of movement-sequence is, in itself,
a sufficient condition for the presence of the behavioural
feature ; only so can he claim that the full power and
prestige of physiological explanation is transmitted from
movements to behaviour, thereby ousting psychological
explanation. But it is very dubious whether particular
movement-sequences are, in general, sufficient conditions
for behavioural features, especially with respect to those
behavioural features the discrimination of which is most
important for the conduct of life. Whether a man with a
club is merely advancing towards me or advancing upon
me may well depend on a condition distinct from the
character of his movements, indeed (it would be natural to
suppose) upon a psychological condition. It might of
course be claimed that any such psychological condition
would, in principle, be re-expressible in psychological
terms, but this claim cannot be made by one who seeks not
the reduction of psychological concepts but their rejection:
if psychological states are not to be reduced to physio
logical states, the proceeds of such reduction cannot be
used to bridge gaps between movements and behaviour.
The truth seems to be that the language of behavioural
description is part and parcel of the system to which
psychological explanation belongs, and cannot be prised
off therefrom. Nor can the eliminator choose to be hanged
for a sheep rather than for a mere lamb, by opting to
jettison behavioural description along with psychological
concepts ; he himself will need behavioural terms like
"describe" and "report" in order to formulate his non
recommendations (or predictions) ; and, perhaps more
importantly, if there is (or is to be) strictly speaking no
such thing as the discrimination of behaviour, then there
are (or are to be) no such things as ways of staying alive,
either to describe or to adopt.
Furthermore, even if the eliminator were to succeed in
bringing psychological concepts beneath the baleful glare
of Principle 1 , he would only have established that other
things being equal psychological concepts and psycho
logical explanation are to be rejected. But other things are
manifestly not equal, in ways which he has not taken into
account. It is one thing to suggest, as I have suggested, that
a proper philosophical understanding of ascriptions of
psychological concepts is a matter of understanding the
roles of such concepts in a psychological theory which
explains behaviour; it is quite another thing to suppose
that creatures to whom such a theory applies will (or
should) be interested in the ascription of psychological
states only because of their interest in having a satisfactory
system for the explanation of behaviour. The psychological
theory which I envisage would be deficient as a theory to
explain behaviour if it did not contain provision for
interests in the ascription of psychological states otherwise
than as tools for explaining and predicting behaviour,
interests (for example) on the part of one creature to be
able to ascribe these rather than those psychological states
to another creature because of a concern for the other
creature. Within such a theory it should be possible to
derive strong motivations on the part of the creatures
subject to the theory against the abandonment of the
central concepts of the theory ( and so of the theory itself),
motivations which the creatures would (or should) regard
as justified.2 Indeed, only from within the framework of
such a theory, I think, can matters of evaluation, and so of
the evaluation of modes of explanation, be raised at all. If I
conjecture aright, then, the entrenched system contains the
materials needed to justify its own entrenchment; whereas
no rival system contains a basis for the justification of
anything at all.
We must be ever watchful against the devil of scientism,
who would lead us into myopic over-concentration on the
nature and importance of knowledge, and of scientific
knowledge in particular; the devil who is even so audacious
as to tempt us to call in question the very system of ideas
required to make intelligible the idea of calling in question
anything at all; and who would even prompt us, in effect,
to suggest that, since we do not really think but only think
that we think, we had better change our minds without
undue delay. 5
4 Let me illustrate with a little fable. The ve ry e minent and very dedicated
neurophysiologist speaks to his wife. " My (for at least a little while longer)
dear, " he says, "I have long thought of myself as an acute and well-informed
interpreter of your actions and behaviour . I think I have been able to
identify nearly every thought that has made you smile and nearly every
desire that has moved you to act. My researches, however, have made such
progress that I shall no longer need to understand you in this way. Instead I
shall be in a position, with the aid of instruments which I shall attach to you,
to assign to each bodily movement which you make in acting a specific
antecedent condition in your cortex. No longer shall I need to concern
myself with your so-called thoughts and feelings. In the meantime, perhaps
you would have dinner with me tonight. I trust that you will not resist if I
bring along some apparatus to help me to determine, as quickly as possible,
the physiological idiosyncracies which obtain in your system. "
I have a feeling that the lady might refuse the proffered invitation.
5 I am indebted to George Myro for the tenor of this remark, as well as
for valuable help with the su bstance of this section.
Index
Aristotle 1 6- 1 7, 72, 8 0 , 9 5 , 99, 1 02 , Foot, Philippa 2.5 , 3 9 , 47-9, 5 3--4,

1 0 9 , 1 1 9, 1 3 2 , 1 42 5 6-60, 63-6
and four Causes 74-5 , 9 6
Armstrong, David 76 genitorial programme 1 40-5
Golding, William 1 1 2
Baker, Judith 63--4, 1 26 Grandy, Richard 7 n . , 93 n .
Blackburn, Simon 4 Grice-Myro Theory o f Identity 8 1 -2 ,
Bootstrap 1 7, 1 03 111
Bradley, F. H. 62
Hare, R . M . 26-3 0
Carnap, Rudolph 70, 1 4 0 Hobbes, Thomas 8 1
category shift 1 09-1 1 Hume, David 4, 8 8 , 1 07, 1 4 6
ceteris paribus laws 1 3 2 Humean Projection 4-.5 , 1 4-17, 8 8 ,
Code, Alan 8 0 , 95 n . 9 0 , 1 07-9 (or content
construction b y a genitor ( o r genitorial internalization) 1 4 6-5 1
programme) 8 3 , 1 40-5 hypothetization and hypostatization
construction routines 1 07 (or scientific vs. metaphysical
constructivist methodology 9 7- 1 0 1 construction) 1 9 , 1 07
and primitiveness 1 00-1
constructivist programme 70-5 incorrigibiliry 1 2.5 , 1 5 4-6
see also constructivist methodology imperatives, categorical-hypothetical
content internalization 1 4 6-5 3 distinction interpreted 49-.5 3
cybernetic notions 6 n . , 76-7 Irwin, Terry 8 0
cybernetics and causal connections
78-9 Kant, Immanuel 9 1 , 97, 1 03 , 1 04
Kantian notion of positive freedo m
Descartes' cogito 1 04 91, 1 1 8
Kantotle 1 1 .5 , 1 3 9
ends, assessment of 6 1 -3 Keyt, David .5 9
essential p roperry 79-80, 1 1 1 - 1 2 Kneale, William 1 1 0
explanation-gap 7 6 Kripke, Saul 1 9
final causes 1 1 2 law-allusiveness 1 3 4, 1 3 7-8

finaliry 72 Leibnizian standards 1 1 8
and absolute value 79 Locke, John 1 43
active and passive 8 0- 1 , 1 1 2-1 3
autonomous 1 1 3 , 1 1 7 Mackie, J. L. 25-6, 28-3 6 , 3 9--4.5,
and cybernetic notions 7 6, 7 8-9 47, 6 6
detached 79, 1 1 2 mechanist vs. vitalist 75-7
and essential property 79-8 0, mechanistically substitutable 8 8 , 90,
1 1 1-12 117
and notion of life 73 metaphysical argument 1 1 8 , 1 3 9--40
and purpose 74, 79 Metaphysical Principle of Supply and
first philosophy 94-5 Demand 1 7, 8 7
1 64 Index
Metaphysical Transubstantiation and genitorial j u stification 1 43
7- 1 4 , 8 1 -7, 90, 1 1 4- 1 5 , 1 1 8 as theoretical concepts 1 25
metier 6-7, 8 0- 1 , 1 1 3- 1 4 unity of 1 42
Mill, j. S . 7 8 psycho-physical identification
Myro, George 6 9 , 1 6 1 n . 1 2 9-3 1
a n d Grice-Myro Theory of Identity
8 1 -2, 1 1 1 Ramsified naming, definition 1 2 8 ,
1 3 0, 1 3 4
rational demand 1 1 7-1 8
obj ectivity :
rationality 6, 1 2, 82-6, 1 1 4
different versions of the idea o f
relativizations of value concepts
35-9, 47
1 1 5- 1 6
and scepticism 3 9-40
Russell, Bertrand 1 40
ontological Marxism 1 3 1
overlaps 77-9, 9 9- 1 00, 1 4 9
Sluga, Hans 3 9
e n d of overlap 8 7 , 1 1 7
Strawson, P. F. 5 8 n.
person vs. h u m a n being 6-7, 8 2 , 1 1 4 Theory-theory 95

priority, di fferent interpretations of
idea of 1 0 2-3 transcendental argument 1 03-7
privileged access 1 25 , 1 5 4-6 see also metaphysical argument
properties, essential vs. necessa ry
79-8 0 Warner, Richard 7 n., 6 9 n., 93 n.
psychological con cepts : Wittgenstein, Ludwig 39, 1 42
expl icit vs. implicit definitions o f
1 23-4 Yerkes-Dodson Law 1 3 3-4

Grice, P. (2001) - The Conception of Value.

Uploaded by

Copyright:

Available Formats

Grice, P. (2001) - The Conception of Value.

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Grice, P. (2001) - The Conception of Value.

Uploaded by

Copyright:

Available Formats

THE CONCEPTION OF VALUE

The Conception of Value

CLARENDON PRESS · OXFORD

Oxford is a trade mark of Oxford University Press

Published in the United States

© Kathleen Grice 1991

All rights reserved. No part of this publication may be reproduced,

British Library Cataloguing in Publication Data

Library of Congress Cataloging in Publication Data

Typeset by Cambrian Typesetters

Paul Grice wrote and delivered his Carus lectures on the

Introduction by ]udith Baker 1

The Carus Lectures on the Conception of Value

1 . Value and Objectivity 23

Reply to Richards, Final Section:

Metaphysics, Philosophical Psychology, and Value 93

Method in Philosophical Psychology (From the

The three works o f Paul Grice collected in this volume

blocked in all areas if there is one area in which attribution

According to Grice, a value judgement--one that is

2 Relative value judgements depend upon finality, but finality must be a

The operation flauntingly named Metaphysical Tran­

we do a person's, in terms of its desire to get its work done.

The title which I have chosen for these lectures embodies,

Mackie seems to m e not to handle very well this attempt

moral views, and indeed ones whose content was

(a) There is a feature W which is objective but

2. The idea, to which Mackie subscribes, of separating

My enterprise in this lecture will, initially at least, be to

The language is not wholly clear; but what is seemingly

(iii) There are, I think, serious doubts of the admissibility

5 . Unlimited cognitive-cum-practical rationalism. No

The main object of this lecture will be to explore the

1 In this lecture I am especially indebted to Richard Warner, without

'sense o f the word' from the activity of a scholar. My idea

2 [APA Address, pp. 12 1-6 1 below; see pp. 140-5 .]

3 [APA Address pp. 12 1-6 below.]

In this final section of my 'Reply to Richards', I shall take

4 [London: Faber & Faber, 1 95 5 . ]

5 [Original Version, 1 9 8 3 ; see pp. 69-9 1 above.]

Such a defence of absolute value is, of course, bristling

In what follows, I shall be presenting some of my ideas

1 Presidential Address delivered at the 49th Annual Meeting of the

I. Some Focal Problems

I shall begin by formulating, in outline, a sequence of four

II. Some General Aspects of the Relation between

First, a preliminary observation : if we are seeking to

i f I adopt the first alternative with its uniqueness claim, I

III. The Form of Psychological "Laws "

In this section I shall briefly discuss two formal features

IV. Semi-realistic Procedure for Introducing some

I turn now to the task of outlining, in a constructive way,

We make the following postulations (if we have not

I have so far been occupying myself with questions about

VI. Type-Progression in Pirotology: Content­

My purpose in this section is to give a little thought to the

VII. Higher- O rder Psychological States

In this section I shall focus on j ust one of the modes of

VIII. Are Psychological Concepts Eliminable ?

The operation flauntingly named Metaphysical Tran

VI. Type-Progression in Pirotology: Content